CN102565718A - Method for judging state of charge (SOC) of lithium battery based on rising and dropping of open-circuit voltage - Google Patents

Abstract

The invention relates to a method for judging a state of charge (SOC) of a lithium battery based on rising and dropping of open-circuit voltage. The method comprises the following steps of: (1) building a charging reference curve and a discharging reference curve of the lithium battery; (2) building short-time change characteristics of rising and dropping of the open-circuit voltage under different SOCs; (3) building a corresponding relation between an Euclidean distance and the SOC according to SOC-Euclidean distance data in the step (2); and (4) obtaining the Euclidean distance between the lithium battery and the charging reference curve or the Euclidean distance between the lithium battery and the discharging reference curve, and obtaining the SOC of the current lithium battery according to the corresponding relation in the step (3). The SOC of the battery is judged according to the short-time change of rising or dropping of the open-circuit voltage after the battery is charged or discharged, and the SOC is not required to be calculated after the open-circuit voltage is stabilized, so that the SOC of the battery can be effectively judged easily within short time.

Description

The lithium battery SOC determination methods that rise descends based on open-circuit voltage

Technical field

The present invention relates to the SOC determination methods in a kind of lithium battery use, especially a kind of lithium battery SOC determination methods of ging up to descend based on open-circuit voltage.

Background technology

Advantages such as lithium ion battery has WV and specific energy density is high, cyclicity is good, memory-less effect are widely used in the portable unit energy-storage battery, and are having a high potential aspect the new-energy automobile electrokinetic cell.The former mainly comprises 3C Product, i.e. computing machine, communication and consumption electronic product.Electrokinetic cell comprises electric bicycle and fields such as new-energy automobile and electric tool.Along with the further raising of lithium battery cost performance, will extend future to large solar system, peak load regulation network, household's electric power storage facility aspect.The accumulator state-of-charge is an important index in the battery use, and promptly the ratio of the actual electric weight that can provide and the electric weight that should be able to provide of charging fully under the current state is represented with SOC, can know the residual capacity of battery.

Now the battery detection equipment on the market generally has only the demonstration of whole Battery pack SOC, and few part has displaying monomer battery SOC information.Computing method mainly contain open-circuit voltage method, ampere-hour integral method etc., estimate through open-circuit voltage and use behind the initial SOC ampere-hour integral method to calculate SOC.But after battery needed one period long period, its open-circuit voltage was just really stable, just can reflect the SOC of battery; Use does not have stable open-circuit voltage to calculate SOC will produce deviation, can't reflect the SOC of battery.

Summary of the invention

Technical matters to be solved by this invention provides a kind of lithium battery SOC determination methods of ging up to descend based on open-circuit voltage; Battery SOC is judged in the variation in short-term that utilizes battery discharge or charging back open-circuit voltage to go up or descend; Need not to carry out SOC after open-circuit voltage is stablized by the time and calculate the SOC situation of accomplishing simple and effective at short notice judgement battery.

For addressing the above problem, technical scheme of the present invention is:

A kind of lithium battery SOC determination methods of ging up to descend based on open-circuit voltage comprises the steps:

(1) sets up lithium cell charging datum curve and discharge datum curve; The charging datum curve is meant that lithium cell charging is the change curve in short-term that 1 back open-circuit voltage descends in time to SOC, and the discharge datum curve is meant that it is the change curve in short-term that 0 back open-circuit voltage gos up in time that lithium battery is discharged to SOC;

(2) set up the variation characteristic in short-term that open-circuit voltage gos up and descends under the different SOC: calculating charges to the Euclidean distance of the change curve in short-term and the charging datum curve of open-circuit voltage decline under the different SOC, calculates the Euclidean distance of the change curve in short-term that is discharged to open-circuit voltage rise under the different SOC and the datum curve that discharges;

(3) based on the SOC-Euclidean distance data of step (2) and set up the corresponding relation of Euclidean distance and SOC;

(4) change curve in short-term that is descended in time by lithium cell charging open-circuit voltage to the current SOC obtains itself and the Euclidean distance of charging datum curve; Or be discharged to open-circuit voltage gos up in time behind the current SOC change curve in short-term by lithium battery and obtain itself and the Euclidean distance of discharge datum curve, obtain the SOC of current lithium battery according to the corresponding relation of step (3).

Preferably, described Euclidean distance computing formula is following:

$L\{V\left(t_j\right),U\left(t_j\right)\}=\sqrt{\frac{1}{m}\underset{j=1}{\overset{m}{\mathrm{\Σ}}}{[V\left(t_j\right)-U\left(t_j\right)]}^2}$

U (t _j): being discharged to SOC and being the change curve in short-term that 0 back open-circuit voltage gos up or charging to SOC is the change curve in short-term that 1 back open-circuit voltage descends;

V (t _j): the change curve in short-term that change curve in short-term that discharge back open-circuit voltage gos up or charging back open-circuit voltage descend;

t _jThe time that expression curve j is ordered, 1≤j≤m, and be positive integer; M representes that curve data counts, m>=2, and be positive integer.

Preferably, adopt the SOC-Euclidean distance data of step (2) to carry out least square fitting in the step (3), Euclidean distance is provided the SOC value of battery as the input of fitting function under the current SOC of the middle employing of step (4) by the fitting function computing.

Preferably, adopt the SOC-Euclidean distance data training of human artificial neural networks of step (2) in the step (3), Euclidean distance is provided the SOC value of battery as the input of artificial neural network under the current SOC of the middle employing of step (4) by the artificial neural network computing.

Preferably, described change curve in short-term is meant the change curve that the change curve that charging back open-circuit voltage descended 3～20 minutes or the back open-circuit voltage that discharges rose 3～20 minutes.

The invention provides a kind of lithium battery SOC determination methods of ging up to descend based on open-circuit voltage; Battery SOC is judged in the variation in short-term that utilizes battery discharge or charging back open-circuit voltage to go up or descend; Need not to carry out SOC after open-circuit voltage is stablized by the time and calculate the SOC situation of accomplishing simple and effective at short notice judgement battery.

Description of drawings

Fig. 1 is the change curve that is discharged to open-circuit voltage behind the different SOC.

Fig. 2 is the change curve that charges to open-circuit voltage behind the different SOC.

Embodiment

Below in conjunction with specific embodiment the present invention is done explanation in further detail, but protection scope of the present invention is not limited to this.

Embodiment 1:

Battery SOC is judged in the variation in short-term that a kind of lithium battery SOC determination methods of ging up to descend based on open-circuit voltage, said lithium battery SOC determination methods are utilized battery discharge or the back open-circuit voltage that charges gos up or descend.SOC-Euclidean distance data are carried out least square fitting, adopt that Euclidean distance is provided the SOC value of battery as the input of fitting function under the current SOC by the fitting function computing.

Concrete steps are following:

1. different SOC are chosen for 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% respectively; Set up the change curve in short-term (referring to Fig. 1) that open-circuit voltage gos up under the different SOC in discharge back; The described duration that changes in short-term is 10 minutes; It is 10 that image data is counted, and each minute gathered data.

2. calculate the change curve in short-term that open-circuit voltage gos up under the different SOC in back that discharge and be discharged to the Euclidean distance that SOC is the change curve in short-term of 0 back open-circuit voltage rise.

Described Euclidean distance formula is following:

$L\{V\left(t_j\right),U\left(t_j\right)\}=\sqrt{\frac{1}{m}\underset{j=1}{\overset{m}{\mathrm{\Σ}}}{[V\left(t_j\right)-U\left(t_j\right)]}^2}$

U (t _j): being discharged to SOC is the change curve in short-term that 0 back open-circuit voltage gos up;

V (t _j): the change curve in short-term that discharge back open-circuit voltage gos up;

t _j: the time that curve j is ordered, 1≤j≤10, and be positive integer; M: curve data is counted, m=10.

3. through above-mentioned data sample, obtained SOC=(0%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%), pairing Euclidean distance L=(0,0.32,0.45,0.55,0.63,0.71,0.77,0.84,0.90,0.95).Use least square fitting to run variation characteristic in short-term that road voltage gos up and the corresponding curve of SOC jointly, (98.9552 * L * L+0.6536 * L-0.1218)/100 equals 0 when SOC＜0 to obtain the function of SOC and L: SOC=; When SOC＞1, equal 1; Decimal place keeps two.

4. the SOC in the lithium battery use judges

After certain discharge, its open-circuit voltage rise curve and SOC are that the Euclidean distance of 0 o'clock open-circuit voltage rise curve is calculated as 0.92, import the calculating of above-mentioned formula, obtain SOC=0.84, promptly 84%.

Embodiment 2:

A kind of lithium battery SOC determination methods of ging up to descend based on open-circuit voltage; Said lithium battery SOC determination methods utilizes battery discharge or charging back open-circuit voltage gos up or battery SOC is judged in the variation in short-term of decline; The characteristic that changes in short-term of calculating the open-circuit voltage rise or descending is provided the performance evaluation result of battery as the input of artificial neural network through computing by artificial neural network.

Concrete steps are following:

1. different SOC are chosen for 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% respectively; Foundation charges to the change curve in short-term (referring to Fig. 2) that open-circuit voltage descends under the different SOC; The described duration that changes in short-term is 10 minutes; It is 10 that image data is counted, and each minute gathered data.

2. calculate the change curve in short-term that open-circuit voltage descends under the different SOC in back that charge and charge to the Euclidean distance that SOC is the change curve in short-term of 1 back open-circuit voltage decline.

Described Euclidean distance computing formula is following:

$L\{V\left(t_j\right),U\left(t_j\right)\}=\sqrt{\frac{1}{m}\underset{j=1}{\overset{m}{\mathrm{\Σ}}}{[V\left(t_j\right)-U\left(t_j\right)]}^2}$

U (t _j): charging to SOC is the change curve in short-term that 1 back open-circuit voltage descends;

V (t _j): the change curve in short-term that charging back open-circuit voltage descends;

t _j: the time that curve j is ordered, 1≤j≤10, and be positive integer; M: curve data is counted, m=10.

3. the Euclidean distance that descends through the different SOC open-circuit voltage corresponding with it, end user's artificial neural networks makes up the SOC function.(can with reference to " Matlab neural network and application " the 4th chapter BP network).

4. the SOC in the lithium battery use judges

Calculate the change curve in short-term of the back open-circuit voltage decline of charging and charge to the Euclidean distance that SOC is the change curve in short-term of 1 back open-circuit voltage decline; Its input through the constructed SOC function of the 3rd step end user's artificial neural networks, can be obtained the lithium battery SOC of this moment.

Claims (5)

1. a lithium battery SOC determination methods of ging up to descend based on open-circuit voltage is characterized in that comprising the steps:

(1) sets up lithium cell charging datum curve and discharge datum curve; The charging datum curve is meant that lithium cell charging is the change curve in short-term that 1 back open-circuit voltage descends in time to SOC, and the discharge datum curve is meant that it is the change curve in short-term that 0 back open-circuit voltage gos up in time that lithium battery is discharged to SOC;

(2) set up the variation characteristic in short-term that open-circuit voltage gos up and descends under the different SOC: calculating charges to the change curve in short-term of open-circuit voltage decline under the different SOC and the Euclidean distance of charging datum curve, calculates the Euclidean distance of the change curve in short-term that is discharged to open-circuit voltage rise under the different SOC and the datum curve that discharges;

(3) based on the SOC-Euclidean distance data of step (2) and set up the corresponding relation of Euclidean distance and SOC;

(4) change curve in short-term that is descended in time by lithium cell charging open-circuit voltage to the current SOC obtains itself and the Euclidean distance of charging datum curve; Or be discharged to open-circuit voltage gos up in time behind the current SOC change curve in short-term by lithium battery and obtain itself and the Euclidean distance of discharge datum curve, obtain the SOC of current lithium battery according to the corresponding relation of step (3).

2. a kind of lithium battery SOC determination methods of ging up to descend according to claim 1 based on open-circuit voltage, it is characterized in that: described Euclidean distance computing formula is following:

$L\{V\left(t_j\right),U\left(t_j\right)\}=\sqrt{\frac{1}{m}\underset{j=1}{\overset{m}{\mathrm{\Σ}}}{[V\left(t_j\right)-U\left(t_j\right)]}^2}$

U (t _j): being discharged to SOC and being the change curve in short-term that 0 back open-circuit voltage gos up or charging to SOC is the change curve in short-term that 1 back open-circuit voltage descends;

V _i(t _j): the change curve in short-term that change curve in short-term that discharge back open-circuit voltage gos up or charging back open-circuit voltage descend;

t _jThe time that expression curve j is ordered, 1≤j≤m, and be positive integer; M representes that curve data counts, m>=2, and be positive integer.

3. a kind of lithium battery SOC determination methods of ging up to descend according to claim 1 based on open-circuit voltage; It is characterized in that: adopt the SOC-Euclidean distance data of step (2) to carry out least square fitting in the step (3); Euclidean distance is provided the SOC value of battery as the input of fitting function under the current SOC of the middle employing of step (4) by the fitting function computing.

4. a kind of lithium battery SOC determination methods of ging up to descend according to claim 1 based on open-circuit voltage; It is characterized in that: the SOC-Euclidean distance data training of human artificial neural networks that adopts step (2) in the step (3); Euclidean distance is provided the SOC value of battery as the input of artificial neural network under the current SOC of the middle employing of step (4) by the artificial neural network computing.

5. according to the described a kind of lithium battery SOC determination methods of ging up to descend based on open-circuit voltage of one of claim 1～4, it is characterized in that: described change curve in short-term is meant the change curve that charging back 3～20 minutes change curve of open-circuit voltage decline or discharge back open-circuit voltage rose 3～20 minutes.

Images