Multi-parameter WPT system metal foreign matter detection method and system based on neural network

Abstract

The invention relates to the technical field of magnetic coupling wireless power transmission, and particularly discloses a multi-parameter WPT system metal foreign matter detection method and system based on a neural network, and the method can uniquely determine the one-to-one relationship of different working states of a WPT system based on different combinations of M identification parameters. A neural network model and a data set are constructed according to the M identification parameters corresponding to the N working states, the neural network model is trained and tested through the constructed data set, the obtained neural network model can output the corresponding working state for any group of M identification parameters, and therefore whether the metal foreign matter exists or not is determined. According to the method, multiple parameters of the WPT system are fused, the neural network model is used for researching metal foreign matter detection under different working conditions of the system, the change of the working conditions can be judged, and the metal foreign matter with a certain size can be recognized at a certain position.

Description

Technical Field

[0001] The present invention relates to the technical field of magnetically coupled wireless power transmission, and in particular to a method and system for detecting metal foreign bodies in a multi-parameter WPT system based on a neural network. Background Art

[0002] Magnetic coupled resonant wireless power transfer (MCR-WPT) is easily affected by metal materials because it generates high-frequency electromagnetic fields in space. Metal materials generally have good electrical conductivity, but the magnetic permeability of different metals varies greatly. According to the different magnetic permeabilities of metals, they can be divided into non-magnetic metals (such as copper, aluminum, gold, etc.) and magnetic metals (such as iron, nickel, cobalt, manganese, etc.). The impact of non-magnetic metals on the system is mainly manifested as eddy current effect, and magnetic metals show enhanced eddy current effect and magnetic effect in the magnetic field.

[0003] In high-frequency electromagnetic fields, the eddy current effect and magnetic effect of non-magnetic metals will change the equivalent parameters of the system, affect system performance and cause safety hazards. At present, most of the research on the detection of non-magnetic metal foreign bodies is based on ideal conditions such as unchanged input power supply and unchanged coupling mechanism position, and system modeling, system parameter and performance index analysis, and system optimization design. However, in most cases, the position of the coupling mechanism, the input voltage or the load size are not constant, and the system operating conditions may change. When directly using power loss or equivalent quality factor method for calculation, the change in system operating conditions may be misjudged as the presence of non-magnetic metal foreign bodies. Summary of the invention

[0004] The present invention provides a multi-parameter WPT system metal foreign body detection method and system based on neural network, and solves the technical problem of how to accurately detect whether there are metal foreign bodies, especially non-magnetic metal foreign bodies, in the WPT system.

[0005] In order to solve the above technical problems, the present invention provides a multi-parameter WPT system metal foreign body detection method based on neural network, comprising the steps of:

[0006] Determine the architecture of the WPT system and its design parameters;

[0007] Constructing a neural network model, wherein the neural network model has M input nodes and N output nodes, the M input nodes are used to input M identification parameters of the WPT system, and the N output nodes are used to output probabilities of N working states of the WPT system;

[0008] Using the determined WPT system, a data set of M identification parameters corresponding to N working states of the WPT system is constructed;

[0009] Use the constructed dataset to train and test the constructed neural network model;

[0010] Inputting a set of M identification parameters of the WPT system into the trained and tested neural network model, the neural network model outputting the probabilities of N working states;

[0011] Determine whether there is a metal foreign body based on the label of the working state with the highest probability.

[0012] Furthermore, M is equal to 2, and the M identification parameters of the WPT system are the ratio Q of U2 to U1 e and power loss factor P eqLoss =P Loss / P OUT , U1 is the equivalent inverter output square wave voltage source, U2 is the effective value of the voltage across the transmitting coil and the equivalent internal resistance of the circuit, P Loss is the power loss caused by metal foreign matter, P OUT is the output power of the system.

[0013] Furthermore, the power loss caused by metal foreign matter P Loss Equal to the total energy P of the transmitting coil PT Subtract the total energy P of the receiving coil PR , the total energy P of the transmitting coil PT Equal to the input power P before the transmitter drives the system in Subtract the system loss P at the transmitter PTLoss , the total energy P of the receiving coil PR Equal to the output power P after the receiving end rectifier circuit OUT Plus the system loss P at the receiving end PRLoss .

[0014] Further, the transmitting end of the WPT system includes a DC power supply, a full-bridge inverter, a transmitting end series compensation capacitor and a transmitting coil connected in sequence, and the receiving end of the WPT system includes a receiving coil, a receiving end series compensation capacitor, a rectifier filter circuit and a load connected in sequence;

[0015] System loss at the transmitter P PTLoss equal I p , R p Respectively represent the current and internal resistance of the transmitting coil, R dson represents the on-state internal resistance of the switch tube in the full-bridge inverter;

[0016] System loss at the receiving end P PRLoss equal I s , R s Respectively represent the current and internal resistance of the receiving coil, U d Represents the conduction voltage drop of the diode in the rectification and filtering circuit.

[0017] Furthermore, N=3, and the N working status packages are normal operation of the system, metal foreign matter exists in the system, and the working condition of the system changes, corresponding to three different labels.

[0018] Furthermore, when the system is working normally, Q e Fixed, P eqLoss is approximately 0;

[0019] When there is metal foreign matter in the system, Q e Not less than the preset slope, P eqLoss Greater than 0;

[0020] When the system working conditions change, Q e , P eqLoss The changes are different when the system is working normally and when there are metal foreign objects in the system.

[0021] Furthermore, the neural network model is a BP neural network.

[0022] The present invention also provides a multi-parameter WPT system metal foreign body detection system based on a neural network, which is applied to the multi-parameter WPT system metal foreign body detection method based on a neural network, and the key lies in: comprising a model generation module, and a transmitting end detection module and a receiving end detection module respectively connected to the transmitting end and the receiving end of the WPT system;

[0023] In the training and testing phase, the transmitter detection module is used to collect information of the transmitter under N working states, the receiver detection module is used to collect information of the receiver under N working states and send it to the transmitter, the transmitter detection module calculates M identification parameters of the WPT system according to the information collected under N working states and sends them to the model generation module; the model generation module is used to construct a neural network model, and construct a data set according to the M identification parameters corresponding to the N working states, and use the constructed data set to train and test the neural network model;

[0024] During the application stage, the transmitting end detection module is used to collect information from the transmitting end, the receiving end detection module is used to collect information from the receiving end and send it to the transmitting end, the transmitting end detection module calculates M identification parameters based on the collected information and sends them to the model generation module; the model generation module outputs the probabilities of N working states; the transmitting end detection module determines whether there is a metal foreign body based on the label of the working state with the highest probability.

[0025] The present invention provides a multi-parameter WPT system metal foreign body detection method and system based on a neural network, which can uniquely determine the one-to-one relationship between different working states of the WPT system based on different combinations of M identification parameters, construct a neural network model and a data set according to the M identification parameters corresponding to N working states, and use the constructed data set to train and test the neural network model. The obtained neural network model can output the corresponding working state for any group of M identification parameters, thereby determining whether there is metal foreign body. The present invention integrates the multiple parameters of the WPT system, uses the neural network model to study the metal foreign body detection under different working conditions of the system, can judge the change of working conditions, and can identify metal foreign bodies of a certain size at a certain position. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 is a model diagram of a magnetic coupling mechanism containing metallic foreign matter provided by an embodiment of the present invention;

[0027] Figure 2 : is a magnetic flux density model distribution diagram of the coupling mechanism provided by an embodiment of the present invention when there is no foreign matter and when there is a metallic foreign matter;

[0028] Figure 3 is a diagram showing the corresponding relationship between the mutual inductance between the transmitting and receiving coils and the coil spacing when there is no foreign matter and when there is a metal foreign matter provided by an embodiment of the present invention;

[0029] Figure 4 is a diagram showing changes in mutual inductance between the transmitting and receiving coils when the position of a metal foreign body is changed, provided by an embodiment of the present invention;

[0030] Figure 5 is a diagram showing changes in mutual inductance between the transmitting and receiving coils when the material of the metallic foreign body is changed, provided by an embodiment of the present invention;

[0031] Figure 6 This is a topological equivalent circuit diagram of a SS-type WPT system without foreign matter provided by an embodiment of the present invention;

[0032] Figure 7 This is a topological equivalent circuit diagram of an SS-type WPT system with non-magnetic metal foreign matter provided by an embodiment of the present invention;

[0033] Figure 8 is a power transmission loss analysis diagram of a magnetic coupling wireless power transmission system provided by an embodiment of the present invention;

[0034] Fig. 9 is a diagram showing changes in system parameters when the position of the coupling mechanism changes according to an embodiment of the present invention;

[0035] Fig.10 is a diagram of system parameter changes when a DC input voltage changes provided by an embodiment of the present invention;

[0036] Fig.11 is a diagram of system parameter changes when load changes provided by an embodiment of the present invention;

[0037] Fig.12 is a diagram of system parameter changes when the metal foreign body coupling mutual inductance changes provided by an embodiment of the present invention;

[0038] Fig.13 is a diagram showing changes in system parameters when the equivalent internal resistance of a metal foreign body changes according to an embodiment of the present invention;

[0039] Fig.14 is a diagram showing changes in system parameters when the equivalent self-inductance of a metal foreign body changes according to an embodiment of the present invention;

[0040] Fig.15 is an example diagram of BP network prediction results provided by an embodiment of the present invention;

[0041] Fig.16 It is a flow chart of a metal foreign body detection method of a multi-parameter WPT system based on a neural network provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0042] The following specifically illustrates the implementation mode of the present invention in conjunction with the accompanying drawings. The embodiments are provided for illustrative purposes only and cannot be understood as limiting the present invention. The accompanying drawings are provided for reference and illustration only and do not constitute a limitation on the scope of patent protection of the present invention, because many changes can be made to the present invention without departing from the spirit and scope of the present invention.

[0043] The influence of non-magnetic metal foreign bodies on the magnetic field of the coupling mechanism is studied using the finite element simulation software COMSOL. Both the transmitting and receiving coils are wound with copper Litz wire into a circular-circular symmetrical structure. The Litz wire has an outer diameter of 1.9 mm, an inner diameter of 80 mm, an outer diameter of 140 mm, and 14 turns. The non-magnetic metal foreign body (hereinafter referred to as the metal foreign body) is a cube (copper) with a side length of 1 cm, such as Figure 1 shown.

[0044] When high frequency excitation is applied to the transmitting coil, the magnetic flux density model with and without foreign matter (copper) at a distance of 5 cm is as follows: Figure 2 shown. Figure 2 (a) does not contain foreign matter, and the magnetic flux distribution between the coils is relatively uniform. Figure 2 (b) The copper block in the middle blocks part of the magnetic flux, and eddy currents are induced on the metal, resulting in losses.

[0045] Analyze the influence of non-ferromagnetic metal foreign matter on the mutual inductance of coils. Fix the foreign matter material to copper, gradually increase the coil spacing, and keep the foreign matter position always in the middle of the coil. The graph of mutual inductance with coil spacing is as follows: Figure 3As shown, M and Mfod represent the mutual inductance between the transmitting coil and the receiving coil when there is no foreign object and when there is a foreign object. Figure 3 It can be seen that as the distance between the coils increases, the mutual inductance shows a nonlinear and sharp decrease trend. In addition, the model containing metal foreign matter will weaken the coupling between the coils at the same distance due to the magnetic field coupling between the foreign matter and the coil, resulting in a greater reduction in mutual inductance.

[0046] Analyze the influence of different positions and types of non-ferromagnetic metal foreign bodies on the magnetic field of the coupling mechanism. Keep the position of the coupling mechanism unchanged and change the position of the metal foreign body to study the influence of different positions of the foreign body on the magnetic field, such as Figure 4 shown. Figure 4 This indicates that the closer the metal foreign body is to the transmitting coil, the stronger the coupling with the transmitting coil magnetic field, and the smaller the mutual inductance between the two coils, that is, the coupling between the metal foreign body and the receiving coil is weaker.

[0047] Keep the size and position of the metal foreign body unchanged, and change the type of metal foreign body (copper, lead, tin) to study the influence of different types of foreign bodies on the magnetic field, such as Figure 5 As shown, Mcu, Mni and Msn represent the mutual inductance between the transmitting coil and the receiving coil under the influence of copper, lead and tin respectively. Figure 5 It shows that different non-ferromagnetic metal foreign bodies have different effects on the magnetic field. Since the equivalent inductance and internal resistance of the metal foreign body circuit are different, they will have different effects on the magnetic field at the same position.

[0048] Considering the SS type magnetic coupling wireless power transmission system without foreign matter, its equivalent circuit is as follows Figure 6 As shown, where L p and L s Respectively represent the self-inductance of the transmitting coil and the receiving coil, R p and R s Respectively represent the internal resistance of the transmitting coil and the receiving coil, C p and C s Represent the capacitance of the transmitting coil and the receiving coil in series, I p and I s Represent the current of the transmitting coil and the receiving coil respectively, U s Represents the equivalent AC source of the front-end DC power supply and the full-bridge inverter circuit, R eq represents the equivalent load resistance of the rectifier filter circuit and the load, and M represents the mutual inductance between the transmitting coil and the receiving coil. According to Kirchhoff's voltage law, the KVL equations in the transmitting and receiving circuits are:

[0049]

[0050] w represents the operating angular frequency of the system.

[0051] The effective value of the voltage U2 across the transmitting coil and the equivalent internal resistance of the circuit in the resonant cavity is:

[0052]

[0053] The equivalent inverter output square wave voltage source U1 is:

[0054]

[0055] Define the equivalent Q value Q e is the ratio of U2 to U1. According to equations (1)-(3), we can get:

[0056]

[0057] It can be seen from formula (4) that in the SS type magnetic coupling wireless power transmission system, once the system parameters such as the transmitting coil, receiving coil, compensation capacitor, mutual inductance and load resistance are determined, the equivalent quality factor Q e is a certain value, that is, the equivalent quality factor is only related to the system parameters.

[0058] In the equivalent circuit of non-ferromagnetic metal foreign bodies, when non-ferromagnetic conductive media appear between or around the coupling mechanism, it can be equivalent to an eddy current calculation loop with an inductor and a resistor in series. m The internal resistance of the eddy current calculation loop represents the energy loss caused by the eddy current, L m The equivalent inductance of the eddy current calculation loop represents the influence of the eddy current magnetic field on the magnetic field of the resonant coil. pm and M sm Respectively represent the coupling mutual inductance between the metal foreign body and the transmitting coil and the receiving coil, such as Figure 7 shown.

[0059] Similarly, the KVL equations in the three loops are:

[0060]

[0061] Where Z m is the equivalent impedance of the metal foreign body circuit. The equivalent quality factor Q is also defined e for:

[0062]

[0063] The parameters a, b, c, and d that are customized to simplify the formula are as follows:

[0064]

[0065] From equations (6) and (7), we can see that when the inherent parameters of the system are determined, the equivalent quality factor Q e Similarly, the parameter R in the equivalent circuit of the metal foreign bodym , L m 、M pm and M sm When other conditions of the system remain unchanged and there is a metal foreign body near the coupling mechanism, the equivalent quality factor Q e The degree of change depends on the size, type and location of the metal foreign body, so the equivalent quality factor Q e As an indicator of whether non-ferromagnetic metallic foreign matter is present in the system.

[0066] Figure 8 It is the power transmission diagram of SS type magnetic coupling wireless power transmission system. PT is the total energy of the transmitting coil, which is equal to the input power P before the transmitting end drives the system in Subtract the system loss P at the transmitter PTLoss .P PR is the total energy of the receiving coil, which is equal to the output power P after the receiving end rectifier circuit OUT Plus the system loss P at the receiving end PRLoss . Power loss caused by metal foreign matter P Loss Satisfies the following formula:

[0067] P Loss =P PT -P PR (8)

[0068] Theoretically, if P PTLoss and P PRLoss Can accurately measure without foreign matter PT Should be equal to P PR , that is, P Loss Equal to 0. If there is a metal foreign body between the two coils, the metal foreign body will induce eddy current in the magnetic field, forming eddy current loss, making P Loss is much greater than 0, so P can be calculated based on Loss value to detect the presence of metallic foreign matter.

[0069] When the SS type magnetic coupling wireless power transmission system is in the presence of non-ferromagnetic metal foreign matter, its equivalent circuit model is as follows: Figure 7 According to the definition of power loss method and formula (5), the total energy P of the transmitting coil can be calculated respectively. PT and the total energy P of the receiving coil PR :

[0070]

[0071] Where R dson and U d are the conduction internal resistance of the full-bridge inverter switch tube and the conduction voltage drop of the full-bridge rectifier diode, R Lis the load resistance, R eq is the equivalent load from the input port of the rectifier bridge, satisfying:

[0072]

[0073] Similarly, looking backward from the input port of the full-bridge inverter, the equivalent impedance of the rear stage of the full-bridge inverter and the equivalent impedance of the load of the full-bridge rectifier are in a conjugate relationship. That is:

[0074]

[0075] It can be seen from equation (5) that the eddy current generated by the metal foreign body in the magnetic field will form a reflection impedance at the transmitting and receiving ends, thereby affecting the transmitting and receiving coil current I p and I s Then, according to equations (8) and (9), the power loss P is calculated. Loss And then determine whether there is any metal foreign matter.

[0076] In most cases, the position of the coupling mechanism, the input voltage or the load size are not constant, and the system operating conditions may change. Equations (6) and (9) show that whether it is the power loss method or the equivalent quality factor method, when the coil mutual inductance, load or input voltage changes, it will also cause power loss P Loss Greater than 0 or equivalent quality factor Q e If the power loss or equivalent quality factor method is used directly for calculation, the system may misjudge the presence of metal foreign matter due to the change in operating conditions. Therefore, it is very important to analyze the changes in system parameters when the operating conditions change.

[0077] According to the previous analysis, when there is a non-ferromagnetic metal foreign body in the system, it can be equivalent to an eddy current calculation loop with an inductor and a resistor in series. In the simulation, the metal foreign body is represented by an equivalent inductor 2 and an equivalent resistor in series, which is coupled with the transmitting coil 1 and the receiving coil 3. If the foreign body is not considered, the series loop is set to an open circuit.

[0078] Based on the above coupling mechanism modeling and analysis, the parameters for the normal operation of the selected computing system are shown in Table 1 below.

[0079] Table 1 Normal working parameters of the system

[0080]

[0081] The input voltage is 10V, the load resistance is 10Ω, and the distance between the two coils is 5cm, that is, the mutual inductance is 4.2uH as the standard working condition. At this time, the simulation obtains the power loss P Loss is 1.79W, the power loss factor P eqLoss The equivalent quality factor Q is 0.021. eIt is 19.03.

[0082] In the actual operation of wireless power transmission system, the coupling mechanism position, input voltage or equivalent load resistance often change. Although the system power loss P Loss Because there is no loss caused by metal foreign matter between the coils, the loss is basically unchanged. However, this ignores the energy leaked when the magnetic field between the transmitting and receiving coils transfers energy. When the two coils are far apart, the system power loss P Loss will increase sharply due to magnetic field energy leakage. Similarly, when the load is too large, it will cause the transmitting coil current I p Rapidly increases, the receiving coil current I s It decreases sharply. According to equations (8) and (9), the system power loss P Loss Only when the distance between the two coils is appropriate and the load is appropriate, the power loss under the above three working conditions will approach 0. Although when the distance between the two coils is far or the load is large, the system power loss P Loss will be much greater than 0, but at this time the system output power P OUT The value is larger, P Loss With P OUT The ratio P eqLoss As the power loss factor, that is:

[0083]

[0084] The power loss factor can be used to measure whether there is metal foreign matter in the system. When there is no foreign matter in the system, even if the coupling mechanism is far away and the load resistance is large, the power loss factor P eqLoss Still close to 0; when there is metal foreign matter, the power loss factor P eqLoss Will still be greater than 0.

[0085] For the SS topology wireless power transmission system, when the distance between the coupling mechanisms becomes closer, that is, the mutual inductance M between the coils becomes larger, the current I of the transmitting and receiving coils will p and I s All of them decrease. From formula (9), we can know that the total energy P of the transmitting coil at this time is PT The total energy P of the receiving coil PR will be reduced. Since there is no loss caused by foreign matter in the transmitting and receiving coils, if the magnetic field energy leakage is not considered, the system power loss P Loss Close to 0. Since the coupling between the metal foreign body and the receiving coil is weak, in order to simplify the model, the electromagnetic coupling M between the metal foreign body and the receiving end coupling mechanism is ignored. sm From equations (6) and (7), we can see that when the mutual inductance increases, both c and d increase, and the equivalent quality factor Q e Similarly, when the coupling mechanism distance becomes farther or offset, that is, when the mutual inductance between coils decreases, the system power loss PLoss Close to 0, but equivalent quality factor Q e Increase.

[0086] Although the power loss P Loss Within a certain range, no matter how the coupling mechanism position changes, it tends to 0. However, due to the excessive distance of the coupling mechanism, the magnetic leakage increases, resulting in a power loss P Loss is much greater than 0, but the power loss factor P eqLoss Still close to 0. By setting the metal foreign body loop to an open circuit and keeping other parameters unchanged, the equivalent quality factor Q is calculated by adjusting the mutual inductance of the coupling mechanism to simulate the position change of the coupling mechanism. e , power loss P Loss and power loss factor P eqLoss The results of these three parameters are as follows Fig. 9 To facilitate the observation of trends, Fig. 9 Medium power loss factor P eqLoss The value is multiplied by 100 and the unit is %. Fig. 9 It shows that the equivalent quality factor Q e As the mutual inductance decreases, the power loss P Loss When the coupling mechanism is close, that is, when the mutual inductance is large, it approaches 0W, but when the coupling mechanism is too far away, it is greater than 0W. To prevent system misjudgment, the power loss factor P proposed in this example eqLoss It remains close to 0 over the entire distance range.

[0087] When the input voltage changes (1-50V), it can be seen from equations (5) and (9) that the transmitting coil transmission power P PT The receiving coil transmits power P PR Each term of the expression contains U in 2, so P PT With P PR The system power loss P between the transmitting and receiving coils increases without considering the magnetic field energy leakage. Loss is close to 0. Equations (6) and (7) show that the equivalent quality factor Q e It has nothing to do with the input voltage. The simulation results are as follows Fig.10 As shown, Q e Basically keep a constant value and consistent with the value under standard working conditions, P Loss When the input voltage is too large, there will be a large deviation due to the large power of the transmitter and receiver. This is why the power loss method is only applicable to low-power systems when detecting metal foreign bodies. eqLoss Still approaching 0, this indicator can be used to measure whether there are metal foreign bodies in the system.

[0088] Similarly, when the equivalent load resistance R eqWhen the value increases within a certain range (1-100Ω), because there is no loss caused by metal foreign matter, the system power loss P between the transmitting and receiving coils Loss Basically unchanged, but equivalent quality factor Q e The simulation results are as follows: Fig.11 As shown, Q e As the load increases, P Loss It approaches 0 under light load, and will deviate greatly under heavy load due to the large power. However, the power loss factor P eqLoss It is still close to 0, and this indicator can be used to measure whether there are metal foreign bodies in the system.

[0089] When there is a non-ferromagnetic metal foreign body in the system, its equivalent foreign body loop coupling mutual inductance, equivalent internal resistance and equivalent self-inductance may change due to various factors. According to the previous magnetic simulation, the position of the metal foreign body from the transmitting coil affects the foreign body loop coupling mutual inductance, and the type and size of the metal foreign body affect the foreign body loop equivalent internal resistance and equivalent self-inductance. Fig.12 , Fig.13 , Fig.14 The equivalent quality factor Q is when the foreign body circuit coupling mutual inductance, equivalent internal resistance and equivalent self-inductance change. e , power loss P Loss and power loss factor P eqLoss The closer the metal foreign body is to the transmitting coil, the greater the mutual inductance of the foreign body loop coupling, and Q e The lower the P eqLoss The larger the foreign body circuit equivalent internal resistance and equivalent self-inductance affect the heat loss and magnetic loss generated by eddy current respectively. When there is a metal foreign body, Q e are different from the values ​​under normal working conditions, and P eqLoss Larger. According to the equivalent quality factor Q e and power loss factor P eqLoss Evaluate the system for the presence of non-ferrous metallic foreign matter.

[0090] The larger the size of the foreign object, the greater the equivalent internal resistance and self-inductance of the circuit, and the greater the impact on the system coupling. e The more obvious the change, the more accurate the system parameter method identification.

[0091] The previous analysis shows that when the working conditions do not change, the power loss factor P can be set according to the condition without foreign matter. eqLoss and equivalent quality factor Q e The power loss factor P eqLoss and equivalent quality factor Q e Changes may occur, resulting in misjudgment, but this change has a certain trend. Based on the previous analysis, the trend can be summarized as shown in Table 2.

[0092] Table 2 Parameter variation trends under different working conditions

[0093]

[0094] By measuring a large amount of experimental data of system parameters under normal working conditions, the presence of foreign matter in the system and changes in working conditions, the neural network (BP neural network is used in this example) can continuously correct the network weights and thresholds through training these sample data so that the error function decreases along the negative gradient direction and approaches the expected output. When training samples, the input layer data is the power loss factor P eqLoss and equivalent quality factor Q e , and record the working status label for each set of data. Label 0 means the system is working normally, 1 means there is non-ferromagnetic metal foreign matter in the system, and 2 means the working conditions of the system have changed. The output layer is the probability of each working state. After the training is completed, the system parameter Q e With P eqLoss Enter the model for prediction, and the results are as follows Fig.15 As shown, the measured Q e With P eqLoss When the values ​​are 22.65 and 0.7795 respectively, the predicted label is 0, which means there is no metal foreign body; Q e With P eqLoss When the values ​​are 20.62 and 54.776 respectively, the predicted label is 1, which means it contains metallic foreign matter, which is consistent with the simulation results.

[0095] Based on the above analysis, the embodiment of the present invention provides a multi-parameter WPT system metal foreign body detection method based on neural network, such as Fig.16 As shown, it includes the steps of:

[0096] S1. Determine the architecture of the WPT system (such as Figure 6 ) and its design parameters (as shown in Table 1);

[0097] S2. Construct a neural network model (BP neural network model is used in this example) with M input nodes and N output nodes. The M input nodes are used to input the identification parameters of M types of WPT systems (in this example, M=2, which refers to the power loss factor P eqLoss and equivalent quality factor Q e ), N output nodes are used to output the probabilities of N working states (in this case, N = 3, which respectively refer to the normal working of the system, the presence of non-ferromagnetic metal foreign matter, and the change of working conditions, and the corresponding labels are 0, 1, and 2 respectively);

[0098] S3, using the determined WPT system to construct a data set of M identification parameters corresponding to N working states of the WPT system;

[0099] S4. Use the constructed data set to train and test the constructed neural network model (the training goal is to minimize the loss between the test value and the true value);

[0100] S5. Input a set of WPT system identification parameters (a set of power loss factors P eqLoss and equivalent quality factor Q e ) to the trained and tested neural network model, the neural network model outputs the probability of N working states;

[0101] S6. Determine whether there is a metal foreign body based on the label of the working state with the highest probability. If the label is 1, it is determined that there is a metal foreign body.

[0102] In this example, a metal foreign body detection experimental prototype platform is also built for experimental verification. The DC input power supply provides energy, and the electronic load meter is used as the output load. Since the output side power information needs to be collected and transmitted to the transmitter, wireless communication is required between the transmitter and the receiver. The prototype uses 2.4GHz WIFI communication to transmit information such as the output power of the receiver to the transmitter controller. The transmitter calculates the equivalent quality factor Q based on the voltage ratio of U2 to U1 in formula (4) e , and calculate the power loss factor P based on the loss information transmitted back by the receiving end eqLoss According to the fitting function trained by the BP neural network, it is determined whether the system has any operating condition changes or there are metal foreign objects and displayed on the screen in real time.

[0103] Under normal working conditions, when there is no foreign matter, the voltage between the transmitting and receiving ends and the transmitting and receiving powers are detected to calculate the power loss factor P eqLoss With Q e After that, the screen displays the calculated data. When there is no foreign matter, it displays "Non FOD". Under normal working conditions, the load resistance is set to 10Ω, and the corresponding circular coil is wound according to the magnetic circuit simulation above. When the coil distance is 5cm, the mutual inductance is 4.2uH, and the voltage gain G is M / L f =1.91. Input voltage is 3V, and the theoretical output voltage is 5.73V.

[0104] Without introducing foreign matter and only changing the working conditions, the system can identify and display different working conditions by changing the load, input voltage, and coil distance. Changing the load and coil distance will affect the equivalent quality factor Q. e and power loss P Loss , but the power loss factor P eqLoss The values ​​are still low.

[0105] Introduce 1cm 3Non-ferromagnetic metal foreign objects such as copper and tin are placed in the center of the two coils. The system can detect the metal foreign objects and display "FOD Detected" on the screen.

[0106] The experimental results show that after the introduction of metallic foreign bodies copper and tin, in addition to the equivalent quality factor Q e In addition to reducing the power loss factor P, due to the eddy current induced in the metal foreign body, the power loss factor P eqLoss The value is obviously increased, and the system can detect and identify foreign objects when they are in the center of the coil. In addition, the system also has a certain detection effect on other non-ferromagnetic metals such as titanium, aluminum, and zinc.

[0107] It should also be noted that the neural network used in this example is not limited to the BP neural network, and any other neural network with the same function as the BP neural network can be used.

[0108] Based on the above-mentioned multi-parameter WPT system metal foreign body detection method based on neural network, the present invention also provides a multi-parameter WPT system metal foreign body detection system based on neural network, which includes a model generation module, and a transmitting end detection module and a receiving end detection module respectively connected to the transmitting end and the receiving end of the WPT system;

[0109] In the training and testing phase, the transmitter detection module is used to collect information about the transmitter under N working states, the receiver detection module is used to collect information about the receiver under N working states and send it to the transmitter, and the transmitter detection module calculates M identification parameters of the WPT system based on the information collected under N working states and sends them to the model generation module; the model generation module is used to construct a neural network model, and construct a data set based on the M identification parameters corresponding to the N working states, and use the constructed data set to train and test the neural network model;

[0110] In the application stage, the transmitter detection module is used to collect information from the transmitter, and the receiver detection module is used to collect information from the receiver and send it to the transmitter. The transmitter detection module calculates M identification parameters based on the collected information and sends them to the model generation module; the model generation module outputs the probabilities of N working states; the transmitter detection module determines whether there is a metal foreign body based on the label of the working state with the highest probability.

[0111] In summary, the neural network-based multi-parameter WPT system metal foreign body detection method and system provided by the embodiment of the present invention can uniquely determine the one-to-one relationship between different working states of the WPT system based on different combinations of M identification parameters, construct a neural network model and data set according to the M identification parameters corresponding to N working states, and use the constructed data set to train and test the neural network model. The obtained neural network model can output the corresponding working state for any group of M identification parameters, thereby determining whether there is metal foreign matter. The present invention integrates the multiple parameters of the WPT system, uses the neural network model to study the metal foreign body detection under different working conditions of the system, can judge the changes in working conditions, and can identify metal foreign bodies of a certain size at a certain position.

[0112] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. A multi-parameter WPT system metal foreign body detection method based on neural network, characterized in that: Includes steps: Determine the architecture of the WPT system and its design parameters; Constructing a neural network model, wherein the neural network model has M input nodes and N output nodes, the M input nodes are used to input M identification parameters of the WPT system, and the N output nodes are used to output probabilities of N working states of the WPT system; Using the determined WPT system, a data set of M identification parameters corresponding to N working states of the WPT system is constructed; Use the constructed dataset to train and test the constructed neural network model; Inputting a set of M identification parameters of the WPT system into the trained and tested neural network model, the neural network model outputting the probabilities of N working states; Determine whether there is a metal foreign body based on the label of the working state with the highest probability.

2. The method for detecting metal foreign bodies in a multi-parameter WPT system based on a neural network according to claim 1 is characterized in that: M is equal to 2, and the M identification parameters of the WPT system are the ratio of U2 to U1, Q e and power loss factor P eqLoss =P Loss / P OUT , U1 is the equivalent inverter output square wave voltage source, U2 is the effective value of the voltage across the transmitting coil and the equivalent internal resistance of the circuit, P Loss is the power loss caused by metal foreign matter, P OUT is the output power of the system.

3. The method for detecting metal foreign bodies in a multi-parameter WPT system based on a neural network according to claim 2 is characterized in that: Power loss caused by metal foreign matter P Loss Equal to the total energy P of the transmitting coil PT Subtract the total energy P of the receiving coil PR , the total energy P of the transmitting coil PT Equal to the input power P before the transmitter drives the system in Subtract the system loss P at the transmitter PTLoss , the total energy P of the receiving coil PR Equal to the output power P after the receiving end rectifier circuit OUT Plus the system loss P at the receiving end PRLoss .

4. The method for detecting metal foreign bodies in a multi-parameter WPT system based on a neural network according to claim 3 is characterized in that: The transmitting end of the WPT system includes a DC power supply, a full-bridge inverter, a transmitting end series compensation capacitor and a transmitting coil connected in sequence, and the receiving end of the WPT system includes a receiving coil, a receiving end series compensation capacitor, a rectifier filter circuit and a load connected in sequence; System loss at the transmitter P PTLoss equal I p , R p Respectively represent the current and internal resistance of the transmitting coil, R dson represents the on-state internal resistance of the switch tube in the full-bridge inverter; System loss at the receiving end P PRLoss equal I s , R s Respectively represent the current and internal resistance of the receiving coil, U d Represents the conduction voltage drop of the diode in the rectification and filtering circuit.

5. The method for detecting metal foreign bodies in a multi-parameter WPT system based on a neural network according to any one of claims 1 to 4, characterized in that: N=3, the N working status packages are normal system operation, metal foreign matter in the system, and changes in system working conditions, corresponding to three different labels.

6. The method for detecting metal foreign bodies in a multi-parameter WPT system based on a neural network according to claim 5 is characterized in that: When the system is working normally, Q e Fixed, P eqLoss is 0; When there is metal foreign matter in the system, Q e Not less than the preset slope, P eqLoss Greater than 0; When the system working conditions change, Q e , P eqLoss The changes are different when the system is working normally and when there are metal foreign objects in the system.

7. The method for detecting metal foreign bodies in a multi-parameter WPT system based on a neural network according to claim 6 is characterized in that: The neural network model is a BP neural network.

8. A multi-parameter WPT system metal foreign body detection system based on a neural network, applied to a multi-parameter WPT system metal foreign body detection method based on a neural network according to any one of claims 1 to 7, characterized in that: It includes a model generation module, and a transmitter detection module and a receiver detection module respectively connected to a transmitter and a receiver of the WPT system; In the training and testing phase, the transmitter detection module is used to collect information of the transmitter under N working states, the receiver detection module is used to collect information of the receiver under N working states and send it to the transmitter, the transmitter detection module calculates M identification parameters of the WPT system according to the information collected under N working states and sends them to the model generation module; the model generation module is used to construct a neural network model, and construct a data set according to the M identification parameters corresponding to the N working states, and use the constructed data set to train and test the neural network model; During the application stage, the transmitting end detection module is used to collect information from the transmitting end, the receiving end detection module is used to collect information from the receiving end and send it to the transmitting end, the transmitting end detection module calculates M identification parameters based on the collected information and sends them to the model generation module; the model generation module outputs the probabilities of N working states; the transmitting end detection module determines whether there is a metal foreign body based on the label of the working state with the highest probability.

Images