Implementation method for dynamically regulating battery charging in a battery swapping station control system

Abstract

The present invention discloses a method for realizing dynamic regulation of battery charging in a battery swapping station control system, which specifically relates to the field of battery charging control in a battery swapping station, and is used to solve the problems of impedance mismatch and current unevenness of charging modules. By analyzing the ripple current waveform of charging modules and extracting impedance phase characteristics, a dynamic impedance difference matrix is generated. Based on this matrix, the generalized entropy of impedance mismatch and the dynamic gain of topology reconstruction are calculated to optimize the connection of charging modules, dynamically form a parallel charging module group, and equalize the current through an inverse compensation signal to alleviate current unevenness; the PWM waveform phase shift adjustment is used to achieve staggered distribution of switch actions and reduce the influence of ripple; the virtual charging unit combines the analysis of bus voltage volatility to derive the charging power demand in real time and optimize the power distribution; in addition, relying on the three-dimensional temperature field model and the topology secondary reconstruction mechanism, it quickly responds to local overheating, improves the adaptability, reliability and operation efficiency of the charging process, and extends the service life of the equipment.

Description

Technical Field

[0001] The present invention relates to the field of battery charging control in a battery swapping station. More specifically, the present invention relates to a method for realizing dynamic regulation of battery charging in a battery swapping station control system. Background Art

[0002] The high-power charging system of a battery swapping station usually adopts a multi-module parallel topology architecture. For example, a 300kW charging unit is composed of multiple 50kW power modules to meet the dynamic charging requirements of battery packs at different times. During the peak battery swapping period, the system needs to call all modules to run at full load to achieve rapid energy replenishment; while during the low valley period, only a small number of battery packs need to be trickle-charged. However, the power modules of the existing chargers adopt a fixed grouping design, and the electrical connection and switch control logic of each group of modules are solidified at the hardware level, resulting in the system being unable to flexibly adjust the module combination according to the real-time load fluctuation. When the load demand is lower than the single-group power capacity, the system is forced to start and stop the modules in whole groups, causing no-load losses of redundant modules. At the same time, due to uneven current distribution, local modules overheat, and the device life is greatly attenuated.

[0003] Limited by the rigid architecture of the hardware topology and the static grouping mode of the control strategy, the existing chargers are unable to dynamically decouple the electrical connection relationship between modules on a millisecond time scale, resulting in the inability to synergistically optimize the system efficiency, current sharing accuracy, and device life, severely restricting the dynamic regulation ability of the battery swapping station for the battery charging process.

[0004] To solve the above problems, a technical solution is provided now. Summary of the Invention

[0005] To overcome the above defects of the prior art, an embodiment of the present invention provides a method for realizing dynamic regulation of battery charging in a battery swapping station control system. By analyzing the ripple current waveform of the charging module and extracting the impedance phase characteristics, a dynamic impedance difference matrix is generated. Based on this matrix, the generalized entropy of impedance mismatch and the dynamic gain of topology reconstruction are calculated to optimize the connection of the charging modules, dynamically form a parallel charging module group, and balance the current through an in-phase compensation signal to relieve current unevenness; the PWM waveform phase shift adjustment is used to realize the staggered distribution of switch actions and reduce the ripple effect; the virtual charging unit combines the analysis of the bus voltage volatility to deduce the charging power demand in real time and optimize the power distribution; in addition, relying on the three-dimensional temperature field model and the topology secondary reconstruction mechanism, it quickly responds to local overheating, improves the adaptability, reliability, and operation efficiency of the charging process, extends the device life, so as to solve the problems proposed in the above background art.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] S1: Continuously collect the ripple current waveforms at the output terminals of the charging modules, extract the impedance phase characteristics of each charging module, and generate a dynamic impedance difference matrix;

[0008] S2: Evaluate the impedance mismatch and reconstruction benefits based on the dynamic impedance difference matrix. When the evaluation results meet the reconstruction conditions, generate a topology reconstruction instruction;

[0009] S3: Control the relay array according to the topology reconstruction instruction, adjust the connections between the charging modules to form a parallel charging module group, and inject a compensation signal with a phase opposite to the impedance characteristics to balance the current;

[0010] S4: Adjust the phase offset of the PWM waveforms of each charging module in the parallel charging module group through a programmable delay unit, so that the switching actions of adjacent charging modules are staggered at a set interval to achieve ripple cancellation and current balance;

[0011] S5: Regard the parallel charging module group as a virtual charging unit, deduce the power demand based on the bus voltage volatility, and allocate the power weight coefficients of each virtual charging unit through a multi-objective optimization algorithm;

[0012] S6: Use a temperature sensing array to monitor the temperature of the charging modules in the parallel charging module group. When local temperature anomalies are detected, trigger a secondary topology reconstruction instruction, adjust the connections and enable standby charging modules.

[0013] In a preferred embodiment, step S1 includes the following:

[0014] Continuously obtain the output ripple current waveforms through current sensors set at the output terminals of each charging module; perform a fast Fourier transform on the collected ripple current waveforms to generate a current spectrum representation, and synchronously collect the voltage waveforms at the output terminals of the charging modules and generate a voltage spectrum representation through a fast Fourier transform; calculate the impedance phase of each charging module based on the phase difference between the current spectrum and the voltage spectrum at the operating frequency; calculate the impedance phase difference between any two charging modules according to the impedance phases of each charging module, take its absolute value, and construct a dynamic impedance difference matrix.

[0015] In a preferred embodiment, step S2 includes the following:

[0016] Normalize the elements of the dynamic impedance difference matrix to form a probability distribution, and then calculate the impedance mismatch generalized entropy index by weighted calculation of the square of the impedance phase difference. Specifically, multiply the normalized matrix elements by the square value of the impedance phase difference and sum them to quantify the complexity and concentration of the impedance phase difference between charging modules. Then, based on the dynamic impedance difference matrix, simulate the impedance difference change before and after topological reconstruction, calculate the change rate of the current standard deviation before and after reconstruction, and combine the nonlinear adjustment factor of power loss to calculate the topological reconstruction dynamic gain index. Specifically, multiply the change rate of the current standard deviation by the nonlinear adjustment factor of power loss to obtain the dynamic benefit of topological reconstruction for current distribution and efficiency improvement.

[0017] In a preferred embodiment, step S2 further includes the following content:

[0018] Weightedly sum the impedance mismatch generalized entropy index and the topological reconstruction dynamic gain index to obtain the topological reconstruction decision coefficient; finally, compare the value of the topological reconstruction decision coefficient with a preset threshold. When the topological reconstruction decision coefficient exceeds the preset threshold, generate a topological reconstruction instruction.

[0019] In a preferred embodiment, step S3 includes the following content:

[0020] After receiving the topological reconstruction instruction, read the dynamic impedance difference matrix. By comparing the impedance phase difference of each pair of charging modules in the matrix with a preset impedance difference threshold, identify the charging module pairs with impedance phase differences exceeding the impedance difference threshold and mark them as target pairs to be disconnected.

[0021] Subsequently, generate a disconnection instruction and send it to the relay array to control the relay array to cut off the direct electrical connection between the marked charging module pairs.

[0022] Then, based on the dynamic impedance difference matrix, use the hierarchical clustering method to group all charging modules according to the impedance phase difference to minimize the average value of the impedance phase difference within the group, and connect the charging modules within each group in parallel through the relay array to form a parallel charging module group.

[0023] Finally, calculate the average value of the impedance phase of the charging modules within each parallel charging module group, determine the impedance phase deviation of each charging module from the average value, generate a compensation signal in the opposite direction of the deviation and superimpose it on the control signal through the signal injection circuit to balance the current distribution within the group.

[0024] In a preferred embodiment, step S4 includes the following content:

[0025] Obtain the topological structure information of the parallel charging module group, and construct an electrical topological graph to record the position of each charging module within the group and its connection relationship with adjacent charging modules.

[0026] Then the total number of charging modules in the parallel charging module group is calculated, and the initial phase offset is determined by dividing the 360 ​​degrees of the complete cycle by the total number of charging modules, and then the phase offset is adjusted based on the deviation of the impedance phase of each charging module from the average value in the group and the dynamic correction factor;

[0027] The adjusted phase offset is then divided by three hundred and sixty degrees and multiplied by the pulse width modulation waveform switching period to be converted into a time delay value, and the pulse width modulation waveform start time is adjusted through the programmable delay unit in each charging module driving circuit to achieve a staggered distribution of the switching action;

[0028] Finally, the ripple coefficient of the output current of the parallel charging module group is monitored in real time. It is calculated by dividing the difference between the peak value and the valley value by the average value. If the ripple coefficient exceeds the preset threshold, the dynamic correction factor is fine-tuned and the phase offset is readjusted until the ripple coefficient meets the requirements to ensure current balance and ripple cancellation effects.

[0029] In a preferred embodiment, step S5 includes the following contents:

[0030] The parallel charging module group is regarded as an integral unit, which is defined as a virtual charging unit. The equivalent power output capacity of the virtual charging unit is calculated by accumulating the rated power of all charging modules in the parallel charging module group.

[0031] Then, by real-time monitoring of the bus voltage changes within a preset time period, the voltage fluctuation rate of the bus voltage is calculated, and the current charging power demand is derived based on the pre-established correspondence between the voltage fluctuation rate and the charging power demand;

[0032] Then, a power weight coefficient is assigned to each virtual charging unit, and all power weight coefficients are adjusted through iterative optimization to minimize the comprehensive evaluation objectives of the balance deviation and efficiency loss of the power weight coefficient allocation, while ensuring that the sum of the power outputs of all virtual charging units is equal to the derived charging power demand;

[0033] Finally, the optimized power weight coefficient is applied to the power control of the virtual charging unit.

[0034] In a preferred embodiment, step S6 includes the following contents:

[0035] The temperature data is collected in real time by multiple temperature sensors installed on the surface of each charging module radiator, and the discrete temperature data is converted into a continuous three-dimensional temperature field model using an interpolation algorithm. The temperature gradient of the three-dimensional temperature field is calculated to identify areas with drastic temperature changes.

[0036] Then, calculate the average temperature and the maximum temperature gradient in the area where each charging module is located, and compare them with the preset temperature anomaly threshold and gradient anomaly threshold. If the average temperature exceeds the temperature anomaly threshold or the maximum temperature gradient exceeds the gradient anomaly threshold, it is determined that there is a local temperature anomaly in the corresponding charging module;

[0037] Next, generate a topology secondary reconstruction instruction for the charging module with local temperature anomaly, and send the topology secondary reconstruction instruction to the control unit;

[0038] Finally, according to the topology secondary reconstruction instruction, disconnect the charging module with local temperature anomaly from the parallel charging module group through the control relay array and switch it to the independent power supply branch, and at the same time, select a spare charging module with an impedance phase characteristic similar to that of the original charging module from the spare charging module pool and connect it to the parallel charging module group, and recalculate the PWM waveform phase offset to ensure ripple cancellation and current balance.

[0039] Technical effects and advantages of the method for realizing dynamic regulation of battery charging in a power exchange station control system according to the present invention:

[0040] The present invention analyzes the ripple current waveform of the charging module in real time, extracts the impedance phase characteristic, generates a dynamic impedance difference matrix, and performs accurate quantitative evaluation on the impedance mismatch in the battery charging system of the power exchange station; based on this matrix, calculates the generalized entropy of the impedance mismatch and the dynamic gain of the topology reconstruction, intelligently optimizes the connection relationship of the charging modules, dynamically forms a parallel charging module group, and precisely balances the current distribution through the inverse compensation signal, effectively alleviating the problem of uneven current caused by impedance differences; at the same time, uses the PWM waveform phase offset adjustment to realize the staggered distribution of switch actions, reduce the ripple effect, and improve the current balance; the virtual charging unit design combines the analysis of the bus voltage volatility, derives the charging power demand in real time, and optimizes the power distribution weight to ensure the power output balance; in addition, relying on the temperature monitoring and topology secondary reconstruction mechanism of the three-dimensional temperature field model to form a closed-loop control, quickly respond to local overheating phenomena, and ensure the continuous stability of the system. Based on the data-driven dynamic management method, it not only improves the adaptability and reliability of the charging process, but also significantly improves the operation efficiency of the power exchange station and prolongs the service life of the equipment. Brief Description of the Drawings

[0041] Figure 1 It is a schematic flow chart of the method for realizing dynamic regulation of battery charging in a power exchange station control system according to the present invention. Detailed Embodiments

[0042] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0043] Embodiment 1: Figure 1 A method for realizing dynamic regulation of battery charging in a power exchange station control system of the present invention is given, including:

[0044] S1: Real-time collect the ripple current waveform at the output end of the charging module, extract the impedance phase characteristics of each charging module, and generate a dynamic impedance difference matrix;

[0045] S2: Evaluate the impedance mismatch and reconstruction benefit according to the dynamic impedance difference matrix. When the evaluation result meets the reconstruction condition, generate a topology reconstruction instruction;

[0046] S3: Control the relay array according to the topology reconstruction instruction, adjust the connection between the charging modules to form a parallel charging module group, and inject a compensation signal with a phase opposite to the impedance characteristic to balance the current;

[0047] S4: Adjust the phase offset of the PWM waveforms of each charging module in the parallel charging module group through a programmable delay unit, so that the switching actions of adjacent charging modules are staggered at a set interval to achieve ripple cancellation and current balance;

[0048] S5: Regard the parallel charging module group as a virtual charging unit, deduce the power demand based on the bus voltage volatility, and allocate the power weight coefficient of each virtual charging unit through a multi-objective optimization algorithm;

[0049] S6: Use the temperature sensing array to monitor the temperature of the charging modules in the parallel charging module group. When local temperature anomalies are detected, trigger a topology secondary reconstruction instruction, adjust the connection and enable the standby charging module.

[0050] In the power exchange station control system, the charging module, as the core power conversion unit, is responsible for converting the grid electric energy into the charging current required by the battery pack. Due to factors such as manufacturing process differences and aging degree, the electrical characteristics (such as impedance) of each charging module are heterogeneous, resulting in uneven current distribution during parallel operation, which in turn affects the overall efficiency and equipment life. The goal of step S1 is to collect and analyze the impedance characteristics of each charging module in real time, generate a dynamic impedance difference matrix, and provide a data basis for step S2 to evaluate the impedance mismatch and reconstruction benefit.

[0051] Step S1 includes the following content:

[0052] S1.1: Real-time collect the ripple current waveform:

[0053] A current sensor is set at the output end of each charging module to continuously obtain the output ripple current waveform. The ripple current waveform is the manifestation of the internal switching action and impedance characteristics of the charging module in terms of current, and serves as the basic signal for analyzing the electrical performance of the charging module. The sampling frequency is set to at least twice the switching frequency of the charging module to ensure that the sampling process can completely capture the waveform details and avoid information loss. This high-frequency sampling method ensures the accuracy and integrity of the waveform data.

[0054] S1.2, Extract the impedance phase characteristics:

[0055] Performing spectrum analysis on the collected ripple current waveform is to convert the waveform data in the time domain into a signal in the frequency domain through fast Fourier transform, generating a spectral representation of the current. At the same time, synchronously collect the voltage waveform at the output end of the charging module, and obtain the spectral representation of the voltage through the same transformation method. At the operating frequency of the charging module, compare the voltage spectrum with the current spectrum, and determine the impedance phase by calculating the phase difference between the two. The impedance phase reflects the impedance characteristics of the charging module at a specific frequency and is used for subsequent analysis of the differences between charging modules. Performing spectrum analysis on the ripple current waveform and the voltage waveform and calculating the impedance phase through the phase difference can deeply reveal the impedance characteristics of the charging module from the perspective of the frequency domain. The reason for choosing the impedance phase as a characteristic parameter is that the phase information is particularly sensitive to the characteristic changes of components such as inductors and capacitors inside the charging module, and can clearly reflect the electrical performance differences between modules. This method is more accurate than only analyzing the amplitude, helps to identify the specific reasons for impedance mismatch, and improves the pertinence of the evaluation.

[0056] S1.3, Generate a dynamic impedance difference matrix:

[0057] According to the impedance phase characteristics extracted from each charging module, construct a matrix to represent the impedance differences between all charging modules. The specific method is that for any two charging modules in the system, calculate the difference in their impedance phases, take the absolute value of the difference, and fill these values into the corresponding positions of the matrix to form a symmetric matrix structure, thereby constructing and obtaining a dynamic impedance difference matrix. The dynamic impedance difference matrix will be updated in real time as the impedance phase changes, reflecting the dynamic changes in the impedance characteristics. Each element in the dynamic impedance difference matrix represents the impedance phase difference between a pair of charging modules.

[0058] Construct a dynamic impedance difference matrix based on impedance phase characteristics to quantify the impedance differences between all charging modules in matrix form, facilitating the intuitive display and analysis of the relative relationships between modules. Calculate the impedance phase difference and take the absolute value as the matrix element, which can be directly related to the unevenness of current distribution between modules, providing a clear basis for subsequent topology adjustment. The real-time update mechanism of the matrix ensures that the system can respond promptly to changes in the status of charging modules, enhancing the adaptability and dynamics of the entire regulation process.

[0059] Step S1 collects the ripple current waveform in real time, extracts the impedance phase characteristics, and generates a dynamic impedance difference matrix. As the analysis result, the dynamic impedance difference matrix will be directly transmitted to Step S2 to evaluate the impedance mismatch degree between charging modules and the potential benefits of topology reconstruction. Step S2 will judge whether to adjust the system topology according to the impedance phase difference data in the matrix and generate corresponding instructions. This continuity of data and logic ensures seamless connection of the entire process from data collection to decision-making, enabling efficient implementation of system regulation.

[0060] Step S2 includes the following:

[0061] S2.1, Evaluate impedance mismatch:

[0062] First, calculate the quantification index of impedance mismatch based on the dynamic impedance difference matrix. Each element in the dynamic impedance difference matrix represents the impedance phase difference between two charging modules. In the processing, these impedance phase differences are first normalized to make their numerical distributions within a unified range, forming a probability distribution. Then, use this probability distribution to calculate the generalized entropy index of impedance mismatch to quantify the degree of impedance mismatch between charging modules. When calculating the generalized entropy index of impedance mismatch, first count the distribution complexity and concentration of impedance phase differences in the probability distribution, then square the value of each impedance phase difference and assign weights to highlight the influence of larger impedance phase differences, and finally summarize and adjust all weighted values to obtain the generalized entropy index of impedance mismatch. For example, the following calculation method can be adopted:

[0063] For the dynamic impedance difference matrix of elements perform normalization to obtain the probability distribution:

[0064]

[0065] where, is the total number of charging modules, is the charging module number corresponding to the row index of the dynamic impedance difference matrix, represents the charging module number corresponding to the column index of the dynamic impedance difference matrix. Calculate the generalized entropy index of impedance mismatch:

[0066]

[0067] Parameter Explanation:

[0068] : The th impedance phase difference between the th charging modules, in radians, dimensionless.

[0069] : The normalized impedance phase difference probability, dimensionless.

[0070] : The dynamic impedance difference matrix transpose multiplied by , the trace is dimensionless.

[0071] All are dimensionless quantities, and the overall impedance mismatch generalized entropy index is dimensionless.

[0072] By calculating the impedance mismatch generalized entropy index, the complexity and concentration of impedance differences between charging modules can be accurately measured. Compared with only calculating the average or maximum value of the impedance phase difference, this method more comprehensively reflects the overall characteristics of the impedance distribution. Especially when the number of charging modules is large, it can effectively capture the impact of local impedance mismatch on the system. Introducing the square weighting process of the impedance phase difference enhances the sensitivity to larger impedance differences, helps to identify module pairs that may cause serious uneven current distribution, thereby improving the pertinence and accuracy of the evaluation. This quantification method provides a reliable data basis for subsequent decisions.

[0073] S2.2, Evaluate the benefits of topology reconstruction:

[0074] Secondly, based on the dynamic impedance difference matrix, evaluate the potential benefits of topology reconstruction on system performance improvement. The processing process first simulates the change in impedance difference after topology reconstruction, and predicts the uniformity of current distribution and the reduction of power loss between charging modules after reconstruction. The specific method is to calculate the current value distribution of each charging module before reconstruction based on the dynamic impedance difference matrix, and statistically analyze its uniformity; then simulate the impedance distribution after reconstruction, recalculate the current value distribution, and compare the change rate of current uniformity before and after reconstruction. At the same time, estimate the numerical change of power loss before and after reconstruction, and combine the change rate of current uniformity. Through a non-linear comprehensive method, the two are comprehensively calculated to obtain the topology reconstruction dynamic gain index. This gain index reflects the dynamic benefits of topology reconstruction on current balance and efficiency improvement. The larger the value, the higher the benefit. For example, the following calculation method can be adopted to obtain it:

[0075] Based on the dynamic impedance difference matrix , simulate the impedance difference matrix after topology reconstruction (assuming the reconstruction reduces the impedance difference).

[0076] Calculate the change rate of the standard deviation of the current before and after reconstruction, and introduce a power loss adjustment factor to calculate the dynamic gain index of topology reconstruction:

[0077]

[0078] : The standard deviation of the current of each charging module before reconstruction, with the unit of ampere (A).

[0079] : The standard deviation of the current of each charging module after simulated reconstruction, with the unit of ampere (A).

[0080] : The difference in power loss before and after reconstruction, with the unit of watt (W).

[0081] Dimensional analysis: Dimensionless, Through the exponential function, It is dimensionless, and the dynamic gain index of topology reconstruction is dimensionless as a whole.

[0082] When evaluating the benefits of topology reconstruction, comprehensively considering the improvement of current distribution and the reduction of power loss can comprehensively reflect the potential improvement of the reconstruction on the system performance. Using a non-linear comprehensive method to fuse these two factors makes the gain index more sensitive to the improvement of power loss. Especially when the power loss changes greatly, it can more accurately reflect the actual benefits of the reconstruction.

[0083] S2.3, Comprehensive decision coefficient:

[0084] Weighted sum of the impedance mismatch generalized entropy index and the dynamic gain index of topology reconstruction to calculate the comprehensive decision coefficient. A higher comprehensive decision coefficient value indicates more serious impedance mismatch and significant reconstruction benefits, while a lower value indicates less mismatch or limited reconstruction effect.

[0085] S2.4, Generate topology reconstruction instructions:

[0086] Finally, a topology reconstruction instruction is generated based on the comprehensive decision coefficient. The value of the comprehensive decision coefficient is compared with a preset threshold. The threshold is determined according to historical operation data and system performance requirements to ensure the rationality of the decision. When the value of the comprehensive decision coefficient exceeds the preset threshold, a topology reconstruction instruction is generated to indicate the system to adjust the connection of the charging modules; if it does not exceed the preset threshold, the current connection state remains unchanged. The generated topology reconstruction instruction will be directly passed to step S3 for subsequent connection adjustment operations. Generating the topology reconstruction instruction based on the comparison between the comprehensive decision coefficient and the preset threshold ensures the objectivity and controllability of the decision-making process. The setting of the threshold can be flexibly adjusted according to specific application scenarios and performance requirements, enhancing the adaptability of the system. This decision-making mechanism based on quantitative evaluation avoids the uncertainty of subjective judgment and helps to achieve the automation and intelligence of charging regulation. At the same time, the instruction is only triggered when the impedance mismatch is significant and the reconstruction benefit is considerable, improving the overall operation efficiency and stability of the swapping station.

[0087] The comprehensive decision coefficient (TRDC) is a dimensionless value obtained by comprehensively calculating the quantitative evaluation of the impedance mismatch between charging modules (i.e., the generalized entropy index of impedance mismatch) and the potential benefit of topology reconstruction (i.e., the dynamic gain index of topology reconstruction), and is used to evaluate the degree of impedance mismatch of charging modules and whether it is worth performing adjustments for topology reconstruction. The topology reconstruction instruction is a trigger signal generated when the comprehensive decision coefficient exceeds the preset threshold, indicating that the connection relationship of the charging modules needs to be adjusted to optimize the system performance, but this instruction only serves as a trigger and does not include specific adjustment schemes.

[0088] In step S2, by evaluating the quantitative index of the impedance mismatch between charging modules and the gain index of topology reconstruction, a comprehensive decision coefficient is generated, and a topology reconstruction instruction is generated when the value of the comprehensive decision coefficient exceeds the preset threshold, providing a clear trigger condition for step S3. This decision-making process based on quantitative analysis ensures that the connection of the charging modules is only adjusted when the degree of impedance mismatch is significant and the benefit of topology reconstruction is considerable, improving the dynamic regulation ability and operation efficiency of the system. Step S3 directly controls the relay array to adjust the connection between the charging modules according to the topology reconstruction instruction generated in step S2.

[0089] The topology reconstruction instruction, as a trigger signal, indicates that the connection relationship between the charging modules needs to be adjusted, but does not include specific adjustment schemes. The task of step S3 is to analyze and execute the connection adjustment between the charging modules according to the topology reconstruction instruction, form a parallel charging module group, and balance the current by injecting a compensation signal. It effectively alleviates the current unevenness caused by impedance mismatch and creates conditions for PWM waveform optimization.

[0090] Step S3 includes the following content:

[0091] S3.1, identifying module pairs with impedance differences exceeding the threshold:

[0092] After receiving the topology reconstruction instruction, first read the dynamic impedance difference matrix, which records the impedance phase differences between all charging modules. Then, set an impedance difference threshold, which is a fixed standard value determined comprehensively based on the impedance tolerance of the charging modules and historical operation data. Check each element in the dynamic impedance difference matrix one by one, and compare the impedance phase difference between any two charging modules with the impedance difference threshold. If the impedance phase difference between two charging modules exceeds the impedance difference threshold, mark these two charging modules as the target pair that needs to disconnect the direct connection, so as to reduce the impact of impedance mismatch on the overall performance of the system in subsequent processing.

[0093] By setting a fixed impedance difference threshold and screening out the charging module pairs with impedance phase differences exceeding this threshold, it is possible to quickly locate those module pairs with relatively serious impedance mismatches, without the need for a comprehensive and complex analysis of the entire dynamic impedance difference matrix. This screening method reduces the computational complexity in the system identification process, improves the pertinence and efficiency of processing. Only performing subsequent operations on the charging module pairs that truly need to be adjusted not only optimizes the response speed of the regulation but also improves the utilization efficiency of system resources and reduces unnecessary processing burdens.

[0094] S3.2, Control the relay array to disconnect:

[0095] After identifying the charging module pairs with impedance differences exceeding the impedance difference threshold, generate specific disconnection instructions for these marked target pairs and send the disconnection instructions to the relay array deployed at the connection nodes between the charging modules. The relay array is a hardware device that can control the electrical on-off state between each pair of charging modules according to the received instructions. After executing the disconnection instructions, the relay array will cut off the direct electrical connection between these marked target pairs, thus providing operation space and flexibility for subsequent connection adjustment and reconstruction. The dynamic control of the connection between charging modules through the relay array enables the electrical connection relationship between charging modules to be adjusted in real time according to actual needs.

[0096] S3.3, Dynamically form parallel charging module groups:

[0097] After completing the disconnection operation, based on the dynamic impedance difference matrix, using the hierarchical clustering method, all charging modules are grouped based on the distance metric of impedance phase difference. Specifically, the grouping process uses the impedance phase difference between charging modules as a measure, and charging modules with similar impedance characteristics are grouped into the same group. The goal of grouping is to minimize the average value of the impedance phase differences of the charging modules within each group. After grouping, the connection relationship is adjusted through a relay array, and the charging modules within each group are connected in parallel with each other, thus forming several parallel charging module groups. Using the hierarchical clustering method to dynamically form parallel charging module groups can ensure a high degree of matching of the impedance characteristics of the charging modules within each parallel charging module group. The adaptive grouping method based on real-time impedance distribution enables the system to flexibly adjust the combination structure of the charging modules according to the current operating state, thereby minimizing the impedance difference within the group to the greatest extent.

[0098] S3.4, Inject compensation signal to equalize current:

[0099] After forming the parallel charging module groups, further processing is performed on the charging modules within each group. First, calculate the average value of the impedance phases of all charging modules within each parallel charging module group. Then, for each charging module within the group, calculate the deviation of its impedance phase from the average value of the group. Based on these deviation amounts, a compensation signal with the opposite deviation direction is generated, and this compensation signal is superimposed on the control signal of each charging module through a dedicated signal injection circuit to cancel the uneven current distribution caused by impedance differences, ultimately achieving current equalization within the group. To assist understanding, the following processing example is given:

[0100] For each parallel charging module group, calculate the average value of the impedance phases of the charging modules within the group , where is the number of modules within the group, is the module 's impedance phase.

[0101] For each charging module within the group , calculate its impedance phase deviation .

[0102] Generate the compensation signal , where 1 is the compensation amplitude (unit: volt), determined by the current deviation requirement; is the system operating frequency (unit: radian / second).

[0103] Through the signal injection circuit, is superimposed on the control signal of the charging module .

[0104] By generating and injecting a compensation signal in the opposite direction of the impedance phase deviation, the problem of uneven current caused by residual impedance differences in the parallel charging module group can be accurately cancelled. Compared with the method that only relies on physical connection adjustment, this signal compensation method provides a more refined current equalization means. Even in the case where the impedance difference cannot be completely eliminated by connection adjustment, it can still effectively improve the uniformity of current distribution, thereby improving the overall operation efficiency and stability of the system, and ensuring the reliability and consistency of the charging process.

[0105] Step S3 completes the optimized adjustment of the connection relationship of the charging modules and the equalization process of the current by identifying the charging module pairs with impedance differences exceeding the threshold, controlling the relay array to disconnect, dynamically forming a parallel charging module group, and injecting a compensation signal. The formed parallel charging module group provides a stable electrical topology structure for the subsequent steps, enabling the subsequent optimization work to be carried out on the basis of current equalization. This processing method effectively alleviates the problem of uneven current distribution caused by impedance mismatch, improves the dynamic regulation ability and operation efficiency of the system, and lays a foundation for the optimization of the entire charging process.

[0106] Step S3 forms a parallel charging module group by adjusting the connection relationship between the charging modules and initially achieves current equalization. However, due to the inconsistency in time of the switching actions of the charging modules, it may lead to the superposition of output current ripples, thereby affecting the stability and charging efficiency of the system. Therefore, step S4 optimizes the phase offset of the PWM waveforms of each charging module in the parallel charging module group, so that the switching actions are staggered at a set interval to further reduce the ripple and improve the current equalization effect.

[0107] Step S4 includes the following contents:

[0108] S4.1, obtaining the electrical topology information of the parallel charging module group:

[0109] First, obtain the topology structure information of the parallel charging module group from the previous step. This information includes the number of each charging module in the group, the parallel connection method between them, and the injection state of the compensation signal. Based on these data, an electrical topology diagram is constructed to detail the specific position of each charging module in the parallel structure and its connection relationship with adjacent charging modules. Through this electrical topology diagram, it is convenient to comprehensively master the overall connection characteristics of the parallel charging module group.

[0110] By obtaining the electrical topology information of the parallel charging module group and constructing the corresponding topology diagram, it provides an accurate structural basis for the phase adjustment of the pulse width modulation waveform. Ensures that the phase adjustment can be optimized for the specific connection relationship, with strong pertinence and effectiveness.

[0111] S4.2, calculating the phase offset of the pulse width modulation waveform:

[0112] After obtaining the electrical topology information of the parallel charging module group, start calculating the phase offset of the pulse width modulation waveform for each charging module in the parallel charging module group. First, determine the total number of charging modules in the parallel charging module group and set the switching period of the pulse width modulation waveform. To achieve an ideal ripple cancellation effect, assign an initial phase offset to each charging module to ensure that the switching actions of adjacent charging modules are evenly staggered. The specific calculation method is to regard a complete cycle as 360 degrees, then divide it by the total number of charging modules to obtain the phase interval between each charging module, and then determine its initial phase offset in sequence according to the serial number of each charging module.

[0113] However, relying solely on evenly distributed phase offsets may not completely eliminate the ripple problem caused by impedance differences. For this reason, introduce a dynamic correction factor and perform further adjustment in combination with the impedance phase characteristics extracted in the foregoing steps. The specific approach is to first calculate the average value of the impedance phases of all charging modules in the parallel charging module group, and then for each charging module, calculate the deviation between its impedance phase and this average value. According to the magnitude of this deviation and a dynamic correction factor, fine-tune the initial phase offset to finally generate the corrected phase offset. The value range of the dynamic correction factor is set between zero and one, and its specific value is jointly determined by the amplitude of the impedance phase deviation and the stability requirements of the system. For example, the adjustment is calculated in the following way:

[0114] Considering the impedance phase characteristics extracted in step S1 Regarding the impact on the switching action, introduce a dynamic correction factor , calculate the corrected phase offset :

[0115]

[0116] Among them, is the average value of the impedance phases of the charging modules in the group, is the dynamic correction factor, which is determined according to the amplitude of the impedance phase deviation and the stability requirements of the parallel charging module group, and the range is [0,1].

[0117] By combining evenly distributed phase allocation with dynamic correction based on impedance phase characteristics, it is convenient to adaptively optimize the phase offset of the pulse width modulation waveform, overcoming the limitations of the traditional evenly distributed method when the impedance difference is large. Based on fully considering the impedance differences between charging modules, the phase adjustment is carried out, greatly improving the accuracy of ripple cancellation. Especially in the scenario where the impedance distribution is uneven, this can effectively reduce the fluctuation amplitude of the output current, thereby enhancing the stability and charging quality of the system.

[0118] S4.3, Implement time control of phase shift:

[0119] After calculating the corrected phase shift amount, precise adjustment of the pulse-width modulation waveform is achieved through hardware control means. Specifically, a programmable delay unit is integrated in the drive circuit of each charging module to adjust the start time of the pulse-width modulation waveform. Converting the corrected phase shift amount into a specific time delay value is done by dividing the phase shift amount by 360 degrees of a complete cycle and then multiplying by the switching period of the pulse-width modulation waveform to obtain the specific time delay value required for each charging module. Subsequently, these time delay values are input into the corresponding programmable delay units, and by adjusting the start time of the pulse-width modulation waveform of each charging module, it is ensured that the switching actions of all charging modules are staggered according to the calculated phase shift amount.

[0120] By implementing time control of phase shift using a programmable delay unit at the hardware level, it can be ensured that the adjustment of the pulse-width modulation waveform fully meets the design requirements, thereby improving the accuracy of the staggered distribution of switching actions.

[0121] S4.4, Verify the ripple cancellation and current equalization effects:

[0122] After completing the adjustment of the pulse-width modulation waveform, the output current waveforms of the parallel charging module group are monitored in real time to evaluate the actual effects of ripple cancellation and current equalization. The specific evaluation method is to calculate the ripple coefficient of the output current, that is, to quantify the ripple amplitude by dividing the difference between the peak and valley values of the output current by the average value of the output current. If the calculated ripple coefficient exceeds a pre-set threshold, the dynamic correction factor will be fine-tuned, and the phase shift amount and the corresponding time delay values will be recalculated, and the adjustment will be repeated until the ripple coefficient drops below the pre-set threshold. Through this closed-loop adjustment mechanism, the phase distribution of the pulse-width modulation waveform can be dynamically optimized to ensure the stability and equalization of the output current.

[0123] By monitoring the ripple in real time and combining with the closed-loop dynamic fine-tuning mechanism, the system can continuously optimize the phase adjustment effect of the pulse-width modulation waveform during actual operation to ensure the achievement of the ripple cancellation and current equalization goals.

[0124] Step S4 completes the optimization adjustment of the pulse-width modulation waveforms of each charging module in the parallel charging module group by obtaining the electrical topology information of the parallel charging module group, calculating and correcting the phase shift amount of the pulse-width modulation waveform, implementing time control of phase shift, and verifying the ripple cancellation and current equalization effects. The adjusted pulse-width modulation waveform makes the switching actions of the charging modules staggered at set intervals, effectively reducing the ripple amplitude of the output current and achieving current equalization. This optimization result provides a stable current environment for the subsequent step of considering the parallel charging module group as a virtual charging unit for power distribution.

[0125] Step S4 achieves the staggered distribution of switching actions by adjusting the PWM waveform phase offset of each charging module in the parallel charging module group, effectively offsetting the ripple and balancing the current distribution. However, on the basis of optimizing the current distribution, the system still needs to further dynamically adjust the power output of each charging module group according to the load demand and operating status to improve the overall charging efficiency and ensure operational stability. To this end, step S5 regards the parallel charging module group as a virtual charging unit, derives the current charging power demand by analyzing the bus voltage fluctuation, and uses an optimization algorithm to assign a power weight coefficient to each virtual charging unit, thereby achieving efficient and balanced power distribution.

[0126] Step S5 includes the following contents:

[0127] S5.1, Mapping of virtual charging units:

[0128] First, the parallel charging module group formed in step S3 is regarded as an integral unit, called a virtual charging unit. Each parallel charging module group is composed of several charging modules, which are connected in parallel and work together. In order to simplify subsequent power management, each parallel charging module group is abstracted as a virtual charging unit, and its equivalent power output capacity is defined as the sum of the rated powers of all charging modules in the parallel charging module group. The specific processing process is to traverse each parallel charging module group, count the rated power of each charging module therein, and then add up these rated powers one by one to obtain the overall power output capacity of the parallel charging module group as the power benchmark of the virtual charging unit.

[0129] This mapping enhances the modular nature of the system, making power allocation and exception handling more uniform and efficient.

[0130] S5.2, derivation of power demand based on bus voltage fluctuation rate:

[0131] After determining the virtual charging unit, the current charging power demand is deduced by monitoring the changes in the bus voltage in real time. The specific processing process includes the following steps: First, continuously collect bus voltage data within a preset time period and record the voltage value at each time point; then, calculate the absolute value of the voltage change between adjacent time points, and average these change values ​​over the entire time period as a measure of the bus voltage fluctuation rate. Next, based on historical operating data and system characteristics, a corresponding relationship between bus voltage fluctuation rate and charging power demand is established. This relationship assumes that the more violent the voltage fluctuation, the lower the power demand, and vice versa. By analyzing a large amount of historical data, the specific changing trend between volatility and power demand is fitted, and based on this, the required charging power value is determined according to the current real-time calculated volatility. For example, the following processing method is used to derive power demand:

[0132] Collect the bus voltage in real time , calculate its volatility , defined as the average absolute value of voltage change per unit time:

[0133]

[0134] Among them, is the observation time window, with the unit of second (s).

[0135] Establish the inverse mapping relationship between the bus voltage volatility and the charging power demand . Assume that the voltage volatility and the power demand are in a non-linear inverse relationship, and it is obtained by fitting historical data:

[0136]

[0137] Parameter explanation:

[0138] Among them, and are fitting parameters, with the unit of watt-second (W·s), with the unit of volt per second (V / s).

[0139] Bus voltage, with the unit of volt (V).

[0140] : Bus voltage volatility, with the unit of volt per second (V / s).

[0141] : Observation time window, with the unit of second (s).

[0142] : Charging power demand, with the unit of watt (W).

[0143] , : Fitting parameters, with watt-second (W·s) and volt per second (V / s) respectively.

[0144] Based on the processing method of deriving the charging power demand from the bus voltage volatility, the system can perceive the changes in the load in real time and make rapid adjustments, avoiding the delay problems caused by relying on fixed-time sampling or preset power values in traditional methods. It is applicable to the operating environment with frequent load fluctuations in the swapping station, and can improve the dynamic adaptability and stability of the system. In addition, through the precise correspondence relationship between the volatility and the power demand, the system can more accurately match the actual load demand.

[0145] S5.3, Power weight allocation of multi-objective optimization algorithm:

[0146] After deriving the total charging power demand, a multi-objective optimization algorithm is used to assign power weight coefficients to each virtual charging unit. Each virtual charging unit is given a weight coefficient between 0 and 1, and the sum of the weight coefficients of all virtual charging units is constantly 1. The specific processing process is as follows: First, a comprehensive evaluation objective is defined, which consists of two parts: one is an index to measure the balance of weight coefficients, determined by calculating the sum of the squared deviations of each weight coefficient from the average of all weight coefficients; the other is an index to measure efficiency, obtained by multiplying the efficiency characteristics of each virtual charging unit by its power output capacity and then accumulating. Subsequently, through iterative calculations, the values of each weight coefficient are gradually adjusted to minimize the comprehensive evaluation objective while ensuring that the sum of the power outputs of all virtual charging units is equal to the derived total charging power demand. To achieve this goal, mature optimization techniques are adopted, such as repeatedly comparing the evaluation results of different weight combinations until the optimal solution is found. For example, the following processing method is adopted:

[0147] Let the power weight coefficients of each virtual charging unit be , satisfying and .

[0148] Define the multi-objective function , comprehensively considering the balance of power distribution and the maximization of efficiency:

[0149]

[0150] where is the average weight, is the efficiency coefficient (dimensionless) of the virtual charging unit , and are weight factors (dimensionless), is the equivalent power output capacity of the virtual charging unit , in watts (W).

[0151] Use the genetic algorithm or particle swarm optimization algorithm to solve , making minimize, while satisfying the constraint conditions:

[0152]

[0153] By using a multi-objective optimization algorithm to allocate power weight coefficients, it is possible to achieve the balance of power distribution and the maximization of efficiency while meeting the total power demand. This overcomes the limitations of traditional average distribution or fixed-ratio distribution methods, and can dynamically adjust the distribution ratio according to the actual operating status and efficiency characteristics of virtual charging units, thereby improving the overall operating efficiency and stability of the system.

[0154] In step S5, by mapping the parallel charging module group to virtual charging units, deriving the charging power demand based on the bus voltage volatility, and using a multi-objective optimization algorithm to dynamically allocate the power weight coefficients of each virtual charging unit, the efficient and balanced distribution of power is achieved. The mapping of virtual charging units simplifies the complexity of the system model, the analysis of bus voltage volatility provides a reliable basis for real-time power demand, and the multi-objective optimization algorithm ensures the optimal balance between balance and efficiency in power distribution. The finally generated power weight coefficients will be directly used to guide the power output control of virtual charging units.

[0155] In step S5, the parallel charging module group is regarded as virtual charging units, the power demand is derived based on the bus voltage volatility, and the power weight coefficients of each virtual charging unit are allocated through an optimization algorithm, achieving the efficient and balanced distribution of power. However, when the charging modules operate at high power, local temperature anomalies may occur due to uneven load or heat dissipation problems, affecting the stability and safety of the parallel charging module group. Therefore, in step S6, a temperature sensing array is used to monitor the temperature of the charging modules in the parallel charging module group, and when a local temperature anomaly is detected, a topology secondary reconstruction instruction is triggered to adjust the connection and enable standby charging modules to ensure the continuous and stable operation of the parallel charging module group.

[0156] Step S6 includes the following:

[0157] S6.1, constructing a three-dimensional temperature field distribution model:

[0158] Through multiple temperature sensors set on the surface of the radiator of each charging module, the temperature data at each position is collected in real time. The number of temperature sensors is determined according to the size and heat dissipation characteristics of the charging module to ensure that the key heat dissipation areas can be fully covered. The collected temperature data are multiple discrete points, and these discrete points are processed through an interpolation algorithm (such as Kriging interpolation method) to convert them into a continuous three-dimensional temperature field model. The specific calculation logic is: taking the position of each temperature sensor as a reference, using the interpolation algorithm to estimate the temperature values between the sensors, thereby generating a continuous model that can reflect the temperature distribution in the internal space of the charging module. On this basis, further analyze the spatial variation of temperature in the three-dimensional temperature field. By calculating the ratio of the difference in temperature values at adjacent positions to the corresponding spatial distance, the rate of change of temperature in space, that is, the temperature gradient, is obtained, which is used to identify the areas where the temperature changes violently.

[0159] The purpose of constructing a three-dimensional temperature field distribution model is to convert discrete temperature data into a continuous spatial distribution representation through an interpolation algorithm, enabling the system to comprehensively grasp the temperature state inside the charging module.

[0160] S6.2, Detect local temperature anomalies:

[0161] After obtaining the three-dimensional temperature field and temperature gradient data, local temperature anomalies of each charging module are detected. The specific processing process is as follows: First, based on the rated operating temperature and heat dissipation characteristics of the charging module, a temperature anomaly threshold and a gradient anomaly threshold are preset as the judgment basis. For each charging module, calculate the average temperature and the maximum temperature gradient within the spatial region where it is located. The average temperature is calculated by summing all the temperature data points in the region and then dividing by the total number of data points to obtain a value representing the overall temperature level; the maximum temperature gradient is calculated by selecting the largest value from the temperature change rates of all adjacent positions in the region. Then, compare the average temperature of each charging module with the preset temperature anomaly threshold, and at the same time compare its maximum temperature gradient with the preset gradient anomaly threshold. If the average temperature of a certain charging module exceeds the temperature anomaly threshold, or its maximum temperature gradient exceeds the gradient anomaly threshold, it is determined that the charging module has a local temperature anomaly.

[0162] The detection of local temperature anomalies uses a dual judgment criterion of average temperature and maximum temperature gradient, aiming to simultaneously evaluate the absolute level of temperature and the severity of spatial changes.

[0163] S6.3, Trigger the topological secondary reconstruction instruction:

[0164] When a local temperature anomaly is detected in a certain charging module, a topological secondary reconstruction instruction is immediately generated for this charging module. The specific processing logic is as follows: According to the anomaly detection result, identify the charging module with temperature anomalies and record its number, and at the same time generate an instruction containing this number and the connection adjustment requirements. After the instruction is generated, it is sent to the control unit as a trigger signal for subsequent connection adjustment operations to reduce the load of the abnormal charging module or isolate its impact on other modules.

[0165] The purpose of triggering the topological secondary reconstruction instruction is to quickly respond to abnormal situations within the parallel charging module group based on the temperature anomaly detection results. This mechanism of dynamically generating instructions enhances the system's adaptability and security, ensuring the efficiency and timeliness of anomaly handling. It avoids the risk of system performance degradation or equipment damage caused by the continuous existence of temperature anomalies.

[0166] S6.4, Adjust the connection and enable the standby charging module:

[0167] After receiving the topology secondary reconstruction instruction, the connection status of the charging modules is adjusted by controlling the relay array. The specific processing procedure is as follows: First, the charging module with abnormal temperature is disconnected from the current parallel charging module group and switched to an independent power supply branch to isolate its impact on other modules. Then, a spare charging module is selected from the spare charging module pool. The selection basis is to compare whether the difference in impedance phase characteristics between the spare module and the abnormal module is less than a preset phase matching threshold to ensure a high degree of electrical characteristic matching between the two. After the spare charging module is selected, it is connected to the parallel charging module group where the abnormal module originally was through the relay array, keeping the number of charging modules in the group unchanged. Finally, according to the adjusted topology structure of the parallel charging module group, the PWM waveform phase offset of each charging module is recalculated.

[0168] The purpose of adjusting the connection and enabling the spare charging module is to quickly restore the normal operation state of the parallel charging module group without interrupting the overall charging service. Switching the abnormal charging module to an independent power supply branch effectively isolates its impact, and the spare charging module selected based on impedance phase characteristic matching ensures the electrical compatibility with other modules in the group, avoiding impedance mismatch problems caused by module replacement. Recalculating the PWM waveform phase offset further guarantees the stability and charging efficiency of the reconstructed system. It improves the fault tolerance and continuous operation ability of the system, providing a solid guarantee for the safety and efficiency of the charging process.

[0169] Step S6 realizes the rapid response and dynamic adjustment of temperature anomalies in the parallel charging module group by constructing a three-dimensional temperature field distribution model, detecting local temperature anomalies, triggering the topology secondary reconstruction instruction, and adjusting the connection and enabling the spare charging module. The application of the three-dimensional temperature field model and the dual anomaly detection standard significantly improves the accuracy and reliability of temperature monitoring; the generation of the topology secondary reconstruction instruction and the enabling of the spare charging module ensure the continuous and stable operation of the system. The adjusted topology structure of the parallel charging module group provides a safe and efficient basis for charging control, ensuring the stability and safety of the entire charging process.

[0170] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the real situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0171] It should be noted that the system of the present invention can be deployed on the device itself to achieve embedded applications, or can also run on a PC or other terminals with a user interface, so as to meet various hardware environments and usage requirements.

[0172] Only some exemplary embodiments of the present invention have been described by way of illustration above. Undoubtedly, for those of ordinary skill in the art, the described embodiments can be modified in various different ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and description are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

[0173] It should be noted that in this text, if there are relational terms such as first and second, they are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the element.

[0174] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. An implementation method for dynamically regulating battery charging in a battery swapping station control system, characterized in that, Including the steps: S1: Continuously collect the ripple current waveforms at the output ends of the charging modules in real time, extract the impedance phase characteristics of each charging module, and generate a dynamic impedance difference matrix; S2: Evaluate the impedance mismatch and reconstruction benefits based on the dynamic impedance difference matrix. When the evaluation result meets the reconstruction conditions, generate a topology reconstruction instruction; S3: Control the relay array according to the topology reconstruction instruction, adjust the connections between the charging modules to form a parallel charging module group, and inject a compensation signal with a phase opposite to the impedance characteristics to balance the current; S4: Adjust the phase offset of the PWM waveforms of each charging module in the parallel charging module group through a programmable delay unit, so that the switching actions of adjacent charging modules are staggered at a set interval to achieve ripple cancellation and current balance; S5: Regard the parallel charging module group as a virtual charging unit, deduce the power demand based on the bus voltage volatility, and allocate the power weight coefficients of each virtual charging unit through a multi-objective optimization algorithm; S6: Use a temperature sensing array to monitor the temperature of the charging modules in the parallel charging module group. When local temperature anomalies are detected, trigger a topology secondary reconstruction instruction, adjust the connections, and enable standby charging modules.

2. The implementation method for dynamically regulating battery charging in a swapping station control system according to claim 1, wherein Step S1 includes the following: Continuously obtain the output ripple current waveforms through current sensors set at the output ends of each charging module; Perform a fast Fourier transform on the collected ripple current waveforms to generate a current spectrum representation, and simultaneously collect the voltage waveforms at the output ends of the charging modules, and generate a voltage spectrum representation through a fast Fourier transform; Based on the phase difference between the current spectrum and the voltage spectrum at the operating frequency, calculate the impedance phase of each charging module; According to the impedance phases of each charging module, calculate the impedance phase difference between any two charging modules, and take its absolute value to construct a dynamic impedance difference matrix.

3. The implementation method for dynamically regulating battery charging in a battery swapping station control system according to claim 2, characterized in that, Step S2 includes the following: Normalize the elements of the dynamic impedance difference matrix to form a probability distribution, and then calculate the generalized entropy index of impedance mismatch through the square weighting of the impedance phase difference. Specifically, multiply the normalized matrix elements by the square value of the impedance phase difference and sum them to quantify the complexity and concentration of the impedance phase differences between the charging modules; Then, based on the dynamic impedance difference matrix, simulate the impedance difference changes before and after topology reconstruction, calculate the change rate of the current standard deviation before and after reconstruction, and calculate the topology reconstruction dynamic gain index in combination with the non-linear adjustment factor of power loss. Specifically, multiply the change rate of the current standard deviation by the non-linear adjustment factor of power loss to obtain the dynamic benefits of topology reconstruction for current distribution and efficiency improvement.

4. The implementation method for dynamically regulating battery charging of an exchange station control system according to claim 3, wherein Step S2 also includes the following: Sum the generalized entropy index of impedance mismatch and the topology reconstruction dynamic gain index with weights to obtain a topology reconstruction decision coefficient; Finally, compare the value of the topology reconstruction decision coefficient with a preset threshold. When the topology reconstruction decision coefficient exceeds the preset threshold, generate a topology reconstruction instruction.

5. The implementation method for dynamically regulating battery charging in a swapping station control system according to claim 4, wherein, Step S3 includes the following: After receiving the topology reconstruction instruction, read the dynamic impedance difference matrix, and by comparing the impedance phase differences of each pair of charging modules in the matrix with a preset impedance difference threshold, identify the charging module pairs with impedance phase differences exceeding the impedance difference threshold and mark them as target pairs to be disconnected; A disconnection command is then generated and sent to the relay array, controlling the relay array to cut off the direct electrical connection between the marked charging module pairs; Then, based on the dynamic impedance difference matrix, a hierarchical clustering method is used to group all charging modules according to the impedance phase difference, so that the average value of the impedance phase difference within the group is minimized, and the charging modules in each group are connected in parallel through a relay array to form a parallel charging module group; Finally, the average impedance phase of the charging modules in each parallel charging module group is calculated, the impedance phase deviation of each charging module from the average value is determined, and a compensation signal in the opposite direction of the deviation is generated and superimposed on the control signal through a signal injection circuit to balance the current distribution within the group.

6. The implementation method for dynamically regulating battery charging in a battery swapping station control system according to claim 5, characterized in that, Step S4 includes the following contents: Obtain topological information of the parallel charging module group and construct an electrical topological diagram to record the position of each charging module in the group and the connection relationship with adjacent charging modules; Then the total number of charging modules in the parallel charging module group is calculated, and the initial phase offset is determined by dividing the 360 ​​degrees of the complete cycle by the total number of charging modules, and then the phase offset is adjusted based on the deviation of the impedance phase of each charging module from the average value in the group and the dynamic correction factor; The adjusted phase offset is then divided by three hundred and sixty degrees and multiplied by the pulse width modulation waveform switching period to be converted into a time delay value, and the pulse width modulation waveform start time is adjusted through the programmable delay unit in each charging module driving circuit to achieve a staggered distribution of the switching action; Finally, the ripple coefficient of the output current of the parallel charging module group is monitored in real time. It is calculated by dividing the difference between the peak value and the valley value by the average value. If the ripple coefficient exceeds the preset threshold, the dynamic correction factor is fine-tuned and the phase offset is readjusted until the ripple coefficient meets the requirements to ensure current balance and ripple cancellation effects.

7. The implementation method for dynamically regulating battery charging of an electric vehicle battery swapping station control system according to claim 6, characterized in that Step S5 includes the following contents: The parallel charging module group is regarded as an integral unit, which is defined as a virtual charging unit. The equivalent power output capacity of the virtual charging unit is calculated by accumulating the rated power of all charging modules in the parallel charging module group. Then, by real-time monitoring of the bus voltage changes within a preset time period, the voltage fluctuation rate of the bus voltage is calculated, and the current charging power demand is derived based on the pre-established correspondence between the voltage fluctuation rate and the charging power demand; Then, a power weight coefficient is assigned to each virtual charging unit, and all power weight coefficients are adjusted through iterative optimization to minimize the comprehensive evaluation objectives of the balance deviation and efficiency loss of the power weight coefficient allocation, while ensuring that the sum of the power outputs of all virtual charging units is equal to the derived charging power demand; Finally, the optimized power weight coefficient is applied to the power control of the virtual charging unit.

8. The implementation method for dynamically regulating battery charging of an exchange substation control system according to claim 7, characterized in that, Step S6 includes the following contents: The temperature data is collected in real time by multiple temperature sensors installed on the surface of each charging module radiator, and the discrete temperature data is converted into a continuous three-dimensional temperature field model using an interpolation algorithm. The temperature gradient of the three-dimensional temperature field is calculated to identify areas with drastic temperature changes. Then, calculate the average temperature and the maximum temperature gradient in the area where each charging module is located, and compare them with the preset temperature anomaly threshold and gradient anomaly threshold. If the average temperature exceeds the temperature anomaly threshold or the maximum temperature gradient exceeds the gradient anomaly threshold, it is determined that there is a local temperature anomaly in the corresponding charging module; Next, generate a topology secondary reconstruction instruction for the charging module with local temperature anomaly, and send the topology secondary reconstruction instruction to the control unit; Finally, according to the topology secondary reconstruction instruction, disconnect the charging module with local temperature anomaly from the parallel charging module group and switch it to the independent power supply branch through the control relay array. At the same time, select a spare charging module with impedance phase characteristics similar to the original charging module from the spare charging module pool and connect it to the parallel charging module group, and recalculate the PWM waveform phase offset to ensure ripple cancellation and current balance.

Images