Battery ECM structure and parameter identification method in maximum inertia support scene

By introducing negative resistance RC link and parameter identification method into the battery equivalent circuit model, the problem that traditional models cannot effectively characterize the battery output characteristics in the scenario of maximum inertia support is solved, and the accuracy of battery parameter identification and grid frequency stability are improved.

CN118914851APending Publication Date: 2024-11-08SICHUAN UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410955816.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-17
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

The traditional battery equivalent circuit model cannot effectively characterize the impact of the high current magnification of the battery on the battery output characteristics under the maximum inertia support scenario, resulting in insufficient parameter identification accuracy and affecting the grid frequency stability.

Method used

The RC link with negative resistance is introduced in the traditional RC equivalent circuit model, and a parameter identification method is proposed. By analyzing the impedance differences of components in the model, a separation method based on impedance characteristics is adopted to improve the accuracy of parameter identification.

Benefits of technology

More accurately characterize the output characteristics of the battery in the maximum inertia support scenario, improve the accuracy of parameter identification, and ensure the safety and resource utilization efficiency of the battery in the inertia support scenario.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118914851A_ABST
    Figure CN118914851A_ABST
Patent Text Reader

Abstract

The invention discloses a battery ECM structure in a maximum inertia support scene and a parameter identification method. An equivalent circuit model comprises a resistor R0, a first RC link, a compensation resistor Rdiff0 and a second RC link which are connected in sequence; the input end of the equivalent circuit model is battery open-circuit voltage UOC, and the output end is battery end voltage U; the first RC link comprises a resistor R1 and a capacitor C1 which are connected in parallel; the second RC link comprises a negative resistor Rdiff1 and a capacitor Cdiff1 which are connected in parallel; wherein the equivalent circuit of the negative resistor Rdiff1 comprises an operational amplifier, a resistor Ra, a resistor Rb and a resistor Rf; the in-phase input end of the operational amplifier is connected with the compensation resistor Rdiff0 and is connected with the output end of the operational amplifier through a resistor Rb; the reverse input end of the operational amplifier is connected with the output end of the operational amplifier through a resistor Ra and is connected with a resistor Rf in series to form the output end of the equivalent circuit model. According to the method, the RC link with the negative resistance is added into the traditional RC equivalent circuit model, so that the influence of the high current rate on the battery under the maximum inertia support scene can be better represented.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of battery charging technology, and in particular to a battery ECM structure and parameter identification method under a maximum inertia support scenario. Background Art

[0002] In order to build a new energy power system, a large number of new energy sources such as wind power and photovoltaic power are connected to the power system, and the overall rotational inertia of the system decreases, which seriously affects the frequency stability of the system. Using energy storage systems to support the frequency of the power grid is one of the efficient and feasible means. Thanks to the vigorous development of battery technology, battery energy storage systems (BESS) have become an important means to solve the low inertia problem of new energy.

[0003] The battery energy storage system (BESS) consists of several batteries connected in series and in parallel. The inertia it can provide is related to the power output capacity of the battery itself. If the maximum inertia current of the battery is overestimated, it may lead to battery safety problems; if the maximum inertia current of the battery is underestimated, it may lead to resource waste. Therefore, it is very important to accurately analyze the maximum output capacity of the battery in the inertia support scenario.

[0004] The equivalent circuit model (ECM) of the battery has a simple structure and clear concepts, and is often used to estimate the power, state of charge (SOC), and health status of the battery. However, the inertia response phase of the power system is characterized by short duration and high power density. The inertia support work condition (ISWC) is a condition with a high current rate and close to the output limit. When the battery is in the inertia support condition ISWC, it will produce a strong polarization phenomenon under the influence of the high current rate, which will cause the battery impedance to increase. The traditional resistor-capacitor (RC) structured ECM cannot characterize this phenomenon. In order to more accurately characterize the battery output characteristics under ISWC, the present invention proposes a structural improvement method and a model parameter identification method for the battery ECM under the maximum inertia support scenario. Summary of the invention

[0005] In view of this, the present invention provides a battery ECM structure and parameter identification method under the maximum inertia support scenario, which can not only better characterize the impact of high current rate on the battery under the maximum inertia support scenario, but also effectively improve the accuracy of parameter identification.

[0006] In order to achieve the above object, the present invention adopts the following technical solution:

[0007] A battery ECM structure in a maximum inertia support scenario includes a resistor R0, a first RC link, a compensation resistor R diff0 and the second RC link; the input end of the equivalent circuit model is the battery open circuit voltage U OC , the output end is the battery terminal voltage U;

[0008] The first RC link includes a resistor R1 and a capacitor C1 connected in parallel;

[0009] The second RC link includes a negative resistor R diff1 and capacitor C diff1 ;

[0010] Among them, the negative resistance R diff1 The equivalent circuit includes an operational amplifier, a resistor R a , resistor R b and resistor R f ;

[0011] The non-inverting input of the operational amplifier and the compensation resistor R diff0 connected, and through the resistor R b Connected to the output terminal of the operational amplifier;

[0012] The inverting input of the operational amplifier is connected through the resistor R a Connect to the output of the operational amplifier and connect a series resistor R f Then it constitutes the output end of the equivalent circuit model.

[0013] A parameter identification method for a battery ECM structure in a maximum inertia support scenario comprises the following steps:

[0014] Step S1, obtaining the maximum inertia current of the battery at different SOC test points;

[0015] Step S2, obtaining the response voltage of the maximum inertia current, and solving the test impedance of the battery according to the maximum inertia current and its response voltage, the solution formula is:

[0016]

[0017] In the formula, Z eq (t) is the test impedance of the battery at time t, U OC (SOC) is the open circuit voltage of the battery, U(t) is the terminal voltage of the battery at time t, I C,max (t) is the maximum inertia current of the battery at time t;

[0018] Step S3, separating the parameters in the test impedance into independent ohmic internal resistance and RC impedance, the separation formula is:

[0019]

[0020] In the formula, R e (t) is the independent ohmic internal resistance at time t, R e (t) = R0(t) + R diff0 (t), R0(t) is the resistance value of resistor R0 at time t, R1(t) is the resistance value of resistor R1 at time t, R diff1 (t) is the negative resistance R at time t diff1 The resistance value of Z p (t) is the total RC impedance at time t; Z p1 (t) is the first RC impedance at time t; Z p2 (t) is the second RC impedance at time t;

[0021] Step S4, processing the total RC impedance to obtain an independent ohmic internal resistance R e The resistance of resistor R1, the negative resistance R diff1 The resistance value, capacitance value of capacitor C1 and capacitance C diff1 Capacitance value;

[0022] Step S41, processing the total RC impedance, determining the inflection point of the total RC impedance, and solving the independent ohmic internal resistance R e The resistance value of RC impedance is:

[0023]

[0024] In the formula, Z eq (t J ) is t J The test impedance at the moment, Z p (t J ) is t J The total RC impedance at time t J is the time corresponding to the inflection point;

[0025] The formula for solving the independent ohmic internal resistance is:

[0026] R e =Z eq (0)

[0027] In the formula, Z eq (0) is the test impedance at time 0;

[0028] Step S42, determining the RC impedance expression after the inflection point, the formula is:

[0029]

[0030] τ2=R diff1 ·C diff1

[0031] Where τ2 is the time constant of the second RC link;

[0032] Step S43, processing the RC impedance expression after the inflection point to obtain the estimated time constant of the second RC link, and then solving the resistor R1 and the negative resistor R diff1 The calculation formula for the estimated value of the time constant of the second RC link is:

[0033]

[0034] In the formula, is the estimated time constant of the second RC link; t end is the end time of the maximum inertia current; Z p (t end ) is t end RC impedance at the moment; Z p (2t J ) is 2t J RC impedance at the moment;

[0035] Resistor R1 and negative resistor R diff1 The formula for solving the resistance value is:

[0036]

[0037] Step S44, obtaining a second RC impedance estimation value according to the time constant estimation value of the second RC link, and obtaining a first RC impedance value at the separation point according to the total RC impedance; wherein the calculation formula of the second RC impedance estimation value is:

[0038]

[0039] In the formula, is the estimated value of the second RC impedance at time t;

[0040] The calculation formula of the first RC impedance value at the separation point is:

[0041]

[0042] In the formula, is the first RC impedance value at the separation point at time t; is the estimated time constant of the first RC link;

[0043] Step S45, processing the first RC impedance value at the separation point to solve for the estimated time constant of the first RC link, the solution formula is:

[0044]

[0045] In the formula, is the first RC impedance value at the separation point at time t1; is the first RC impedance value at the separation point at time t2;

[0046] Step S46, based on the estimated time constant of the first RC link and the estimated time constant of the second RC link, as well as the resistor R1 and the negative resistor R diff1 The resistance value of capacitors C1 and C diff1 The capacitance value is solved as follows:

[0047]

[0048] Compared with the prior art, the present invention has the following advantages:

[0049] The present invention can better characterize the impact of high current rate on the battery under the maximum inertia support scenario by adding an RC link with negative resistance to the traditional RC equivalent circuit model. Secondly, the present invention proposes an identification method for the parameters of the proposed improved model structure. By analyzing the impedance differences of the components in the model, a separation method based on impedance characteristics is proposed, which effectively improves the accuracy of parameter separation and ensures the accuracy of parameter identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and should not be regarded as limiting the scope. For those of ordinary skill in the art, other relevant drawings can be obtained based on these drawings without creative work. In the drawings:

[0051] Figure 1 This is a schematic diagram of the battery ECM structure under the maximum inertia support scenario provided by the present invention.

[0052] Figure 2 This is a schematic diagram of a negative resistance equivalent circuit provided by the present invention.

[0053] Figure 3 Schematic diagram of a BESS grid-connected system with inertia support capability provided by the present invention.

[0054] Figure 4 The present invention provides a waveform of the frequency and its rate of change when the power grid is subjected to frequency disturbance.

[0055] Figure 5 The BESS output current curve of a cluster of ISWCs obtained by simulation provided by the present invention.

[0056] Figure 6 The maximum inertia current I of the battery provided by the present invention C,max(t) and the corresponding relationship with the response voltage.

[0057] Figure 7 This is a schematic diagram of a traditional battery n-order RC equivalent circuit model provided by the present invention.

[0058] Figure 8 A comparison chart of the battery equivalent impedance and the battery test impedance provided by the present invention.

[0059] Fig. 9 A schematic diagram of the decomposition of impedance parameters of a conventional battery test provided by the present invention.

[0060] Fig.10 A schematic diagram of the decomposition of impedance parameters for improved battery testing based on impedance characteristics provided by the present invention.

[0061] Fig.11 This is a correlation analysis diagram obtained based on the improved parameter identification method provided by the present invention.

[0062] Fig.12 This is a trend comparison chart of the improved ECM provided by the present invention and the traditional model MAE. DETAILED DESCRIPTION

[0063] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with embodiments and drawings. The exemplary embodiments of the present invention and their description are only used to explain the present invention and are not intended to limit the present invention.

[0064] Example 1

[0065] This embodiment discloses a battery ECM structure in a maximum inertia support scenario, such as Figure 1 As shown, it includes a resistor R0, a first RC link, a compensation resistor R diff0 and the second RC link; the input end of the equivalent circuit model is the battery open circuit voltage U OC , the output end is the battery terminal voltage U;

[0066] The first RC link includes a resistor R1 and a capacitor C1 connected in parallel;

[0067] The second RC link includes a negative resistor R diff1 and capacitor C diff1 ;

[0068] Among them, the negative resistance R diff1 The equivalent circuit includes an operational amplifier, a resistor R a , resistor R b and resistor R f ;

[0069] The non-inverting input of the operational amplifier and the compensation resistor Rdiff0 connected, and through the resistor R b Connected to the output terminal of the operational amplifier;

[0070] The inverting input of the operational amplifier is connected through the resistor R a Connect to the output of the operational amplifier and connect a series resistor R f Then it constitutes the output end of the equivalent circuit model.

[0071] Example 2

[0072] This embodiment discloses a parameter identification method of a battery ECM structure in a maximum inertia support scenario, comprising the following steps:

[0073] Step S1, obtaining the maximum inertia current of the battery at different SOC test points.

[0074] Specifically, first, a grid-connected model of a battery energy storage system with inertia support capability is established, and a benchmark current curve is obtained through simulation.

[0075] like Figure 3 As shown in the figure, a BESS grid-connected system with inertia support capability is constructed. The grid-connected system is a system with constant source-load balance. When the load increases, according to the principle of primary frequency regulation, the generator will reduce the speed and increase the output power to maintain power balance, but the frequency of the system will drop at this time, causing adverse effects on the power grid. When the damping effect is ignored, the power-frequency relationship in the power grid can be expressed as:

[0076]

[0077] In the formula, ΔP e The power provided to the generator in the system; ΔP L is the load power; ΔP B The power provided by the battery energy storage system. When the battery energy storage system outputs power, ΔP B is positive; otherwise, it is negative; H is the inertia time constant of the generator, and f is the frequency in the power grid.

[0078] When there is a power shortage in the power grid, the frequency changes as follows Figure 4 As shown in the figure, the stage from the initial frequency disturbance to the lowest frequency point is the inertia response stage. The frequency response is affected by many factors such as the parameters of the power grid generator and the size of the load disturbance. Different power grids have different responses to the load. In this embodiment, based on the known power grid parameters, the maximum frequency support capability provided by the BESS to the power grid is obtained under a given load disturbance. The output formula of the BESS when providing inertia is:

[0079]

[0080] In the formula, HB is the inertia time constant of BESS. According to the above formula, the output of BESS when providing inertia support is related to the system frequency change rate and H B The waveform is similar to the frequency change rate, and the amplitude is similar to H B Here, this embodiment has the following assumptions:

[0081] ① Ignoring the change of BESS voltage, it can be considered that the current is similar to the power, the waveform is similar to the frequency change rate, and the amplitude is affected by H. B Therefore, the current can be used as input to observe the response voltage and obtain the impedance when the inertia is the maximum.

[0082] ② In large power grids, the energy provided by BESS has little impact on system frequency deviation, so it can be considered that different H B The inertial response time under the setting is the same, and the current curve I is changed by equal amplitude C The amplitude of (t) is used to simulate different H B The following situation;

[0083] ③ This embodiment does not focus on the balance problem inside the battery, so it is assumed that the output of each battery in the BESS is balanced, and the output result of the BESS in the simulation can be changed with equal amplitude for the single battery.

[0084] According to the principle, after constructing a simulation model with inertia support capability, the benchmark current curve I of BESS under ISWC can be obtained. C,ref (t), that is Figure 5 The current curve that changes with time and whose standard value is 1. In this embodiment, the SOC of each battery in the BESS is set to 50%, the HB is 10s, and a 10% disturbance is added to the load. The current curve obtained is the reference current curve I C,ref (t).

[0085] Secondly, the battery is stimulated using the reference current curve to obtain the maximum inertia current of the battery at different SOC test points.

[0086] Specifically, in this embodiment, the battery used in the BESS is a 18650 lithium-ion battery with a rated capacity of 1.5 Ah, a rated voltage of 3.6 V, a voltage range of 2.5 to 4.2 V, and a SOC range of 10-90%.

[0087] The simulated reference current curve I C,ref (t) is adjusted with equal amplitude to obtain a cluster of current waveforms, such as Figure 5 As shown, the input current is applied to the battery after the equal amplitude change, and the response voltage of the battery is observed after the test is completed. The input current at this time is recorded as I C,i (t), t ineis the duration of the maximum inertia current. C,i (t) can just make the lowest point of the battery response voltage reach the lower limit cut-off voltage, then this current is considered to be the maximum inertia current I of the battery. C,max (t), maximum battery inertia current I C,max The corresponding relationship between (t) and the response voltage is as follows: Figure 6 As shown, I C,peak is the peak value of the maximum inertia current; if the battery fails to reach the lower limit cut-off voltage under the excitation of a simulated current waveform, the amplitude is adjusted to select a waveform with a large peak current to continue testing; if the cut-off voltage is reached in advance, a smaller current is selected to continue testing until the above conditions are met. The specific implementation steps are as follows:

[0088] a) Discharge the battery at 0.3A current to the lower cut-off voltage and leave it for 1 hour;

[0089] b) Charge the battery to 10% SOC at 0.75A and leave it alone for 1 hour;

[0090] c) Using input current I C,i (t) Exciting the battery to obtain the battery response voltage. If the lowest point of the battery response voltage is near the lower limit cut-off voltage, proceed to step e); otherwise, proceed to step d);

[0091] d) Change I C,i (t), at this time i=i+1, and the new I C,i (t), repeat step a) to step c);

[0092] e) Changing the SOC value in step b), repeating steps a) to e) to obtain the maximum inertia current at 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80% and 90% SOC respectively.

[0093] This embodiment is tested at 10°C, 25°C and 45°C. Taking 25°C as an example, it can be seen that the maximum inertia current I under 10% to 90% SOC is obtained through the test. C,max (t), the lowest point of the battery response voltage drops to near the lower cut-off voltage of 2.5V.

[0094] Step S2, obtaining the response voltage of the maximum inertia current, and solving the test impedance of the battery according to the maximum inertia current and its response voltage.

[0095] (I) Proposal of improved battery RC equivalent circuit model

[0096] The traditional battery n-order RC equivalent circuit model is as follows Figure 7 As shown, the voltage-current relationship of the battery is expressed as follows:

[0097]

[0098] Where U p,i (t) is the polarization voltage of the i-th group of RC links, R i is the resistance of the RC link of the i-th group, C i is the capacitance of the i-th group of RC links, and I is the current flowing through the loop of the n-th order RC equivalent circuit model of the battery.

[0099] If the capacitor C i The initial state is 0, then the polarization voltage on the RC link is:

[0100]

[0101] The terminal voltage of the battery is expressed as:

[0102]

[0103] Where U(t) is the terminal voltage of the battery; n is the total number of RC links in the equivalent circuit model; U OC (SOC) is the open circuit voltage of the battery, which is a function related to the battery SOC and is often expressed by a polynomial.

[0104] According to the equivalent circuit model, the equivalent impedance of the battery can be expressed as:

[0105]

[0106] In the formula, is the equivalent impedance of the battery.

[0107] In addition, the test impedance of the battery can also be directly calculated from the excitation current and response voltage data. According to the maximum inertia current obtained in the test link, the battery's maximum inertia current I can be obtained. C,max (t) The test impedance under excitation is:

[0108]

[0109] In this embodiment, the parameters of the n-th order RC equivalent circuit model of the battery can be identified by using the HPPC method. Specifically, the battery equivalent impedance obtained according to formula (6) The battery test impedance Z obtained according to formula (7) eq (t) is compared, and the comparison results are as follows Figure 8 As shown, where Z diff is the difference between the two impedances. Figure 8The equivalent impedance of the battery 1st, 2nd, and 3rd order RC equivalent circuit models is given in the figure. It can be seen from the figure that the battery equivalent impedance increases monotonically with a gradually decreasing slope. The battery test impedance initially exhibits similar behavior to the battery equivalent impedance, but as the frequency event develops to the middle and late stages, the behavioral difference between the two becomes more obvious. This is because in the early stages, due to the short current action time, the current has not yet produced a strong polarization effect on the battery. As the high current action time increases, the solid-phase diffusion reaction inside the battery intensifies, resulting in a large difference between the battery surface SOC and the average SOC. The polarization of the battery becomes more and more obvious, so the error of the battery equivalent impedance increases.

[0110] Through the above analysis, it is not difficult to find that the battery equivalent impedance is an upward curve with a decreasing slope; while the battery test impedance is an upward curve with a decreasing slope in the early stage and an upward curve with a increasing slope in the later stage. There is an obvious difference between the two, namely and Z eq There is an obvious difference between them. By subtracting the two, we can get:

[0111]

[0112] Where U RC (t) is the total voltage of the n-order RC link, Z diff (t) is the differential impedance, U diff (t) is the difference voltage.

[0113] Combining formulas (8) and (9), we get:

[0114]

[0115] In order to facilitate quantitative analysis, the battery impedance Z is tested eq (t) is decomposed, such as Fig. 9 Specifically, according to the battery test impedance Z eq The characteristics of (t) are decomposed into three parts, namely A0, A1 and A2, where A0 represents an independent ohmic resistance, which can be further divided into A 0,1 and A 0,2 ; A 0,1 Represents the ohmic internal resistance (i.e. R0) in the equivalent circuit model, A 0,2 is the compensation resistor R used to process parameter identification errors. diff0 , its value is very small and does not change much during the entire test cycle, so it can be used in I C,max (t) When the stimulus starts, Z eq (t) The difference between 0,2 Determine R diff0 ; A1 represents the total impedance of the n-order RC link; A2 is the differential impedance Zdiff In the main part of (t), the slope of A2 changes oppositely to that of A1.

[0116] For part A2, it can be expressed similarly to A1, but needs to be modified to reflect the difference in impedance characteristics. This embodiment uses "negative resistance" to obtain a "negative time constant", so that the impedance characteristics of the RC link are concave. It is worth noting that negative resistance does not mean that there is a physical negative resistance inside the battery. On the contrary, it means that the high current I C,max (t) amplifies the solid-phase diffusion reaction in the battery, resulting in an increase in impedance.

[0117] (II) Equivalence of Negative Resistance

[0118] Negative resistance is generated by equivalent transformation of circuit elements, such as Figure 2 As shown. According to the virtual short and virtual open characteristics of the operational amplifier:

[0119] i + =i _ =u d (11)

[0120] In the formula, i + and i - are the currents at the positive input and negative input of the operational amplifier, u d is the voltage between the positive and negative input terminals of the operational amplifier.

[0121] According to Kirchhoff Voltage Laws (KVL), the voltage at point C and the value of negative resistance are:

[0122]

[0123] Where, u is the voltage between AB; u CB is the voltage at point C relative to point B; R a , R b Both are process resistances, and R a =R b ; R f is the positive resistance to be equivalently transformed; i1 is the current flowing through loop 1; i2 is the current flowing through loop 2; i3 is the current flowing through loop 3.

[0124] According to formula (12), we can get:

[0125]

[0126] Through the above analysis, we can know that R f After the negative resistance equivalent model conversion, the final value of the resistor R is negative. diff1 .

[0127] Step S3, separating the parameters in the test impedance into independent ohmic internal resistance and RC impedance.

[0128] Specifically, Fig.10 As shown, according to the structural characteristics of the battery equivalent circuit model, the battery test impedance Z eq (t) is separated according to the following formula:

[0129]

[0130] In the formula, R e (t) is an independent ohmic internal resistance, and R e (t) = R0(t) + R diff0 (t), R0(t) is the resistance value of resistor R0 at time t; Z p (t) is the total RC impedance; Z p1 (t) is the first RC impedance; Z p2 (t) is the second RC impedance; R1(t) is the resistance value of resistor R1 at time t, R diff1 (t) is the negative resistance R at time t diff1 resistance value.

[0131] Step S4, processing the total RC impedance to obtain an independent ohmic internal resistance R e The resistance of resistor R1, the negative resistance R diff1 The resistance value, capacitance value of capacitor C1 and capacitance C diff1 Capacitance value.

[0132] Step S41, processing the total RC impedance, determining the inflection point of the total RC impedance, and solving the independent ohmic internal resistance R e resistance value.

[0133] Depend on Fig.10 It can be seen that there are two obvious changes in the test impedance of the battery, which are the form of convex function and the form of concave function, corresponding to the impedance of the first RC link and the impedance of the second RC link respectively. The dividing point between the two is the inflection point J, which is mathematically defined as:

[0134]

[0135] In the formula, Z eq (t J ) is t J The test impedance at the moment, Z p (t J ) is t J RC impedance at time t J is the time corresponding to the inflection point.

[0136] In this embodiment, the analysis of the RC impedance characteristics is as follows:

[0137] ①0-0.5t J :This stage Z p2 Smaller, battery Z p Mainly by Z p1 Decision, Z p The intersection with the y-axis is the independent ohmic internal resistance R of the battery. e =R0+R diff0 ,Right now:

[0138] R e =Z eq (0)(16)

[0139] ②0.5t J -t J :This stage Z p2 The value and growth rate gradually increase, while Z p1 The growth slows down at t = t J When , the growth rates of the two are the same;

[0140] ③t J -2t J :This stage Z p By Z p1 and Z p2 Joint impact, Z p1 The growth rate gradually approaches 0;

[0141] ④2t J -t end :This stage Z p1 Can be regarded as a constant, Z p The rate of change of Z p2 is determined by the rate of change.

[0142] Step S42, determining the RC impedance expression after the inflection point.

[0143] According to the above analysis, in 2t J -t end During this period, Z p1 tends to be stable; when t>2t J When Z p1 = R1, then Z after the inflection point p The main part is Z p2 and Z, which has stabilized p1 , at this time Z p The expression is:

[0144]

[0145] τ2=R diff1 ·C diff1 (18)

[0146] Where τ2 is the time constant of the second RC link.

[0147] Step S43, processing the RC impedance expression after the inflection point to obtain the estimated time constant of the second RC link, and then solving the resistor R1 and the negative resistor R diff1 resistance value.

[0148] Specifically, by taking the derivative of formula (17), we can get:

[0149]

[0150] Set t=2t J and t = t end Substituting into formula (19) we get:

[0151]

[0152] Dividing the two equations in formula (20) yields:

[0153]

[0154] Solving formula (21) yields:

[0155]

[0156] In the formula, is the estimated time constant of the second RC link.

[0157] Set t=2t J and t = t end Substituting into formula (17) we get:

[0158]

[0159] Solving the two equations in formula (23) together yields:

[0160]

[0161] Step S44, obtaining a second RC impedance estimation value according to the time constant estimation value of the second RC link, and obtaining a first RC impedance value at the separation point according to the total RC impedance.

[0162] Specifically, according to the estimated time constant of the second RC link, combined with equation (14), the separated for:

[0163]

[0164] In the formula, is the estimated value of the second RC impedance.

[0165] Then Z p (t) and Available for:

[0166]

[0167] In the formula, is the first RC impedance value at the separation point; is the estimated time constant of the first RC link.

[0168] Step S45, processing the first RC impedance value at the separation point to solve for an estimated time constant value of the first RC link.

[0169] Specifically, by taking the derivative of formula (26), we can get:

[0170]

[0171] According to the above analysis, 0-0.5t J Period Z p1 If the rate of change is large, then select t=t1 and t=t2 (0 <t1,t2<0.5t J ) for analysis, and substituting these two points into formula (27) we can obtain:

[0172]

[0173] Solving formula (28) yields:

[0174]

[0175] Step S46, based on the estimated time constant of the first RC link and the estimated time constant of the second RC link, as well as the resistor R1 and the negative resistor R diff1 The resistance value of capacitors C1 and C diff1 Capacitance value.

[0176]

[0177] This embodiment is further described by taking a specific case.

[0178] 1. Data Introduction

[0179] In this embodiment, the grid-connected model of the energy storage system is built using MATLAB / Simulink software to obtain the ISWC current curve, and the Chroma 17011 programmable battery tester is used to perform charge and discharge tests on the lithium-ion battery. The battery uses a 18650 lithium-ion battery produced by a certain manufacturer, with a rated capacity of 1500mAh, a rated voltage of 3.6V, and an upper and lower cut-off voltage of 2.5 / 4.2V. By conducting experiments on the operating data obtained in the simulation model, the impedance data of the battery at maximum inertia support can be obtained.

[0180] 2. Evaluation indicators

[0181] ①Determination coefficient

[0182] Coefficient of determination R 2 It is a parameter that measures the fit between two sets of data. Its value close to 1 means that the model prediction value can fully explain the experimental value. Its expression is:

[0183]

[0184] In the formula, y i , are the experimental value, model predicted value and experimental average value respectively, and m is the number of data.

[0185] ② Voltage error

[0186] The voltage error is the model voltage U model With the experimental value U test Contrast is used to compare the voltage errors between different models. The formula is as follows:

[0187] ξ j =100·(U test -U model ) / U test (32)

[0188] In the formula, ξ i is the voltage error at the jth sampling point.

[0189] ③ Mean absolute error

[0190] Through the Mean Absolute Error (MAE), we can intuitively see the overall error between the data. The formula is as follows:

[0191]

[0192] Where MAE is the average value of the voltage error and N is the total number of sampling points.

[0193] 3. Analysis of simulation results

[0194] According to formula (15) and formula (16), we can get C,max (t) Ohmic resistance R under excitation e The inflection point corresponds to time t J As shown in Table 1.

[0195] Table 120~90% SOC I C,max (t) R during excitation e and t J

[0196]

[0197] In obtaining R e and t J Then, from equation (14) and equation (17), we can get t>2t J Z p (t), and then according to formula (22), the estimated time constant of the second RC link can be obtained As shown in Table 2.

[0198] Table 220~90% SOC I C,max (t) Time constant of the second RC link during excitation

[0199]

[0200] After obtaining the time constant of the negative resistance RC link, the simultaneous equations (23) can be obtained as R1 and R diff1 , as shown in Table 3.

[0201] Table 320~90% SOC I C,max (t) R1 and R during excitation diff1

[0202]

[0203] According to equations (25) and (26), the model estimates and From equation (29), we can get the estimated time constant of the first RC link: As shown in Table 4. It can be seen from the table that the time constant of the first RC link is basically less than 4s. In the parameter identification method based on the inflection point proposed in this embodiment, 2t is selected. J After Z p According to the analysis, the growth rate of the impedance of the first RC link has decayed by 2 times the unit time constant, and its rate of change is close to 0. Therefore, in this stage (t≥2t J )Z p The rate of change of Z p2 The rate of change of Z is the same p1It is regarded as a constant, which indirectly verifies the rationality of this method.

[0204] Table 4 20~90% SOC I C,max (t) Time constant of the first RC link during excitation

[0205]

[0206] According to the above parameter results, the parameter identification method proposed in this embodiment obtains With the experimental value Z p The comparison chart between Fig.11 As shown, Fig.11 (a) to (h) correspond to the results of 20-90% SOC respectively. It can be seen from the figure that at 2t J (basically 8s) later, With Z p The rate of change is the same, and can be regarded as a constant. Therefore, according to the method proposed in this embodiment, p The rate of change of τ2 is first determined, according to t≥2t J The data of two points during the period can be used to obtain R1 and R diff1 ; According to τ2, we can get By Z p and Make a difference to get the separated By analyzing its rate of change, we can obtain τ1. After obtaining R1 and τ1, we can get the first RC link estimated by the model. The short dashed line represents the impedance sum of the first RC link and the second RC link estimated by the model.

[0207] In addition, from Fig.11 It can also be seen that the impedance at each SOC point obtained by the parameter separation method proposed in this embodiment is highly consistent with the experimental value, and the determination coefficient R 2 is greater than 0.9995. This proves the accuracy of the proposed parameter separation method based on impedance characteristics.

[0208] In order to compare the advantages of the improved ECM, the improved ECM is compared with the traditional first-order RC circuit (1RC) and the second-order RC circuit (2RC). Fig.12 The trend comparison chart of MAE of improved ECM and traditional model shows that the MAE of improved ECM is below 0.5%, and the MAE of 2RC is also less than 1RC, which is reasonable because 2RC is higher than 1RC, so the circuit accuracy will be higher, but the MAE of 2RC is larger than that of improved ECM, basically exceeding 2%. To show the fit between the circuit output voltage and the experimental value, Fig.12 The R of the three models is shown 2Comparison with experimental results. At each temperature, the R 2 The R values ​​of the Thevenin model and the DP model obtained by the hybrid power pulse characteristic test method HPPC (Hybrid Pulse Power Characteristic, HPPC) are very close to 1, indicating that the regression accuracy is very high. 2 In addition, the average R 2 It is 0.9995, which is much higher than the Thevenin model (0.9132) and the DP model (0.9420). Therefore, the accuracy of the improved ECM is greatly improved compared with the traditional model.

[0209] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A battery ECM structure in a maximum inertia support scenario, characterized in that: It includes the resistor R0, the first RC link, the compensation resistor R diff0 and the second RC link; the input end of the equivalent circuit model is the battery open circuit voltage U OC , the output end is the battery terminal voltage U; The first RC link includes a resistor R1 and a capacitor C1 connected in parallel; The second RC link includes a negative resistor R diff1 and capacitor C diff1 ; Among them, the negative resistance R diff1 The equivalent circuit includes an operational amplifier, a resistor R a , resistor R b and resistor R f ; The non-inverting input of the operational amplifier and the compensation resistor R diff0 connected, and through the resistor R b Connected to the output terminal of the operational amplifier; The inverting input of the operational amplifier is connected through the resistor R a Connected to the output of the operational amplifier and connected in series with resistor R f Then it constitutes the output end of the equivalent circuit model.

2. The parameter identification method of a battery ECM structure in a maximum inertia support scenario according to claim 1 is characterized in that: The following steps are involved: Step S1, obtaining the maximum inertia current of the battery at different SOC test points; Step S2, obtaining the response voltage of the maximum inertia current, and solving the test impedance of the battery according to the maximum inertia current and its response voltage, the solution formula is: In the formula, Z eq (t) is the test impedance of the battery at time t, U OC (SOC) is the open circuit voltage of the battery, U(t) is the terminal voltage of the battery at time t, I C,max (t) is the maximum inertia current of the battery at time t; Step S3, separating the parameters in the test impedance into independent ohmic internal resistance and RC impedance, the separation formula is: In the formula, R e (t) is the independent ohmic internal resistance at time t, R e (t) = R0(t) + R diff0 (t), R0(t) is the resistance value of resistor R0 at time t, R1(t) is the resistance value of resistor R1 at time t, R diff1 (t) is the negative resistance R at time t diff1 The resistance value of Z p (t) is the total RC impedance at time t; Z p1 (t) is the first RC impedance at time t; Z p2 (t) is the second RC impedance at time t; Step S4, processing the total RC impedance to obtain an independent ohmic internal resistance R e The resistance of resistor R1, the negative resistance R diff1 The resistance value, capacitance value of capacitor C1 and capacitor C diff1 Capacitance value; Step S41, processing the total RC impedance, determining the inflection point of the total RC impedance, and solving the independent ohmic internal resistance R e The resistance value of RC impedance is: In the formula, Z eq (t J ) is t J The test impedance at the moment, Z p (t J ) is t J The total RC impedance at time t J is the time corresponding to the inflection point; The formula for solving the independent ohmic internal resistance is: R e =Z eq (0) In the formula, Z eq (0) is the test impedance at time 0; Step S42, determining the RC impedance expression after the inflection point, the formula is: τ2=R diff1 ·C diff1 Where τ2 is the time constant of the second RC link; Step S43, processing the RC impedance expression after the inflection point to obtain the estimated time constant of the second RC link, and then solving the resistor R1 and the negative resistor R diff1 The calculation formula for the estimated value of the time constant of the second RC link is: In the formula, is the estimated time constant of the second RC link; t end is the end time of the maximum inertia current; Z p (t end ) is t end RC impedance at the moment; Z p (2t J ) is 2t J RC impedance at the moment; Resistor R1 and negative resistor R diff1 The formula for solving the resistance value is: Step S44, obtaining a second RC impedance estimation value according to the time constant estimation value of the second RC link, and obtaining a first RC impedance value at the separation point according to the total RC impedance; wherein the calculation formula of the second RC impedance estimation value is: In the formula, is the estimated value of the second RC impedance at time t; The calculation formula of the first RC impedance value at the separation point is: In the formula, is the first RC impedance value at the separation point at time t; is the estimated value of the time constant of the first RC link; Step S45, processing the first RC impedance value at the separation point to solve for the estimated time constant of the first RC link, the solution formula is: In the formula, is the first RC impedance value at the separation point at time t1; is the first RC impedance value at the separation point at time t2; Step S46, based on the estimated time constant value of the first RC link and the estimated time constant value of the second RC link, as well as the resistor R1 and the negative resistor R diff1 The resistance value of capacitors C1 and C diff1 The capacitance value is solved as follows: