A method for positioning a receiving coil of a wireless power transmission system

Abstract

The invention discloses a receiving coil positioning method for a wireless power transmission system, which belongs to the field of wireless power transmission and comprises: a wireless power transmission planar array transmitting coil and receiving coil coupling mechanism, a BP neural network coil positioning model and a BP neural network coil positioning model optimization method; the transmitting coils are stacked and placed at a position where the cross-coupling coefficient of the transmitting coil is minimum; the current data of the transmitting coil is used as the input layer of the BP neural network coil positioning model, the coordinates (x, y) of the output receiving coil center point are used as the output layer of the BP neural network coil positioning model, and the optimal area coefficient Kthopt is obtained, thereby obtaining the optimal position of the receiving coil.

Description

Technical Field

[0001] The invention belongs to the field of wireless power transmission, and in particular relates to a receiving coil positioning method of a wireless power transmission system. Background Art

[0002] The wireless power transmission technology of electric vehicles is gradually developing. When the electric vehicle is wirelessly charged, the position change of the receiving coil and the transmitting coil will affect the fluctuation of the transmission power during the wireless power transmission process. Therefore, the positioning of the receiving coil installed in the electric vehicle plays a vital role. Summary of the invention

[0003] The purpose of the present invention is to provide a method for positioning a receiving coil of a wireless power transmission system, so as to solve the problem of positioning the relative positions of a receiving coil and a transmitting coil of an electric vehicle.

[0004] The technical solution to achieve the purpose of the present invention is:

[0005] A method for positioning a receiving coil of a wireless power transmission system, characterized by comprising: a wireless power transmission planar array transmitting coil and a receiving coil coupling mechanism, a BP neural network coil positioning model, and a BP neural network coil positioning model optimization method;

[0006] A wireless power transmission planar array transmitting coil and receiving coil coupling mechanism, comprising: the transmitting coils are stacked and placed at a position where the cross-coupling coefficient of the transmitting coils is minimum, and the parameters of the transmitting coils are the same, including: the frequency, amplitude, and internal resistance of the power supply; the internal resistance, mutual inductance, and compensation capacitance of the transmitting coil; and the receiving coil is placed above the transmitting coil;

[0007] BP neural network coil positioning model, including:

[0008] The current data of the transmitting coil is used as the input layer of the BP neural network coil positioning model, and the coordinates (x, y) of the center point of the output receiving coil are used as the output layer of the BP neural network coil positioning model;

[0009] The BP neural network coil positioning model optimization method includes:

[0010] According to the current data of the transmitting coil, the planar array transmitting coil area is divided into two areas, two different BP neural network training models are established, and two important parameters of the partition modeling optimization method are defined: the current critical value I thi and the area coefficient K thi ; For the current critical value I thi , when the coordinates of the receiving coil center point of the full-area BP neural network coil positioning model corresponding to the transmitting coil current are all greater than the current critical value I thiWhen the current critical value I thi The number of th , N th The value of is related to the current value range of the BP neural network coil positioning model in the whole area and the amount of training data; the optimal current critical value I is defined thopt , if i=N th / 2, N th is an even number, with the current critical value I thNth / 2 When the accuracy of the BP neural network coil positioning model established by the two regions is greater than the accuracy of the BP neural network coil positioning model established by other current critical values, the current critical value I is defined at this time. thNth / 2 is the optimal current critical value I thopt , define the optimal area coefficient K thopt , optimal current critical value, I thopt The corresponding area coefficient K thi is the optimal regional coefficient, i.e. K thopt =K thi ; Optimal area coefficient K thopt Characterization parameters of the optimal regional BP neural network coil positioning model optimization method.

[0011] Optionally, the receiving coil is placed above the transmitting coil and includes:

[0012] When the spatial position of the receiving coil changes, the mutual inductance between the receiving coil and the transmitting coil changes, causing the current data of the transmitting coil to change.

[0013] Optionally, using the current data of the transmitting coil as the input layer of the BP neural network coil positioning model includes:

[0014] The current data of the transmitting coil is divided into a training set and a test set in a ratio of 9:1;

[0015] The training set is used to train the BP neural network coil positioning model;

[0016] The test set is used to verify the output results of the BP neural network coil positioning model.

[0017] Optionally, the specific process of the BP neural network coil positioning model optimization method is as follows:

[0018] (1) Determine the current data of the receiving coil of the full-area BP neural network coil positioning model, the maximum current I max and the minimum current I min The distribution of the difference between the current value and the current value is used to preliminarily determine the number of current critical values ​​N th .

[0019] (2) Calculate the current critical value of the BP neural network coil positioning model in the entire area, and establish the BP neural network training models in different areas based on the current critical value.

[0020] (3) Analyze and compare the training model results of the two regions under different current critical values, and further adjust the number of current critical values ​​N according to the results of the BP neural network coil positioning model, the current value difference of the BP neural network coil positioning model, and the current data of the receiving coil of the full-region BP neural network coil positioning model. th , until the best BP neural network coil positioning model accuracy is obtained.

[0021] (4) According to the best training results of the BP neural network coil positioning model, the optimal current critical value I is selected in the BP neural network coil positioning model of the whole area. thopt .

[0022] (5) According to the optimal current critical value I thopt Calculate the optimal area coefficient K thopt .

[0023] The present invention discloses a receiving coil positioning method for a wireless power transmission system, which belongs to the field of wireless power transmission, and comprises: a wireless power transmission planar array transmitting coil and a receiving coil coupling mechanism, a BP neural network coil positioning model and a BP neural network coil positioning model optimization method; the transmitting coils are stacked and placed at a position where the cross-coupling coefficient of the transmitting coil is minimum; the current data of the transmitting coil is used as the input layer of the BP neural network coil positioning model, and the coordinates (x, y) of the output receiving coil center point are used as the output layer of the BP neural network coil positioning model to obtain the optimal area coefficient K thopt , and then get the best position of the receiving coil. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a flow chart of the BP neural network coil positioning model optimization method provided by the present invention;

[0025] Figure 2 It is a structural diagram of a wireless power transmission planar array transmitting coil and receiving coil coupling mechanism provided by the present invention;

[0026] Figure 3 It is an equivalent circuit diagram corresponding to the coupling mechanism of the planar array transmitting coil and receiving coil of the wireless power transmission provided by the present invention;

[0027] Figure 4 It is a flow chart of the partition modeling optimization method based on BP neural network. DETAILED DESCRIPTION

[0028] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

[0029] See also Figure 1 , Figure 2 and Figure 3 , a wireless power transmission system receiving coil positioning method provided by the present invention includes: a wireless power transmission planar array transmitting coil and a receiving coil coupling mechanism, a BP neural network coil positioning model and a BP neural network coil positioning model optimization method;

[0030] A wireless power transmission planar array transmitting coil and receiving coil coupling mechanism, comprising: the transmitting coils are stacked and placed at a position where the cross-coupling coefficient of the transmitting coils is minimum, and the parameters of the transmitting coils are the same, including: the frequency, amplitude, and internal resistance of the power supply; the internal resistance, mutual inductance, and compensation capacitance of the transmitting coil; and the receiving coil is placed above the transmitting coil;

[0031] Optionally, in some embodiments, the transmitting coil is stacked by four square coils, and the placement position can achieve the minimum coil cross-coupling coefficient. The four transmitting coils are stacked, and the cross-coupling coefficient can be minimized by reasonably placing the transmitting coils. The receiving coil is at a certain height from the four transmitting coils.

[0032] In this embodiment, the KVL equations are written for four transmitting coils and one receiving coil respectively, as follows:

[0033]

[0034] Where U 1 , U 2 , U 3 , U 4 is the AC voltage source for the four transmitting coils. The frequencies of the four AC voltage sources are the same. 1 ,I 2 ,I 3 ,I 4 are the loop currents of the four transmitting coils, I 5 is the current of the receiving coil loop, R S1 , R S2 , R S3 , R S4 are the internal resistances of the four voltage sources, R 1 , R 2 , R 3 , R 4 , R 5 are the internal resistances of the first transmitting coil, the second transmitting coil, the third transmitting coil, the fourth transmitting coil and the receiving coil, L 1 , L 2 , L 3, L 4 , L 5 are the inductances of the first transmitting coil, the second transmitting coil, the third transmitting coil, the fourth transmitting coil and the receiving coil, C 1 , C 2 , C 3 , C 4 , C 5 are the compensation capacitors of the first transmitting coil, the second transmitting coil, the third transmitting coil, the fourth transmitting coil and the receiving coil, M 15 、M 25 、M 35 、M 45 The mutual inductances of the first transmitting coil and receiving coil, the second transmitting coil and receiving coil, the third transmitting coil and receiving coil, and the fourth transmitting coil and receiving coil are respectively, M 12 、M 13 、M 14 、M 23 、M 24 、M 34 are the cross-coupling mutual inductances of the first transmitting coil and the second transmitting coil, the first transmitting coil and the third transmitting coil, the first transmitting coil and the fourth transmitting coil, the second transmitting coil and the third transmitting coil, the second transmitting coil and the fourth transmitting coil, and the third transmitting coil and the fourth transmitting coil, respectively; R load It is the load of the receiving coil loop.

[0035] After the four transmitting coils are properly placed, the cross-coupling mutual inductance between the transmitting coils can be ignored in theoretical analysis, that is, M 12 =M 13 =M 14 =M 23 =M 24 =M 34 = 0. In addition, the parameters, structure and function of each transmitting coil element are exactly the same, that is, R S1 =R S2 =R S3 =R S4 =R S , the internal resistance of each transmitting coil is approximately equal to the internal resistance of the receiving coil, that is, R 1 =R 2 =R 3 =R 4 =R 5 =R 0 , formula (1) can be simplified to

[0036]

[0037] make

[0038]

[0039] At this time, formula (2) can be simplified to

[0040]

[0041] The receiving coil loop current I can be obtained 5 The expression is:

[0042]

[0043] The loop current expressions of the four transmitting coils can be obtained as follows:

[0044]

[0045] It can be seen from equation (5) that there is a corresponding relationship between the current of each transmitting coil and the mutual inductance between the receiving coils. In the planar array transmitting coil WPT system, when the spatial position of the receiving coil changes, the mutual inductance M between each transmitting coil and the receiving coil 15 、M 25 、M 35 、M 45 will change, that is, the current I 1 ,I 2 ,I 3 ,I 4 It corresponds to the center coordinates (x, y) of the receiving coil.

[0046] BP neural network coil positioning model, including:

[0047] The current data of the transmitting coil is used as the input layer of the BP neural network coil positioning model, and the coordinates (x, y) of the center point of the output receiving coil are used as the output layer of the BP neural network coil positioning model;

[0048] Optionally, in some embodiments, an input layer of 4 is established, i.e., the current I of 4 transmitting coils 1 ,I 2 ,I 3 ,I 4 ,The output layer is 2, which is the neural network training model of the center point coordinates x and y of the receiving coil.

[0049] The BP neural network coil positioning model optimization method includes:

[0050] According to the current data of the transmitting coil, the planar array transmitting coil area is divided into two areas, two different BP neural network training models are established, and two important parameters of the partition modeling optimization method are defined: the current critical value I thi and the area coefficient K thi ; For the current critical value Ithi , when the coordinates of the receiving coil center point of the full-area BP neural network coil positioning model corresponding to the transmitting coil current are all greater than the current critical value I thi When the current critical value I thi The number of th , N th The value of is related to the current value range of the BP neural network coil positioning model in the whole area and the amount of training data; the optimal current critical value I is defined thopt , if i=N th / 2, N th is an even number, with the current critical value I thNth / 2 When the accuracy of the BP neural network coil positioning model established by the two regions is greater than the accuracy of the BP neural network coil positioning model established by other current critical values, the current critical value I is defined at this time. thNth / 2 is the optimal current critical value I thopt , define the optimal area coefficient K thopt , optimal current critical value, I thopt The corresponding area coefficient K thi is the optimal regional coefficient, i.e. K thopt =K thi ; Optimal area coefficient K thopt Characterization parameters of the optimal regional BP neural network coil positioning model optimization method.

[0051] Optionally, in some embodiments, the current threshold I thi The expression is:

[0052]

[0053] Where i = 1, 2, ···, N th ;I max Indicates the maximum value of each transmitting coil current in the network model, I min Represents the minimum value of the current of each transmitting coil in the training network.

[0054] Optionally, in some embodiments, the area coefficient K thi , which means the maximum current I in the network model max and the current critical value I thi The difference between the maximum current I max and the minimum current I min The expression of the difference between is:

[0055] Optionally, as a specific implementation of a method for positioning a receiving coil of a wireless power transmission system provided by the present invention, the receiving coil is placed above the transmitting coil and includes:

[0056] When the spatial position of the receiving coil changes, the mutual inductance between the receiving coil and the transmitting coil changes, causing the current data of the transmitting coil to change.

[0057] Optionally, as a specific implementation of a wireless power transmission system receiving coil positioning method provided by the present invention, using the current data of the transmitting coil as the input layer of the BP neural network coil positioning model includes:

[0058] The current data of the transmitting coil is divided into a training set and a test set in a ratio of 9:1;

[0059] The training set is used to train the BP neural network coil positioning model;

[0060] The test set is used to verify the output results of the BP neural network coil positioning model.

[0061] See also Figure 4 As a specific implementation method of a wireless power transmission system receiving coil positioning method provided by the present invention, the specific process of the BP neural network coil positioning model optimization method is as follows:

[0062] (1) Determine the current data of the receiving coil of the full-area BP neural network coil positioning model, the maximum current I max and the minimum current I min The distribution of the difference between the current value and the current value is used to preliminarily determine the number of current critical values ​​N th .

[0063] (2) Calculate the current critical value of the BP neural network coil positioning model in the entire area, and establish the BP neural network training models in different areas based on the current critical value.

[0064] (3) Analyze and compare the training model results of the two regions under different current critical values, and further adjust the number of current critical values ​​N according to the results of the BP neural network coil positioning model, the current value difference of the BP neural network coil positioning model, and the current data of the receiving coil of the full-region BP neural network coil positioning model. th , until the best BP neural network coil positioning model accuracy is obtained.

[0065] (4) According to the best training results of the BP neural network coil positioning model, the optimal current critical value I is selected in the BP neural network coil positioning model of the whole area. thopt .

[0066] (5) According to the optimal current critical value I thopt Calculate the optimal area coefficient K thopt .

Claims

1. A method for positioning a receiving coil of a wireless power transmission system, comprising: Wireless power transmission planar array transmitting coil and receiving coil coupling mechanism, BP neural network coil positioning model and BP neural network coil positioning model optimization method; A wireless power transmission planar array transmitting coil and receiving coil coupling mechanism, comprising: the transmitting coils are stacked and placed at a position where the cross-coupling coefficient of the transmitting coils is minimum, and the parameters of the transmitting coils are the same, including: the frequency, amplitude, and internal resistance of the power supply; the internal resistance, mutual inductance, and compensation capacitance of the transmitting coil; and the receiving coil is placed above the transmitting coil; BP neural network coil positioning model, including: The current data of the transmitting coil is used as the input layer of the BP neural network coil positioning model, and the coordinates (x, y) of the center point of the output receiving coil are used as the output layer of the BP neural network coil positioning model; The BP neural network coil positioning model optimization method includes: According to the current data of the transmitting coil, the planar array transmitting coil area is divided into two areas, two different BP neural network training models are established, and two important parameters of the partition modeling optimization method are defined: the current critical value I thi and the area coefficient K thi ; For the current critical value I thi , when the coordinates of the receiving coil center point of the full-area BP neural network coil positioning model corresponding to the transmitting coil current are all greater than the current critical value I thi When the current critical value I thi The number of th , N th The value of is related to the current value range of the BP neural network coil positioning model in the whole area and the amount of training data; the optimal current critical value I is defined thopt , if i=N th / 2, N th is an even number, with the current critical value I thNth / 2 When the accuracy of the BP neural network coil positioning model established by the two regions is greater than the accuracy of the BP neural network coil positioning model established by other current critical values, the current critical value I is defined at this time. thNth / 2 is the optimal current critical value I thopt , define the optimal area coefficient K thopt , the optimal current critical value I thopt The corresponding area coefficient K thi is the optimal regional coefficient, i.e. K thopt =K thi ; Optimal area coefficient K thopt Characterization parameters of the optimal regional BP neural network coil positioning model optimization method.

2. The wireless power transmission system receiving coil positioning method according to claim 1, characterized in that: The receiving coil is placed above the transmitting coil and includes: When the spatial position of the receiving coil changes, the mutual inductance between the receiving coil and the transmitting coil changes, resulting in a change in the current data of the transmitting coil.

3. The wireless power transmission system receiving coil positioning method according to claim 1, characterized in that: The method of using the current data of the transmitting coil as the input layer of the BP neural network coil positioning model includes: The current data of the transmitting coil is divided into a training set and a test set in a ratio of 9:1; The training set is used to train the BP neural network coil positioning model; The test set is used to verify the output results of the BP neural network coil positioning model.

4. The wireless power transmission system receiving coil positioning method according to claim 1, characterized in that: The specific process of the BP neural network coil positioning model optimization method is as follows: (1) Determine the current data of the receiving coil of the full-area BP neural network coil positioning model, the maximum current I max and the minimum current I min The distribution of the difference between the current value and the current value is used to preliminarily determine the number of current critical values ​​N th ; (2) Calculate the current critical value of the BP neural network coil positioning model in the entire region, and establish the BP neural network training models in different regions according to the current critical value; (3) Analyze and compare the training model results of the two regions under different current critical values, and further adjust the number N of the current critical values ​​according to the results of the BP neural network coil positioning model, the current value difference of the BP neural network coil positioning model, and the current data of the receiving coil of the full-region BP neural network coil positioning model. th , until the best BP neural network coil positioning model accuracy is obtained; (4) According to the best training result of the BP neural network coil positioning model, the best current critical value I is selected in the BP neural network coil positioning model of the whole area. thopt ; (5) According to the optimal current critical value I thopt Calculate the optimal area coefficient K thopt .

Images