Fault diagnosis system and method for wireless power transmission system

Abstract

The invention discloses a wireless power transmission system fault diagnosis system and method, and relates to the technical field of wireless power transmission. The wireless power transmission system fault diagnosis system comprises a front end, a diagnosis end and a storage end. The front end comprises an electromagnetic monitoring unit and a sensing processing unit; the diagnosis end comprises a feature extraction unit, a fault monitoring unit and a diagnosis processing unit; according to the invention, through continuous correction and compensation of the front end on the state of the sensor and deep fusion and dynamic response of the diagnosis end on the multi-source features, accurate fault diagnosis and safe adaptive control of the wireless power transmission system can be realized in a complex environment with a high-intensity magnetic field and multiple noise interferences.

Description

Technical Field

[0001] The present application relates to the field of wireless power transmission technology, and in particular to a wireless power transmission system fault diagnosis system and method. Background Art

[0002] Wireless Power Transfer (WPT) is a technology that can transmit electrical energy without physical wires. Its core is to transfer energy from a transmitter to a receiver through electromagnetic fields (such as electromagnetic induction, magnetic resonance, microwaves, or lasers). It is widely used in medical (powering implantable medical devices such as pacemakers), industrial (powering high-altitude environmental equipment), electric vehicles (car charging), and drones (wireless charging). However, failures in wireless power transfer systems can lead to performance degradation and even serious safety accidents.

[0003] Current wireless power transmission systems (especially magnetic resonance WPT) generate high-intensity alternating magnetic fields (typically in the frequency range of tens to hundreds of kilohertz) during operation. This strong magnetic environment has a serious impact on various sensors in the sensing layer, causing significant sensor calibration drift over long-term operation. Furthermore, noise and electromagnetic interference in the strong magnetic environment affect the accuracy of feature extraction, thereby affecting the accuracy of wireless power transmission system fault diagnosis.

[0004] In order to solve the above-mentioned defects, a wireless power transmission system fault diagnosis system and method are provided. Summary of the Invention

[0005] According to one aspect of the present application, a wireless power transmission system fault diagnosis system is provided, the method comprising: a front end, a diagnosis end, and a storage end;

[0006] The front end communicates with the sensor to obtain sensing information, wherein the sensing information includes field strength vector, sensitivity, noise density and bandwidth; data integration is performed based on the sensing information to obtain output offset and noise increment, which are then normalized and weighted fusion calculated to obtain interference index; thereby, the interference index of the sensor in each monitoring period can be obtained, and a two-dimensional rectangular coordinate system is constructed with the monitoring period as the horizontal axis and the interference index as the vertical axis. The interference index is depicted in the coordinate system, and then a curve is used to connect them in sequence to obtain a curve graph of interference index-monitoring period; based on the curve graph of interference index-monitoring period, in-depth analysis is performed to obtain a sensing judgment index, and sensing judgment intervals Q1, Q2, Q3 and Q4 are preset. If the sensing judgment index is in the sensing judgment interval Q1, no adjustment is required; if the sensing judgment index is in the sensing judgment interval Q2, a gain compensation strategy is generated; if the sensing judgment index is in the sensing judgment interval Q3, a calibration zeroing strategy is generated; if the sensing judgment index is in the sensing judgment interval Q4, a replacement strategy is generated; corresponding operations are performed according to the execution strategy of each sensor;

[0007] The diagnostic end is used to obtain processed sensor-collected system information of the wireless power transmission system, and extract features in the time domain, frequency domain, and multi-physical fields based on the system information to obtain characteristic parameters; the characteristic parameters include voltage waveform distortion rate, current waveform distortion rate, detuning amount, quality factor, temperature rise, and transmission efficiency; then, normalization processing and weighted fusion are performed based on the characteristic parameters to obtain health indicators, and a health indicator-monitoring cycle curve diagram is constructed, which is deeply analyzed to obtain fault judgment indicators. This process is the same as the deep analysis of the interference indicator-monitoring cycle curve diagram, and corresponding safety measures are generated according to the fault judgment indicators.

[0008] Optionally, the specific process of integrating data to obtain output offset and noise increment is:

[0009] Let the field strength vector be {B x (t),B y (t),B z (t)}, at each sampling time t, according to the formula Calculate the magnetic field vector amplitude B inst (t), the moment of maximum magnetic field vector amplitude within a monitoring period is selected as the peak moment, and the magnetic field vector amplitude corresponding to the peak moment is taken as the modulus of the peak vector and recorded as B0;

[0010] Then according to the formula Calculate the DC component B DC , where T = 1 / f, f is the operating frequency of the wireless power transmission system;

[0011] Extract the sensitivity, noise density, and bandwidth of the sensor and record the noise density and bandwidth as N0 and BW; multiply the sensitivity of the sensor by the DC component to obtain the output offset and record it as ΔV off , according to the formula Calculate the noise increment σ noise .

[0012] Optionally, the process of in-depth analysis of the interference indicator-monitoring period graph is as follows:

[0013] The entire monitoring period interval in the interference index curve is recorded as [T1,T n ], where n represents the total number of monitoring periods in the interference index curve; the interference index γ is calculated using calculus in the entire monitoring period interval [T1,T n ] and compared with the maximum possible area to obtain the normalized area index The calculation formula is:

[0014] Calculate the maximum rising rate of the interference indicator γ in the interference indicator curve, and then divide it by the preset maximum rate threshold Get the normalized rate index The calculation formula is: Where i∈n represents the index of any monitoring period in the entire monitoring period interval [T1,Tn] of the interference index curve graph;

[0015] According to the interference index curve, the monitoring period in which the interference index γ reaches its maximum value is calculated relative to the entire monitoring period interval [T1,T n ] position ratio P γ , the calculation formula is: And calculated according to the formula The interference index γ is higher than the preset threshold L γ The cumulative time percentage within the interval D γ ;

[0016] The area index Rate Indicator Position ratio P γ and the cumulative time percentage D γ Perform weighted fusion to obtain the sensing judgment index.

[0017] Optionally, the policy execution process is:

[0018] When the received execution strategy is a replacement strategy, the sensor replacement and magnetic shielding inspection instructions are sent to the corresponding engineer;

[0019] When the received execution strategy is the calibration zeroing strategy, the output zero point is reset by measuring and clearing the DC bias of the sensor, so that the output of the sensor returns to the ideal zero point in the absence of field or in the reference field. Specifically:

[0020] Step 1: Enter calibration mode: The sensor enters the calibration state, in which external fault alarms are ignored;

[0021] Step 2: Maintain a static magnetic environment: This requires the device to be kept in a static calibration environment with no external magnetic field or a known reference field for a certain period of time to stabilize the output.

[0022] Step 3: Update the calibration register: Extract the original output voltage V at M acquisition moments in the monitoring cycle closest to the current moment raw , calculate the average value of the original output voltage of M samples as the measurement zero offset V of the sensor off and write it into the sensor register and storage end as the zero point reference value in future data processing;

[0023] Step 4: Exit calibration mode: clear the calibration instruction flag, resume online monitoring, and start using the new zero offset to compensate all subsequent outputs;

[0024] Step 5: Calibration verification: After calibration, measure the output under zero magnetic field again to ensure that |Vraw-V off |≤ε, if the calibration verification fails for three consecutive times, the replacement strategy will be implemented;

[0025] When the received execution strategy is the gain compensation strategy, the slope coefficient of the sensor output is adjusted to restore it to the nominal response curve when there is slight drift or noise disturbance; specifically:

[0026] Step 1: Obtain the reference point: Under no-load or standard field strength conditions, measure the original output curve of the sensor and record the reference slope S ref and zero point output;

[0027] Step 2: Calculate the current slope S cur :Online calculation or simulation to estimate the current sensitivity S cur =ΔV / ΔB, where ΔV is the output increment under a known field strength step change at both ends, and ΔB is the field strength step change size;

[0028] Step 3: Calculate the slope compensation factor: Slope compensation factor K g Is the gain factor used for digital compensation. In the signal processing pipeline, the original reading is first subtracted from the zero offset and then multiplied by K g , it can restore to the "equivalent" reference slope and eliminate the sensitivity drift caused by temperature drift or magnetic interference;

[0029] Step 4: Digital compensation: In the sensor data output pipeline, the original voltage V raw First subtract the current zero offset, then multiply by K g :Vc omp =K g ·(V raw -V off )+V mind , where V raw To measure zero offset, V mind is the midpoint voltage, K g Store it in the storage for the next initialization load.

[0030] Optionally, the feature parameter extraction process is:

[0031] Time domain analysis:

[0032] The harmonic components of the output voltage are extracted to obtain the fundamental amplitude V1, the 2nd harmonic amplitude V2, the 3rd harmonic amplitude V3, and so on until the hth harmonic amplitude V h , where h is a positive integer; according to the formula Calculate the voltage waveform distortion rate V THD ; From this, the voltage waveform distortion rate V at each moment in the monitoring period can be obtained THD , and calculate the mean value to get the voltage waveform distortion rate of the monitoring period. The voltage waveform distortion rate of each monitoring period can be recorded as

[0033] For output current I out Harmonic components are extracted to obtain the fundamental amplitude I1, the second harmonic amplitude I2, the third harmonic amplitude I3, and so on until the hth harmonic amplitude I h , where h is a positive integer; according to the formula Calculate the distortion rate I of the current waveform THD ; From this, the current waveform distortion rate I at each moment in the monitoring period can be obtained THD , and calculate the mean value to get the current waveform distortion rate of the monitoring period. The current waveform distortion rate of each monitoring period can be recorded as

[0034] Frequency domain analysis:

[0035] Subtract the nominal resonant frequency from the actual resonant frequency and take the absolute value to get the detuning amount, which is recorded as Δf r , thus the detuning amount corresponding to each moment in the monitoring period can be obtained, and the mean value is calculated as the detuning amount of the monitoring period and recorded as The quality factor G is obtained by dividing the actual resonant frequency at each moment in the monitoring period by the -3dB bandwidth of the resonance curve, and the quality factor G at each moment is averaged to obtain the quality factor of the monitoring period, which is recorded as G. i ;

[0036] Multiphysics Analysis:

[0037] The relationship between magnetic field intensity and temperature rise is established through finite element simulation: ΔT = k·B inst (t) 2 ·f·t, where k is the thermal conductivity coefficient of the material. From this, the temperature rise at each moment in the monitoring period can be obtained, and the average value is calculated to obtain the temperature rise during the monitoring period, which is recorded as ΔT i ;

[0038] The input power is obtained by multiplying the input voltage by the input current, and the output power is obtained by multiplying the output voltage by the output current. The output power is then divided by the input power and multiplied by 100% to obtain the transmission efficiency η. The transmission efficiency at each moment in the monitoring period can be obtained by averaging the transmission efficiency of the monitoring period, which is recorded as η. i .

[0039] Optionally, corresponding safety measures are taken based on the received fault judgment indicators, specifically:

[0040] Fault judgment intervals R1, R2, R3, and R4 are set. If the fault judgment index is in the fault judgment interval R1, no adjustment is required and continuous monitoring is sufficient. If the fault judgment index is in the fault judgment interval R2, the transmission power is reduced by 20%. If the fault judgment index is in the fault judgment interval Q3, the temperature rise ΔT of the monitoring period closest to the current time of the system is extracted. n :(1) When ΔT n <ΔT crit When , according to the formula Calculate the transmission power adjustment value P reduce , and follow P reduce Adjust the transmission power; P rated Indicates rated power; ΔT crit Indicates the critical temperature rise; (2) When ΔT n ≥ΔT crit If the fault judgment index is in the fault judgment interval Q4, the system will be shut down.

[0041] According to one aspect of the present application, a method for diagnosing a fault in a wireless power transmission system is provided, comprising the following steps:

[0042] U1: Communicates with the sensor to obtain sensing information, including field strength vector, sensitivity, noise density and bandwidth;

[0043] U2: Based on the sensor information, data integration is performed to obtain the output offset and noise increment, which are then normalized and weighted fused to obtain the interference index. This can be used to obtain the interference index of the sensor in each monitoring period.

[0044] U3: A two-dimensional rectangular coordinate system is constructed with the monitoring period as the horizontal coordinate and the interference index as the vertical coordinate. The interference index is depicted in the coordinate system, and then a curve is used to connect them in sequence to obtain a curve graph of the interference index-monitoring period. The sensing judgment index is obtained by performing in-depth analysis based on the curve graph of the interference index-monitoring period. There are preset sensing judgment intervals Q1, Q2, Q3 and Q4. If the sensing judgment index is in the sensing judgment interval Q1, no adjustment is required; if the sensing judgment index is in the sensing judgment interval Q2, a gain compensation strategy is generated; if the sensing judgment index is in the sensing judgment interval Q3, a calibration zeroing strategy is generated; if the sensing judgment index is in the sensing judgment interval Q4, a replacement strategy is generated; corresponding operations are performed according to the execution strategy of each sensor;

[0045] U3-1: When the received execution strategy is the replacement strategy, it sends the sensor replacement and magnetic shielding inspection instructions to the corresponding engineer;

[0046] U3-2: When the received execution strategy is the calibration zeroing strategy, the output zero point is reset by measuring and clearing the DC bias of the sensor, so that the output of the sensor returns to the ideal zero point in the absence of field or in the reference field;

[0047] U3-3: When the received execution strategy is the gain compensation strategy, the slope coefficient of the sensor output is adjusted to restore it to the nominal response curve when there is a slight drift or noise disturbance;

[0048] U4: Obtains processed sensor-collected system information of the wireless power transmission system and extracts characteristic parameters based on the system information in the time domain, frequency domain, and multi-physics field. The characteristic parameters include voltage waveform distortion rate, current waveform distortion rate, detuning, quality factor, temperature rise, and transmission efficiency.

[0049] U5: Normalize the characteristic parameters and perform weighted fusion to obtain the health index. Construct a health index-monitoring period curve graph, and perform in-depth analysis on it to obtain the fault judgment index. This process is the same as the in-depth analysis of the interference index-monitoring period curve graph. Generate corresponding safety measures based on the fault judgment index.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] (1) This application uses the interference index-monitoring cycle curve to conduct in-depth analysis to accurately quantify the alternating magnetic field and noise interference to the sensor, and combined with the sensor's own characteristics, it can dynamically determine the optimal processing strategy. Even in a high-intensity alternating magnetic field and electromagnetic noise environment, the front end can maintain the accuracy of each sensor's readings, ensuring the data quality of the entire fault diagnosis chain, thereby significantly improving the fault diagnosis accuracy and reliability of the WPT system in a strong magnetic environment; compared with the complex equivalent circuit multi-stage coupling model, it only captures the two easily measurable interference dimensions of "DC pull-off" and "AC noise", achieving online diagnosis and engineering feasibility;

[0052] (2) This application comprehensively monitors the wireless power transmission system through the fusion of time domain, frequency domain, and multi-physical fields to obtain a health indicator-monitoring cycle curve diagram, forming a continuous tracking of the entire life cycle of the wireless power transmission system, and a deep fusion and dynamic response to multi-source features; based on the deep analysis of the health indicator-monitoring cycle curve diagram, a fault judgment index is obtained, and corresponding safety measures are executed based on this, realizing a closed-loop process of feature extraction, indicator synthesis, trend judgment, and safety response. The system can automatically evaluate the health status in each monitoring cycle and dynamically trigger the most appropriate protection or maintenance measures according to the trend to ensure the safety and reliability of the wireless power transmission system. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The accompanying drawings, which constitute part of this application, are intended to provide a further understanding of this application. The exemplary embodiments and descriptions of this application are intended to explain this application and do not constitute an improper limitation on this application. In the accompanying drawings:

[0054] Figure 1 It is a system connection block diagram of the present invention;

[0055] Figure 2 It is a flowchart of the method of the present invention. DETAILED DESCRIPTION

[0056] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0057] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0058] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0059] like Figure 1 As shown, the embodiment of the present application provides a wireless power transmission system fault diagnosis system, which includes: a front end, a diagnosis end and a storage end; the front end includes an electromagnetic monitoring unit and a sensor processing unit; the diagnosis end includes a feature extraction unit, a fault monitoring unit, and a diagnosis processing unit;

[0060] The front end accurately measures the electromagnetic interference of each sensor during wireless power transmission, performs a comprehensive analysis to determine the sensor status, and implements corresponding execution strategies for each sensor to achieve automatic adjustment of the sensor; specifically:

[0061] The electromagnetic monitoring unit monitors and analyzes the electromagnetic interference of each sensor in the wireless power transmission system to accurately quantify the interference suffered by the sensor, specifically:

[0062] 3-5 small three-axis Hall arrays are installed around each sensor at ±50mm to collect the field intensity vector of the sensor in real time and record it as {B x (t),B y (t),B z (t)}; All Hall sensors are synchronized through hardware triggering or time stamping to ensure alignment of the three-axis components and avoid phase error; at each sampling time t, according to the formula Calculate the magnetic field vector amplitude B inst (t), the moment of maximum magnetic field vector amplitude within a monitoring period is selected as the peak moment, and the magnetic field vector amplitude corresponding to the peak moment is taken as the modulus of the peak vector and recorded as B0;

[0063] Then according to the formula Calculate the DC component B DC , where T = 1 / f, f is the operating frequency of the wireless power transmission system;

[0064] It should be explained that those skilled in the art have also attempted to measure magnetic induction coupling interference by calculating the induced electromotive force in the equivalent circuit of the sensor during operation of the wireless power transmission system. However, this method involves multi-stage coupling analysis of tiny loops inside the sensor, PCB traces, and shielding. The model is extremely complex and difficult to apply online, and it does not have the feasibility of online diagnosis. It also lacks generalization, and the model development and maintenance costs are extremely high, far exceeding the capabilities of most industrial field controllers. In comparison, those skilled in the art ultimately chose to capture the two major interference effects of DC bias and AC noise on the sensor output. This method is more intuitive, simple, and engineering-feasible, and is more suitable for online diagnosis and maintenance processes.

[0065] Obtain the sensitivity, noise density, and bandwidth of the sensor and record the noise density and bandwidth as N0 and BW; obtain the output offset of the sensor by multiplying it by the DC component and record it as ΔV off , according to the formula Calculate the noise increment σ noise ;

[0066] Shift the output by ΔV off and noise increment σ noise Normalization is performed to obtain the DC bias coefficient α and the noise coefficient β, and then the DC bias coefficient α and the noise coefficient β are weighted and fused to obtain the interference index γ. The normalization formula is: Where V FS is the full-scale voltage, that is, the maximum voltage swing that the sensor can output within its rated measurement range; it should be explained that the DC bias coefficient α reflects the severity of the zero drift. When there is a DC magnetic field component B in the environment DC When the DC magnetic field is applied, the output voltage of the sensor will shift linearly according to its sensitivity (unit: mV / mT). The output offset reflects the DC bias effect of the DC magnetic field on the internal circuit of the sensor (bias current, amplifier, etc.).

[0067] Thus, the interference index of the sensor in each monitoring period can be obtained. A two-dimensional rectangular coordinate system is constructed with the monitoring period as the horizontal coordinate and the interference index as the vertical coordinate. The interference index is depicted in the coordinate system, and then the curve is connected in sequence to obtain the interference index-monitoring period curve graph; the entire monitoring period interval in the interference index curve graph is recorded as [T1,T n ], where n represents the total number of monitoring periods in the interference index curve; the interference index γ is calculated using calculus in the entire monitoring period interval [T1,T n ] and the maximum possible area (i.e. 1× total period length) to obtain the normalized area index The calculation formula is: From the formula, we can see that It reflects the cumulative amount of the overall interference level, The larger the value, the more serious the long-term interference (electromagnetic and noise);

[0068] Calculate the maximum rising rate of the interference indicator γ in the interference indicator curve, and then divide it by the preset maximum rate threshold Get the normalized rate index Used to capture sudden interference events; the calculation formula is: Where i∈n represents the index of any monitoring period in the entire monitoring period interval [T1,Tn] of the interference index curve graph;

[0069] According to the interference index curve, the monitoring period in which the interference index γ reaches its maximum value is calculated relative to the entire monitoring period interval [T1,T n ] position ratio P γ , the calculation formula is: And calculated according to the formula The interference index γ is higher than the preset threshold L γ (e.g. 0.6) The cumulative time percentage within the interval D γ ;

[0070] The area index Rate Indicator Position ratio P γ and the cumulative time percentage D γ Weighted fusion is performed to obtain a sensing judgment index; sensing judgment intervals Q1, Q2, Q3, and Q4 are set, and those skilled in the art set them to (0, 0.2], (0.2, 0.4], (0.4, 0.7], and (0.7, 1], respectively. Those skilled in the art can fine-tune the sensing judgment intervals according to actual needs; if the sensing judgment index is in the sensing judgment interval Q1, it indicates that the sensor is in a normal state and no adjustment is required; if the sensing judgment index is in the sensing judgment interval Q2, it indicates that the sensor has a slight drift and needs to be gain compensated, and a gain compensation strategy is generated; if the sensing judgment index is in the sensing judgment interval Q3, it indicates that the sensor has a certain degree of drift and needs to be recalibrated and zeroed, and a calibration and zeroing strategy is generated; if the sensing judgment index is in the sensing judgment interval Q4, it indicates that the sensor has suffered serious interference and a replacement strategy is generated; thereby, each sensor generates a corresponding execution strategy and sends the execution strategy to the sensor processing unit;

[0071] The sensor processing unit performs corresponding processing based on the execution strategy generated by the sensor, specifically:

[0072] When the received execution strategy is a replacement strategy, the sensor replacement and magnetic shielding inspection instructions are sent to the corresponding engineer;

[0073] When the received execution strategy is the calibration zeroing strategy, the output zero point is reset by measuring and clearing the DC bias of the sensor, so that the output of the sensor returns to the ideal zero point in the absence of field or in the reference field. The specific process is as follows:

[0074] Step 1: Enter calibration mode: The sensor enters the calibration state, in which external fault alarms are ignored;

[0075] Step 2: Maintain a static magnetic environment: This requires maintaining a static calibration time in an environment without an external magnetic field or with a known reference field (e.g., turning off the WPT power or enabling the coil static field) to stabilize the output.

[0076] Step 3: Update the calibration register: Extract the original output voltage V at M acquisition moments in the monitoring cycle closest to the current moment raw , calculate the average value of the original output voltage of M samples as the measurement zero offset V of the sensor off and write it into the sensor register and storage end as the zero point reference value in future data processing;

[0077] Step 4: Exit calibration mode: clear the calibration instruction flag, resume online monitoring, and start using the new zero offset to compensate all subsequent outputs;

[0078] Step 5: Calibration verification: After calibration is completed, measure the output under zero magnetic field again to ensure that |V raw -V off |≤ε, if the calibration verification fails for three consecutive times, the replacement strategy will be adopted, and ε is the maximum deviation allowed by this system;

[0079] When the received execution strategy is the gain compensation strategy, the slope coefficient of the sensor output is adjusted to restore it to the nominal response curve when there is slight drift or noise disturbance. The specific process is as follows:

[0080] Step 1: Obtain the reference point: Under no-load or standard field strength conditions (such as the known magnetic field output by a standard Helmholtz coil), measure the original output curve of the sensor and record the reference slope S ref and zero point output;

[0081] Step 2: Calculate the current slope S cur :Online calculation or simulation to estimate the current sensitivity S cur =ΔV / ΔB, where ΔV is the output increment under a known field strength step change at both ends, and ΔB is the field strength step change size;

[0082] Step 3: Calculate the slope compensation factor: Slope compensation factor K g is the gain factor used for digital compensation. When the sensor sensitivity decreases (S cur<S ref ), K g >1 will amplify the output. When the sensitivity increases unexpectedly, K g <1 will be attenuated; in the signal processing pipeline, the original reading is first subtracted from the zero offset and then multiplied by K g , it can restore to the "equivalent" reference slope and eliminate the sensitivity drift caused by temperature drift or magnetic interference;

[0083] Step 4: Digital compensation: In the sensor data output pipeline, the original voltage V raw First subtract the current zero offset, then multiply by K g :Vc omp =K g ·(V raw -V off )+V mind , where V raw To measure zero offset, V mind is the midpoint voltage, K g Store it in the storage end for the next initialization load;

[0084] It should be explained that the compensation process should be performed when the system is idle or under low load to prevent external interference from affecting the slope measurement. g When it exceeds a reasonable range (e.g., 0.8–1.2), the calibration zeroing strategy is switched;

[0085] Through in-depth analysis of the interference index-monitoring cycle curve, the alternating magnetic field and noise interference to the sensor can be accurately quantified. Combined with the sensor's own characteristics, the optimal processing strategy can be dynamically determined. Even in high-intensity alternating magnetic fields and electromagnetic noise environments, the front end can maintain the accuracy of each sensor's readings, ensuring the data quality of the entire fault diagnosis chain, thereby significantly improving the fault diagnosis accuracy and reliability of the WPT system in strong magnetic environments. Compared with the complex equivalent circuit multi-stage coupling model, it only captures the two easily measurable interference dimensions of "DC pull-up" and "AC noise", realizing online diagnosis and engineering feasibility.

[0086] The diagnostic end includes a feature extraction unit, a fault monitoring unit, and a diagnostic processing unit. It collects system information of the wireless power transmission system through sensors after processing (gain compensation, calibration, zeroing, or replacement). Based on this information, it monitors and analyzes the wireless power transmission system to diagnose faults and implement safety measures in a timely manner to ensure the safety of wireless power transmission faults. Specifically:

[0087] System information includes the magnetic field distribution around the receiving coil, the resonant frequency f r , harmonic components, electrical parameters, thermal parameters, and mechanical parameters; the magnetic field distribution includes amplitude B and frequency f, and the electrical parameters include input voltage (input voltage at the transmitter) Vin , Input current (input current at the transmitter) I out , output voltage (output voltage at the receiving end) V out and output current; thermal parameters include coil temperature and reference temperature (ambient temperature or starting temperature);

[0088] The feature extraction unit extracts time domain, frequency domain and multi-physical field features through the sensing information of the wireless power transmission system to obtain the characteristic parameters of the monitoring period, where the characteristic parameters include the voltage waveform distortion rate Current waveform distortion rate Detuning Quality factor G i , temperature rise ΔT i and transmission efficiency η i Specifically:

[0089] Time domain analysis:

[0090] The output voltage V out Harmonic components are extracted to obtain the fundamental amplitude V1, the second harmonic amplitude V2, the third harmonic amplitude V3, and so on until the hth harmonic amplitude V h , where h is a positive integer; according to the formula Calculate the voltage waveform distortion rate V THD ; It needs to be explained that the voltage waveform distortion rate V THD The larger the value, the greater the risk of rectifier / inverter failure or load abnormality (nonlinear distortion). From this, the voltage waveform distortion rate V at each moment in the monitoring period can be obtained. THD , and calculate the mean value to get the voltage waveform distortion rate of the monitoring period. The voltage waveform distortion rate of each monitoring period can be recorded as

[0091] For output current I out Harmonic components are extracted to obtain the fundamental amplitude I1, the second harmonic amplitude I2, the third harmonic amplitude I3, and so on until the hth harmonic amplitude I h , where h is a positive integer; according to the formula Calculate the distortion rate I of the current waveform THD ; It needs to be explained that the current waveform distortion rate I THD The larger the value, the greater the risk of load-side filter, diode, or switch component failure. From this, the current waveform distortion rate I at each moment in the monitoring cycle can be obtained. THD , and calculate the mean value to get the current waveform distortion rate of the monitoring period. The current waveform distortion rate of each monitoring period can be recorded as

[0092] Frequency domain analysis:

[0093] Subtract the nominal resonant frequency from the actual resonant frequency and take the absolute value to get the detuning amount, which is recorded as Δf r , thus the detuning amount corresponding to each moment in the monitoring period can be obtained, and the mean value is calculated as the detuning amount of the monitoring period and recorded as The quality factor G is obtained by dividing the actual resonant frequency at each moment in the monitoring period by the -3dB bandwidth of the resonance curve, and the quality factor G at each moment is averaged to obtain the quality factor of the monitoring period, which is recorded as G. i ;

[0094] Multiphysics Analysis:

[0095] The relationship between magnetic field intensity and temperature rise is established through finite element simulation (such as COMSOL): ΔT = k·B inst (t) 2 ·f·t, where k is the thermal conductivity coefficient of the material. From this, the temperature rise at each moment in the monitoring period can be obtained, and the average value is calculated to obtain the temperature rise during the monitoring period, which is recorded as ΔT i ;

[0096] Multiply the input voltage by the input current to get the input power P in , the output voltage multiplied by the output current is the output power P out , then divide the output power by the input power and multiply by 100% to get the transmission efficiency η; thus, the transmission efficiency at each moment in the monitoring period can be obtained, and the average value is calculated to get the transmission efficiency of the monitoring period, which is recorded as η i ;

[0097] The fault monitoring unit evaluates the health status of the wireless power transmission system by comprehensively analyzing the characteristic parameters, and continuously tracks the health status of the power transmission system to output fault judgment indicators of the wireless power transmission system; specifically:

[0098] The characteristic parameters are processed by max-min linear normalization, and the processed characteristic parameters are weighted fused to obtain the health index Γ i ; Specifically, the max-min linear normalization processing formula is: in The maximum voltage distortion value and the minimum voltage distortion value are set by default in this system (usually the maximum voltage distortion value and the minimum voltage distortion value within the monitoring period); The maximum current distortion value and the minimum current distortion value are set by default in this system (usually the maximum current distortion value and the minimum current distortion value within the monitoring period); is the maximum detuning value and minimum detuning value set by default in this system (usually the maximum detuning value and minimum detuning value within the monitoring period); max(G i )、min(G i) is the maximum quality factor and minimum quality factor set by default in this system (usually the maximum quality factor and minimum quality factor within the monitoring period); max(ΔT i )、min(ΔT i ) are the maximum and minimum temperature rises set by default in this system (usually taken within the monitoring period); max(η i ), min(η i ) are the maximum transmission efficiency and minimum transmission efficiency set by default in this system (usually the maximum transmission efficiency and minimum transmission efficiency within the monitoring period);

[0099] With monitoring period T i As the horizontal axis, the health index Γ i Construct a two-dimensional rectangular coordinate system for the vertical coordinate and transform the health index Γ i Depict it in the coordinate system and connect it with smooth curves to get the curve diagram of health index-monitoring period; calculate the health index Γ using calculus i During the entire monitoring period [T1,T n ] and the maximum possible area (i.e. 1 × total period length T1 to T n ), and obtain the normalized area index The calculation formula is: From the formula, we can see that The overall failure risk of the reaction is cumulative;

[0100] Calculate the health index Γ in the health index curve i The maximum rate of ascent is then divided by the preset maximum rate threshold. Get the normalized rate index Used to capture sudden failure events; the calculation formula is: Where i∈n, rate indication The larger the value, the more serious the risk of failure.

[0101] Calculate the health index Γ i The monitoring period that reaches the maximum value is relative to the entire monitoring period interval [T1,T n ] position ratio P Γ , the calculation formula is: And calculated according to the formula Health Index P Γ Above the preset threshold L Γ The cumulative time percentage within the interval D Γ ;

[0102] The area index Rate Indicator Position ratio P Γ and the cumulative time percentage D ΓPerform weighted fusion to obtain fault judgment indicators and send them to the diagnosis processing unit;

[0103] The diagnostic processing unit takes corresponding safety measures based on the received fault judgment indicators to ensure the safe operation of the wireless power transmission system; specifically:

[0104] Fault judgment intervals R1, R2, R3 and R4 are set, and those skilled in the art take their values ​​as (0, 0.2], (0.2, 0.4], (0.3, 0.6] and (0.6, 1] respectively, and their values ​​can be manually fine-tuned; if the fault judgment index is in the fault judgment interval R1, it means that the wireless power transmission system is operating normally, and no adjustment is required, and continuous monitoring is sufficient; if the fault judgment index is in the fault judgment interval R2, it means that there is a slight fault risk, and the transmission power is reduced by 20% (that is, 80% of the rated power is used for wireless power transmission); if the fault judgment index is in the fault judgment interval Q3, it means that there is a moderate fault risk, and the temperature rise ΔT of the monitoring period closest to the current moment of the system is extracted. n , (1) When ΔT n <ΔT crit (the default critical temperature rise of this system), then according to the formula Calculate the transmission power adjustment value P reduce , and follow P reduce Adjust the transmission power; P rated Indicates rated power; ΔT crit It represents the critical temperature rise, which is usually determined by the device thermal limit (such as the upper limit of the semiconductor junction temperature) and the temperature limit of the insulation material. For details, please refer to the specification or thermal design parameters. From the formula, we can see that P reduce Maximum is 50% of rated power, P reduce The minimum is 30% of the rated power; (2) When ΔT n ≥ΔT crit When the fault judgment index is in the fault judgment interval Q4, it indicates that there is a serious fault risk in the wireless power transmission system, and the system is shut down for processing;

[0105] By comprehensively monitoring the wireless power transmission system through the fusion of time domain, frequency domain, and multi-physical fields, a health indicator-monitoring cycle curve is obtained, forming a continuous tracking of the entire life cycle of the wireless power transmission system, and deep integration and dynamic response to multi-source characteristics; based on the in-depth analysis of the health indicator-monitoring cycle curve, a fault judgment index is obtained, and corresponding safety measures are implemented based on this, realizing a closed-loop process of feature extraction, indicator synthesis, trend judgment, and safety response. The system can automatically evaluate the health status within each monitoring cycle and dynamically trigger the most appropriate protection or maintenance measures based on the trend to ensure the safety and reliability of the wireless power transmission system.

[0106] like Figure 2 As shown, the embodiment of the present application also provides a method for diagnosing faults in a wireless power transmission system, the method comprising the following steps:

[0107] U1: Communicates with the sensor to obtain sensing information, including field strength vector, sensitivity, noise density and bandwidth;

[0108] U2: Based on the sensor information, data integration is performed to obtain the output offset and noise increment, which are then normalized and weighted fused to obtain the interference index. This can be used to obtain the interference index of the sensor in each monitoring period.

[0109] U3: A two-dimensional rectangular coordinate system is constructed with the monitoring period as the horizontal coordinate and the interference index as the vertical coordinate. The interference index is depicted in the coordinate system, and then a curve is used to connect them in sequence to obtain a curve graph of the interference index-monitoring period. The sensing judgment index is obtained by performing in-depth analysis based on the curve graph of the interference index-monitoring period. There are preset sensing judgment intervals Q1, Q2, Q3 and Q4. If the sensing judgment index is in the sensing judgment interval Q1, no adjustment is required; if the sensing judgment index is in the sensing judgment interval Q2, a gain compensation strategy is generated; if the sensing judgment index is in the sensing judgment interval Q3, a calibration zeroing strategy is generated; if the sensing judgment index is in the sensing judgment interval Q4, a replacement strategy is generated; corresponding operations are performed according to the execution strategy of each sensor;

[0110] U3-1: When the received execution strategy is the replacement strategy, it sends the sensor replacement and magnetic shielding inspection instructions to the corresponding engineer;

[0111] U3-2: When the received execution strategy is the calibration zeroing strategy, the output zero point is reset by measuring and clearing the DC bias of the sensor, so that the output of the sensor returns to the ideal zero point in the absence of field or in the reference field;

[0112] U3-3: When the received execution strategy is the gain compensation strategy, the slope coefficient of the sensor output is adjusted to restore it to the nominal response curve when there is a slight drift or noise disturbance;

[0113] U4: Obtains processed sensor-collected system information of the wireless power transmission system and extracts characteristic parameters based on the system information in the time domain, frequency domain, and multi-physics field. The characteristic parameters include voltage waveform distortion rate, current waveform distortion rate, detuning, quality factor, temperature rise, and transmission efficiency.

[0114] U5: Normalize the characteristic parameters and perform weighted fusion to obtain the health index. Construct a health index-monitoring period curve graph, and perform in-depth analysis on it to obtain the fault judgment index. This process is the same as the in-depth analysis of the interference index-monitoring period curve graph. Generate corresponding safety measures based on the fault judgment index.

[0115] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0116] The above are merely embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should all be included within the scope of the claims of the present application.

Claims

1. A wireless power transmission system fault diagnosis system, comprising: Front end, diagnostic end and storage end; characterized in that the front end is connected to the sensor for communication to obtain sensor information, performs data integration based on the sensor information to obtain output offset and noise increment, and performs normalization and weighted fusion calculation to obtain interference index; then obtains the interference index of the sensor in each monitoring period, constructs a two-dimensional rectangular coordinate system with the monitoring period as the horizontal coordinate and the interference index as the vertical coordinate, depicts the interference index in the coordinate system, and then uses curves to connect in sequence to obtain a curve graph of interference index-monitoring period; performs in-depth analysis based on the curve graph of interference index-monitoring period to obtain a sensor judgment index, compares it with a preset sensor judgment interval to generate a gain compensation strategy or a correction zeroing strategy or a replacement strategy; performs corresponding operations according to the execution strategy of each sensor; The diagnostic end is used to obtain processed sensor-collected system information of the wireless power transmission system, perform feature extraction in the time domain, frequency domain, and multi-physical fields based on the system information to obtain feature parameters, and then perform normalization and weighted fusion based on the feature parameters to obtain health indicators. A health indicator-monitoring cycle curve graph is constructed, and an in-depth analysis is performed on it to obtain fault judgment indicators. This process is the same as the in-depth analysis of the interference indicator-monitoring cycle curve graph, and corresponding safety measures are generated based on the fault judgment indicators.

2. A wireless power transmission system fault diagnosis system according to claim 1, characterized in that: The specific process of data integration to obtain output offset and noise increment is: Let the field strength vector be {B x (t),B y (t),B z (t)}, at each sampling time t, according to the formula Calculate the magnetic field vector amplitude B inst (t), the moment of maximum magnetic field vector amplitude within a monitoring period is selected as the peak moment, and the magnetic field vector amplitude corresponding to the peak moment is taken as the modulus of the peak vector and recorded as B0; Then according to the formula Calculate the DC component B DC , where T = 1 / f, f is the operating frequency of the wireless power transmission system; Extract the sensitivity, noise density, and bandwidth of the sensor and record the noise density and bandwidth as N0 and BW; multiply the sensitivity of the sensor by the DC component to obtain the output offset and record it as ΔV off , according to the formula Calculate the noise increment σ noise .

3. A wireless power transmission system fault diagnosis system according to claim 2, characterized in that: The process of in-depth analysis of the interference indicator-monitoring period graph is as follows: The entire monitoring period interval in the interference index curve is recorded as [T1,T n ], where n represents the total number of monitoring periods in the interference index curve; the interference index γ is calculated using calculus in the entire monitoring period interval [T1,T n ] and compared with the maximum possible area to obtain the normalized area index The calculation formula is: Calculate the maximum rising rate of the interference indicator γ in the interference indicator curve, and then divide it by the preset maximum rate threshold Get the normalized rate index The calculation formula is: Where i∈n represents the index of any monitoring period in the entire monitoring period interval [T1,Tn] of the interference index curve graph; According to the interference index curve, the monitoring period in which the interference index γ reaches its maximum value is calculated relative to the entire monitoring period interval [T1,T n ] position ratio P γ , the calculation formula is: And calculated according to the formula The interference index γ is higher than the preset threshold L γ The cumulative time percentage within the interval D γ ; The area index Rate Indicator Position ratio P γ and the cumulative time percentage D γ Perform weighted fusion to obtain sensing judgment indicators.

4. A wireless power transmission system fault diagnosis system according to claim 3, characterized in that: The execution strategy process is: When the received execution strategy is a replacement strategy, the sensor replacement and magnetic shielding inspection instructions are sent to the corresponding engineer; When the received execution strategy is the calibration zeroing strategy, the output zero point is reset by measuring and clearing the DC bias of the sensor, so that the output of the sensor returns to the ideal zero point in the absence of field or in the reference field; When the received execution strategy is a gain compensation strategy, the slope coefficient of the sensor output is adjusted so that the sensor is restored to the nominal response curve when there is a slight drift or noise disturbance.

5. A wireless power transmission system fault diagnosis system according to claim 4, characterized in that: The execution process of the calibration zeroing strategy is as follows: Step 1: Enter calibration mode: The sensor enters the calibration state, in which external fault alarms are ignored; Step 2: Maintain a static magnetic environment: This requires the device to be kept in a static calibration environment with no external magnetic field or a known reference field for a certain period of time to stabilize the output. Step 3: Update the calibration register: Extract the original output voltage V at M acquisition moments in the monitoring cycle closest to the current moment raw , calculate the average value of the original output voltage of M samples as the measurement zero offset V of the sensor off and write it into the sensor register and storage end; Step 4: Exit calibration mode: clear the calibration instruction flag, resume online monitoring, and start using the new zero offset to compensate all subsequent outputs; Step 5: Calibration verification: After calibration is completed, measure the output under zero magnetic field again to ensure that |V raw -V off |≤ε, if the calibration verification fails for three consecutive times, the replacement strategy will be implemented.

6. A wireless power transmission system fault diagnosis system according to claim 5, characterized in that: The execution process of the gain compensation strategy is: Step 1: Obtain the reference point: Under no-load or standard field strength conditions, measure the original output curve of the sensor and extract the reference slope S ref and zero point output; Step 2: Calculate the current slope S cur :Online calculation or simulation to estimate the current sensitivity S cur =ΔV / ΔB, where ΔV is the output increment under a known field strength step change at both ends, and ΔB is the field strength step change size; Step 3: Calculate the slope compensation factor: Slope compensation factor K g Is the gain factor used for digital compensation. In the signal processing pipeline, the original reading is first subtracted from the zero offset and then multiplied by K g , it can restore to the "equivalent" reference slope and eliminate the sensitivity drift caused by temperature drift or magnetic interference; Step 4: Digital compensation: In the sensor data output pipeline, the original voltage V raw First subtract the current zero offset, then multiply by K g :V comp =K g ·(V raw -V off )+V mind , where V raw To measure zero offset, V mind is the midpoint voltage, K g Store in storage.

7. A wireless power transmission system fault diagnosis system according to claim 6, characterized in that: The feature parameter extraction process is: Time domain analysis: The harmonic components of the output voltage are extracted to obtain the fundamental amplitude V1, the 2nd harmonic amplitude V2, the 3rd harmonic amplitude V3, and so on until the hth harmonic amplitude V h , where h is a positive integer; according to the formula Calculate the voltage waveform distortion rate V THD ; From this, the voltage waveform distortion rate V at each moment in the monitoring period can be obtained THD , and calculate the mean value to get the voltage waveform distortion rate of the monitoring period. The voltage waveform distortion rate of each monitoring period can be recorded as For output current I out Harmonic components are extracted to obtain the fundamental amplitude I1, the second harmonic amplitude I2, the third harmonic amplitude I3, and so on until the hth harmonic amplitude I h , where h is a positive integer; according to the formula Calculate the distortion rate I of the current waveform THD ; From this, the current waveform distortion rate I at each moment in the monitoring period can be obtained THD , and calculate the mean value to get the current waveform distortion rate of the monitoring period. The current waveform distortion rate of each monitoring period can be recorded as Frequency domain analysis: Subtract the nominal resonant frequency from the actual resonant frequency and take the absolute value to get the detuning amount, which is recorded as Δf r , thus the detuning amount corresponding to each moment in the monitoring period can be obtained, and the mean value is calculated as the detuning amount of the monitoring period and recorded as The quality factor G is obtained by dividing the actual resonant frequency at each moment in the monitoring period by the -3dB bandwidth of the resonance curve, and the quality factor G at each moment is averaged to obtain the quality factor of the monitoring period, which is recorded as G. i ; Multiphysics Analysis: The relationship between magnetic field intensity and temperature rise is established through finite element simulation: ΔT = k·B inst (t) 2 ·f·t, where k is the thermal conductivity coefficient of the material. From this, the temperature rise at each moment in the monitoring period can be obtained, and the average value is calculated to obtain the temperature rise during the monitoring period, which is recorded as ΔT i ; The input power is obtained by multiplying the input voltage by the input current, and the output power is obtained by multiplying the output voltage by the output current. The output power is then divided by the input power and multiplied by 100% to obtain the transmission efficiency η. The transmission efficiency at each moment in the monitoring period can be obtained by averaging the transmission efficiency of the monitoring period, which is recorded as η. i .

8. A wireless power transmission system fault diagnosis system according to claim 7, characterized in that: Take corresponding safety measures for the received fault judgment indicators, specifically: If the fault judgment indicator is in the fault judgment interval R1, no adjustment is required and monitoring is continued; if the fault judgment indicator is in the fault judgment interval R2, the transmission power is reduced by 20%; if the fault judgment indicator is in the fault judgment interval Q3, the temperature rise ΔT of the monitoring period closest to the current time of the system is extracted n :(1) When ΔT n <ΔT crit When , according to the formula Calculate the transmission power adjustment value P reduce , and follow P reduce Adjust the transmission power; P rated Indicates rated power; ΔT crit Indicates the critical temperature rise; (2) When ΔT n ≥ΔT crit If the fault judgment index is in the fault judgment interval Q4, the system will be shut down.

9. A method for diagnosing faults in a wireless power transmission system, characterized in that Applied to a wireless power transmission system fault diagnosis system according to any one of claims 1 to 8, the method comprises the following steps: U1: Communicates with the sensor to obtain sensing information, including field strength vector, sensitivity, noise density, and bandwidth; U2: Based on the sensor information, data integration is performed to obtain the output offset and noise increment, which are then normalized and weighted fused to obtain the interference index. This can be used to obtain the interference index of the sensor in each monitoring period. U3: A two-dimensional rectangular coordinate system is constructed with the monitoring period as the horizontal coordinate and the interference index as the vertical coordinate. The interference index is depicted in the coordinate system, and then a curve is used to connect them in sequence to obtain a curve graph of the interference index-monitoring period. Based on the curve graph of the interference index-monitoring period, an in-depth analysis is performed to obtain the sensing judgment index; if the sensing judgment index is in the sensing judgment interval Q1, no adjustment is required; if the sensing judgment index is in the sensing judgment interval Q2, a gain compensation strategy is generated; if the sensing judgment index is in the sensing judgment interval Q3, a calibration zeroing strategy is generated; if the sensing judgment index is in the sensing judgment interval Q4, a replacement strategy is generated; corresponding operations are performed according to the execution strategy of each sensor; U4: Obtains processed sensor-collected system information of the wireless power transmission system and extracts characteristic parameters based on the system information in the time domain, frequency domain, and multi-physics field. The characteristic parameters include voltage waveform distortion rate, current waveform distortion rate, detuning, quality factor, temperature rise, and transmission efficiency. U5: Normalize the characteristic parameters and perform weighted fusion to obtain the health index. Construct a health index-monitoring period curve graph, and perform in-depth analysis on it to obtain the fault judgment index. This process is the same as the in-depth analysis of the interference index-monitoring period curve graph. Generate corresponding safety measures based on the fault judgment index.

Images