Self-adaptive control method of wireless charging pile

Abstract

The invention relates to the technical field of high-power wireless charging control, and discloses a self-adaptive control method of a wireless charging pile, which comprises the following steps of: periodically driving a transmitting coil to generate a micro-power detection electric pulse, calculating a complex impedance value of a loop and comparing a change track of the complex impedance value with a pre-stored target impedance fingerprint model to obtain a target impedance fingerprint model; according to the invention, a power transmitting coil is multiplexed as a high-precision position sensor, so that the design inertia of separation of positioning and charging functions in the prior art is avoided, and the reliability of the positioning and charging system is improved. The physical interaction of the charging system is used for replacing external auxiliary positioning hardware which is easily interfered by the environment, so that the cost and complexity of the system are reduced, and meanwhile, the high reliability and all-weather working capability are obtained.

Description

Technical Field

[0001] This invention relates to an adaptive control method for wireless charging piles, belonging to the field of high-power wireless charging control technology. Background Technology

[0002] Currently, to ensure the efficiency and safety of energy transmission, an auxiliary positioning system consisting of optical, ultrasonic, or geomagnetic sensors is typically deployed separately from the charging system. This system determines whether the device to be charged has entered the preset charging area and activates the charging system only after confirming alignment. However, when such charging facilities are widely used in complex public environments such as outdoors, the aforementioned system architecture, which separates positioning and charging functions, exposes its inherent limitations. Because the sensors used in the auxiliary positioning system are susceptible to interference from environmental factors such as rain, snow, mud, or sunlight, the stability of their operation directly determines the availability of the entire charging service. That is, when the positioning function fails, even if the energy transmission unit is functioning properly, the system cannot be activated normally. This dependency determined by the system architecture has become a major factor affecting the reliability of the large-scale application of this technology.

[0003] To address this issue, a direct approach is to upgrade the hardware specifications of the assisted positioning system or increase the complexity of the algorithm. However, this raises a trade-off between cost and reliability. Higher-specification external hardware not only increases the initial investment and subsequent maintenance costs but also introduces more components and more complex integration relationships, without changing the fact that the charging function is still dependent on an external independent system. In addition to the aforementioned design inertia at the hardware level, some existing technologies also fail to break free from their dependence on external environment sensing hardware in their control methods, thus introducing new uncertainties. For example, Chinese invention patent CN112906433A discloses an adaptive wireless charging... The core logic of the electric control system and method patent is to use scene capture equipment to perform image recognition to determine whether a mobile terminal object exists, and then control the start and stop of the charging operation. However, this control method still relies on external auxiliary sensors in essence, only replacing ultrasonic or geomagnetic sensors with optical sensors. Therefore, its reliability in complex environments such as rain, snow, mud, or sunlight still faces similar challenges as the aforementioned technologies. More importantly, this method can only determine the existence of the target and cannot obtain the precise position coupling state information necessary to achieve efficient energy transmission. Therefore, it lacks the ability to perform closed-loop adjustment of dynamic position deviations during the charging process.

[0004] Specifically, existing technologies suffer from the following shortcomings: 1. The operational reliability of charging services is limited by external, independent hardware that is sensitive to environmental changes; 2. The overall hardware cost and structural complexity of the system restrict its large-scale, economical deployment; 3. The positioning and charging subsystems are functionally separate, lacking a collaborative working mechanism, and cannot perform closed-loop adjustment of dynamic position deviations during the charging process. Therefore, the technical problem this invention aims to solve is how to utilize the charging system's own energy interaction components to directly acquire and analyze the position status information of the device to be charged, thereby constructing an adaptive control method that can autonomously complete target identification, position judgment, and closed-loop guidance without relying on external independent sensing hardware. Summary of the Invention

[0005] This invention provides an adaptive control method for wireless charging piles, the main purpose of which is to solve the problems of insufficient reliability, high cost and inability to perform closed-loop collaborative control in existing charging systems due to their reliance on external independent positioning hardware.

[0006] To achieve the above objectives, the present invention provides an adaptive control method for a wireless charging pile, comprising the following steps: Step a: In the standby state of the wireless charging pile, the transmitting coil is periodically driven to generate a series of micro-power detection pulses with a parameter combination that is lower than the system resonant frequency and the instantaneous power is less than a preset power threshold. Step b: Within each detection cycle of a low-power detection pulse, the voltage and current waveform data of the transmitting coil circuit excited by the low-power detection pulse are acquired in real time. Then, using a digital signal processing algorithm, a complex impedance value containing real and imaginary parameters is calculated based on the amplitude ratio and phase difference of the voltage and current waveform data. Specifically, the calculation includes: identifying and extracting the peak or effective values ​​of the voltage and current waveforms within one or more detection cycles to determine their amplitude ratio; and calculating the time difference between characteristic points of the voltage and current waveforms, such as peak points or zero-crossing points, to determine their phase difference. Step c: Compare the trajectory of the change in the complex impedance value with a multidimensional impedance fingerprint feature model of the target receiving coil that is pre-stored in the memory. The multidimensional impedance fingerprint feature model of the target receiving coil contains an ideal impedance target value that is pre-calibrated through experiments and corresponds to the target receiving coil when it is in the best coupling position. Step d: When the trajectory of the change in the complex impedance value matches the multidimensional impedance fingerprint feature model of the target receiving coil within the preset tolerance range, the alignment guidance process is initiated. In step e, during the alignment guidance process, the complex impedance value is continuously compared with the ideal impedance target value, and the physical position offset command is calculated based on the deviation between the two and output through the communication module until the complex impedance value enters within a preset error threshold of the ideal impedance target value, and then the power inverter is finally commanded to switch to high-power wireless charging mode.

[0007] Preferably, before step a, the method further includes the following steps: in an unloaded environment where there are no devices to be charged around the wireless charging pile, steps a and b are performed, and the calculated complex impedance value is stored as an open space impedance baseline parameter; the matching in step d further includes the change in the current complex impedance value relative to the open space impedance baseline parameter, which conforms to the preset change trend of the target receiving coil multidimensional impedance fingerprint feature model.

[0008] Preferably, in step d, the matching of the change trajectory of the complex impedance value with the multidimensional impedance fingerprint feature model of the target receiving coil is achieved by determining whether the rate of change of the real part parameter and the imaginary part parameter of the complex impedance value over time both fall within a preset feature window of the model.

[0009] Preferably, in step a, the duty cycle of the low-power detection pulse is set to less than 1% of the normal duty cycle under high-power wireless charging mode.

[0010] Preferably, the communication module is a 485 communication module or a CAN bus communication module, and the physical position offset instruction is encoded as a visual guidance instruction that can be parsed by the battery management system of the device to be charged.

[0011] Preferably, in step e, the rule for calculating the physical position offset command based on the deviation between the complex impedance value and the ideal impedance target value includes: calculating the complex impedance deviation. ,in, This is the complex impedance value. The target value of the ideal impedance; when the imaginary part of the complex impedance deviation... The absolute value is greater than a preset first threshold. At that time, a command is generated to adjust the vertical coupling distance; when the real part of the complex impedance deviation... The absolute value is greater than a preset second threshold. At that time, a command is generated to adjust the horizontal alignment.

[0012] Preferably, the first threshold With the second threshold All are positive real numbers, and their values ​​are preset based on the statistical distribution of complex impedance deviation measured under different known physical offsets during the system debugging phase.

[0013] Preferably, when a mismatch is determined in step d, if the trajectory of the change in the complex impedance value does not conform to the multidimensional impedance fingerprint feature model of the target receiving coil, but its change exceeds a preset interference threshold, a metal foreign object interference alarm is generated, and step a is suspended within a preset time.

[0014] Preferably, during the alignment and guidance process, if the complex impedance value fails to enter the preset error threshold of the ideal impedance target value within the preset guidance timeout period, the alignment and guidance process is automatically terminated and the process returns to the standby state in step a.

[0015] Preferably, the memory is a non-volatile memory, and the multi-dimensional impedance fingerprint feature model of the target receiving coil can be updated via a remote communication interface to adapt to different models of devices to be charged.

[0016] Compared with the prior art, the beneficial effects of the present invention are: 1. The method of this invention establishes a way for a high-power wireless charging system to directly sense the electromagnetic environment of its working area. In standby mode, the system does not passively wait for external commands, but its control unit periodically drives the transmitting coil to generate micro-power detection pulses and continuously monitors the voltage and current response of the coil circuit under the excitation of the pulse. In this way, the charging system gets rid of its dependence on external independent positioning hardware and transforms the physical medium of energy transmission into an active real-time environmental state sensing unit. This allows subsequent charging interaction behavior to be based on the first-hand physical information obtained by the system itself, avoiding the problem of the entire system control logic being erroneously triggered due to inaccurate or malfunctioning information from external sensors in complex environments.

[0017] 2. To enable the charging system to wake up and activate, a precondition based on target electromagnetic feature recognition is set. After the control unit detects that an object has entered the magnetic field range through low-power detection, it does not immediately start charging. Instead, it continuously calculates the change trajectory of the equivalent impedance of the transmitting coil after the object's intervention and compares it with the pre-stored target receiving coil impedance feature model. Only when the measured impedance change characteristics match the model will the system confirm the validity of the target. This process transforms the system's activation permission from a vague judgment of spatial location to a confirmation of electromagnetic identity, enabling the system to effectively distinguish legitimate devices to be charged from other metallic interference objects, avoiding energy mis-output or safety hazards caused by misidentification.

[0018] 3. A closed-loop guidance and collaborative alignment process was constructed between the transmitter and the device to be charged, with real-time coupling status as the core. After confirming the identity of the device to be charged, the control unit immediately compares the real-time measured impedance parameters with the ideal impedance parameters representing the optimal coupling position in the database, and interprets the deviation between the two as a specific physical position offset command. This command is sent to the device to be charged through the onboard communication bus to provide adjustment guidance for the driver. In this process, the three independent links of positioning perception, control decision and communication execution in the original charging system are integrated into a dynamic convergent control closed loop with physical layer coupling status as the sole objective. This transforms the vehicle parking process, which originally required repeated attempts, into a deterministic and efficient guided alignment operation.

[0019] 4. The initiation of high-power energy transfer is structurally linked to the system reaching a preset optimal physical coupling state. The only trigger condition for the system to switch from low-power detection mode to high-power charging mode is when the real-time monitored impedance value of the transmitting coil enters a small error window representing optimal alignment. This means that high-efficiency energy transfer in this method is not a result pursued after the vehicle is aligned, but rather a prerequisite for the system to start high-power operation. This design internalizes operational efficiency and safety into the inherent logic of system startup, so that energy transfer behavior is physically restricted to the most efficient and safest coupling state, thereby avoiding energy leakage and efficiency loss that may be caused by charging attempts under large positional offsets. Attached Figure Description

[0020] Figure 1 This is a flowchart of the closed-loop alignment control based on impedance feedback according to the present invention. Figure 2 This is a graph showing the relationship between the complex impedance components and the coil center offset of the present invention. Figure 3 This is a flowchart illustrating the full-cycle workflow of the adaptive control of the wireless charging pile according to the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be described in further detail below. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0022] This invention provides an adaptive control method for wireless charging piles. The control flow is configured as a multi-stage closed-loop control system, including a standby identification stage based on low-power detection, a target confirmation stage based on impedance fingerprint matching, a closed-loop alignment guidance stage based on real-time impedance feedback, and an execution stage for switching to a high-power charging mode. This method multiplexes the power transmitting coil under different operating modes, utilizing its electromagnetic interaction with the environment to directly obtain the system state, thereby achieving autonomous identification and precise alignment of the device to be charged. In the actual deployment of charging facilities, environmental factors such as changes in light intensity, rain, snow, and dirt can affect the reliability of external optical or ultrasonic sensors. To reliably identify the charging area with low power consumption in standby mode, this method is configured to execute a periodic low-power detection process. This process is scheduled by an onboard control unit, such as an ATmega88 microcontroller integrating a timer, pulse width modulation (PWM) module, and analog-to-digital converter (ADC). In standby mode, the control unit periodically, for example, every 100ms, instructs the power switching transistors in the power inverter to operate with a specific parameter combination. A driving signal is generated, and the driving frequency of this parameter combination is set at a position deviating from the system's resonant frequency. For example, if the resonant frequency of the system for high-power energy transfer is 85kHz, the driving frequency of the micro-power detection pulse can be set to 20kHz. Simultaneously, its duty cycle is set to a value lower than 1% of the normal duty cycle under high-power charging mode, for example, 0.4%. Thus, the instantaneous power generated by the transmitting coil is controlled within a preset power threshold, for example, below 1W. The deviating operating state from the resonant point prevents efficient energy transfer, thereby transforming the transmitting coil's function from an energy transmission unit to an electromagnetic probe for environmental state sensing. To convert the intervention of physical entities in the environment into a state variable that can be quantitatively analyzed by the control system, this method synchronously acquires the voltage and current waveforms on the transmitting coil circuit excited by the detection pulse within each detection cycle of the micro-power detection pulse, using a voltage and current sampling circuit connected to the ADC input channel of the control unit. The processing program within the control unit calculates the phase difference between the two waveforms by executing a digital signal processing algorithm that integrates error suppression measures and zero-crossing detection and peak optimization. Ratio of amplitude To address the impact of noise and quantization errors in actual measurements, this algorithm first performs digital low-pass filtering on the acquired raw waveform data to suppress noise. Then, it uses linear interpolation to accurately locate the zero-crossing points in the filtered data and calculates the phase difference. Simultaneously, the peak values ​​of voltage and current are optimally identified within the cycle to obtain the amplitude ratio. Finally, the results calculated over multiple consecutive periods are averaged to obtain stable and reliable measurement values, thus obtaining the phase difference after error suppression. Ratio of amplitude Then, according to This relationship yields a parameter containing the real part. With imaginary part parameter Complex impedance value For example, if the peak voltage measured at a certain moment is 10.0V and the peak current is 0.5A, and the current phase lags behind the voltage... The complex impedance value calculated by the control unit is... The complex impedance value As a multi-dimensional system state variable, it can quantitatively characterize the equivalent load characteristics of the transmitting coil under the current electromagnetic environment.

[0023] To distinguish the device to be charged from other metallic interference objects, this method establishes a pre-confirmation condition based on target electromagnetic feature identification. In the initial stage of system deployment, under an unloaded environment where no devices to be charged exist around the wireless charging station, steps a and b are performed as described above. The statistical average of the calculated complex impedance values ​​is then used as an open space impedance baseline parameter. The data is stored in the non-volatile memory of the control unit. This memory also pre-stores one or more multi-dimensional impedance fingerprint feature models of the target receiving coil. This model is a data structure that records the variation of the complex impedance value of the device to be charged at different coupling positions. It is established during the system debugging phase by using a three-axis moving platform to carry a standard target receiving coil, moving it above the transmitting coil along a preset path, and collecting the complex impedance data throughout the entire process. It includes an ideal impedance target value corresponding to the optimal coupling position of the target receiving coil. And a feature window for determining the identity of the target, which defines the rate of change of the real and imaginary parts of the complex impedance over time. and The effective range; when the control unit detects the current complex impedance value Relative open space impedance baseline parameters When the change exceeds the preset noise threshold, the matching determination program is initiated and calculations are performed continuously. and Only when both rates of change fall within the preset feature window of the model will the system determine that the multidimensional impedance fingerprint feature model of the target receiving coil matches within the preset tolerance range and start the alignment guidance process.

[0024] After confirming the identity of the device to be charged, this method constructs a closed-loop guidance and collaborative alignment process based on real-time coupling status to guide the vehicle to the optimal charging position. After the process is initiated, the control unit continuously acquires the real-time complex impedance value through low-power detection electrical pulses. And calculate its value compared with the target value of the ideal impedance in the memory. Complex impedance deviation ,in, This is the complex impedance value. The target impedance value is the ideal impedance. Based on the mapping relationship between physical offset and impedance deviation established during the system calibration phase, this deviation is interpreted as a physical position offset command. The interpretation rules include: when the imaginary part of the complex impedance deviation... The absolute value is greater than a preset first threshold. At that time, a command is generated to adjust the vertical coupling distance, when the real part of the complex impedance deviation is... The absolute value is greater than a preset second threshold. When the horizontal alignment is adjusted, a command is generated to adjust the horizontal alignment. In the offline calibration procedure, to establish a deterministic mapping relationship between the complex impedance components and the physical position offset, the control unit drives the triaxial displacement platform to perform the following data acquisition and model generation steps: First, with the receiving coil and transmitting coil horizontally aligned, the receiving coil is moved only along the vertical Z-axis in 1cm increments, and the complex impedance corresponding to each height point is recorded. Sequences are used to analyze the imaginary part of complex impedances. The variation law with vertical distance was studied; subsequently, the receiving coil was fixed at a preset reference vertical height, and moved only along the horizontal X-axis and Y-axis in 1cm increments, recording the corresponding complex impedance. A matrix used to analyze the real part of complex impedances. The variation pattern with horizontal offset; the control unit, based on the collected dataset, if determined... and If the vertical and horizontal offsets are approximately independent monotonic functions, then control rules are directly generated. If analysis reveals a coupling effect between the two, then a control rule is generated based on the real-time complex impedance deviation. As input, the corrected physical position offset command The output is a two-dimensional lookup table and stored in memory; first threshold With the second threshold All values ​​are positive real numbers, pre-set based on the statistical distribution of complex impedance deviations measured under different known physical offsets during system debugging. These physical position offset commands are encoded into visual guidance commands that can be parsed by the battery management system of the device to be charged, and output to the device to be charged via an onboard 485 communication module or CAN bus communication module. The driver adjusts the vehicle position according to the guidance on the display interface of the device to be charged, and the control unit recalculates the deviation and issues a new command in the next detection cycle. This cycle is repeated to form a dynamically convergent control closed loop until the complex impedance value is reached. Entering the ideal impedance target value Within a preset error threshold.

[0025] To ensure that the initiation of high-power energy transfer is tied to the system reaching a preset optimal physical coupling state, this method uses the end of the closed-loop alignment guidance process as the trigger condition for switching to high-power charging mode; only when the complex impedance value... Entering the ideal impedance target value Only after the preset error threshold is reached will the control unit finally issue a command to switch the power inverter to high-power wireless charging mode. At this time, the system's operating frequency switches to the resonant frequency, and the duty cycle is also increased to the normal operating level. To ensure the robustness of the control system, the method is also equipped with an anomaly handling mechanism. During the alignment guidance process, if the complex impedance value fails to enter the preset error threshold of the ideal impedance target value within the preset guidance timeout period, such as 3 minutes, the alignment guidance process is automatically terminated, and the system returns to the standby state. In addition, if the trajectory of the complex impedance value does not conform to the multi-dimensional impedance fingerprint feature model of the target receiving coil during standby detection, but its change exceeds a preset interference threshold, the control unit generates a metal foreign object interference alarm and suspends the execution of step a within a preset time, such as 5 minutes. It should be noted that the multi-dimensional impedance fingerprint feature model of the target receiving coil can be updated through a remote communication interface to adapt to different models of devices to be charged.

[0026] Example 1: The adaptive control method of the present invention operates as follows in an unattended charging station deployed in an outdoor public transportation hub in a high-latitude region. This station faces frequent rain, snow, and freezing weather in winter, with ground markings often covered by snow and the presence of metal debris in the environment. When a bus waiting to be charged enters a charging bay covered by a thin layer of snow, the driver cannot clearly see the stop line on the ground, resulting in a horizontal offset of more than 50cm between the vehicle's initial parking position and the center of the transmitting coil. Simultaneously, a metal can is left at the edge of the transmitting coil. Under these conditions, the charging pile's control unit is in standby mode and, according to the specific implementation procedure, periodically drives the transmitting coil to generate micro-power detection pulses at a frequency of 20kHz and a duty cycle of 0.4%. When the receiving coil under the vehicle chassis enters the magnetic field range of the transmitting coil, the control unit collects the complex impedance value of the transmitting coil circuit. It begins to deviate from its open space impedance baseline parameter calibrated under no-load conditions. At this point, the control unit not only monitors the impedance change trend caused by the coupling of the receiving coil, which conforms to the pre-stored multidimensional impedance fingerprint characteristic model of the target receiving coil, but also... and All fall within the feature window, and local, discontinuous impedance jumps caused by the metal can are also detected. However, due to the impedance fingerprint-based identification mechanism, the matching basis is the continuous trajectory of the complex impedance value changing with distance, rather than a single numerical change. Therefore, the system can identify the impedance component that conforms to the model's trend as the device to be charged, and determine the isolated impedance anomaly point that does not conform to the model trajectory as metal foreign object interference. Thus, a metal foreign object interference alarm is generated at the same time as the alignment guidance process is initiated. The matching process between the trajectory of the complex impedance value change and the multidimensional impedance fingerprint feature model of the target receiving coil can be achieved through one of the following two specific methods: The first method is to use the Dynamic Time Warping (DTW) algorithm, which is used to calculate the similarity between the real-time trajectory and the standard template trajectory. The matching threshold used to determine whether a match is successful is set as follows: During the offline calibration phase of the system, positive sample trajectories of multiple combination methods are collected, i.e., the target vehicle. The system identifies two approaches to matching positive and negative sample trajectories: 1) Entering the trajectory in different postures and 2) Illegal negative sample trajectories, such as other metal objects entering or vehicles moving abnormally; 2) DTW distances between all positive and negative sample trajectories and the standard template trajectory, forming two distance distribution databases. The final matching threshold is set to a value that maximizes the distinction between positive and negative samples, for example, three standard deviations above the mean of the positive sample distance distribution, to ensure high-reliability identification. The second approach is an alternative feature window matching method, which consumes fewer computational resources. The effective range of the model's preset feature window is determined as follows: During the offline calibration phase, the system simulates one or more typical vehicle entry trajectories and extracts the complex impedance values ​​of all nodes on the trajectory to form a time series. Subsequently, the maximum and minimum rates of change of the real and imaginary parts of the complex impedance in this series over time are calculated, and the range defined by these extreme values ​​is solidified as a feature window for storage. When the statistical value of the rate of change of the real-time trajectory falls within this window, a successful match is determined.

[0027] After confirming the identity of the device to be charged, the system immediately enters a closed-loop alignment guidance phase based on real-time impedance feedback. The control unit will then use the complex impedance value calculated in real time. With the ideal impedance target value in memory In comparison, at the initial moment, due to the large horizontal offset, the calculated complex impedance deviation... real part The absolute value is greater than the preset second threshold. Based on this, the control unit sends a physical position offset command to the left to the vehicle's battery management system via the CAN bus. The driver slowly moves the vehicle according to the graphical guidance on the in-vehicle screen. During the vehicle's movement, the control unit continuously detects and calculates the deviation at 100ms intervals, sending updated adjustment commands to the vehicle in real time. As the horizontal position deviation gradually decreases... Entry threshold If, after reaching the optimal range, the vertical coupling distance is still not optimal, then Still greater than the first threshold The system will continue to send instructions to adjust forward and backward until... Both the real and imaginary parts of the value reach the ideal target value. Within the preset error threshold, the system determines that the alignment process is complete and finally instructs the power inverter to switch to the 85kHz high-power wireless charging mode. This process transforms the two separate steps of positioning and charging in the traditional charging operation into a continuous control process with the physical layer coupling state as the convergence target. That is, by changing the way the control target is changed, the spatial positioning problem that originally relied on external sensors is redefined as a parameter optimization problem based on the electromagnetic interaction of the system.

[0028] Example 2: To objectively verify the effectiveness of the closed-loop alignment guidance process in improving the final performance of the system in the adaptive control method of the present invention, the following experimental platform was built. This platform consists of a wireless charging pile transmitter integrating the control method of the present invention and a standard receiving coil mounted on a high-precision three-axis electric displacement platform. The displacement platform has [missing information - likely related to X, Y, and Z axes]. The positioning and reading accuracy was used to simulate the positional offset of the device to be charged. The entire experiment was conducted in an electromagnetically shielded environment to eliminate the influence of external radio frequency interference on the measurement of complex impedance values. The experimental data was synchronously recorded by the host computer via the CAN bus, including the calculated complex impedance value of the transmitter, the output physical position offset command, and the real-time coordinates of the position sensor of the displacement platform. This experiment set up an experimental group using the method of this invention and a control group with the closed-loop alignment guidance process disabled. The operation mode of the control group was as follows: after confirming the target through low-power detection and fingerprint matching, it directly switched to the high-power charging mode at the initial position to simulate the parking scenario without guidance assistance. The experimental process was as follows: eight initial physical position offsets were set. For each initial offset point, the control group was run first and its stable energy transmission efficiency in the offset state was recorded. Then, the receiving coil was reset to the same initial offset point and the experimental group was run. The displacement platform moved according to the physical position offset command issued by the charging pile until the charging pile determined that the alignment was completed and switched to the high-power charging mode. The time consumed by the guidance process, the final residual physical position offset, and the stable energy transmission efficiency at this time were recorded. The specific experimental data are shown in Table 1.

[0029] Table 1: Performance comparison data of the experimental group and the control group under different initial offsets.

[0030] ; Table 1 shows that the energy transfer efficiency of the control group was affected by a large initial physical offset, reaching a minimum of 59.5%. In contrast, the experimental group, starting from the same initial offset position, had a higher energy transfer efficiency based on the deviation between the measured initial complex impedance value and the ideal impedance target value. The system generates and outputs a series of convergent physical position offset commands to drive the displacement platform to perform position correction. Ultimately, the residual physical position offset of all test points is controlled within 1 cm. Correspondingly, the final energy transmission efficiency is stable at over 91%, and the time taken for the guidance process is within 15 seconds. This data confirms that the closed-loop alignment guidance process can effectively control the physical position based on the complex impedance state variables obtained by the system itself, bringing a large initial state error to within the allowable range. This allows the system to operate in the preset high-efficiency range when switching to high-power working mode.

[0031] To further demonstrate that the technical solution of the present invention using complex impedance for closed-loop guidance has substantial technical advantages and unexpected technical effects compared with the simplified solution constructed by those skilled in the art based on conventional thinking, the following comparative example 1 is set up.

[0032] Comparative Example 1: This comparative example aims to simulate a simplified technical path that is most readily conceived by those skilled in the art when attempting to reuse a transmitting coil for position guidance. Unlike the method of this invention in Example 2, which uses multi-dimensional closed-loop guidance based on complex impedance (including real and imaginary parts), the control method in this comparative example, after confirming the target, only utilizes the real part parameter of the complex impedance. This scalar is used as the sole criterion for determining the coupling position and attempts to guide the receiving coil to a preset ideal real part target value. Apart from the core algorithm difference, the experimental platform, hardware configuration, initial physical position offset, and test environment used in this comparative example are strictly consistent with the experimental group settings in Example 2. The experimental process is as follows: Set an initial physical position offset that is the same as that of a certain test point in Example 2. The comparative control program is initiated, which, after confirming the target through fingerprint matching, continuously calculates the real part of the real-time complex impedance. and compare it with the real part of the ideal impedance target value. (Based on the data from Example 2, this value is approximately) The control unit compares the two and bases its decisions on the deviation between them. The size of the value generates a unified, directionless adjustment command to drive the displacement platform to correct its position until... Once the system enters the preset error threshold, it switches to high-power charging mode. The time taken for the guidance process, the final residual physical position offset, and the stable energy transfer efficiency are recorded. During the guidance process, the following phenomenon was observed: Because the control system relies solely on a single real parameter for judgment, it cannot decouple the horizontal and vertical offsets. The coupling effect occurs when the displacement platform attempts to move in the horizontal direction to reduce... At the same time, this movement itself may cause the vertical coupling distance to enter a state that can produce a similar effect. The value range caused the guiding command to oscillate meaninglessly between forward and backward. The system consumed far more time than in Example 2 and still could not stably converge to the optimal physical position. To forcibly terminate the test, after the guiding time reached 60 seconds, if If the threshold is reached, the system will immediately switch to high-power mode. See Table 2 for specific test data.

[0033] Table 2: Performance comparison data between Comparative Example 1 and Example 2 of the present invention.

[0034] ; The experimental results show that the simplified scheme, which uses only the real part of the complex impedance as the guiding basis, cannot effectively decouple the horizontal and vertical positional offsets in its physical principle. This leads to instability and non-convergence in the guiding process, resulting in a large residual physical positional offset. Therefore, after switching to high-power operating mode, its energy transmission efficiency is much lower than that of the method claimed in this invention. This result confirms that the technical solution of this invention, which uses the real and imaginary parts of the complex impedance to correspond to and guide the horizontal and vertical positional adjustments respectively, is the key to overcoming the inherent defects of conventional technical paths and has non-obviousness.

[0035] Example 3: This example combines Figures 1 to 3 An adaptive control method for a wireless charging station is described, such as... Figure 1 As shown, this process begins with the impedance acquisition and calculation module. It calculates the real-time complex impedance value Z by acquiring voltage and current waveform data from the transmitting coil. This Z value is then sent to the target identification and judgment module. This module reads the open space impedance baseline in database A2 and matches it with the impedance fingerprint feature model in database A1. If an abnormal interference signal is detected, an alarm is triggered. If the match is successful, the process proceeds to the calculation guidance instruction module, which calculates the target ideal impedance value based on the target value in database A1. The physical position offset command is calculated and sent to the device to be charged. After obtaining a precise alignment signal, the command controls the charging mode module to send a power mode switching command to the transmitting coil, thereby completing the entire closed-loop control.

[0036] like Figure 2As shown, this figure is a characteristic curve of a complex impedance as a function of position offset. The horizontal axis represents the offset from the center of the transmitting coil in centimeters (cm), and the vertical axis represents the impedance in ohms. The figure clearly shows the real part of the complex impedance as the offset increases from 0 cm to 35 cm. The value of decreases, while the imaginary part of the complex impedance... The value is trending upwards, while the ideal target value is... This is represented by a constant reference line, which serves as the basis for determining the alignment of the subsequent closed-loop guidance.

[0037] like Figure 3 As shown, the method starts in a standby detection state. In this state, the system periodically sends low-power detection pulses and continuously calculates the open space impedance. When the detected impedance change exceeds the noise threshold, the process enters the target confirmation state, comparing the collected impedance change trajectory with the pre-stored impedance fingerprint model. If the impedance fingerprint matches successfully, the closed-loop alignment guidance state is initiated. If the fingerprint does not match and the change exceeds the preset interference threshold, the system enters the alarm and pause state. If no matching target leaves, the system returns to the standby detection state. In the closed-loop alignment guidance state, the system will continuously send adjustment commands due to the continuous position shift. The commands are based on the continuously calculated deviation between the real-time impedance and the ideal value. When the complex impedance enters the preset error threshold of the ideal target, the process switches to the high-power charging state, and the system operating frequency switches to the system resonant frequency for energy transmission. If a guidance timeout occurs during the guidance process, the process also enters the alarm and pause state. After the high-power charging is completed or the vehicle leaves, the system returns to the standby detection state. In alarm and pause states, the system will issue a metal foreign object interference alarm or handle timeout events, and return to standby detection state after the pause time ends or the alarm is cleared.

[0038] Example 4: To ensure the consistency and reproducibility of the adaptive control method of the present invention in different hardware batches and application scenarios, a standardized offline calibration and model generation procedure needs to be executed before the wireless charging pile leaves the factory or after on-site deployment. This procedure aims to deterministically generate the core data that the control unit relies on for decision-making in subsequent work, namely the multidimensional impedance fingerprint feature model of the target receiving coil and related control thresholds. The calibration procedure is executed on a controlled test platform, which includes a wireless charging pile transmitter to be calibrated, a target receiving coil as a reference, and a three-axis electric displacement platform. The travel range of the displacement platform is not less than ±40cm in the horizontal X and Y axis directions and not less than 20cm in the vertical Z axis direction, and its three-axis positioning accuracy is not less than 0.1cm. Before calibration begins, the target receiving coil is fixed to the execution end of the displacement platform, and its geometric center is aligned with the geometric center of the transmitting coil in the XY plane as the origin (0, 0) of the coordinate system. The vertical distance at this time is the Z-axis coordinate.

[0039] The calibration process begins with the acquisition of three-dimensional spatial impedance data. The control system drives the displacement platform, causing the receiving coil to move point-by-point within a preset three-dimensional grid space. This grid space is set to range from -30cm to +30cm in the X and Y axes, and from 15cm to 30cm in the Z axis. The step size is 1cm in all three directions. The coordinates of each grid node are... At this point, the control unit performs 10 excitation and measurement operations using micro-power detection pulses, and averages the 10 calculated complex impedance values ​​to obtain the stable complex impedance value for that spatial point. The system stores the mapping data, which includes spatial coordinates and corresponding complex impedance values, in a temporary database. After traversing all grid nodes, a high-resolution database describing the impedance positional relationship of the specific transceiver coil pair within a preset space is established. Subsequently, the control unit extracts the complex impedance value corresponding to the coordinate origin (0, 0) at the optimal charging height (e.g., 20 cm) from this database and uses it as the ideal impedance target value. The data is stored in non-volatile memory, while the entire original database is transformed into the main part of the target receiving coil's multidimensional impedance fingerprint feature model through a specific data modeling method and stored together. One specific implementation method is to generate a downsampled three-dimensional lookup table: that is, the original high-density grid data is downsampled by a preset step size (e.g., 5cm) to form a smaller sparse grid stored in memory. In the actual guidance process, for any real-time position that is not on the grid point, its corresponding complex impedance value is obtained by performing trilinear interpolation on the data of the nearest surrounding grid point. As an alternative, a polynomial fitting method can also be used, that is, a ternary quadratic or cubic polynomial function is used to fit the distribution of the real and imaginary parts of the complex impedance in three-dimensional space, and the fitted polynomial coefficients are stored as model parameters.

[0040] Furthermore, to determine the key control thresholds in the closed-loop alignment guidance process, the control unit performs parameter calculations based on the generated original database. For the first threshold... With the second threshold The determination of the control unit analysis at the ideal impedance target value The local impedance gradient in the vicinity, calculated near the point (0, 0, 20 cm), is the real part of the complex impedance when the location shifts by a small 1 cm in the X-axis or Y-axis direction. The average change, and the imaginary part of the complex impedance when the position shifts by a small 1 cm in the Z-axis direction. The average change of these two average changes is taken as a preset proportion, such as 30%, and used as the second threshold. With the first threshold To determine the feature window for target recognition, the control unit simulates one or more typical vehicle entry trajectories in the database, for example, moving from the point (-30cm, -30cm, 25cm) at a simulated speed of 5km / h to the origin (0, 0, 20cm). The control unit extracts the complex impedance values ​​of all nodes on the trajectory to form a time series. Then, it calculates the maximum and minimum rates of change of the real and imaginary parts of the complex impedance over time in the series, and stores the range defined by these extreme values ​​as the feature window for determining whether the subsequent target is the device to be charged. After all model data and threshold parameters have been calculated and written to non-volatile memory, the offline calibration process of the wireless charging pile ends.

[0041] Example 5: To maintain the consistency of the measurement reference during long-term operation of the control method of the present invention, the control unit of the wireless charging pile is configured to execute a periodic open space impedance baseline parameter self-calibration procedure. The trigger condition for this procedure is set so that when the system does not detect any valid target receiving coil entry event within a preset time period, such as 4 consecutive hours, the control unit automatically enters the baseline recalibration mode. In this mode, the system performs the same low-power detection and complex impedance value calculation steps as the initial calibration to obtain the complex impedance measurement value under the current no-load environment. And compare it with the original open space impedance baseline parameters stored in non-volatile memory. Compare the magnitudes of the deviations between the two. If the drift exceeds a preset threshold, which can be set according to the component aging curve, then... If the modulus is 2%, the control unit will use a new one. Cover the original The system automatically updates the baseline parameters. If the deviation modulus exceeds a fault threshold of, for example, 5%, the system will lock into standby mode and send a maintenance request for abnormal hardware status to the background management system through the communication module.

[0042] In the closed-loop alignment guidance process, to address potential nonlinear or non-cooperative movement of the device to be charged, the control logic of this method also includes a monitoring and handling mechanism for abnormal states during the guidance process. During alignment guidance, the control unit not only calculates the current complex impedance deviation... The deviation modulus value is also monitored at the same time. The trend of change over multiple consecutive detection cycles should, during a normal convergence process, show that the modulus value monotonically decreases with time. If the control unit detects... If the detection rate increases instead of decreasing for more than three consecutive detection cycles, it indicates that the actual movement direction of the device to be charged does not match the commanded direction. The system then suspends sending new physical position offset commands and sends a command through the communication module requesting the driver to confirm or maintain the position. If the non-convergence trend is corrected within a preset time, such as 10 seconds, the guidance process continues. If the trend persists or fails to recover within the time limit, the control unit determines that the guidance has failed and actively terminates the alignment guidance process, returning the system to the standby state in step a, while recording a guidance anomaly event.

[0043] Example 6: After the offline calibration and model generation procedures are completed, in order to further solidify the system's operating logic and improve its adaptability under complex working conditions, the following standardized setting process for key operating parameters needs to be executed. This process first quantifies the triggering conditions for the metal foreign object interference alarm, and the control unit analyzes the complex impedance values ​​on all grid nodes collected during the offline calibration process. Relative to open space impedance baseline parameters The deviations are calculated, and the statistical distribution of these deviations is determined. The interference threshold used to trigger the metal foreign object interference alarm is set as a boundary value outside the normal fluctuation range of the complex impedance value caused by environmental electromagnetic noise and the noise of the measurement circuit itself when no target receiving coil is present. A specific setting method is to set it to 6 times the standard deviation obtained by continuously measuring the complex impedance value 1000 times under no-load conditions. In this way, a reproducible judgment benchmark can be established for the identification of non-target metal objects.

[0044] Accordingly, to ensure that the timeout determination during the closed-loop alignment guidance process has an engineering basis, the preset guidance timeout period is calculated based on the simulated trajectory data from the offline calibration phase. Specifically, the control unit analyzes the movement from the boundary point of entering the magnetic field's effective range to the ideal impedance target value when simulating a typical vehicle entry trajectory. The longest time required to locate This maximum timeout is calculated based on a preset minimum guide movement speed, such as 0.1 m / s, and the final guide timeout period is set to this maximum timeout. A multiple, such as 1.5 times, is used to provide a time window for the guidance process that can accommodate normal operation time while effectively terminating abnormal stalls or non-convergence. Furthermore, to adapt to different models of devices to be charged, each target receiving coil multidimensional impedance fingerprint feature model generated according to the procedure of Embodiment 3 is assigned a unique model identifier and bound to the vehicle model identifier of the device to be charged. This identifier is stored in a mapping table located in non-volatile memory. When a device to be charged enters the charging area, it sends its vehicle model identifier to the charging pile through an initial low-power communication handshake protocol. The control unit of the charging pile then loads the corresponding fingerprint feature model and associated... These parameters are used in the subsequent identification and guidance process.

[0045] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0046] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An adaptive control method for a wireless charging station, characterized in that, Includes the following steps: Step a: In the standby state of the wireless charging pile, the transmitting coil is periodically driven to generate a series of micro-power detection pulses with a parameter combination that is lower than the system resonant frequency and the instantaneous power is less than a preset power threshold. Step b: In each detection cycle of the low-power detection pulse, the voltage waveform data and current waveform data of the transmitting coil circuit excited by the low-power detection pulse are collected in real time, and a complex impedance value containing real part parameters and imaginary part parameters is calculated based on the amplitude ratio and phase difference of the voltage waveform data and current waveform data. Step c: Compare the trajectory of the change in the complex impedance value with a multidimensional impedance fingerprint feature model of the target receiving coil that is pre-stored in the memory. The multidimensional impedance fingerprint feature model of the target receiving coil contains an ideal impedance target value that is pre-calibrated through experiments and corresponds to the target receiving coil when it is in the best coupling position. Step d: When the trajectory of the change in the complex impedance value matches the multidimensional impedance fingerprint feature model of the target receiving coil within the preset tolerance range, the alignment guidance process is initiated. In step e, during the alignment guidance process, the complex impedance value is continuously compared with the ideal impedance target value, and the physical position offset command is calculated based on the deviation between the two and output through the communication module until the complex impedance value enters within a preset error threshold of the ideal impedance target value, and then the power inverter is finally commanded to switch to high-power wireless charging mode.

2. The adaptive control method for a wireless charging pile according to claim 1, characterized in that, Before step a, there are also steps: in an unloaded environment where there are no devices to be charged around the wireless charging pile, steps a and b are performed, and the calculated complex impedance value is stored as the open space impedance baseline parameter; the matching in step d further includes the change in the current complex impedance value relative to the open space impedance baseline parameter, which conforms to the preset change trend of the target receiving coil multidimensional impedance fingerprint feature model.

3. The adaptive control method for a wireless charging pile according to claim 2, characterized in that, In step d, the matching of the trajectory of the complex impedance value with the multidimensional impedance fingerprint feature model of the target receiving coil is achieved by determining whether the rate of change of the real part and the imaginary part of the complex impedance value over time falls within a preset feature window of the model.

4. The adaptive control method for a wireless charging pile according to claim 1, characterized in that, In step a, the duty cycle of the low-power detection pulse is set to less than 1% of the normal duty cycle under high-power wireless charging mode.

5. The adaptive control method for a wireless charging pile according to claim 1, characterized in that, The communication module is either a 485 communication module or a CAN bus communication module, and the physical position offset command is encoded as a visual guide command that can be parsed by the battery management system of the device to be charged.

6. The adaptive control method for a wireless charging pile according to claim 1, characterized in that, In step e, the rules for calculating the physical position offset command based on the deviation between the complex impedance value and the ideal impedance target value include: calculating the complex impedance deviation. ,in, This is the complex impedance value. The target value of the ideal impedance; when the imaginary part of the complex impedance deviation... The absolute value is greater than a preset first threshold. At that time, a command is generated to adjust the vertical coupling distance; when the real part of the complex impedance deviation... The absolute value is greater than a preset second threshold. At that time, a command is generated to adjust the horizontal alignment.

7. The adaptive control method for a wireless charging pile according to claim 6, characterized in that, First threshold With the second threshold All are positive real numbers, and their values ​​are preset based on the statistical distribution of complex impedance deviation measured under different known physical offsets during the system debugging phase.

8. The adaptive control method for a wireless charging pile according to claim 1, characterized in that, If the determination in step d is a mismatch, and the trajectory of the change in the complex impedance value does not conform to the multidimensional impedance fingerprint feature model of the target receiving coil, but its change exceeds a preset interference threshold, then a metal foreign object interference alarm is generated, and step a is suspended within a preset time.

9. The adaptive control method for a wireless charging pile according to claim 1, characterized in that, If the complex impedance value fails to enter the preset error threshold of the ideal impedance target value within the preset guidance timeout period during the alignment guidance process, the alignment guidance process will be automatically terminated and the process will return to the standby state in step a.

10. The adaptive control method for a wireless charging pile according to claim 1, characterized in that, The memory is non-volatile, and the multidimensional impedance fingerprint feature model of the target receiving coil can be updated via a remote communication interface.

Images