CN103135066A - Measuring method of electric quantity of segmented iron phosphate lithium battery - Google Patents

Abstract

The invention discloses a measuring method of electric quantity of segmented iron phosphate lithium battery based on load voltage and Kalman filtering. In parts of starting and finishing of charging and discharging, namely, the parts of smaller than 10% of state of charge (SOC) and bigger than 90% of the SOC, due to the fact that a resolution ratio of a congruent relationship of voltage and the SOC is high, measurement is carried out directly by the adoption of a method of the load voltage, and the electric quantity of the battery is directly obtained at the moment by the adoption of the congruent relationship of the voltage and the SOC. In the middle part of the charging and the discharging, namely the parts of bigger than 10% of the SOC and smaller than 90% of the SOC, a fast Kalman filtering SOC estimation method is adopted. By the adoption of the measuring method of the electric quantity of the segmented iron phosphate lithium battery, terms, related to the SOC, in a battery terminal voltage model are calculated in advance, revised in certain combined with an experiment result and stored in a data sheet. In an iterative process of the Kalman filtering, a large number of calculations in every time can be avoided, and online testing speed of a singlechip is improved.

Description

A kind of measuring method of segmentation ferric phosphate lithium cell electric weight

Technical field

The present invention relates to a kind of segmentation ferric phosphate lithium cell electric weight online test method based on load voltage and Kalman filtering.

Background technology

Ferric phosphate lithium cell is as the electrokinetic cell of a kind of large capacity, high conversion efficiency, long-life and low Memorability, has been widely used in a large amount of fields such as the energy, environmental protection, chemical industry.The on-line quick detection of battery electric quantity estimates it is core technology in the management of battery also referred to as state of charge (State of Charge, SOC), is also one of focus of this area research.Traditional electric weight detection technique mainly comprises open-circuit voltage method, ampere-hour method, look-up table etc., but these method ubiquities are to the relatively poor problem of ferric phosphate lithium cell SOC curve flat region estimated accuracy.

For the deficiency that classic method exists, the ferric phosphate lithium cell SOC algorithm for estimating of employing Kalman filtering more and more becomes the focus of this area research.But traditional Kalman filtering SOC algorithm is too complicated, and data operation quantity is too large, is not suitable for based on the minicell management system under single-chip microcomputer or Micro Chip environment.

The technology similar to the present invention has at present: 1. Cheng Yan green grass or young crops is waited the paper " based on the electric automobile remaining capacity estimation of Kalman filtering " that is published in " Electronic University Of Science ﹠ Technology Of Hangzhou's journal "; 2. the paper that is published in " Tsing-Hua University's journal " such as woods Cheng Tao " is estimated electric automobile power battery SOC with improved Ah counting method "; 3. Electronic University Of Science ﹠ Technology Of Hangzhou, the patent of invention of He Zhiwei etc., " a kind of estimation method of battery dump energy of sample-based point Kalman filtering " (CN101598769B), and " a kind of estimation method of battery dump energy based on combined sampling point Kalman filtering " is (CN101604005B); 4. University Of Chongqing, the patent of invention of Deng Li etc. " evaluation method of residual capacity of iron-lithium phosphate power cell ".

The algorithm that these technology adopt is substantially similar, is all to adopt following Kalman filtering optimal estimation recurrence equation group:

$\left\{\begin{array}{c}{\mathrm{<figure-callout\; id="0"\; label="Z"\; filenames="HDA00002779916600011.png"\; state="\{\{state\}\}">Z</figure-callout>}}_{0|0}={\mathrm{<figure-callout\; id="0"\; label="SOC"\; filenames="HDA00002779916600011.png"\; state="\{\{state\}\}">SOC</figure-callout>}}_0,{\mathrm{<figure-callout\; id="0"\; label="P"\; filenames="HDA00002779916600011.png"\; state="\{\{state\}\}">P</figure-callout>}}_{0|0}=\mathrm{var}\left(z_0\right)\\ Z_{k|k-1}=Z_{k-1|k-1}-{\mathrm{\&eta;}}_i{i}_{k-1}\mathrm{\&Delta;t}/C_n\\ P_{k|k}=A_{k-1}{P}_{k-1|k-1}{A}_{k-1}+Q\\ Y_{k-1}=K_0-{\mathrm{Ri}}_{k-1}-{\mathrm{<figure-callout\; id="1"\; label="K"\; filenames="HDA00002779916600011.png,HDA00002779916600012.png"\; state="\{\{state\}\}">K</figure-callout>}}_1/Z_{k|k-1}-K_2{z}_{k|k-1}+{\mathrm{<figure-callout\; id="3"\; label="K"\; filenames="HDA00002779916600012.png"\; state="\{\{state\}\}">K</figure-callout>}}_3\ln z_{k|k-1}+K_4\ln (1-z_{k|k-1})\\ L_{k-1}=P_{k|k-1}{C}_k^T{(C_k{P}_{k|k-1}{C}_k^T+R)}^{-1}\\ Z_{k|k}=z_{k|k-1}+L_{k-1}(U_{k-1}-Y_{k-1})\\ P_{k|k}=(I-L_k{C}_k)P_{k|k-1}\\ k=\mathrm{1,2},...\end{array}\right.---\left(1\right)$

The key distinction is that selected battery terminal voltage model is different.

As: the paper of Cheng Yanqing etc. " based on the electric automobile remaining capacity estimation of Kalman filtering ", " estimate electric automobile power battery SOC with improved Ah counting method " with the paper of Lin Chengtao etc., and the patent of invention " a kind of estimation method of battery dump energy of sample-based point Kalman filtering " of He Zhiwei etc. and the battery terminal voltage model of " a kind of estimation method of battery dump energy based on combined sampling point Kalman filtering " employing are all

Y _k-1=K ₀-Ri _k-1-K ₁/Z _k|k-1-K ₂z _k|k-1+K ₃lnz _k|k-1+K ₄ln(1-z _k|k-1)+v _k（2）

In formula, Yk for calculate with battery model the load voltage of battery, ik is load current, R is the internal resistance of cell, z _K|k-1K SOC estimated value constantly, K0, K1, K2, K3 and K4 are model coefficients, with so that model better is complementary with data, Vk is the measurement noise.Think in this model that internal resistance of cell R is constant, is not inconsistent with actual conditions.

The battery terminal voltage model that the patent of invention of Deng Li etc. " evaluation method of residual capacity of iron-lithium phosphate power cell " adopts is:

$V\left(t\right)={\mathrm{\&eta;}}_p-{\mathrm{\&eta;}}_n+{\mathrm{\&phi;}}_{e,p}-{\mathrm{\&phi;}}_{e,n}+U_{\mathrm{oc}}\left(\mathrm{SOC}\right)-\frac{R_f}{A}I---\left(3\right)$

In formula, V (t) for calculate with battery model the load voltage of battery, U _oc(SOC) be open-circuit voltage, η _p-η _nBe superpotential, φ _{E, p}-φ _e,nPoor for the liquid phase current potential,

Be ohm superpotential.

Obviously the battery terminal voltage model of said method employing is too complicated, and data operation quantity is too large, is not suitable for based on the minicell management system under single-chip microcomputer or Micro Chip environment.

Summary of the invention

The object of the invention is to propose a kind of segmentation ferric phosphate lithium cell electric weight online test method based on load voltage and Kalman filtering, operand is little, and measuring speed is fast, is applicable to adopt the minicell management system of single-chip microcomputer or Micro Chip.

A kind of segmentation ferric phosphate lithium cell electric weight online test method based on load voltage and Kalman filtering comprises the steps:

Step 1, record the cell load voltage U by experiment _NegativeWith SOC relation table M(U _Negative), and the battery model parameter:

(1) battery is carried out once complete charging and discharging, utilize a minute looks to record cell load voltage U in this process _NegativeThe corresponding relation of value and SOC value, and deposit corresponding numerical relation in table M (U _Negative, SOC):

Y wherein ₀, y ₁..., y _nBe the cell load voltage U _Negative, U is classified on the right side as _NegativeCorresponding SOC value;

(2) each that represents battery with battery state model and observation model SOC value x and voltage y, current i relation constantly:

State model: $x_k=x_{k-1}+\frac{i_k\mathrm{\&Delta;t}}{{\mathrm{\&eta;}}_i{\mathrm{\&eta;}}_{T}Q}+w_k---\left(1\right)$

Observation model: y _k=E ₀+ g (x _k)-R (i _k) i _k+ v _k(2)

X wherein _kBe battery k SOC constantly, i _kBe k moment electric current, Q is the battery rated capacity, η _iAnd η _TBe respectively the discharge and recharge coefficient relevant with electric current and temperature, Δ t is measuring intervals of TIME, w _kState-noise, y _kK cell load voltage estimated value constantly, E ₀The constant relevant with cell voltage potential, R(i _k) be the internal resistance of cell, with current i _kRelevant, " the HPPC method of testing in FreedomCAR battery testing handbook is calculated, v to adopt the U.S. _kFor measuring noise, g(x _k) be according to x _kRelevant variable is by N(x, the g of tabling look-up) obtain;

This N(x, g) by measuring, assay method is: measure k cell load voltage y constantly _kWith load current i _k, by table M(U _Negative, SOC) obtain this moment battery y _kCorresponding x _k, by the model of formula (2) and ignore and measure noise v _k, calculate g(x _k):

g(x _k)=y _k+R(i _k)i _k-E ₀（3）

Result is charged to table N(x, g)

$N(x,g)=\begin{array}{}\end{array}\left[\begin{array}{cc}{x}_0& g\left(x_0\right)\\ x_1& g\left(x_1\right)\\ .,& .\\ .& .\\ .& .\\ x_n& g\left(x_n\right)\end{array}\right]$

Step 2, measure the cell load current i with current conversion module _kAnd direction, and be in charging or discharge condition according to direction of current judgement battery this moment, measure the cell load voltage U of this moment with voltage transformation module _Negative

Step 3, according to the cell load voltage U _NegativeLook into the cell load voltage U _NegativeWith SOC relation table M(U _Negative, SOC) judge whether the battery SOC state is in beginning or the end region that discharges and recharges;

If step 4 battery SOC state is in beginning or the end region that discharges and recharges, adopt load method, by looking into the cell load voltage U _NegativeWith SOC relation table M(U _Negative, SOC), obtain battery SOC, get back to step 2 after completing;

If step 5 battery SOC state is not in beginning or the end region that discharges and recharges, adopt the Fast Kalman filtering SOC estimation technique, obtain SOC value of battery, specific as follows:

(1) according to load voltage U _NegativeEstimate the battery SOC of this moment ₀As initial value

Namely

Be battery 0 SOC constantly., with load voltage U _NegativeThe detection variance

As the Kalman Filter Estimation initial variance;

(2) measure the cell load voltage U constantly at k _kWith the cell load current i _k, k=1,2,3,

(3) adopt Fast Kalman filtering iteration recursive operation, calculate the battery SOC estimated value:

K is the battery SOC predicted value constantly

K is cell load voltage estimated value y constantly _k:

Battery SOC estimated value after K filtering constantly

K=1,2,3…

Wherein, kalman gain $L_k=\frac{P_k^{-}{C}_k}{P_k^{-}{C}_k^2+D_v},$ $P_k^{-}=P_{k-1}^{-}+D_w,$ The Kalman filtering variance, $C_k=-\frac{0.00001}{x_k}-0.6469+\frac{0.0891}{x_k}+\frac{0.0946}{1-x_k},$ D _vAnd D _wBe respectively the state-noise variance and measure noise variance.

a kind of segmentation ferric phosphate lithium cell electric weight online test method based on load voltage and Kalman filtering of the present invention, discharging and recharging beginning and latter end, be SOC＜10% and SOC〉90% part, because slope of a curve is larger, as shown in Figure 1, this moment, voltage and SOC corresponding relation resolution were higher, can directly measure by load method, namely utilize the corresponding relation of voltage and SOC directly to obtain the electric weight of battery at this moment, and the center section that is discharging and recharging, the i.e. part of 10%＜SOC＜90%, adopt a kind of Fast Kalman filtering SOC estimation technique, be characterized in item relevant with SOC in the battery terminal voltage model is calculated in advance, and carry out certain correction in conjunction with experimental result, deposit in tables of data, like this in the Kalman filtering iterative process, that avoids all will do a large amount of computings at every turn, improve the online detection speed of single-chip microcomputer.

Description of drawings

Fig. 1 ferric phosphate lithium cell charging and discharging curve figure;

Fig. 2 ferric phosphate lithium cell electric weight of the present invention on-line measuring device schematic diagram;

Fig. 3 is schematic flow sheet of the present invention.

The invention will be further described below in conjunction with the drawings and specific embodiments.

Embodiment

As shown in Figure 2; a kind of segmentation ferric phosphate lithium cell electric weight on-line measuring device based on load voltage and Kalman filtering of the present invention comprises single-chip microcomputer 1; temperature polling instrument 2; constant current/constant voltage charging circuit 3; constant-current discharge circuit 4; battery protection module 5, voltage transformation module 6, current conversion module 7.Wherein, constant current/constant voltage charging circuit 3 is responsible for battery is charged; constant-current discharge circuit 4 is responsible for battery is carried out impulse electricity; temperature polling instrument 2 detects in real time battery temperature and send single-chip microcomputer 1 to carry out computing; battery protection module 5 prevents from the battery overshoot and crosses putting; voltage transformation module 6 detects cell load voltage and send single-chip microcomputer 1 to carry out computing, and current conversion module detects the cell load electric current and send single-chip microcomputer 1 to carry out computing, and single-chip microcomputer 1 utilizes temperature, voltage and current to calculate the SOC of battery.

As shown in Figure 3, a kind of segmentation ferric phosphate lithium cell electric weight online test method based on load voltage and Kalman filtering of the present invention specifically comprises the steps:

Step 1, record the cell load voltage U by experiment _NegativeWith SOC relation table M(U _Negative), and the battery model parameter:

(1) battery is carried out once complete charging and discharging, utilize a minute looks to record cell load voltage U in this process _NegativeThe corresponding relation of value and SOC value, and deposit corresponding numerical relation in table M (U _Negative, SOC):

Y wherein ₀, y ₁..., y _nBe the cell load voltage U _Negative, U is classified on the right side as _NegativeCorresponding SOC value;

(2) each that represents battery with battery state model and observation model SOC value x and voltage y, current i relation constantly:

State model: $x_k=x_{k-1}+\frac{i_k\mathrm{\&Delta;t}}{{\mathrm{\&eta;}}_i{\mathrm{\&eta;}}_{T}Q}+w_k---\left(1\right)$

Observation model: y _k=E ₀+ g (x _k)-R (i _k) i _k+ v _k(2)

X wherein _kBe battery k SOC constantly, i _kBe k moment electric current, Q is the battery rated capacity, η _iAnd η _TBe respectively the discharge and recharge coefficient relevant with electric current and temperature, Δ t is measuring intervals of TIME, w _kState-noise, y _kK cell load voltage estimated value constantly, E ₀The constant relevant with cell voltage potential, R(i _k) be the internal resistance of cell, with current i _kRelevant, " the HPPC method of testing in FreedomCAR battery testing handbook is calculated, v to adopt the U.S. _kFor measuring noise, g(x _k) be according to x _kRelevant variable is by N(x, the g of tabling look-up) obtain;

This N(x, g) by measuring, assay method is: measure k cell load voltage y constantly _kWith load current i _k, by table M(U _Negative, SOC) obtain this moment battery y _kCorresponding x _k, by the model of formula (2) and ignore and measure noise v _k, calculate g(x _k):

g(x _k)=y _k+R(i _k)i _k-E ₀（3）

Result is charged to table N(x, g)

$N(x,g)=\begin{array}{}\end{array}\left[\begin{array}{cc}{x}_0& g\left(x_0\right)\\ x_1& g\left(x_1\right)\\ .,& .\\ .& .\\ .& .\\ x_n& g\left(x_n\right)\end{array}\right]$

Step 2, measure the cell load current i with current conversion module _kAnd direction, and be in charging or discharge condition according to direction of current judgement battery this moment, measure the cell load voltage U of this moment with voltage transformation module _Negative

Step 3, according to the cell load voltage U _NegativeLook into the cell load voltage U _NegativeWith SOC relation table M(U _Negative, SOC) judge whether the battery SOC state is in beginning or the end region that discharges and recharges;

If step 4 battery SOC state is in beginning or the end region that discharges and recharges, adopt load method, by looking into the cell load voltage U _NegativeWith SOC relation table M(U _Negative, SOC), obtain battery SOC, get back to step 2 after completing;

If step 5 battery SOC state is not in beginning or the end region that discharges and recharges, adopt the Fast Kalman filtering SOC estimation technique, obtain SOC value of battery, specific as follows:

(1) according to load voltage U _NegativeEstimate the battery SOC of this moment ₀As initial value

Namely Be battery 0 SOC constantly., with load voltage U _NegativeThe detection variance

As the Kalman Filter Estimation initial variance;

(2) measure the cell load voltage U constantly at k _kWith the cell load current i _k, k=1,2,3,

(3) adopt Fast Kalman filtering iteration recursive operation, calculate the battery SOC estimated value:

K is the battery SOC predicted value constantly

K is cell load voltage estimated value y constantly _k:

Battery SOC estimated value after K filtering constantly

K=1,2,3…

Wherein, kalman gain $L_k=\frac{P_k^{-}{C}_k}{P_k^{-}{\mathrm{<figure-callout\; id="2"\; label="C\; k"\; filenames="HDA00002779916600011.png,HDA00002779916600012.png"\; state="\{\{state\}\}">C</figure-callout>}}_k^{}+D_v},$ $P_k^{-}=P_{k-1}^{-}+D_w,$

The Kalman filtering variance, $C_k=-\frac{0.00001}{x_k}-0.6469+\frac{0.0891}{x_k}+\frac{0.0946}{1-x_k},$ D _vAnd D _wBe respectively the state-noise variance and measure noise variance.

The above, it is only preferred embodiment of the present invention, be not that technical scope of the present invention is imposed any restrictions, therefore every foundation technical spirit of the present invention all still belongs in the scope of technical solution of the present invention any trickle modification, equivalent variations and modification that above embodiment does.

Claims (1)

1. the segmentation ferric phosphate lithium cell electric weight online test method based on load voltage and Kalman filtering, is characterized in that comprising the steps:

Step 1, record the cell load voltage U by experiment _NegativeWith SOC relation table M(U _Negative), and the battery model parameter:

(1) battery is carried out once complete charging and discharging, utilize a minute looks to record cell load voltage U in this process _NegativeThe corresponding relation of value and SOC value, and deposit corresponding numerical relation in table M (U _Negative, SOC):

Y wherein ₀, y ₁..., y _nBe the cell load voltage U _Negative, U is classified on the right side as _NegativeCorresponding SOC value;

(2) each that represents battery with battery state model and observation model SOC value x and voltage y, current i relation constantly:

State model: $x_k=x_{k-1}+\frac{i_k\mathrm{\&Delta;t}}{{\mathrm{\&eta;}}_i{\mathrm{\&eta;}}_{T}Q}+w_k---\left(1\right)$

Observation model: y _k=E ₀+ g (x _k)-R (i _k) i _k+ v _k(2)

X wherein _kBe battery k SOC constantly, i _kBe k moment electric current, Q is the battery rated capacity, η _iAnd η _TBe respectively the discharge and recharge coefficient relevant with electric current and temperature, Δ t is measuring intervals of TIME, w _kState-noise, y _kK cell load voltage estimated value constantly, E ₀The constant relevant with cell voltage potential, R(i _k) be the internal resistance of cell, with current i _kRelevant, " the HPPC method of testing in FreedomCAR battery testing handbook is calculated, v to adopt the U.S. _kFor measuring noise, g(x _k) be according to x _kRelevant variable is by N(x, the g of tabling look-up) obtain;

This N(x, g) by measuring, assay method is: measure k cell load voltage y constantly _kWith load current i _k, by table M(U _Negative, SOC) obtain this moment battery y _kCorresponding x _k, by the model of formula (2) and ignore and measure noise v _k, calculate g(x _k):

g(x _k)=y _k+R(i _k)i _k-E ₀（3）

Result is charged to table N(x, g)

$N(x,g)=\begin{array}{}\end{array}\left[\begin{array}{cc}{x}_0& g\left(x_0\right)\\ x_1& g\left(x_1\right)\\ .,& .\\ .& .\\ .& .\\ x_n& g\left(x_n\right)\end{array}\right]$

Step 2, measure the cell load current i with current conversion module _kAnd direction, and be in charging or discharge condition according to direction of current judgement battery this moment, measure the cell load voltage U of this moment with voltage transformation module _Negative

Step 3, according to the cell load voltage U _NegativeLook into the cell load voltage U _NegativeWith SOC relation table M(U _Negative, SOC) judge whether the battery SOC state is in beginning or the end region that discharges and recharges;

If step 4 battery SOC state is in beginning or the end region that discharges and recharges, adopt load method, by looking into the cell load voltage U _NegativeWith SOC relation table M(U _Negative, SOC), obtain battery SOC, get back to step 2 after completing;

If step 5 battery SOC state is not in beginning or the end region that discharges and recharges, adopt the Fast Kalman filtering SOC estimation technique, obtain SOC value of battery, specific as follows:

(1) according to load voltage U _NegativeEstimate the battery SOC of this moment ₀As initial value

Namely

Be battery 0 SOC constantly., with load voltage U _NegativeThe detection variance

As the Kalman Filter Estimation initial variance;

(2) measure the cell load voltage U constantly at k _kWith the cell load current i _k, k=1,2,3,

(3) adopt Fast Kalman filtering iteration recursive operation, calculate the battery SOC estimated value:

K is the battery SOC predicted value constantly

K is cell load voltage estimated value y constantly _k:

Battery SOC estimated value after K filtering constantly

K=1,2,3…

Wherein, kalman gain $L_k=\frac{P_k^{-}{C}_k}{P_k^{-}{C}_k^2+D_v},$ $P_k^{-}=P_{k-1}^{-}+D_w,$

The Kalman filtering variance, $C_k=-\frac{0.00001}{x_k}-0.6469+\frac{0.0891}{x_k}+\frac{0.0946}{1-x_k},$ D _vAnd D _wBe respectively the state-noise variance and measure noise variance.

Images