An Online Detection Method for the AC Impedance of the Battery of a Charging Power Supply

Abstract

The present invention discloses an online detection method for the AC impedance of a battery of a charging power supply. During the charging process of the battery with a steady-state charging current, the PWM modulation of the power supply is switched to multi-objective optimized RPWM modulation for online AC impedance detection. The carrier frequency of the multi-objective optimized RPWM modulation is obtained through optimized design, and the optimization design objective is set to make the output current spectrum of the charging system meet the requirements of the current harmonic content in the frequency band of the measured AC impedance during perturbation. The carrier frequency under multi-objective optimized RPWM modulation changes in the form of a carrier frequency sequence, and the sequence period depends on the impedance measurement frequency band requirements. The present invention realizes wide-band battery impedance detection, accurately detects the impedance in real time, does not require additional hardware devices, saves costs, only needs to change the carrier frequency, and its implementation method is simple and has little impact on the battery charging system.

Description

Technical Field

[0001] The present invention relates to a charging power supply, and more specifically to an online detection method for the battery AC impedance of a charging power supply. Background Art

[0002] In recent years, the energy crisis and environmental pollution problems have become increasingly serious. Worldwide, more and more people are concerned about the development and application of new energy. In the new energy industry, batteries, as energy storage devices and power supply sources, have outstanding advantages in aspects such as specific energy, volume, lifespan, and environmental friendliness. Battery parameters reflect their performance and health status, etc. However, the internal state of the battery cannot be directly detected, which will affect the safe and stable operation of the battery.

[0003] Electrochemical Impedance Spectroscopy (EIS), as a non-destructive parameter measurement and effective battery kinetic behavior measurement method, can decouple processes such as internal interface reactions, charge transfer, and ion diffusion in the battery at different frequency bands in the form of complex impedance to obtain important parameters for battery modeling and predicting state information such as SOC, SOH, and SOF.

[0004] Battery EIS is also known as AC impedance spectroscopy. The main methods for AC impedance detection can be divided into offline detection and online detection. Offline detection requires special equipment, has a high cost, and cannot reflect the information of internal changes in the battery in real time. Online detection can reflect the internal characteristics of the battery in real time. Usually, the online detection of battery AC impedance is carried out during the charging and discharging process of the battery. A small signal perturbation is injected into the reference value of the charging power supply control loop, and the battery AC impedance is obtained by sampling the voltage response and charging current response at both ends of the battery. The perturbation signals include single-frequency sine perturbation, multi-frequency sine perturbation, square wave perturbation, and pulse perturbation, etc. Injecting a single-frequency sine perturbation can ensure the impedance detection accuracy, but to understand the internal characteristics of the battery, impedance information at multiple frequencies is required, and injecting single-frequency perturbations sequentially will take a lot of time. While injecting perturbations containing multiple frequencies can reduce the detection time, the impedance detection accuracy will also decrease. At the same time, when injecting perturbations for online detection in the controller section, the impedance detection frequency band will be restricted by the controller bandwidth. The existing impedance detection method based on RPWM can obtain high-frequency impedance information. However, its carrier frequency is randomly generated, and the distribution of corresponding current harmonics in each frequency band is uncontrollable, affecting the impedance detection accuracy. Summary of the Invention

[0005] In order to avoid the deficiencies of the above-mentioned prior art, the present invention provides an online detection method for the battery AC impedance of a charging power supply that does not require additional equipment and can effectively reduce costs. The multi-objective optimization RPWM modulation method is selected to effectively excite perturbations and accurately detect the battery AC impedance in real time.

[0006] The present invention adopts the following technical solutions to solve the technical problems:

[0007] The battery AC impedance on-line detection method of the charging power supply of the present invention is characterized in that: during the charging process of the battery with a steady-state charging current, the PWM modulation of the power supply is switched to multi-objective optimized RPWM modulation for on-line AC impedance detection; the carrier frequency of the multi-objective optimized RPWM modulation is obtained through optimized design, and the optimization design objective is set to make the output current spectrum of the charging system meet the current harmonic requirement of the measured AC impedance frequency band during perturbation; the carrier frequency under multi-objective optimized RPWM modulation changes in the form of a carrier frequency sequence, and the sequence period depends on the impedance measurement frequency band requirement.

[0008] The battery AC impedance on-line detection method of the charging power supply of the present invention is also characterized by being carried out according to the following steps:

[0009] Step 1: Generate a carrier signal

[0010] Constrained by the characteristic quantities of the carrier frequency sequence, an optimized carrier frequency sequence f s is obtained through multi-objective optimized design. The optimized carrier frequency sequence f s is a carrier frequency sequence that makes the output current spectrum of the charging system meet the current harmonic requirement of the measured AC impedance frequency band during perturbation; a carrier signal is generated from the optimized carrier frequency sequence f s ;

[0011] The characteristic quantities of the carrier frequency sequence include: the carrier frequency sequence period T c , the upper limit of the carrier frequency f max , the lower limit of the carrier frequency f min and the number of carrier frequencies N in the carrier frequency sequence;

[0012] Step 2: Perform pulse width modulation on the modulation wave signal and the carrier signal to obtain a pulse signal for driving the switching tube of the charging power supply during perturbation;

[0013] Step 3: During the charging process of the battery with a steady-state charging current, switch the switching tube of the charging power supply from PWM modulation to multi-objective optimized RPWM modulation to implement perturbation; sample the voltage across the battery and the charging current during perturbation, and calculate the battery AC impedance value through FFT analysis of the sampled voltage across the battery and the charging current during perturbation to realize on-line detection of the battery AC impedance.

[0014] The battery AC impedance on-line detection method of the charging power supply of the present invention is also characterized in that:

[0015] Determine the carrier frequency sequence period T c, 1 / T c is the minimum frequency required for battery AC impedance detection.

[0016] The characteristics of the on-line battery AC impedance detection method for the charging power supply of the present invention also lie in:

[0017] The carrier frequency sequence f s satisfies: f b < f min ≤ f s ≤ f max < f init

[0018] where: f b is the controller bandwidth, and f init is the carrier frequency of PWM modulation.

[0019] The characteristics of the on-line battery AC impedance detection method for the charging power supply of the present invention also lie in:

[0020] The number of carrier frequencies N in the carrier frequency sequence is set to: N = T c × f min .

[0021] The characteristics of the on-line battery AC impedance detection method for the charging power supply of the present invention also lie in:

[0022] Set the optimization objectives of the optimized carrier frequency sequence according to the objective functions ζ1 and ζ2 shown in Equation (1);

[0023]

[0024] In Equation (1):

[0025] Let m represent the harmonic order. m max is the harmonic order of the harmonic with a frequency of f max , and m max = f max × T c ;

[0026] T is the selected THD threshold, and T = 8%;

[0027] H1 is the harmonic amplitude of the harmonic with a frequency of 1 / T c ;

[0028] H s is the selected harmonic amplitude, and H s = 0.3%H1;

[0029] H m is the harmonic amplitude of the mth harmonic, and there is:

[0030]

[0031] In Equation (2):

[0032] a m and b m are the coefficients of the Fourier series when performing FFT on the charging current, and there is:

[0033]

[0034]

[0035] Let f y represent the y-th frequency in the carrier frequency sequence, y = 1, 2, 3..., N; p y = 2πf y / T c ;

[0036] Let f i represent the i-th frequency in the carrier frequency sequence, i = 0, 1, 2..., y - 1; p i = 2πf i / T c , and define p0 = 0;

[0037] d is the amplitude of the modulation wave. When m = 1, H1 is obtained from Equation (2).

[0038] Under multi-objective optimized RPWM modulation, the frequency sequence that minimizes the objective function shown in Equation (1) for the output current spectrum is the optimized carrier frequency sequence.

[0039] The characteristics of the online detection method for the battery AC impedance of the charging power supply of the present invention also lie in:

[0040] The steady-state charging current refers to the current change amount △I bat satisfying Equation (3):

[0041] △I bat <10%I bat (3)

[0042] In Equation (3): I bat is the battery charging current.

[0043] Compared with the prior art, the beneficial effects of the present invention are reflected in:

[0044] 1. The present invention does not require additional hardware devices. For the carrier frequency under multi-objective optimized design RPWM modulation, after perturbation, not only can the charging current spectrum of the charging system meet the requirements of battery impedance measurement accuracy, but also the impact on the normal charging process of the battery is greatly reduced;

[0045] 2. The present invention realizes controllability of the output current spectrum of the charging system, designs the cycle of the carrier frequency sequence according to the required frequency band for battery AC impedance detection, and detects the broadband impedance in real time. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 is the block diagram of the on-line detection of battery AC impedance in the present invention;

[0047] Figure 2 is the schematic diagram of the charging power supply circuit with the function of on-line detection of battery AC impedance in the present invention;

[0048] Figure 3 is the carrier signal with variable frequency generated in the present invention;

[0049] Figure 4 is the FFT result of the charging current during multi-objective optimized RPWM modulation in the present invention;

[0050] Figure 5a and Figure 5b are respectively the amplitude-frequency characteristic curve and phase-frequency characteristic curve of the impedance during multi-objective optimized RPWM modulation in the present invention;

[0051] Figure 6 is the Nyquist curve of the battery impedance in the present invention;

[0052] Figure 7 is the FFT result of the charging current during non-optimized RPWM modulation in the present invention;

[0053] Figure 8a and Figure 8b are respectively the amplitude-frequency characteristic curve and phase-frequency characteristic curve of the impedance during non-optimized RPWM modulation in the present invention;

[0054] Table 1 shows the given parameters of the DC-DC converter in the embodiment.

[0055] Table 2 shows the parameter selection of the multi-objective optimized RPWM in the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0056] Referring to Figure 1 and Figure 2 , in this embodiment, the method for on-line detection of battery AC impedance of the charging power supply is that during the process of charging the battery with a steady-state charging current, the PWM modulation of the power supply is switched to multi-objective optimized RPWM modulation for on-line detection of AC impedance; the carrier frequency of the multi-objective optimized RPWM modulation is obtained through optimized design, and the optimization design goal is set to make the output current spectrum of the charging system meet the requirement of the current harmonic content in the frequency band of the measured AC impedance during perturbation; the carrier frequency under multi-objective optimized RPWM modulation changes in the form of a carrier frequency sequence, and the sequence period depends on the impedance measurement frequency band requirement.

[0057] In this embodiment, the on-line detection method for the battery AC impedance of the charging power supply is carried out according to the following steps:

[0058] Step 1: Generate a carrier signal

[0059] Constrained by the characteristic quantities of the carrier frequency sequence, an optimized carrier frequency sequence f is obtained through multi-objective optimization design s , the optimized carrier frequency sequence f s is the carrier frequency sequence that enables the current spectrum output by the charging system to meet the requirements of the current harmonic content in the frequency band of the measured AC impedance during perturbation; the carrier signal is generated from the optimized carrier frequency sequence f s ;

[0060] The characteristic quantities of the carrier frequency sequence include: the period T of the carrier frequency sequence c , the upper limit f of the carrier frequency max , the lower limit f of the carrier frequency min and the number N of carrier frequencies in the carrier frequency sequence;

[0061] Step 2: Perform pulse width modulation on the modulation wave signal and the carrier signal to obtain a pulse signal for driving the switching tube of the charging power supply during perturbation;

[0062] Step 3: During the process of charging the battery with a steady-state charging current, switch the switching tube of the charging power supply from PWM modulation to multi-objective optimized RPWM modulation to implement perturbation; sample the voltage across the battery and the charging current during perturbation, and calculate the battery AC impedance value through FFT analysis of the sampled voltage across the battery and the charging current during perturbation, so as to realize the on-line detection of the battery AC impedance.

[0063] In specific implementation, the corresponding technical measures also include:

[0064] Determine the period T of the carrier frequency sequence according to the frequency band of the measured AC impedance c . By using Fourier transform to analyze the current output by the charging system, it can be known that the current harmonics are mainly distributed at integer multiples of 1 / T c . In order to meet the requirements of the battery AC impedance measurement frequency band, 1 / T c should be the minimum frequency required for battery AC impedance detection.

[0065] The carrier frequency sequence f s satisfies: f b < f min ≤ f s ≤ f max < f init

[0066] where: f b is the controller bandwidth, f initis the carrier frequency of PWM modulation.

[0067] The lower limit f of the carrier frequency in the carrier frequency sequence min should be greater than the controller bandwidth f b , and at the same time, the larger the switching frequency, the greater the loss of the charging power supply. Therefore, the upper limit f of the carrier frequency max does not exceed the carrier frequency f of PWM modulation init ;

[0068] The controller bandwidth f b is calculated according to the following formula:

[0069]

[0070] where: L is the inductor of the DC-DC converter in this embodiment, R is the load, U d is the input power supply, P is the proportional coefficient of the PI controller, I is the integral coefficient of the PI controller, and d is the amplitude of the modulation wave.

[0071] Set the number N of carrier frequencies in the carrier frequency sequence to: N = T c ×f min .

[0072] Set the optimization objectives for optimizing the carrier frequency sequence according to the objective functions ζ1 and ζ2 shown in Equation (1);

[0073]

[0074] ζ1 is used to characterize that the harmonic distribution of the output current of the charging system is uniform to meet the accuracy requirements of battery AC impedance detection;

[0075] ζ2 is used to characterize that the total harmonic distortion rate THD of the output current of the charging system does not exceed the threshold T, so as to meet the requirement that the disturbance has little impact on the normal charging of the battery.

[0076] In Equation (1):

[0077] Let m represent the harmonic order, m max is the harmonic order of the frequency f max , m max = f max ×T c ;

[0078] T is the selected THD threshold, T = 8%;

[0079] H1 is the harmonic amplitude of the frequency 1 / T c ;

[0080] H s is the selected harmonic amplitude, H s = 0.3%H1;

[0081] H m is the harmonic amplitude of the m-th harmonic, and there is:

[0082]

[0083] In formula (2):

[0084] a m and b m are the coefficients of the Fourier series when performing FFT on the charging current, and there is:

[0085]

[0086]

[0087] Let f y represent the y-th frequency in the carrier frequency sequence, y = 1, 2, 3..., N; p y = 2πf y / T c ;

[0088] Let f i represent the i-th frequency in the carrier frequency sequence, i = 0, 1, 2..., y - 1; p i = 2πf i / T c , and define p0 = 0;

[0089] When m = 1, H1 is obtained from formula (2);

[0090] Under the multi-objective optimized RPWM modulation, the frequency sequence that makes the objective function shown in formula (1) the smallest for the output current spectrum is the optimized carrier frequency sequence, and intelligent algorithms such as genetic algorithms can be used to solve the optimal carrier frequency sequence that satisfies the objective function.

[0091] The steady-state charging current refers to the current change amount △I bat flowing through the battery and satisfying formula (3):

[0092] △I bat <10%I bat (3)

[0093] In formula (3): I bat is the battery charging current.

[0094] To verify the effectiveness of the method of the present invention, a battery power circuit model is built in Matlab / simulink. Usually, the battery power supply is a DC-DC converter connected to the battery, and the given parameters of the DC-DC converter are shown in Table 1; first, the battery impedance detection frequency band is selected as 12.5 Hz - 4 kHz, and the other parameters of the multi-objective optimized RPWM are shown in Table 2.

[0095] The carrier signal of the multi-objective optimized RPWM modulation is as Figure 3 shown. Figure 4 Shown is the FFT result of the output current of the charging system when the multi-objective optimized RPWM modulation is adopted in the DC-DC converter. In this embodiment, the charging current waveform is optimized by designing the switching frequency in the RPWM modulation. The charging current spectrum is dispersed and the amplitudes of each harmonic are uniform. At the same time, the influence of the disturbance on the charging system is small, and the THD of the charging current is only 5.46%.

[0096] Figure 5a Shown is the theoretical amplitude-frequency characteristic curve and the measured point plot of the battery AC impedance under the multi-objective optimized RPWM modulation. Figure 5b Shown is the theoretical phase-frequency characteristic curve and the measured point plot of the battery AC impedance under the multi-objective optimized RPWM modulation. The measured points of the impedance are distributed near the theoretical curve, that is, the impedance detection accuracy of the method in this embodiment is high. Converting the measured Bode point plot of the AC impedance into a Nyquist plot with the real part of the impedance on the horizontal axis and the negative imaginary part of the impedance on the vertical axis, it is Figure 6 the battery electrochemical impedance spectrum shown. It can be seen that the middle frequency band is a semi-circle and the high frequency band is resistive-inductive, being a straight line parallel to the imaginary axis.

[0097] Figure 7 Shown is the FFT result of the output current of the charging system when the RPWM modulation is not optimized, and the THD of the charging current is 21.07%. Comparing Figure 4 and Figure 7 the FFT results of the output current, it can be known that the THD of the charging current under the multi-objective optimized RPWM modulation in this embodiment is smaller and the distribution of each harmonic in the spectrum is more uniform. Figure 8a is the theoretical amplitude-frequency characteristic curve and the measured point plot of the battery AC impedance under the non-optimized RPWM modulation. Figure 8b is the theoretical phase-frequency characteristic curve and the measured point plot of the battery AC impedance under the non-optimized RPWM modulation. Comparing Figure 5a and Figure 5b and Figure 8a and Figure 8b it can be known that the battery impedance detection result under the multi-objective optimized RPWM modulation in this embodiment is more accurate than that under the non-optimized RPWM. The maximum error of the detected impedance amplitude under the multi-objective optimized RPWM modulation is about 8 mΩ, while the maximum error of the detected impedance amplitude under the non-optimized RPWM modulation is about 21 mΩ. Therefore, this embodiment has an advantage in the battery AC impedance detection.

[0098] Table 1 shows the given parameters of the DC-DC converter in this embodiment

[0099] Parameter description Parameter value <![CDATA[Input power supply U d (V)]]> 40 Inductance L (mH) 10 Output capacitance C (mF) 10 Carrier frequency of PWM modulation (Hz) 10K P 0.05 I 1

[0100] Table 2 shows the parameter selection of multi-objective optimized RPWM in this embodiment

[0101] Parameter description Parameter value Upper limit of carrier frequency of multi-objective optimized PWM (Hz) 10K Lower limit of carrier frequency of multi-objective optimized PWM (Hz) 2K Carrier frequency sequence period (s) 0.08

[0102] The method of the present invention makes up for the deficiencies of the existing online battery AC impedance detection technology, which is limited by the controller bandwidth and has low detection accuracy. It has the advantages of simple and easy perturbation application method, high impedance value detection accuracy, and little impact on the battery charging power supply. The effectiveness of the method of the present invention is verified by simulation.

Claims

1. An online detection method for the battery AC impedance of a charging power supply, characterized in that: at During the charging process of the battery with a steady-state charging current, the PWM modulation of the power supply is switched to multi-objective optimized RPWM modulation for on-line detection of the AC impedance; the carrier frequency of the multi-objective optimized RPWM modulation is obtained through optimized design, and the optimization design objective is set to make the current spectrum output by the charging system meet the requirement of the current harmonic content in the frequency band of the measured AC impedance during perturbation; the carrier frequency under multi-objective optimized RPWM modulation changes in the form of a carrier frequency sequence, and the sequence period depends on the impedance measurement frequency band requirement; the on-line detection method is carried out according to the following steps: Step 1, generate a carrier signal Constrained by the characteristic quantity of the carrier frequency sequence, an optimized carrier frequency sequence f is obtained through multi-objective optimization design s , the optimized carrier frequency sequence f s is a carrier frequency sequence that enables the output current spectrum of the charging system to meet the requirements of the current harmonic quantity in the frequency band of the measured AC impedance during perturbation; a carrier signal is generated from the optimized carrier frequency sequence f s ​ The characteristic quantities of the carrier frequency sequence include: the carrier frequency sequence period T c , the upper limit of the carrier frequency f max , the lower limit of the carrier frequency f min and the number of carrier frequencies N in the carrier frequency sequence; Step 2, perform pulse width modulation on the modulation wave signal and the carrier signal to obtain a pulse signal for driving the charging power switch tube during perturbation; Step 3, during the charging process of the battery with a steady-state charging current, switch the charging power switch tube from PWM modulation to multi-objective optimized RPWM modulation to implement perturbation; sample the voltage across the battery and the charging current during perturbation, and calculate the AC impedance value of the battery through FFT analysis for the sampled voltage across the battery and the charging current during perturbation, so as to realize the on-line detection of the battery AC impedance.

2. The on-line detection method for the battery AC impedance of the charging power supply according to claim 1, characterized in that: Determine the carrier frequency sequence period T according to the frequency band of the measured AC impedance c , 1 / T c is the minimum frequency required for battery AC impedance detection.

3. The on-line detection method for the battery AC impedance of the charging power supply according to claim 1, characterized in that: Carrier frequency sequence f s Satisfies: f b < f min ≤ f s ≤ f max < f init Where: f b is the controller bandwidth, and f init is the carrier frequency of PWM modulation.

4. The on-line detection method for the battery AC impedance of the charging power supply according to claim 1, characterized in that: Set the number of carrier frequencies N in the carrier frequency sequence to: N = T c × f min .

5. The on-line detection method for the battery AC impedance of the charging power supply according to claim 1, characterized in that: Set the optimization objective of the optimized carrier frequency sequence according to the objective functions ζ1 and ζ2 shown in formula (1); In formula (1): Let \(m\) denote the harmonic order, \(m\) max is the harmonic order with a frequency of \(f\) max , and \(m\) max = \(f\) max × \(T\) c ; T is the selected THD threshold, T = 8%; H1 is the harmonic amplitude with a frequency of 1 / T c ; H s is the selected harmonic amplitude, H s = 0.3% H1; H m is the harmonic amplitude of the m-th harmonic, and there is: In formula (2): a m and b m are the coefficients of the Fourier series when performing FFT on the charging current, and there is: With f y represents the y-th frequency in the carrier frequency sequence, where y = 1, 2, 3..., N; p y = 2πf y / T c ; Let f i represent the i-th frequency in the carrier frequency sequence, where i = 0, 1, 2..., y - 1; p i = 2πf i / T c , and define p0 = 0; d is the modulation wave amplitude, and when m = 1, H1 is obtained from formula (2); The frequency sequence that minimizes the objective function shown in formula (1) in the output current spectrum under multi-objective optimized RPWM modulation is the optimized carrier frequency sequence.

6. The on-line detection method for the battery AC impedance of the charging power supply according to claim 1, characterized in that: The steady-state charging current refers to the current change amount △I flowing through the battery bat satisfies Equation (3): △I bat <10%I bat (3) In Equation (3): I bat is the battery charging current.

Images