CN103439603A - Method and device for detecting charge state of super-capacitor energy storage device - Google Patents

Abstract

The invention discloses a method for detecting the charge state of a super-capacitor energy storage device. The method comprises the steps of building a single body state space model according to a super-capacitor singe body circuit model, collecting the current value and the terminal voltage value in real time when each super-capacitor singe body works, and working out the SOC value of each super-capacitor singe body through an extended Kalman filtering algorithm according to the single body state space model and the current value and the terminal voltage value when each super-capacitor singe body works. The invention further discloses a device for detecting the charge state of the super-capacitor energy storage device, and the detection accuracy of the SOC values can be improved through the technology.

Description

A kind of super capacitor energy storage device charge state detection method and device

Technical field

The present invention relates to the super capacitor energy-storage technical field, relate in particular to a kind of super capacitor energy storage device charge state detection method and device.

Background technology

Ultracapacitor is again double layer capacitor (Electrical Double-Layer Capacitor), by the polarization electrolyte, carrys out energy storage, is a kind of Novel energy storage apparatus.Can repeated charge hundreds thousand of times of ultracapacitor, have that short, long service life of duration of charging, power density are high, characteristics such as good temp characteristic, the saving energy and environmental protection.Because ultracapacitor has more outstanding advantage, thus its in increasing field, be widely applied, for example hybrid vehicle, accumulator system, intelligent grid, Aero-Space etc.

Because ultracapacitor monomer voltage lower (generally being no more than 4V), energy density are low, therefore the bank of super capacitors for accumulator system is usually to be combined by several ultracapacitor monomer connection in series-parallel.In practical application, if the improper use to ultracapacitor, such as overcharging or the abnormal conditions such as unreasonable are controlled in overdischarge, temperature, can cause the phenomenon of the interior electrolyte leakage of ultracapacitor and even blast to occur, for avoiding dangerous situation, the state-of-charge (State of Charge, SOC) of necessary Real-Time Monitoring ultracapacitor monomer and bank of super capacitors, prevent that it from overcharging or overdischarge.

The detection of traditional ultracapacitor monomer SOC mainly adopts the ampere-hour method, the ampere-hour method must know that the SOC initial value of ultracapacitor monomer just can record ultracapacitor SOC actual value in advance, but can't accurately determine again the SOC initial value of ultracapacitor monomer, ultracapacitor monomer capacity is little in addition, the capacity integral error is remarkable, can cause the SOC computational accuracy lower, and for whole bank of super capacitors, the accumulative total of this error can enlarge, and makes the SOC result of final calculating and actual conditions that very large difference be arranged.

Summary of the invention

The invention provides a kind of super capacitor energy storage device charge state detection method and device, can improve the computational accuracy of SOC.

The invention provides a kind of super capacitor energy storage device charge state detection method, comprising:

Set up the free state spatial model according to ultracapacitor monomer circuit model;

Current value and terminal voltage value during each ultracapacitor monomer work of Real-time Collection;

Current value and terminal voltage value during according to described free state spatial model, described each ultracapacitor monomer work, utilize expanded Kalman filtration algorithm to calculate the SOC value of each ultracapacitor monomer.

Preferably, describedly according to ultracapacitor monomer circuit model, set up the free state spatial model, be specially:

Ultracapacitor monomer circuit model is carried out to parameter identification, obtain state equation and the output equation of ultracapacitor monomer according to ultracapacitor monomer circuit model, wherein, described ultracapacitor monomer circuit model is the first order nonlinear equivalent-circuit model.

Preferably, described state equation and output equation are specially:

$\left[\frac{\mathrm{SOC}(t+1)}{U_c(t+1)}\right]=\left[\begin{array}{cc}1& 0\\ 0& \exp (-\frac{\mathrm{\Δt}}{R_b(C_a+C_b*U_c)})\end{array}\right]\left[\begin{array}{c}\mathrm{SOC}\left(t\right)\\ U_c\left(t\right)\end{array}\right]+\left[\begin{array}{c}-\frac{\mathrm{\Δt}}{Q_0}\\ R_b(1-\exp (-\frac{\mathrm{\Δt}}{R_b(C_a+C_b*U_c)}))\end{array}\right]I\left(t\right)+w\left(t\right)---\left(5\right);$

U(t)＝F[SOC(t)]+R _aI(t)+v(t) (6)；

Wherein, SOC (t) means discrete state t ultracapacitor state-of-charge constantly, U _c(t) mean the discrete state t voltage of the inner equivalent capacity of ultracapacitor monomer constantly, Δ t is sampling interval, equivalent internal resistance R _a, parallel resistance R _b, constant capacitor C _a, variable capacitance C _b, Q ₀for the ultracapacitor rated capacity of dispatching from the factory, I (t) is discrete state t ultracapacitor monomer charging and discharging currents constantly, U (t) is discrete state t ultracapacitor monomer terminal voltage constantly, F[SOC (t)] be the nonlinear function of the voltage of ultracapacitor SOC and equivalent capacity, w (t) is the interference of immesurable stochastic variable to quantity of state, and v (t) is the measurement noise of ultracapacitor terminal voltage.

Preferably, described expanded Kalman filtration algorithm is specially:

Order

$A_t=\left[\begin{array}{cc}1& 0\\ 0& \exp (-\frac{\mathrm{\Δt}}{R_b(C_a+C_b*U_c)})\end{array}\right];$

$x_t=\left[\begin{array}{c}\mathrm{SOC}\left(t\right)\\ U_c\left(t\right)\end{array}\right];$

$B_t=\left[\begin{array}{c}-\frac{\mathrm{\Δt}}{Q_0}\\ R_b(1-\exp (-\frac{\mathrm{\Δt}}{R_b(C_a+C_b*U_c)}))\end{array}\right];$

u _t＝I(t)；

$C_t=\left[\begin{array}{cc}{\frac{\mathrm{dF}\left({\mathrm{SOC}}_t\right)}{\mathrm{dSOC}}|}_{\mathrm{SOC}={\mathrm{SOC}}_t}& 0\end{array}\right];$

y _t＝U(t)；

Wherein,

mean F[SOC (t)] to the derivative of ultracapacitor monomer state-of-charge SOC, then get SOC=SOC (t), the result finally calculated;

Described formula (5) and described formula (6) can be rewritten as:

State equation: x _t+1=A _tx _t+ B _tu _t+ w (t) (7);

Output equation: y _t=C _tx _t+ R _au _t+ v (t) (8);

Order

f(x _t，u _t)＝A _tx _t+B _tu _t，g(x _t，u _t)＝C _tx _t+R _au _t，

Described formula (7) and described formula (8) can be rewritten as:

State equation: x _t+1=f (x _t, u _t)+w (t) (9);

Output equation: y _t=g (x _t, u _t)+v (t) (10);

Step 1: described formula (9) and described formula (10) state variable x and mean square deviation error P are carried out to initialization;

Step 2: state variable x in step 1 is carried out to the prediction of expanded Kalman filtration algorithm state variable and estimate to calculate;

Step 3: mean square deviation error P in step 1 is carried out to expanded Kalman filtration algorithm mean square deviation error prediction and estimate to calculate;

Step 4: mean square deviation error P in step 3 is carried out to kalman gain calculating;

Step 5: the prediction of state variable x in step 2 is estimated to carry out to the optimal estimation calculating of expanded Kalman filtration algorithm state variable x;

Step 6: the prediction of mean square deviation error P in step 3 is estimated to carry out to the optimal estimation calculating of expanded Kalman filtration algorithm mean square deviation error P;

Step 7: judge whether the SOC value restrains and the invariant state that tends towards stability, if so, finish the SOC computation process of this monomer; If not, return to the SOC value of next moment monomer of step 2 iterative computation until result of calculation converges to SOC value the state that tends towards stability of ultracapacitor monomer accurately the most at last.

Preferably, also comprise:

Judge that whole bank of super capacitors is in charging duty or electric discharge duty;

According to the SOC value of duty and each ultracapacitor monomer, obtain the SOC value of each series super capacitor bank;

According to the SOC value of described each series super capacitor bank, obtain the SOC value of whole bank of super capacitors.

Preferably, the described SOC value according to duty and each ultracapacitor monomer, obtain the SOC value of each series super capacitor bank, is specially:

If current bank of super capacitors is the charging duty, in the SOC value series connection group for this reason of current series super capacitor bank one of the SOC value of ultracapacitor monomer maximum; If current bank of super capacitors is the electric discharge duty, in the SOC value series connection group for this reason of current series super capacitor bank one of the SOC value of ultracapacitor monomer minimum.

Preferably, after current value and terminal voltage value when each ultracapacitor monomer work of Real-time Collection, also comprise:

When the time of terminal voltage value in default low state A has surpassed predefined time threshold T1, or, when the time of terminal voltage value in default high state of value B has surpassed predefined time threshold T2, start early warning mechanism.

The present invention also provides a kind of super capacitor energy storage device state-of-charge pick-up unit, comprising:

Set up free state spatial model module, for according to ultracapacitor monomer circuit model, setting up the free state spatial model;

Acquisition module, current value and terminal voltage value during for each ultracapacitor monomer work of Real-time Collection;

Computing module, with the described free state spatial model module of setting up, with described acquisition module, be connected, current value and terminal voltage value while being used for according to free state spatial model, each ultracapacitor monomer work, utilize expanded Kalman filtration algorithm to calculate the SOC value of each ultracapacitor monomer.

Preferably, also comprise:

The judgement block of state, be connected with described computing module, for judging whole bank of super capacitors, is in charging duty or electric discharge duty;

Computing module, also for the SOC value according to duty and each ultracapacitor monomer, obtain the SOC value of each series super capacitor bank, according to the SOC value of each series super capacitor bank, obtains the SOC value of whole bank of super capacitors.

Preferably, also comprise:

Alarm module, with described acquisition module, be connected, for when the time of terminal voltage value in default low state A has surpassed predefined time threshold T1, or, when the time of terminal voltage value in default high state of value B has surpassed predefined time threshold T2, start early warning mechanism.

A kind of super capacitor energy storage device charge state detection method provided by the invention and device, by according to ultracapacitor monomer circuit model, setting up the free state spatial model, then utilize expanded Kalman filtration algorithm to calculate the SOC value of each ultracapacitor monomer, such computing method precision is high, and the SOC value of each the ultracapacitor monomer calculated is more accurate.And adopt expanded Kalman filtration algorithm to carry out ultracapacitor SOC calculating, by " revise-prediction of estimate-gain of prediction is estimated;; the closed loop that can realize ultracapacitor SOC detects; overcome the shortcoming that cumulative errors is arranged in the prior art; result of calculation is more accurate; and also each monomer in whole bank of super capacitors all can not overcharge or the bad phenomenon such as overdischarge, thereby extend the serviceable life of super capacitor energy storage device, guarantee the use safety of whole super capacitor energy storage device.

The accompanying drawing explanation

In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, below will the accompanying drawing of required use in embodiment or description of the Prior Art be briefly described, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skills, under the prerequisite of not paying creative work, can also obtain according to these accompanying drawings other accompanying drawing.

Fig. 1 is a kind of super capacitor energy storage device charge state detection method schematic flow sheet of the embodiment of the present invention;

Fig. 2 is the schematic flow sheet of another embodiment of a kind of super capacitor energy storage device charge state detection method of the present invention;

Fig. 3 is the first order nonlinear circuit model figure of the ultracapacitor monomer of the embodiment of the present invention;

Fig. 4 is the ultracapacitor connection in series-parallel circuit model figure in groups of the embodiment of the present invention;

Fig. 5 is the structural representation of a kind of super capacitor energy storage device state-of-charge pick-up unit of the embodiment of the present invention.

Embodiment

In order to make technical matters to be solved by this invention, technical scheme and beneficial effect clearer, below in conjunction with drawings and Examples, the present invention is further elaborated.

Refer to a kind of super capacitor energy storage device charge state detection method schematic flow sheet of the embodiment of the present invention shown in Fig. 1, comprising:

Step S101: according to ultracapacitor monomer circuit model, set up the free state spatial model.

Concrete, at first ultracapacitor monomer circuit model is carried out to parameter identification, obtain state equation and the output equation of ultracapacitor monomer according to ultracapacitor monomer circuit model, wherein, ultracapacitor monomer circuit model is the first order nonlinear equivalent-circuit model.

Step S102: current value and terminal voltage value during each ultracapacitor monomer work of Real-time Collection.

Step S103: current value and terminal voltage value during according to free state spatial model, each ultracapacitor monomer work, utilize expanded Kalman filtration algorithm to calculate the SOC value of each ultracapacitor monomer.

Concrete, current value and terminal voltage value during according to the state equation of ultracapacitor monomer and output equation, each ultracapacitor monomer work, adopt expanded Kalman filtration algorithm to calculate the SOC value of each ultracapacitor monomer.

Implement above-described embodiment, by according to ultracapacitor monomer circuit model, setting up the free state spatial model, then utilize expanded Kalman filtration algorithm to calculate the SOC value of each ultracapacitor monomer, such computing method precision is high, and the SOC value of each the ultracapacitor monomer calculated is more accurate.

Below in conjunction with the schematic flow sheet of another embodiment of a kind of super capacitor energy storage device charge state detection method of the present invention shown in Fig. 2, further describe a kind of super capacitor energy storage device charge state detection method of the embodiment of the present invention, comprising:

Step S201: ultracapacitor monomer circuit model is carried out to parameter identification.

Concrete, circuit as shown in Figure 3, this circuit is the first order nonlinear equivalent-circuit model.This model has not only been considered the impact on SOC value computational accuracy that the variation of the equivalent capacity that causes because the ultracapacitor terminal voltage is different brings, also considered the impact on SOC value computational accuracy that ultracapacitor self discharge effect phenomenon is brought, such design is more accurate when describing the ultracapacitor charge and discharge process, and the SOC value of utilizing this model to calculate the ultracapacitor monomer will be more accurate.Data that can provide by producer or oneself measure equivalent internal resistance R _a, parallel resistance R _b, constant capacitor C _a, variable capacitance C _bparameter, U _cthe voltage of the inner equivalent capacity of ultracapacitor monomer, wherein variable capacitance C _bvalue according to U _cvariation and change.

Step S202: set up the free state spatial model according to ultracapacitor monomer circuit model, and formula is arranged and carry out discretize, obtain state equation and the output equation of ultracapacitor monomer.

Concrete, after identification of Model Parameters, set up the free state spatial model, need to obtain state equation and the output equation of ultracapacitor monomer.The circuit relationships equation of describing according to Fig. 3 is as follows:

U(t)＝R _aI(t)+U _c(t) (1)

$I\left(t\right)=(C_a+C_b*U_c)\frac{{\mathrm{dU}}_c\left(t\right)}{\mathrm{dt}}+\frac{U_c\left(t\right)}{R_b}---\left(2\right)$

Wherein, I (t) is discrete state t ultracapacitor monomer charging and discharging currents constantly; At C _b* U _cin, unit is the unit of electric capacity, U _cnumerical value only is provided, means C _bwith U _cvariation and change.

The nonlinear equation that can also obtain ultracapacitor monomer SOC is as follows:

$\mathrm{SOC}\left(t\right)=\mathrm{SOC}\left(t_0\right)-\frac{1}{Q_0}{\∫}_0^{t}I\left(t\right)\mathrm{dt}---\left(3\right)$

U _c(t)＝F[SOC(t)] (4)

Wherein, SOC (t ₀) expression initial time ultracapacitor state-of-charge; SOC (t) means discrete state t ultracapacitor state-of-charge constantly; Q ₀it is the ultracapacitor rated capacity of dispatching from the factory; F[SOC (t)] be the nonlinear function of the voltage of ultracapacitor SOC and equivalent capacity.

Concrete, formula is arranged and to carry out after discretize obtaining the state equation of ultracapacitor monomer as follows:

$\left[\frac{\mathrm{SOC}(t+1)}{U_c(t+1)}\right]=\left[\begin{array}{cc}1& 0\\ 0& \exp (-\frac{\mathrm{\Δt}}{R_b(C_a+C_b*U_c)})\end{array}\right]\left[\begin{array}{c}\mathrm{SOC}\left(t\right)\\ U_c\left(t\right)\end{array}\right]+\left[\begin{array}{c}-\frac{\mathrm{\Δt}}{Q_0}\\ R_b(1-\exp (-\frac{\mathrm{\Δt}}{R_b(C_a+C_b*U_c)}))\end{array}\right]I\left(t\right)+w\left(t\right)---\left(5\right)$

Output equation is as follows:

U(t)＝F[SOC(t)]+R _aI(t)+v(t) (6)

Wherein, Δ t is sampling interval; R _b(C _a+ C _b* U _c) be to discharge and recharge link capacitance resistance time parameter; W (t) is the interference of immesurable stochastic variable to quantity of state, and v (t) is the measurement noise of ultracapacitor terminal voltage, supposes that disturbance variable w (t) and v (t) are white Gaussian noise.

Order

$A_t=\left[\begin{array}{cc}1& 0\\ 0& \exp (-\frac{\mathrm{\Δt}}{R_b(C_a+C_b*U_c)})\end{array}\right];$

$x_t=\left[\begin{array}{c}\mathrm{SOC}\left(t\right)\\ U_c\left(t\right)\end{array}\right];$

$B_t=\left[\begin{array}{c}-\frac{\mathrm{\Δt}}{Q_0}\\ R_b(1-\exp (-\frac{\mathrm{\Δt}}{R_b(C_a+C_b*U_c)}))\end{array}\right];$

u _t＝I(t)；

$C_t=\left[\begin{array}{cc}{\frac{\mathrm{dF}\left({\mathrm{SOC}}_t\right)}{\mathrm{dSOC}}|}_{\mathrm{SOC}={\mathrm{SOC}}_t}& 0\end{array}\right];$

y _t＝U(t)；

Wherein,

mean F[SOC (t)] to the derivative of ultracapacitor monomer state-of-charge SOC, then get SOC=SOC (t), the result finally calculated.

The state equation of ultracapacitor monomer and output equation, formula (5) and formula (6) can be rewritten as:

State equation: x _t+1=A _tx _t+ B _tu _t+ w (t) (7)

Output equation: y _t=C _tx _t+ R _au _t+ v (t) (8)

Wherein, system input variable u _tthe charging and discharging currents of ultracapacitor, i.e. current value during certain ultracapacitor monomer work; y _tfor the output variable of system, the terminal voltage value while being certain ultracapacitor monomer work; x _tfor the state variable of system, ultracapacitor SOC value is contained in wherein, obtains x _tobtain the SOC value of certain ultracapacitor.

Step S203: current value and terminal voltage value during each ultracapacitor monomer work of Real-time Collection.

Concrete, I (t) and U (t) while gathering each ultracapacitor monomer work.When the time of U (t) magnitude of voltage in low state A (such as the A=0 volt) has surpassed predefined time threshold T1, or when the time of U (t) magnitude of voltage in high state of value B (such as the B specified ceiling voltage that is monomer) has surpassed predefined time threshold T2, start early warning mechanism, can carry out the power-off service operation by the mode notification technique personnel of alarm bell or note.Can further make like this each monomer in whole bank of super capacitors all can not overcharge or the bad phenomenon such as overdischarge.

Step S204: current value and terminal voltage value during according to the state equation of ultracapacitor monomer and output equation, each ultracapacitor monomer work adopt expanded Kalman filtration algorithm to calculate the SOC value of each ultracapacitor monomer.

Concrete,

Order

f(x _t，u _t)＝A _tx _t+B _tu _t，g(x _t，u _t)＝C _t x _t+R _au _t，

Respectively that system is with (x _t, u _t) be nonlinear state variable relation function and the non-linear measurement relation function of independent variable.

System state space model, formula (7) and formula (8) can be rewritten as:

State equation: x _t+1=f (x _t, u _t)+w (t) (9)

Output equation: y _t=g (x _t, u _t)+v (t) (10)

For each sampling t state variable x and mean square deviation error P constantly, expanded Kalman filtration algorithm all can be made twice different estimation.The mean square deviation error P of take is example, before the spreading kalman gain is calculated, predicts for the first time estimated value

complete and (predict for the first time estimated value at previous moment (being the t-1 moment) mean square deviation error amount

basis on, utilize mean square deviation error prediction estimate equation downward constantly (be t constantly) recursion obtain), with upper right mark "-", mean; After the spreading kalman gain is calculated, to predicting for the first time estimated value

the correction that gains, then obtain optimal estimation value for the second time

with upper right mark "+", mean.

The EKF equation is as follows, and wherein E is unit matrix, upper right mark " T " representing matrix transposition, and upper right mark " 1 " representing matrix is inverted.

Step 1: state variable x and mean square deviation error P are carried out to initialization.

$x_0^{+}=E\left[x_0\right]---\left(11\right)$

$P_0^{+}=E\left[(x_0-x_0^{+}){(x_0-x_0^{+})}^T\right]---\left(12\right)$

Step 2: state variable x in step 1 is carried out to the prediction of expanded Kalman filtration algorithm state variable and estimate to calculate.

$x_t^{-}=f(x_{t-1}^{+},u_{t-1})=A_{t-1}{x}_{t-1}^{+}+B_{t-1}{u}_{t-1}---\left(13\right)$

Step 3: mean square deviation error P in step 1 is carried out to expanded Kalman filtration algorithm mean square deviation error prediction and estimate to calculate.Wherein, D _wthe variance that means w (t).

$P_t^{-}=A_{t-1}{P}_{t-1}^{+}{A}_{t-1}^T+D_w---\left(14\right)$

Step 4: mean square deviation error P in step 3 is carried out to kalman gain calculating.Wherein, M _tmean the result that gain is calculated, D _vthe variance that means v (t).

$M_t=P_t^{-}{C}_t^T{(C_t{P}_t^{-}{C}_t^T+D_v)}^{-1}---\left(15\right)$

Step 5: the prediction of state variable x in step 2 is estimated to carry out to the optimal estimation calculating of expanded Kalman filtration algorithm state variable x.

$x_t^{+}=x_t^{-}+M_t[y_t-g(x_t^{-},u_t)]---\left(16\right)$

Step 6: the prediction of mean square deviation error P in step 3 is estimated to carry out to the optimal estimation calculating of expanded Kalman filtration algorithm mean square deviation error P.

$P_t^{+}=(E-M_t{C}_t)P_t^{-}---\left(17\right)$

Step 7: judge whether the SOC value restrains and the invariant state that tends towards stability.If so, finish the SOC computation process of this monomer; If not, return to the SOC value of next moment monomer of step 2 iterative computation, t=t+1, until result of calculation converges to SOC value the state that tends towards stability of ultracapacitor monomer accurately the most at last.

Step S205: judge that whole bank of super capacitors is in charging duty or electric discharge duty.

Concrete, circuit diagram as shown in Figure 4, total m the parallel branch of this integral body bank of super capacitors one, each parallel branch the inside has n ultracapacitor monomer series-connected, amounts to m*n ultracapacitor monomer, and the embodiment of the present invention does not limit the concrete numerical value of m and n.

Determine that duty is to adopt the judgement switching value to realize, i.e. the charged state switch charged state that is in place; The discharge condition switch discharge condition that is in place.

Step S206: according to the SOC value of duty and each ultracapacitor monomer, obtain the SOC value of each series super capacitor bank.

Concrete, or the circuit diagram of take as shown in Figure 4 is example, with C _xymean arbitrary ultracapacitor monomer, SOC _xymean the SOC value of arbitrary ultracapacitor monomer, use respectively SOC ₁, SOC ₂..., SOC _mthe SOC value of expression bank of super capacitors from 1 to m parallel branch.

If current bank of super capacitors is the charging duty, the SOC value of current series super capacitor bank is SOC _max, C for example ₁₁to C _1nthis series connection group, if be now the charging duty, find out SOC ₁₁to SOC _1nmiddle maximal value is the SOC of this series connection group ₁; If current ultracapacitor is the electric discharge duty, the SOC value of current series super capacitor bank is SOC _min, C for example _m1to C _mnthis series connection group, if be now the electric discharge duty, find out SOC _m1to SOC _mnmiddle minimum value is the SOC of this series connection group _m.

Step S207: according to the SOC value of each series super capacitor bank, obtain the SOC value of whole bank of super capacitors.

Concrete, the SOC value of each series super capacitor bank is averaged to summation, can obtain the SOC value of whole bank of super capacitors, $\mathrm{SOC}=\frac{{\mathrm{SOC}}_1+{\mathrm{SOC}}_2+......+{\mathrm{SOC}}_m}{m}.$

The invention process above-described embodiment, adopt expanded Kalman filtration algorithm to carry out ultracapacitor SOC calculating, can realize that by " revise-prediction of estimate-gain of prediction is estimated " closed loop of ultracapacitor SOC detects, overcome the shortcoming that cumulative errors is arranged in the prior art, result of calculation is more accurate, and each monomer in whole bank of super capacitors all can not overcharge or the bad phenomenon such as overdischarge, thereby extend the serviceable life of super capacitor energy storage device, guarantee the use safety of whole super capacitor energy storage device.

Below in conjunction with the structural representation of a kind of super capacitor energy storage device state-of-charge pick-up unit of the embodiment of the present invention shown in Fig. 5, further describe the structure of this super capacitor energy storage device state-of-charge pick-up unit of the embodiment of the present invention.

Set up free state spatial model module 501, for according to ultracapacitor monomer circuit model, setting up the free state spatial model.

Concrete, at first ultracapacitor monomer circuit model is carried out to parameter identification, obtain state equation and the output equation of ultracapacitor monomer according to ultracapacitor monomer circuit model.

Circuit as shown in Figure 3, this circuit is the first order nonlinear equivalent-circuit model.This model has not only been considered the impact on SOC value computational accuracy that the variation of the equivalent capacity that causes because the ultracapacitor terminal voltage is different brings, also considered the impact on SOC value computational accuracy that ultracapacitor self discharge effect phenomenon is brought, such design is more accurate when describing the ultracapacitor charge and discharge process, and the SOC value of utilizing this model to calculate the ultracapacitor monomer will be more accurate.Data that can provide by producer or oneself measure equivalent internal resistance R _a, parallel resistance R _b, constant capacitor C _a, variable capacitance C _bparameter, U _cthe voltage of the inner equivalent capacity of ultracapacitor monomer, wherein variable capacitance C _bvalue according to U _cvariation and change.

After identification of Model Parameters, set up the free state spatial model, need to obtain state equation and the output equation of ultracapacitor monomer.The circuit relationships equation of describing according to Fig. 3 is as follows:

U(t)＝R _aI(t)+U _c(t) (1)

$I\left(t\right)=(C_a+C_b*U_C)\frac{{\mathrm{dU}}_c\left(t\right)}{\mathrm{dt}}+\frac{U_c\left(t\right)}{R_b}---\left(2\right)$

Wherein, I (t) is discrete state t ultracapacitor monomer charging and discharging currents constantly; At C _b* U _cin, unit is the unit of electric capacity, U _cnumerical value only is provided, means C _bwith U _cvariation and change.

The nonlinear equation that can also obtain ultracapacitor monomer SOC is as follows:

$\mathrm{SOC}\left(t\right)=\mathrm{SOC}\left(t_0\right)-\frac{1}{Q_0}{\∫}_0^{t}I\left(t\right)\mathrm{dt}---\left(3\right)$

U _C(t)＝F[SOC(t)] (4)

Wherein, SOC (t ₀) expression initial time ultracapacitor state-of-charge; SOC (t) means discrete state t ultracapacitor state-of-charge constantly; Q ₀it is the ultracapacitor rated capacity of dispatching from the factory; F[SOC (t)] be the nonlinear function of the voltage of ultracapacitor SOC and equivalent capacity.

Formula is arranged and to carry out after discretize obtaining the state equation of ultracapacitor monomer as follows:

$\left[\frac{\mathrm{SOC}(t+1)}{U_c(t+1)}\right]=\left[\begin{array}{cc}1& 0\\ 0& \exp (-\frac{\mathrm{\Δt}}{R_b(C_a+C_b*U_c)})\end{array}\right]\left[\begin{array}{c}\mathrm{SOC}\left(t\right)\\ U_c\left(t\right)\end{array}\right]+\left[\begin{array}{c}-\frac{\mathrm{\Δt}}{Q_0}\\ R_b(1-\exp (-\frac{\mathrm{\Δt}}{R_b(C_a+C_b*U_c)}))\end{array}\right]I\left(t\right)+w\left(t\right)---\left(5\right)$

Output equation is as follows:

U(t)＝F[SOC(t)]+R _aI(t)+v(t) (6)

Wherein, Δ t is sampling interval; R _b(C _a+ C _b* U _c) be to discharge and recharge link capacitance resistance time parameter; W (t) is the interference of immesurable stochastic variable to quantity of state, and v (t) is the measurement noise of ultracapacitor terminal voltage, supposes that disturbance variable w (t) and v (t) are white Gaussian noise.

Order

$A_t=\left[\begin{array}{cc}1& 0\\ 0& \exp (-\frac{\mathrm{\Δt}}{R_b(C_a+C_b*U_c)})\end{array}\right];$

$x_t=\left[\begin{array}{c}\mathrm{SOC}\left(t\right)\\ U_c\left(t\right)\end{array}\right];$

$B_t=\left[\begin{array}{c}-\frac{\mathrm{\Δt}}{Q_0}\\ R_b(1-\exp (-\frac{\mathrm{\Δt}}{R_b(C_a+C_b*U_c)}))\end{array}\right];$

u _t＝I(t)；

$C_t=\left[\begin{array}{cc}{\frac{\mathrm{dF}\left({\mathrm{SOC}}_t\right)}{\mathrm{dSOC}}|}_{\mathrm{SOC}={\mathrm{SOC}}_t}& 0\end{array}\right];$

y _t＝U(t)；

Wherein,

mean F[SOC (t)] to the derivative of ultracapacitor monomer state-of-charge SOC, then get SOC=SOC (t), the result finally calculated.

The state equation of ultracapacitor monomer and output equation, formula (5) and formula (6) can be rewritten as:

State equation: x _t+1=A _tx _t+ B _tu _t+ w (t) (7)

Output equation: y _t=C _tx _t+ R _au _t+ v (t) (8)

Wherein, system input variable u _tthe charging and discharging currents of ultracapacitor, i.e. current value during certain ultracapacitor monomer work; y _tfor the output variable of system, the terminal voltage value while being certain ultracapacitor monomer work; x _tfor the state variable of system, ultracapacitor SOC value is contained in wherein, obtains x _tobtain the SOC value of certain ultracapacitor.

Acquisition module 502, current value and terminal voltage value during for each ultracapacitor monomer work of Real-time Collection.

Concrete, I (t) and U (t) while gathering each ultracapacitor monomer work.

Computing module 503, with set up free state spatial model module 501 and be connected with acquisition module 502, current value and terminal voltage value while being used for according to free state spatial model, each ultracapacitor monomer work, utilize expanded Kalman filtration algorithm to calculate the SOC value of each ultracapacitor monomer.

Concrete, current value and terminal voltage value during according to the state equation of ultracapacitor monomer and output equation, each ultracapacitor monomer work, adopt expanded Kalman filtration algorithm to calculate the SOC value of each ultracapacitor monomer.

Concrete,

Order

f(x _t，u _t)＝A _tx _t+B _tu _t，g(x _t，u _t)＝C _t x _t+R _au _t，

Respectively that system is with (x _t, u _t) be nonlinear state variable relation function and the non-linear measurement relation function of independent variable.

System state space model, formula (7) and formula (8) can be rewritten as:

State equation: x _t+1=f (x _t, u _t)+w (t) (9)

Output equation: y _t=g (x _t, u _t)+v (t) (10)

For each sampling t state variable x and mean square deviation error P constantly, expanded Kalman filtration algorithm all can be made twice different estimation.The mean square deviation error P of take is example, before the spreading kalman gain is calculated, predicts for the first time estimated value complete and (predict for the first time estimated value

at previous moment (being the t-1 moment) mean square deviation error amount

basis on, utilize mean square deviation error prediction estimate equation downward constantly (be t constantly) recursion obtain), with upper right mark "-", mean; After the spreading kalman gain is calculated, to predicting for the first time estimated value

the correction that gains, then obtain optimal estimation value for the second time

with upper right mark "+", mean.

The EKF equation is as follows, and wherein E is unit matrix, upper right mark " T " representing matrix transposition, and upper right mark " 1 " representing matrix is inverted.

State variable x and mean square deviation error P are carried out to initialization.

$x_0^{+}=E\left[x_0\right]---\left(11\right)$

$P_0^{+}=E\left[(x_0-x_0^{+}){(x_0-x_0^{+})}^T\right]---\left(12\right)$

State variable x in formula (11) is carried out to the prediction of expanded Kalman filtration algorithm state variable to be estimated to calculate.

$x_t^{-}=f(x_{t-1}^{+},u_{t-1})=A_{t-1}{x}_{t-1}^{+}+B_{t-1}{u}_{t-1}---\left(13\right)$

Mean square deviation error P in formula (12) is carried out to expanded Kalman filtration algorithm mean square deviation error prediction to be estimated to calculate.Wherein, D _wthe variance that means w (t).

$P_t^{-}=A_{t-1}{P}_{t-1}^{+}{A}_{t-1}^T+D_w---\left(14\right)$

Mean square deviation error P in formula (14) is carried out to kalman gain calculating.Wherein, M _tmean the result that gain is calculated, D _vthe variance that means v (t).

$M_t=P_t^{-}{C}_t^T{(C_t{P}_t^{-}{C}_t^T+D_v)}^{-1}---\left(15\right)$

The prediction of state variable x in formula (13) is estimated to carry out to the optimal estimation calculating of expanded Kalman filtration algorithm state variable x.

$x_t^{+}=x_t^{-}+M_t[y_t-g(x_t^{-},u_t)]---\left(16\right)$

The prediction of mean square deviation error P in formula (14) is estimated to carry out to the optimal estimation calculating of expanded Kalman filtration algorithm mean square deviation error P.

$P_t^{+}=(E-M_t{C}_t)P_t^{-}---\left(17\right)$

Judge whether the SOC value restrains and the invariant state that tends towards stability.If so, finish the SOC computation process of this monomer; If not, return to the SOC value of next moment monomer of step 2 iterative computation, t=t+1, until result of calculation converges to SOC value the state that tends towards stability of ultracapacitor monomer accurately the most at last.

Preferably, also comprise:

Judgement block of state 504, be connected with computing module 503, for judging whole bank of super capacitors, is in charging duty or electric discharge duty.

Concrete, circuit diagram as shown in Figure 4, total m the parallel branch of this integral body bank of super capacitors one, each parallel branch the inside has n ultracapacitor monomer series-connected, amounts to m*n ultracapacitor monomer, and the embodiment of the present invention does not limit the concrete numerical value of m and n.

Concrete, determine that duty is to adopt the judgement switching value to realize, i.e. the charged state switch charged state that is in place; The discharge condition switch discharge condition that is in place.

Computing module 503, also for the SOC value according to duty and each ultracapacitor monomer, obtain the SOC value of each series super capacitor bank.

Concrete, or the circuit diagram of take as shown in Figure 4 is example, with C _xymean arbitrary ultracapacitor monomer, SOC _xymean the SOC value of arbitrary ultracapacitor monomer, use respectively SOC ₁, SOC ₂..., SOC _mthe SOC value of expression bank of super capacitors from 1 to m parallel branch.

If current bank of super capacitors is the charging duty, the SOC value of current series super capacitor bank is SOC _max, C for example ₁₁to C _1nthis series connection group, if be now the charging duty, find out SOC ₁₁to SOC _1nmiddle maximal value is the SOC of this series connection group ₁; If current ultracapacitor is the electric discharge duty, the SOC value of current series super capacitor bank is SOC _min, C for example _m1to C _mnthis series connection group, if be now the electric discharge duty, find out SOC _m1to SOC _mnmiddle minimum value is the SOC of this series connection group _m.

Computing module 503, also for the SOC value according to each series super capacitor bank, obtain the SOC value of whole bank of super capacitors.

Concrete, the SOC value of each series super capacitor bank is averaged to summation, can obtain the SOC value of whole bank of super capacitors, $\mathrm{SOC}=\frac{{\mathrm{SOC}}_1+{\mathrm{SOC}}_2+......+{\mathrm{SOC}}_m}{m}.$

This device also comprises:

Alarm module 505, with acquisition module 502, be connected, for when the time of U (t) magnitude of voltage in low state (such as 0 volt) has surpassed predefined time threshold T1, or, when the time of U (t) magnitude of voltage in high state of value (such as the specified ceiling voltage of monomer) has surpassed predefined time threshold T2, start early warning mechanism.

The invention process above-described embodiment, adopt expanded Kalman filtration algorithm to carry out ultracapacitor SOC calculating, can realize that by " revise-prediction of estimate-gain of prediction is estimated " closed loop of ultracapacitor SOC detects, overcome the shortcoming that cumulative errors is arranged in the prior art, result of calculation is more accurate, and each monomer in whole bank of super capacitors all can not overcharge or the bad phenomenon such as overdischarge, thereby extend the serviceable life of super capacitor energy storage device, guarantee the use safety of whole super capacitor energy storage device.

It should be noted that, through the above description of the embodiments, those skilled in the art can be well understood to the mode that the present invention can add essential hardware platform by software and realize, can certainly all by hardware, implement.Understanding based on such, what technical scheme of the present invention contributed to background technology can embody with the form of software product in whole or in part, this computer software product can be stored in storage medium, as ROM/RAM, magnetic disc, CD etc., comprise that some instructions are with so that a computer equipment (can be personal computer, server, or the network equipment etc.) carry out the described method of some part of each embodiment of the present invention or embodiment.

Above disclosed is only the preferred embodiment in the embodiment of the present invention, certainly can not limit with this interest field of the present invention, and the equivalent variations of therefore doing according to the claims in the present invention, still belong to the scope that the present invention is contained.

Claims (10)

1. a super capacitor energy storage device charge state detection method, is characterized in that, comprising:

Set up the free state spatial model according to ultracapacitor monomer circuit model;

Current value and terminal voltage value during each ultracapacitor monomer work of Real-time Collection;

Current value and terminal voltage value during according to described free state spatial model, described each ultracapacitor monomer work, utilize expanded Kalman filtration algorithm to calculate the SOC value of each ultracapacitor monomer.

2. the method for claim 1, is characterized in that, describedly according to ultracapacitor monomer circuit model, sets up the free state spatial model, is specially:

Ultracapacitor monomer circuit model is carried out to parameter identification, obtain state equation and the output equation of ultracapacitor monomer according to ultracapacitor monomer circuit model, wherein, described ultracapacitor monomer circuit model is the first order nonlinear equivalent-circuit model.

3. method as claimed in claim 2, is characterized in that, described state equation and output equation are specially:

$\left[\frac{\mathrm{SOC}(t+1)}{U_c(t+1)}\right]=\left[\begin{array}{cc}1& 0\\ 0& \exp (-\frac{\mathrm{\Δt}}{R_b(C_a+C_b*U_c)})\end{array}\right]\left[\begin{array}{c}\mathrm{SOC}\left(t\right)\\ U_c\left(t\right)\end{array}\right]+\left[\begin{array}{c}-\frac{\mathrm{\Δt}}{Q_0}\\ R_b(1-\exp (-\frac{\mathrm{\Δt}}{R_b(C_a+C_b*U_c)}))\end{array}\right]I\left(t\right)+w\left(t\right)---\left(5\right);$

U(t)=F[SOC(t)]+R _aI(t)+v(t) (6)；

Wherein, SOC (t) means discrete state t ultracapacitor state-of-charge constantly, U _c(t) mean the discrete state t voltage of the inner equivalent capacity of ultracapacitor monomer constantly, Δ t is sampling interval, equivalent internal resistance R _a, parallel resistance R _b, constant capacitor C _a, variable capacitance C _b, Q ₀for the ultracapacitor rated capacity of dispatching from the factory, I (t) is discrete state t ultracapacitor monomer charging and discharging currents constantly, U (t) is discrete state t ultracapacitor monomer terminal voltage constantly, F[SOC (t)] be the nonlinear function of the voltage of ultracapacitor SOC and equivalent capacity, w (t) is the interference of immesurable stochastic variable to quantity of state, and v (t) is the measurement noise of ultracapacitor terminal voltage.

4. method as claimed in claim 3, is characterized in that, described expanded Kalman filtration algorithm is specially:

Order

$A_t=\left[\begin{array}{cc}1& 0\\ 0& \exp (-\frac{\mathrm{\Δt}}{R_b(C_a+C_b*U_c)})\end{array}\right];$

$x_t=\left[\begin{array}{c}\mathrm{SOC}\left(t\right)\\ U_c\left(t\right)\end{array}\right];$

$B_t=\left[\begin{array}{c}-\frac{\mathrm{\Δt}}{Q_0}\\ R_b(1-\exp (-\frac{\mathrm{\Δt}}{R_b(C_a+C_b*U_c)}))\end{array}\right];$

u _t＝I(t)；

$C_t=\left[\begin{array}{cc}{\frac{\mathrm{dF}\left({\mathrm{SOC}}_t\right)}{\mathrm{dSOC}}|}_{\mathrm{SOC}={\mathrm{SOC}}_t}& 0\end{array}\right];$

y _t＝U(t)；

Wherein,

mean F[SOC (t)] to the derivative of ultracapacitor monomer state-of-charge SOC, then get SOC=SOC (t), the result finally calculated;

Described formula (5) and described formula (6) can be rewritten as:

State equation: x _t+1=A _tx _t+ B _tu _t+ w (t) (7);

Output equation: y _t=C _tx _t+ R _au _t+ v (t) (8);

Order

f(x _t，u _t)=A _tx _t+B _tu _t，g(x _t，u _t)=C _tx _t+R _au _t，

Described formula (7) and described formula (8) can be rewritten as:

State equation: x _t+1=f (x _t, u _t)+w (t) (9);

Output equation: y _t=g (x _t, u _t)+v (t) (10);

Step 1: described formula (9) and described formula (10) state variable x and mean square deviation error P are carried out to initialization;

Step 2: state variable x in step 1 is carried out to the prediction of expanded Kalman filtration algorithm state variable and estimate to calculate;

Step 3: mean square deviation error P in step 1 is carried out to expanded Kalman filtration algorithm mean square deviation error prediction and estimate to calculate;

Step 4: mean square deviation error P in step 3 is carried out to kalman gain calculating;

Step 5: the prediction of state variable x in step 2 is estimated to carry out to the optimal estimation calculating of expanded Kalman filtration algorithm state variable x;

Step 6: the prediction of mean square deviation error P in step 3 is estimated to carry out to the optimal estimation calculating of expanded Kalman filtration algorithm mean square deviation error P;

Step 7: judge whether the SOC value restrains and the invariant state that tends towards stability, if so, finish the SOC computation process of this monomer; If not, return to the SOC value of next moment monomer of step 2 iterative computation until result of calculation converges to SOC value the state that tends towards stability of ultracapacitor monomer accurately the most at last.

5. as the described method of claim 1-4 any one claim, it is characterized in that, also comprise:

Judge that whole bank of super capacitors is in charging duty or electric discharge duty;

According to the SOC value of duty and each ultracapacitor monomer, obtain the SOC value of each series super capacitor bank;

According to the SOC value of described each series super capacitor bank, obtain the SOC value of whole bank of super capacitors.

6. method as claimed in claim 5, is characterized in that, the described SOC value according to duty and each ultracapacitor monomer obtains the SOC value of each series super capacitor bank, is specially:

If current bank of super capacitors is the charging duty, in the SOC value series connection group for this reason of current series super capacitor bank one of the SOC value of ultracapacitor monomer maximum; If current bank of super capacitors is the electric discharge duty, in the SOC value series connection group for this reason of current series super capacitor bank one of the SOC value of ultracapacitor monomer minimum.

7. the method for claim 1, is characterized in that, after current value and terminal voltage value when each ultracapacitor monomer work of Real-time Collection, also comprises:

When the time of terminal voltage value in default low state A has surpassed predefined time threshold T1, or, when the time of terminal voltage value in default high state of value B has surpassed predefined time threshold T2, start early warning mechanism.

8. a super capacitor energy storage device state-of-charge pick-up unit, is characterized in that, comprising:

Set up free state spatial model module, for according to ultracapacitor monomer circuit model, setting up the free state spatial model;

Acquisition module, current value and terminal voltage value during for each ultracapacitor monomer work of Real-time Collection;

Computing module, with the described free state spatial model module of setting up, with described acquisition module, be connected, current value and terminal voltage value while being used for according to free state spatial model, each ultracapacitor monomer work, utilize expanded Kalman filtration algorithm to calculate the SOC value of each ultracapacitor monomer.

9. device as claimed in claim 8, is characterized in that, also comprises:

The judgement block of state, be connected with described computing module, for judging whole bank of super capacitors, is in charging duty or electric discharge duty;

Computing module, also for the SOC value according to duty and each ultracapacitor monomer, obtain the SOC value of each series super capacitor bank, according to the SOC value of each series super capacitor bank, obtains the SOC value of whole bank of super capacitors.

10. device as claimed in claim 9, is characterized in that, also comprises:

Alarm module, with described acquisition module, be connected, for when the time of terminal voltage value in default low state A has surpassed predefined time threshold T1, or, when the time of terminal voltage value in default high state of value B has surpassed predefined time threshold T2, start early warning mechanism.

Images