CN105487016A - SOC value estimation method and system thereof - Google Patents

Abstract

The invention provides an SOC value estimation method comprising the steps that an SOC initial value is calculated by utilizing a Kalman filtering algorithm; SOC initial value assignment is performed and then an SOC value is calculated via an ampere-hour integral algorithm and the SOC value is saved, and the SOC value is also calculated by utilizing the Kalman filtering algorithm; whether the SOC values calculated by the two algorithms are within a preset error value is judged; and if the SOC values calculated by the two algorithms are within the preset error value, the SOC value calculated by the ampere-hour integral algorithm is outputted. The invention also provides an SOC value estimation system. SOC estimation of a battery is performed through combination of the ampere-hour integral algorithm and the Kalman filtering algorithm so that accumulative error caused by the ampere-hour integral algorithm can be overcome, the jump phenomenon of the Kalman filtering algorithm can also be overcome, the SOC values can be saved through EEPROM and thus the accurate SOC value can be stably and reliably obtained.

Description

A kind of SOC estimation method and system thereof

Technical field

The present invention relates to field of batteries, particularly relate to a kind of SOC estimation method and system thereof.

Background technology

The dump energy (StateofCharge, SOC) of battery directly reacts the course continuation mileage of electric automobile in a certain aspect, be a module important in battery management system, the accurate estimation therefore for battery SOC just seems very important.

At present, the method for estimating remaining capacity of battery is mainly divided into two large classes: direct method and indirect method.Direct method refers to that equipment directly measures present battery residual capacity by experiment; Indirect method, mainly through the physicochemical characteristic of inside battery, needs high-precision equipment to be therefore difficult in practice realize in estimation procedure.Ampere-hour integral method (Ahintegrationmethod is called for short Ah method), open-circuit voltage method (Open-circuitvoltagemethod is called for short OCV method), internal resistance method (Resistancemethod) etc. belong to indirect method.

But ampere-hour integral method can produce cumulative errors in computation process, cause the SOC calculated to increase error with the discharge and recharge time and increase, the accuracy of the SOC of ampere-hour integral method calculating simultaneously initial value is difficult to determine; Open-circuit voltage method needs to leave standstill for a long time to reach inside battery voltage stabilization, is difficult to realize in actual car running process; Internal resistance method also exists the difficulty of estimation internal resistance, hardware is also difficult to realize.In addition, also by artificial neural network algorithm (ArtificialNeuralNetworkAlgorithm), Kalman filtering algorithm (Kalmanfilteralgorithm, be called for short KF) etc. indirect method carry out estimating battery SOC, but neural network algorithm is due to its Operation system setting difficulty, and application cost is high in battery management system, do not possess advantage; And Kalman filtering algorithm there will be hopping phenomenon and can not preserve in the process calculating SOC, this algorithm can not ensure the Stability and veracity of SOC.

Therefore, a kind of SOC estimation method of design is needed badly, to improve the Stability and veracity of SOC.

Summary of the invention

In view of this, the object of the present invention is to provide a kind of SOC estimation method and system thereof, be intended to the problem that accuracy is not high and stability is lower solving SOC in prior art.

The present invention proposes a kind of SOC estimation method, comprising:

Kalman filtering algorithm is utilized to calculate SOC initial value;

Described SOC initial value assignment is calculated SOC value as ampere-hour integral algorithm, and preserves described SOC value, utilize Kalman filtering algorithm to calculate SOC value simultaneously;

Judge the SOC value that calculated by above-mentioned two kinds of algorithms whether in preset error value;

If in preset error value, then export the SOC value calculated by ampere-hour integral algorithm.

Preferably, described preset error value is ± 2.5%, is kept at the SOC value calculated by ampere-hour integral algorithm after initial value assignment by EEPROM.

Preferably, described method also comprises:

If not in prediction error value, then continue described SOC initial value assignment to calculate SOC value as ampere-hour integral algorithm, and preserve described SOC value, utilize Kalman filtering algorithm to calculate SOC value simultaneously;

Continue to judge the SOC value that calculated by above-mentioned two kinds of algorithms whether in preset error value;

If in preset error value, then export the SOC value calculated by ampere-hour integral algorithm.

On the other hand, the present invention also provides a kind of SOC valuation system, comprising:

Initial Value module, calculates SOC initial value for utilizing Kalman filtering algorithm;

Assignment module, for described SOC initial value assignment is calculated SOC value as ampere-hour integral algorithm, and preserves described SOC value, utilizes Kalman filtering algorithm to calculate SOC value simultaneously;

Judge module, for judging the SOC value that calculated by above-mentioned two kinds of algorithms whether in preset error value;

Output module, if in preset error value, then exports the SOC value calculated by ampere-hour integral algorithm.

Preferably, described preset error value is ± 2.5%, is kept at the SOC value calculated by ampere-hour integral algorithm after initial value assignment by EEPROM.

Preferably, described SOC valuation system also comprises:

Loop module, if for not in prediction error value, then continues described SOC initial value assignment to calculate SOC value as ampere-hour integral algorithm, and preserves described SOC value, utilize Kalman filtering algorithm to calculate SOC value simultaneously; Continue to judge the SOC value that calculated by above-mentioned two kinds of algorithms whether in preset error value; If in preset error value, then export the SOC value calculated by ampere-hour integral algorithm.

Technical scheme provided by the invention adopts ampere-hour integral method and Kalman filtering algorithm to combine to carry out SOC estimation to battery, both the cumulative errors that ampere-hour integral method causes had been overcome, overcome again Kalman filtering algorithm and occur hopping phenomenon, and by EEPROM, SOC value is preserved, can be reliable and stable obtain accurate SOC value.It is memoryless that ampere-hour integral method calculating SOC does not have historical inheritance, and SOC calculates with battery current and initially SOC is closely bound up; It is support with historical data that Kalman filtering algorithm calculating SOC has historical inheritance, and SOC calculates with the voltage of battery closely bound up, can rapidly converge to exact value with initial value is irrelevant.The present invention in conjunction with these two kinds of methods relative merits and carry out improving and can obtain the SOC value of battery of accurate stable.The method is applicable to the SOC estimation of various electrokinetic cell, can the tracking SOC actual value of dynamic stability compared to additive method the present invention, and the SOC accurate stable being more suitable for electric automobile exports.

Accompanying drawing explanation

Fig. 1 is SOC estimation method process flow diagram in an embodiment of the present invention;

Fig. 2 is SOC valuation system architecture schematic diagram in an embodiment of the present invention.

Embodiment

In order to make object of the present invention, technical scheme and advantage clearly understand, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, be not intended to limit the present invention.

The specific embodiment of the invention provides a kind of SOC estimation method, mainly comprises the steps:

S11, Kalman filtering algorithm is utilized to calculate SOC initial value;

S12, described SOC initial value assignment is calculated SOC value as ampere-hour integral algorithm, and preserve described SOC value, utilize Kalman filtering algorithm to calculate SOC value simultaneously;

S13, judge the SOC value that calculated by above-mentioned two kinds of algorithms whether in preset error value;

If S14 is in preset error value, then export the SOC value calculated by ampere-hour integral algorithm.

A kind of SOC estimation method provided by the invention adopts ampere-hour integral method (Ahintegrationmethod, be called for short Ah method) and Kalman filtering algorithm (Kalmanfilteralgorithm, be called for short KF) combine and SOC estimation is carried out to battery, both the cumulative errors that ampere-hour integral method causes had been overcome, overcome again Kalman filtering algorithm and occur hopping phenomenon, and by EEPROM, SOC value is preserved, can be reliable and stable obtain accurate SOC value.

Below will be described in detail to a kind of SOC estimation method provided by the present invention.

Referring to Fig. 1, is SOC estimation method process flow diagram in an embodiment of the present invention.

In step s 11, Kalman filtering algorithm is utilized to calculate SOC initial value.

In the present embodiment, in engineer applied general Kalman filtering is carried out discretize after process again.Linear discrete system state space equation mainly comprises state equation and output equation, relation between state equation descriptive system state variable and input variable, output equation descriptive system output quantity and state variable, output quantity and input quantity, equation is shown below:

$\left\{\begin{array}{c}{\stackrel{\·}{x}}_{k+1}=A_k{x}_k+B_k{u}_k+{\mathrm{\Γw}}_k\\ y_k=C_k{x}_k+D_k{u}_k+v_k\end{array}\right.---\left(1\right)$

Wherein, x in formula _k, u _k, y _kbe respectively the state variable of etching system during k, input quantity and matrix, B _krepresent input matrix, C _krepresent output matrix, D _krepresent feedforward matrix.

In the present embodiment, the information that Kalman Filter Estimation utilizes output quantity yk and input quantity uk to obtain upgrades unknown state x by calculating _knonlinear IEM model for the predicted value of estimated state.H _kfor Kalman filtering gain matrix, P _kfor error co-variance matrix, I is unit matrix.Concrete Kalman filtering algorithm stepping type is as follows:

Filtering equations starting condition:

Wherein E is the variance of state variable, and var represents the covariance of system.

The state estimation time upgrades:

y _k＝C _kx _k+D _ku _k(4)

The error covariance time upgrades:

$P_{k+1/k}=A_k{P}_{k/k}{A}_k^T+\mathrm{\Γ}{Q}_k{\mathrm{\Γ}}^T---\left(5\right)$

Kalman gain matrix:

$H_{k+1}=P_{k+1/k}{C}_{k+1}^T/(C_{k+1}{P}_{k+1/k}{C}_{k+1}^T+R_k)---\left(6\right)$

State estimation measurement updaue:

Error covariance measurement updaue:

P _k+1/k+1＝(I-H _k+1C _k+1)P _k+1/k(8)

In the present embodiment, the above variable of normal conditions can not accurately obtain, and debugs after generally arranging initial value according to system testing requirement.Kalman filtering algorithm has good convergence for the uncertain of initial value, can approach within the very short time interval real-valued near.

In step s 12, described SOC initial value assignment is calculated SOC value as ampere-hour integral algorithm, and preserves described SOC value, utilize Kalman filtering algorithm to calculate SOC value simultaneously.

In the present embodiment, the SOC value calculated by ampere-hour integral algorithm after initial value assignment is kept at by EEPROM.In the present embodiment, can not restrain because ampere-hour integral method calculates SOC initial value, therefore using the SOC of Kalman filtering algorithm calculating as ampere-hour integration SOC initial value, after initial assignment, calculate SOC with Kalman filtering algorithm and ampere-hour integral method respectively.

In the present embodiment, electric automobile is applied in battery management system that to calculate the general method therefor of SOC be ampere-hour integral method.Ampere-hour integral method is the most frequently used SOC method of estimation.If discharge and recharge initial state can be determined and be designated as SOC0, so the SOC of current state is:

$SOC={\mathrm{SOC}}_0\±\frac{1}{C_N}\underset{0}{\overset{t}{\∫}}\mathrm{\η}Id\mathrm{\τ}---\left(9\right)$

Wherein C _nfor battery rated capacity, I is battery current, and η is that efficiency for charge-discharge (also claiming coulombic efficiency) charging set efficiency is relevant.If current measurement is forbidden in the application of Ah integral method, will cause the SOC error of calculation, long term accumulation, error is increasing.To battery efficiency be considered in computation process, when the condition of high temperature and current fluctuation violent, error is larger.Ah integral method can be used for all batteries of electric automobile, if current measurement is accurate, has the data of enough estimation initial states, and it is a kind of simple, reliable SOC method of estimation.Relative to open-circuit voltage method long-term storage or leave standstill and obtain open-circuit voltage, ampere-hour integral method is more reliable.

In step s 13, the SOC value that calculated by above-mentioned two kinds of algorithms is judged whether in preset error value.

In the present embodiment, described preset error value is ± 2.5%.In the present invention, Kalman filtering algorithm calculating SOC convergence is good, its SOC value calculated is revised the SOC that ampere-hour integral method calculates, eliminate cumulative errors, measuring error and the noise etc. in ampere-hour integral method computation process, two kinds of methods combine and obtain the SOC value of battery of accurate stable.The present invention can estimate battery SOC accurately, more be conducive to the management of electric automobile to battery, SOC estimates the course continuation mileage that can calculate electric automobile accurately accurately, is convenient to the control of driver for vehicle, is also applicable to the electric automobile applied environment that current fluctuation is violent.

In step S14, if in preset error value, then export the SOC value calculated by ampere-hour integral algorithm.

In the present embodiment, when the SOC value that Kalman filtering algorithm obtains is assigned to ampere-hour integral method SOC value by its error within ± 2.5%, separate computations is continued after assignment, do not stop to judge to revise SOC value in whole computation process, the SOC that output SOC obtains with ampere-hour integral method is as the criterion, and is kept in EEPROM.

In the present embodiment, a kind of SOC estimation method provided by the invention, also comprise circulation step (not shown), namely comprise: if not in prediction error value, described SOC initial value assignment is then continued to calculate SOC value as ampere-hour integral algorithm, and preserve described SOC value, utilize Kalman filtering algorithm to calculate SOC value simultaneously; Continue to judge the SOC value that calculated by above-mentioned two kinds of algorithms whether in preset error value; If in preset error value, then export the SOC value calculated by ampere-hour integral algorithm.

Technical scheme provided by the invention adopts ampere-hour integral method and Kalman filtering algorithm to combine to carry out SOC estimation to battery, both the cumulative errors that ampere-hour integral method causes had been overcome, overcome again Kalman filtering algorithm and occur hopping phenomenon, and by EEPROM, SOC value is preserved, can be reliable and stable obtain accurate SOC value.It is memoryless that ampere-hour integral method calculating SOC does not have historical inheritance, and SOC calculates with battery current and initially SOC is closely bound up; It is support with historical data that Kalman filtering algorithm calculating SOC has historical inheritance, and SOC calculates with the voltage of battery closely bound up, can rapidly converge to exact value with initial value is irrelevant.The present invention in conjunction with these two kinds of methods relative merits and carry out improving and can obtain the SOC value of battery of accurate stable.The method is applicable to the SOC estimation of various electrokinetic cell, can the tracking SOC actual value of dynamic stability compared to additive method the present invention, and the SOC accurate stable being more suitable for electric automobile exports.

The specific embodiment of the invention also provides a kind of SOC valuation system 10, mainly comprises:

Initial Value module 11, calculates SOC initial value for utilizing Kalman filtering algorithm;

Assignment module 12, for described SOC initial value assignment is calculated SOC value as ampere-hour integral algorithm, and preserves described SOC value, utilizes Kalman filtering algorithm to calculate SOC value simultaneously;

Judge module 13, for judging the SOC value that calculated by above-mentioned two kinds of algorithms whether in preset error value;

Output module 14, if in preset error value, then exports the SOC value calculated by ampere-hour integral algorithm.

A kind of SOC valuation system 10 provided by the invention, adopt ampere-hour integral method and Kalman filtering algorithm to combine and SOC estimation is carried out to battery, both the cumulative errors that ampere-hour integral method causes had been overcome, overcome again Kalman filtering algorithm and occur hopping phenomenon, and by EEPROM, SOC value is preserved, can be reliable and stable obtain accurate SOC value.

Refer to Fig. 2, be depicted as the structural representation of SOC valuation system 10 in an embodiment of the present invention.In the present embodiment, SOC valuation system 10 comprises Initial Value module 11, assignment module 12, judge module 13, output module 14 and loop module 15.

Initial Value module 11, calculates SOC initial value for utilizing Kalman filtering algorithm.

Assignment module 12, for described SOC initial value assignment is calculated SOC value as ampere-hour integral algorithm, and preserves described SOC value, utilizes Kalman filtering algorithm to calculate SOC value simultaneously.

In the present embodiment, the SOC value calculated by ampere-hour integral algorithm after initial value assignment is kept at by EEPROM.In the present embodiment, can not restrain because ampere-hour integral method calculates SOC initial value, therefore using the SOC of Kalman filtering algorithm calculating as ampere-hour integration SOC initial value, after initial assignment, calculate SOC with Kalman filtering algorithm and ampere-hour integral method respectively.

Judge module 13, for judging the SOC value that calculated by above-mentioned two kinds of algorithms whether in preset error value.

In the present embodiment, described preset error value is ± 2.5%.In the present invention, Kalman filtering algorithm calculating SOC convergence is good, its SOC value calculated is revised the SOC that ampere-hour integral method calculates, eliminate cumulative errors, measuring error and the noise etc. in ampere-hour integral method computation process, two kinds of methods combine and obtain the SOC value of battery of accurate stable.The present invention can estimate battery SOC accurately, more be conducive to the management of electric automobile to battery, SOC estimates the course continuation mileage that can calculate electric automobile accurately accurately, is convenient to the control of driver for vehicle, is also applicable to the electric automobile applied environment that current fluctuation is violent.

Output module 14, if in preset error value, then exports the SOC value calculated by ampere-hour integral algorithm.

In the present embodiment, when the SOC value that Kalman filtering algorithm obtains is assigned to ampere-hour integral method SOC value by its error within ± 2.5%, separate computations is continued after assignment, do not stop to judge to revise SOC value in whole computation process, the SOC that output SOC obtains with ampere-hour integral method is as the criterion, and is kept in EEPROM.

Loop module 15, if for not in prediction error value, then continues described SOC initial value assignment to calculate SOC value as ampere-hour integral algorithm, and preserves described SOC value, utilize Kalman filtering algorithm to calculate SOC value simultaneously; Continue to judge the SOC value that calculated by above-mentioned two kinds of algorithms whether in preset error value; If in preset error value, then export the SOC value calculated by ampere-hour integral algorithm.

Technical scheme provided by the invention adopts ampere-hour integral method and Kalman filtering algorithm to combine to carry out SOC estimation to battery, both the cumulative errors that ampere-hour integral method causes had been overcome, overcome again Kalman filtering algorithm and occur hopping phenomenon, and by EEPROM, SOC value is preserved, can be reliable and stable obtain accurate SOC value.It is memoryless that ampere-hour integral method calculating SOC does not have historical inheritance, and SOC calculates with battery current and initially SOC is closely bound up; It is support with historical data that Kalman filtering algorithm calculating SOC has historical inheritance, and SOC calculates with the voltage of battery closely bound up, can rapidly converge to exact value with initial value is irrelevant.The present invention in conjunction with these two kinds of methods relative merits and carry out improving and can obtain the SOC value of battery of accurate stable.The method is applicable to the SOC estimation of various electrokinetic cell, can the tracking SOC actual value of dynamic stability compared to additive method the present invention, and the SOC accurate stable being more suitable for electric automobile exports.

It should be noted that in above-described embodiment, included unit is carry out dividing according to function logic, but is not limited to above-mentioned division, as long as can realize corresponding function; In addition, the concrete title of each functional unit, also just for the ease of mutual differentiation, is not limited to protection scope of the present invention.

In addition, one of ordinary skill in the art will appreciate that all or part of step realized in the various embodiments described above method is that the hardware that can carry out instruction relevant by program has come, corresponding program can be stored in a computer read/write memory medium, described storage medium, as ROM/RAM, disk or CD etc.

The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, all any amendments done within the spirit and principles in the present invention, equivalent replacement and improvement etc., all should be included within protection scope of the present invention.

Claims (6)

1. a SOC estimation method, is characterized in that, described method comprises:

Kalman filtering algorithm is utilized to calculate SOC initial value;

Described SOC initial value assignment is calculated SOC value as ampere-hour integral algorithm, and preserves described SOC value, utilize Kalman filtering algorithm to calculate SOC value simultaneously;

Judge the SOC value that calculated by above-mentioned two kinds of algorithms whether in preset error value;

If in preset error value, then export the SOC value calculated by ampere-hour integral algorithm.

2. SOC estimation method as claimed in claim 1, it is characterized in that, described preset error value is ± 2.5%, is kept at the SOC value calculated by ampere-hour integral algorithm after initial value assignment by EEPROM.

3. SOC estimation method as claimed in claim 2, it is characterized in that, described method also comprises:

If not in prediction error value, then continue described SOC initial value assignment to calculate SOC value as ampere-hour integral algorithm, and preserve described SOC value, utilize Kalman filtering algorithm to calculate SOC value simultaneously;

Continue to judge the SOC value that calculated by above-mentioned two kinds of algorithms whether in preset error value;

If in preset error value, then export the SOC value calculated by ampere-hour integral algorithm.

4. a SOC valuation system, is characterized in that, described SOC valuation system comprises:

Initial Value module, calculates SOC initial value for utilizing Kalman filtering algorithm;

Assignment module, for described SOC initial value assignment is calculated SOC value as ampere-hour integral algorithm, and preserves described SOC value, utilizes Kalman filtering algorithm to calculate SOC value simultaneously;

Judge module, for judging the SOC value that calculated by above-mentioned two kinds of algorithms whether in preset error value;

Output module, if in preset error value, then exports the SOC value calculated by ampere-hour integral algorithm.

5. SOC valuation system as claimed in claim 4, it is characterized in that, described preset error value is ± 2.5%, is kept at the SOC value calculated by ampere-hour integral algorithm after initial value assignment by EEPROM.

6. SOC valuation system as claimed in claim 5, it is characterized in that, described SOC valuation system also comprises:

Loop module, if for not in prediction error value, then continues described SOC initial value assignment to calculate SOC value as ampere-hour integral algorithm, and preserves described SOC value, utilize Kalman filtering algorithm to calculate SOC value simultaneously; Continue to judge the SOC value that calculated by above-mentioned two kinds of algorithms whether in preset error value; If in preset error value, then export the SOC value calculated by ampere-hour integral algorithm.