Lithium battery ohmic impedance online measurement method based on charging condition

Abstract

The invention discloses a lithium battery ohmic impedance on-line measurement method based on a charging working condition, and the method comprises the steps: carrying out the charging current request of a time segment: requesting different charging currents according to the time segment when a BMS is in a fast charging constant current stage; data acquisition and queue management: acquiring voltage and current data of the battery cells in real time, and respectively storing the voltage and current data in a voltage sequence queue and a current sequence queue; performing time-phased FFT analysis and impedance calculation by using an impedance weighted average method: performing FFT analysis on data of each frequency band, calculating alternating-current impedance under a target frequency, and taking a real part of the alternating-current impedance as ohmic impedance; and obtaining ohmic impedance by adopting an impedance weighted average method: carrying out weighted average on the ohmic impedance of each frequency band to obtain final ohmic impedance. According to the method, the ohmic impedance of the lithium battery can be dynamically monitored on line, the precision is high, and additional hardware does not need to be added.

Description

Technical Field

[0001] The present invention belongs to the technical field of online measurement of lithium battery impedance, and specifically refers to an online measurement method of lithium battery ohmic impedance under charging conditions. Background Art

[0002] The ohmic impedance of lithium batteries is an important parameter for evaluating the battery health state (SOH), battery capacity (SOC) and performance based on the equivalent circuit model. At present, the commonly used ohmic impedance measurement methods in BMS mainly include offline methods and online methods. The offline method includes the hybrid pulse power characteristic curve method (HPPC), which calculates the ohmic impedance by applying a pulse current and measuring the voltage response. This method requires complex experimental equipment, a long experimental cycle, and cannot reflect the battery status in real time. Electrochemical impedance spectroscopy (EIS) measures the impedance spectrum through a frequency domain excitation signal, but requires special equipment, is costly and is not suitable for actual vehicle conditions. The online method includes the Ohm's law method, which calculates the ohmic impedance by measuring the voltage change ΔU and the current change ΔI. This method requires additional hardware circuits, has high requirements for the accuracy and synchronization of voltage and current sensors, and the estimation results are prone to divergence under dynamic conditions. Therefore, the prior art has the following problems: the offline method cannot measure in real time and has high cost; the online method has low accuracy under dynamic conditions and high hardware requirements. Summary of the invention

[0003] The object of the present invention is to provide an online measurement method for the ohmic impedance of a lithium battery under charging conditions, which can dynamically monitor the ohmic impedance of a lithium battery online with high accuracy without adding additional hardware.

[0004] To achieve the above object, the present invention provides an online measurement method for the ohmic impedance of a lithium battery under charging conditions, comprising the following steps:

[0005] Step 1: Time-segmented charging current request: When the BMS is in the constant current stage of fast charging, it requests charging currents of different sizes in time segments;

[0006] Step 2: Data collection and queue management: collect the voltage and current data of the battery cells in real time and store them in the voltage sequence queue and current sequence queue respectively;

[0007] Step 3: Impedance weighted average method for time-segment FFT analysis and impedance calculation: Perform FFT analysis on the data of each frequency segment, calculate the AC impedance at the target frequency, and take its real part as the ohmic impedance;

[0008] Step 4: Obtain ohmic impedance using impedance weighted average method: Perform weighted average on the ohmic impedance of each frequency band to obtain the final ohmic impedance.

[0009] As a further solution of the present invention: in step 1, a charging current request is performed according to the following steps:

[0010] 1) 0.1Hz frequency segment (first 40 seconds):

[0011] Time period 1: request current value reqCurr1, lasting t1 = 5 seconds;

[0012] Time period 2: request current value reqCurr2, lasting t2 = 5 seconds;

[0013] Repeat the above time period 4 times, for a total of 40 seconds;

[0014] 2) 0.2Hz frequency segment (second 40 seconds):

[0015] Time period 3: request current value reqCurr1, lasting t3 = 2.5 seconds;

[0016] Time period 4: request current value reqCurr2, lasting t4 = 2.5 seconds;

[0017] Repeat the above time period 8 times, for a total of 40 seconds;

[0018] 3) 0.4Hz frequency segment (the third 40 seconds):

[0019] Time period 5: request current value reqCurr1, lasting t5 = 1.25 seconds;

[0020] Time period 6: request current value reqCurr2, lasting t6 = 1.25 seconds;

[0021] Repeat the above time period 16 times, for a total of 40 seconds;

[0022] The above frequency bands are switched cyclically until charging enters the constant voltage stage; reqCurr1 and reqCurr2 are close in size but not equal, and |reqCurr1-reqCurr2| is not greater than 20A.

[0023] As a further solution of the present invention: the step 2 specifically includes the following steps: the queue adopts the first-in-first-out (FIFO) principle. If the queue is full, the head element is removed from the queue first and then inserted into the tail. The queue size is determined by the sampling frequency F. s and sampling window time T win The queue length is calculated as: F s ·T win .

[0024] As a further solution of the present invention: in step 3, within every 40 seconds, the following operations are performed:

[0025] 1) Data extraction: Extract the voltage and current data of the corresponding frequency segment according to the time period of the current request;

[0026] 2) FFT analysis: Calculate the average current value I_avg for the current data, subtract I_avg from each current value and multiply it by the Hanning window, perform fast Fourier transform (FFT) to obtain the current frequency domain signal I(f); directly multiply the voltage data by the Hanning window, perform fast Fourier transform, and obtain the voltage frequency domain signal U(f);

[0027] 3) Impedance calculation:

[0028] Calculate the AC impedance Z(f) at the target frequency:

[0029] Z(f)=U(f) / I(f)

[0030] Get the impedance Z(f_target) at the target frequency f_target, where f_target is the center frequency of the current frequency segment (such as 0.1Hz, 0.2Hz, 0.4Hz);

[0031] The ohmic impedance R_ohm is the real part of Z(f_target):

[0032] R_ohm=Re(Z(f_target)).

[0033] As a further solution of the present invention: in the step 4, the weighted average method specifically includes the following steps:

[0034] 1) Weight distribution:

[0035] The weight of the 0.1Hz frequency segment is 50%;

[0036] The weight of the 0.2Hz frequency segment is 30%;

[0037] The weight of the 0.4Hz frequency segment is 20%;

[0038] 2) Weighted average calculation: The ohmic impedance of each frequency band is weighted averaged to obtain the final ohmic impedance R_ohm:

[0039] R_ohm=w1*R_ohm0.1Hz+w2*R_ohm0.2Hz+w4*R_ohm0.4Hz

[0040] Among them, w1+w2+w3=1.

[0041] Compared with the prior art, the present invention updates the ohmic impedance in real time during the charging process, without the need for terminal charging or offline experiments. The time-segmented current request design avoids the addition of additional voltage or current excitation hardware. The FFT analysis and Hanning window processing are used to effectively suppress noise and improve measurement accuracy. The impedance information of each frequency band is comprehensively considered through a weighted evaluation algorithm to further improve measurement accuracy and robustness. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a flow chart of the invention. DETAILED DESCRIPTION

[0043] The present invention will be further described below in conjunction with the accompanying drawings.

[0044] like Figure 1 As shown, a method for online measurement of ohmic impedance of a lithium battery under charging conditions comprises the following steps:

[0045] Step 1: Time-segmented charging current request: When the BMS is in the constant current stage of fast charging, it requests charging currents of different sizes in time segments;

[0046] Step 2: Data collection and queue management: collect the voltage and current data of the battery cells in real time and store them in the voltage sequence queue and current sequence queue respectively;

[0047] Step 3: Impedance weighted average method for time-segment FFT analysis and impedance calculation: Perform FFT analysis on the data of each frequency segment, calculate the AC impedance at the target frequency, and take its real part as the ohmic impedance;

[0048] Step 4: Impedance weighted average method: perform weighted average of the ohmic impedance of each frequency band to obtain the final ohmic impedance.

[0049] In step 1, follow the steps below to request a charging current:

[0050] 1) 0.1Hz frequency segment (first 40 seconds):

[0051] Time period 1: Request current value reqCurr1, lasting t1 = 5 seconds.

[0052] Time period 2: request current value reqCurr2, lasting t2=5 seconds.

[0053] Repeat the above time period 4 times, for a total of 40 seconds.

[0054] 2) 0.2Hz frequency segment (second 40 seconds):

[0055] Time period 3: request current value reqCurr1, lasting t3 = 2.5 seconds.

[0056] Time period 4: request current value reqCurr2, lasting t4=2.5 seconds.

[0057] Repeat the above time period 8 times, for a total of 40 seconds.

[0058] 3) 0.4Hz frequency segment (the third 40 seconds):

[0059] Time period 5: request current value reqCurr1, lasting t5=1.25 seconds.

[0060] Time period 6: request current value reqCurr2, lasting t6 = 1.25 seconds.

[0061] Repeat the above time period 16 times, for a total of 40 seconds.

[0062] The above frequency bands are switched cyclically until charging enters the constant voltage stage. Considering the current change rate of the fast charging pile, reqCurr1 and reqCurr2 are close but not equal, and |reqCurr1-reqCurr2| is not greater than 20A, for example, reqCurr1=100A, reqCurr2=105A, and the difference ΔI=5A.

[0063] Step 2 specifically includes the following steps: The queue adopts the first-in-first-out (FIFO) principle. If the queue is full, the head element is removed from the queue first and then inserted into the tail. The queue size is determined by the sampling frequency F s and sampling window time T win The queue length is calculated as: F s ·T win Among them, if F s =10Hz, T win =40s, the queue size is 400.

[0064] In step 3, perform the following operations every 40 seconds:

[0065] 1) Data extraction: According to the time period of the current request, the voltage and current data of the corresponding frequency segment are extracted.

[0066] 2) FFT analysis:

[0067] The average current value I_avg is calculated for the current data, and each current value is subtracted from I_avg and multiplied by the Hanning window, and a fast Fourier transform (FFT) is performed to obtain the current frequency domain signal I(f).

[0068] Current calculation:

[0069] I(k)=I S (k)-I -avg ,k=0,1,......,N-1

[0070] Among them I s (k) is the current value of the kth sampling current, I -avg is the arithmetic mean of the Nth sample, and its calculation formula is:

[0071]

[0072] The calculation formula of the Hanning window is:

[0073] I win (k)=I(k)·W(k), k=0,1,...N-1.

[0074] Where W(k) represents the Hanning window function, which is expressed as follows:

[0075]

[0076] After adding the Hanning window, perform FFT on the signal: First, perform DFT operation:

[0077]

[0078] Combine the above DFT with odd-even grouping, recursive operation and merging operation to quickly calculate X(k).

[0079] I(f)=X([f·T win ])

[0080] Where f is the frequency and Twin is the sampling window time.

[0081] Similarly, the voltage data is directly multiplied by the Hanning window and fast Fourier transformed to obtain the voltage frequency domain signal U(f).

[0082] 3) Impedance calculation:

[0083] Calculate the AC impedance Z(f) at the target frequency:

[0084] Z(f)=U(f) / I(f)

[0085] The impedance Z(f_target) at the target frequency f_target is obtained, where f_target is the center frequency of the current frequency segment (eg, 0.1 Hz, 0.2 Hz, 0.4 Hz).

[0086] The ohmic impedance R_ohm is the real part of Z(f_target):

[0087] R_ohm=Re(Z(f_target))

[0088] In step 4, in order to further improve the accuracy of ohmic impedance measurement, the impedance weighted average method is used to weight the impedance of each frequency band. The specific steps are as follows:

[0089] 1) Weight distribution:

[0090] The 0.1 Hz frequency segment has a weight of 50%.

[0091] The 0.2 Hz frequency segment has a weight of 30%.

[0092] The 0.4 Hz frequency segment has a weight of 20%.

[0093] 2) Weighted average calculation: The ohmic impedance of each frequency band is weighted averaged to obtain the final ohmic impedance R_ohm:

[0094] R_ohm=w1*R_ohm0.1Hz+w2*R_ohm0.2Hz+w4*R_ohm0.4Hz

[0095] Among them, w1+w2+w3=1.

[0096] Repeat steps 1 to 4 until the BMS enters constant voltage charging mode or exits charging.

[0097] Example: Step 1: Time-segmented charging current request:

[0098] When the BMS is in the constant current charging stage of fast charging, the charging current is requested according to the following design:

[0099] 0.1Hz frequency segment (first 40 seconds):

[0100] Time period 1: requested current value reqCurr1 = 100 A, lasting t1 = 5 seconds.

[0101] Time period 2: requested current value reqCurr2 = 110 A, lasting t2 = 5 seconds.

[0102] Repeat the above time period 4 times, for a total of 40 seconds.

[0103] 0.2Hz frequency segment (second 40 seconds):

[0104] Time period 3: requested current value reqCurr1 = 100 A, lasting t3 = 2.5 seconds.

[0105] Time period 4: requested current value reqCurr2 = 110 A, lasting t4 = 2.5 seconds.

[0106] Repeat the above time period 8 times, for a total of 30 seconds.

[0107] 0.4Hz frequency segment (the third 40 seconds):

[0108] Time period 5: requested current value reqCurr1 = 100 A, lasting t5 = 1.25 seconds.

[0109] Time period 6: requested current value reqCurr2 = 110 A, lasting t6 = 1.25 seconds.

[0110] Repeat the above time period 16 times, for a total of 40 seconds.

[0111] The above frequency bands are switched cyclically until charging enters the constant voltage stage.

[0112] Step 2: Data Collection and Queue Management

[0113] The voltage and current data of the battery cells are collected in real time and stored at the end of the corresponding voltage sequence queue and current sequence queue respectively. The queue size is 400.

[0114] Step 3: Time-segment FFT analysis and impedance calculation

[0115] An FFT analysis is performed every 40 seconds to calculate the impedance in the current frequency band and take its real part as the ohmic impedance.

[0116] Step 4: Impedance Weighted Average Method

[0117] The ohmic impedance of each frequency band is weighted averaged to obtain the final ohmic impedance:

[0118] R_ohm=0.5*R_ohm0.1Hz+0.3*R_ohm0.2Hz+0.2*R_ohm0.4Hz

[0119] Repeat steps 1 to 4 until the BMS enters constant voltage charging mode or exits charging.

Claims

1. A method for online measurement of ohmic impedance of a lithium battery under charging conditions, characterized in that: The following steps are involved: Step 1: Time-segmented charging current request: When the BMS is in the constant current stage of fast charging, it requests charging currents of different sizes in time segments; Step 2: Data collection and queue management: collect the voltage and current data of the battery cells in real time and store them in the voltage sequence queue and current sequence queue respectively; Step 3: Impedance weighted average method for time-segment FFT analysis and impedance calculation: Perform FFT analysis on the data of each frequency segment, calculate the AC impedance at the target frequency, and take its real part as the ohmic impedance; Step 4: Obtain ohmic impedance using impedance weighted average method: Perform weighted average on the ohmic impedance of each frequency band to obtain the final ohmic impedance.

2. The method for online measurement of ohmic impedance of a lithium battery under charging conditions according to claim 1, characterized in that: In step 1, a charging current request is made according to the following steps: 1) 0.1Hz frequency segment (first 40 seconds): Time period 1: request current value reqCurr1, lasting t1 = 5 seconds; Time period 2: request current value reqCurr2, lasting t2 = 5 seconds; Repeat the above time period 4 times, for a total of 40 seconds; 2) 0.2Hz frequency segment (second 40 seconds): Time period 3: request current value reqCurr1, lasting t3 = 2.5 seconds; Time period 4: request current value reqCurr2, lasting t4 = 2.5 seconds; Repeat the above time period 8 times, for a total of 40 seconds; 3) 0.4Hz frequency segment (the third 40 seconds): Time period 5: request current value reqCurr1, lasting t5 = 1.25 seconds; Time period 6: request current value reqCurr2, lasting t6 = 1.25 seconds; Repeat the above time period 16 times, for a total of 40 seconds; The above frequency bands are switched cyclically until charging enters the constant voltage stage; reqCurr1 and reqCurr2 are close in size but not equal, and |reqCurr1-reqCurr2| is not greater than 20A.

3. The method for online measurement of ohmic impedance of a lithium battery under charging conditions according to claim 1, characterized in that: The step 2 specifically includes the following steps: the queue adopts the first-in-first-out (FIFO) principle. If the queue is full, the head element is removed from the queue first and then inserted into the tail. The queue size is determined by the sampling frequency F. s and sampling window time T win The queue length is calculated as: F s ·T win .

4. A method for online measurement of ohmic impedance of a lithium battery under charging conditions according to claim 1, characterized in that: In step 3, within every 40 seconds, the following operations are performed: 1) Data extraction: Extract the voltage and current data of the corresponding frequency segment according to the time period of the current request; 2) FFT analysis: Calculate the average current value I_avg for the current data, subtract I_avg from each current value and multiply it by the Hanning window, perform fast Fourier transform (FFT) to obtain the current frequency domain signal I(f); directly multiply the voltage data by the Hanning window. Perform fast Fourier transform to obtain the voltage frequency domain signal U(f); 3) Impedance calculation: Calculate the AC impedance Z(f) at the target frequency: Z(f)=U(f) / I(f) Get the impedance Z(f_target) at the target frequency f_target, where f_target is the center frequency of the current frequency segment (such as 0.1Hz, 0.2Hz, 0.4Hz); The ohmic impedance R_ohm is the real part of Z(f_target): R_ohm=Re(Z(f_target)).

5. A method for online measurement of ohmic impedance of a lithium battery under charging conditions according to claim 1, characterized in that: In step 4, the weighted average method specifically includes the following steps: 1) Weight distribution: The weight of the 0.1Hz frequency segment is 50%; The weight of the 0.2Hz frequency segment is 30%; The weight of the 0.4Hz frequency segment is 20%; 2) Weighted average calculation: The ohmic impedance of each frequency band is weighted averaged to obtain the final ohmic impedance R_ohm: R_ohm=w1*R_ohm0.1Hz+w2*R_ohm0.2Hz+w4*R_ohm0.4Hz Among them, w1+w2+w3=1.

Images