Fitting simulation method for characteristic curve of power battery

Abstract

The invention discloses a fitting simulation method for a characteristic curve of a power battery. The fitting simulation method comprises the following steps of S1, estimating theoretical modeling of amplitude based on a least square method; s3, a ternary lithium battery, a load and a universal meter are set up for the experiment, voltage measurement data and a scatter diagram are obtained, fitting simulation is conducted on the data through the least square method, and a battery voltage characteristic curve is drawn through an accurate method.The principle and the implementation method are simple, hardware cost is saved, and the accuracy of the battery voltage characteristic curve is improved. According to the method, the verification period is shortened, the development time is saved, the amplitude data can be fitted with higher precision, the described curve is more accurate, and the method has a wide application prospect and is beneficial to popularization and application.

Description

Technical Field

[0001] The present invention relates to the technical field of fitting simulation, and particularly to a method for fitting and simulating the characteristic curve of a power battery. Background Art

[0002] The battery model is of great significance to the development of the battery management system. A good battery power model can not only provide a basis for the estimation algorithm of the remaining power, but also simulate and imitate the characteristics and behavior process of the battery, and assist in formulating the energy management strategy. The significance of battery modeling lies in determining the mathematical relationship between the environmental factors of the battery and various characteristic quantities. The objects considered include terminal voltage, working current, SOC, temperature, internal resistance, electromotive force, and SOH, etc. Finding the connection between them is of great significance to the development of the battery management system. On the one hand, through the established battery model, various performances of the battery during operation can be estimated, so that various battery management strategies (such as energy control strategy, battery equalization strategy, etc.) can be simulated, and the effectiveness of the strategy can be verified by software methods. Compared with the method of using experimental verification, such simulation verification does not require the use of a hardware circuit board and actual battery samples, which not only saves the hardware cost, but also shortens the verification cycle and saves the development time. In the simulation of the equalization control strategy based on energy transfer, it only takes a few minutes to perform process simulation on a specific initial state in software using the battery model, while it may take several hours to verify the effectiveness of the strategy using an actual circuit board. For the case of more samples to be tested, the advantages of simulation verification are more obvious. On the other hand, an accurate battery model is of great significance to the remaining power (SOC) evaluation algorithm. SOC evaluation is a difficult point in the battery management system and is of great significance to the development of other functions of the battery management system. In practice, the internal SOC is often estimated through external physical quantities such as voltage, current, and temperature. If a more accurate external characteristic model can be established, it is extremely beneficial to find the numerical relationship between SOC and various directly measurable physical quantities, so that the internal SOC can be evaluated through the monitored external performance. Therefore, there is an urgent need to develop a method for fitting and simulating the characteristic curve of a power battery to solve the above technical problems.

[0003] In view of this, the present invention is specifically proposed. Summary of the Invention

[0004] The object of the present invention is to provide a method for fitting and simulating the characteristic curve of a power battery. The principle and implementation method are simple. It not only saves the hardware cost, shortens the verification cycle, and saves the development time, but also can fit the amplitude data with higher precision, and the depicted curve is more accurate, having a broad application prospect and being conducive to popularization and application.

[0005] To achieve the above object, a method for fitting and simulating the characteristic curve of a power battery provided by the present invention includes the following steps:

[0006] S1: Theoretical modeling for estimating the amplitude based on the least squares method;

[0007] S2: Experiment for verifying the accuracy of fitting the amplitude by the least squares method:

[0008] S3: Set up a ternary lithium battery, a load, and a multimeter for experiments to obtain voltage measurement data and a scatter plot, fit and simulate the data by the least squares method, and draw the battery voltage characteristic curve by an accurate method.

[0009] Preferably, the specific steps of S1 are as follows:

[0010] Let the ideal cosine signal be:

[0011] y(t) = E1cos2πft + E2sin2πft + Q = Ecos(2πft + φ) + Q ①

[0012] The data recording sequence is the acquisition samples y1, y2,..., y at known times t1, t2,..., t n n , and the fitting process of the three-parameter cosine curve is that the frequency f of the input signal is known, and A1, B1, and C are selected or found to minimize the sum of the squared residuals ε,

[0013]

[0014] From the minimum of ε, we can obtain:

[0015]

[0016] In formula ②, A1 = A N / A D ; B1 = B N / B D ;

[0017] The parameters A1, B1, and C are the least squares fitting values of E1, E2, and Q,

[0018] where, α i = cos2πft i ; β i = sin2πft i ;

[0019]

[0020]

[0021] ​

[0022]

[0023] In Equations (4), (5), (6), and (7),

[0024] The fitting function is:

[0025]

[0026] Another expression form is:

[0027]

[0028] In Equation (9),

[0029]

[0030] The effective value of the fitting residual is:

[0031]

[0032] A fitting and simulation method for the characteristic curve of a power battery provided by the present invention has the following beneficial effects.

[0033] 1. The principle and implementation method of the present invention are simple. The simulation verification does not require the use of a hardware circuit board and actual battery samples, which not only saves the hardware cost but also shortens the verification cycle and saves the development time.

[0034] 2. The present invention can fit the amplitude data with higher precision, depict a more accurate curve, and establish a relatively accurate external characteristic model, which is extremely beneficial for finding the numerical relationship between the SOC and various directly measurable physical quantities, so as to evaluate the internal SOC through the monitored external performance. Brief Description of the Drawings

[0035] Figure 1 is the flowchart of the least squares method;

[0036] Figure 2 is the simulation image of the least squares method with SNR = 0;

[0037] Figure 3 is the simulation image of the least squares method with SNR = 6;

[0038] Figure 4 is the simulation image of the least squares method with SNR = 10;

[0039] Figure 5 is the simulation image of the least squares method with SNR = 40;

[0040] Figure 6is a voltage scatter plot;

[0041] Figure 7 is the voltage characteristic curve simulated by the least squares method. Specific embodiments

[0042] The present invention will be further described below in conjunction with specific embodiments and drawings to facilitate understanding of the content of the present invention.

[0043] A method for fitting and simulating the characteristic curve of a power battery provided by the present invention includes the following steps:

[0044] S1: Theoretical modeling for estimating the amplitude based on the least squares method; the flowchart of the least squares method is as Figure 1 shown.

[0045] S2: Experiment for verifying the accuracy of fitting the amplitude by the least squares method:

[0046] S3: Set up a ternary lithium battery, a load, and a multimeter for experiments to obtain voltage measurement data and a scatter plot, fit and simulate the data by the least squares method, and draw the battery voltage characteristic curve by an accurate method.

[0047] The specific steps of S1 are as follows:

[0048] Let the ideal cosine signal be:

[0049] y(t) = E1cos2πft + E2sin2πft + Q = Ecos(2πft + φ) + Q ①

[0050] The data recording sequence is the acquisition samples y1, y2,..., y at known times t1, t2,..., t n of n , and the fitting process of the three-parameter cosine curve is that the frequency f of the input signal is known, and A1, B1, and C are selected or found to minimize the sum of the squares of the residuals ε,

[0051]

[0052] From the minimum of ε, we can get:

[0053]

[0054] In formula ②, A1 = A N / A D ; B1 = B N / B D ;

[0055] The parameters A1, B1, and C are the least squares fitting values of E1, E2, and Q,

[0056] where αi = cos(2πft) i ; β i = sin(2πft) i ;

[0057]

[0058]

[0059]

[0060]

[0061] In equations ④, ⑤, ⑥, and ⑦,

[0062] The fitting function is:

[0063]

[0064] Another expression is:

[0065]

[0066] In equation ⑨,

[0067]

[0068] The root mean square value of the fitting residual is:

[0069]

[0070] Under the condition of the same signal-to-noise ratio, the above method is used to simulate the ideal signal. The signal-to-noise ratio (SNR) or signal-to-noise ratio (S / N) refers to the signal-to-noise ratio of an electronic device or electronic system. The unit of the signal-to-noise ratio measurement device is dB, and the calculation method is 10lg(Ps / Pn). The effective powers of the signal and noise are calculated for Ps and Pn respectively, and can be converted to the voltage amplitude 20lg(Vs / Vn). Vn is the effective value of the signal voltage and noise. The following is the simulation result: Assume the signal is X(t) = 9cos(2qπf0t + π / 6) + 3, f0 = 1000Hz, add Gaussian white noise with different signal-to-noise ratios to the signal, simulate, restore the signal, estimate the amplitude, and calculate the error.

[0071] The simulation results of the least squares method and the error calculation under different signal-to-noise ratios are shown in Table 1:

[0072] Table 1 Simulation Results of the Least Squares Method

[0073]

[0074] The simulation images of the least squares method under different signal-to-noise ratios are asFigures 2 - 5 As shown, it can be seen from the simulation image that noise will affect the fitting accuracy of the least squares method. The larger the signal-to-noise ratio, the smaller the noise, and the higher the fitting accuracy. The relative errors are all controlled within 0.16%, meeting the fitting conditions for the voltage curve.

[0075] The battery discharges under full load conditions, and the changes in battery voltage are collected in real time. The voltage scatter plot obtained from the experiment is as Figure 6 shown. The algorithm of the least squares method is used to Figure 6 process it, effectively reducing the noise of the unevenly distributed scatter points and reducing errors. The voltage characteristic curve simulated by the least squares method is as Figure 7 shown.

[0076] The principle and implementation method of the present invention are simple. The simulation verification does not require the use of a hardware circuit board and actual battery samples, which not only saves hardware costs but also shortens the verification cycle and saves development time. The present invention can fit amplitude data with higher accuracy, depict a more accurate curve, establish a relatively accurate external characteristic model, which is extremely beneficial for finding the numerical relationship between SOC and various directly measurable physical quantities, so as to evaluate the internal SOC through the monitored external performance.

[0077] In this article, specific examples are used to elaborate on the inventive concept in detail. The description of the above embodiments is only used to help understand the core idea of the present invention. It should be pointed out that for those of ordinary skill in the art in this technical field, any obvious modification, equivalent replacement or other improvement made without departing from the inventive concept shall be included in the protection scope of the present invention.

Claims

1. A fitting simulation method for the characteristic curve of a power battery, characterized in that, It includes the following steps: S1: Theoretical modeling for estimating amplitude based on the least squares method; S2: Experiment for verifying the accuracy of fitting amplitude by the least squares method: S3: Set up a ternary lithium battery, a load, and a multimeter to conduct experiments, obtain voltage measurement data and scatter plots, fit and simulate the data using the least squares method, and draw the battery voltage characteristic curve by an accurate method.

2. The fitting simulation method for the characteristic curve of a power battery according to claim 1, wherein The specific steps of S1 are as follows: Assume the ideal cosine signal is: y(t) = E1cos2πft + E2sin2πft + Q = E cos(2πft + φ) + Q ① The data recording sequence is the acquisition samples y1, y2, …, y at known times t1, t2, …, t n , and the fitting process of the three-parameter cosine curve is that the frequency f of the input signal is known, and A1, B1, and C are selected or found to minimize the sum of squared residuals ε n , From the minimum of ε, it can be obtained that: In formula ②, A1 = A N / A D ; B1 = B N / B D ; The parameters A1, B1, and C are the least squares fitting values of E1, E2, and Q, where α i = cos2πft i ; β i = sin2πft i ; In Formulas ④, ⑤, ⑥, and ⑦, The fitting function is: Another expression form is: In Formula ⑨, The effective value of the fitting residual is:

Images