CN104242329A - Micro-grid hybrid energy storage system power distribution method based on fuzzy control rules - Google Patents

Abstract

The invention relates to a micro-grid hybrid energy storage system power distribution method based on fuzzy control rules. A power distribution strategy of variable-filter time constant in a micro-grid hybrid energy storage system is achieved. The micro-grid hybrid energy storage system power distribution method regulates the time constant of a low pass filter according to a charge state of an energy storage unit, dynamically regulates charging and discharging power of the energy storage unit, enables the charge state of the energy storage unit to be kept in a certain range, avoids overcharging and over-discharging of the energy storage unit and facilitates power balance of the micro-grid hybrid energy storage system.

Description

Based on the micro-capacitance sensor mixed energy storage system power distribution method of fuzzy control rule

Technical field

The present invention relates to a kind of grid control method, particularly a kind of micro-capacitance sensor mixed energy storage system power distribution method based on fuzzy control rule.

Background technology

In order to ensure the stability that micro-capacitance sensor runs, keeping realtime power for electric equilibrium, reasonably need be controlled mixed energy storage system.Microgrid power distribution as shown in Figure 1.In system, energy-storage units (storage battery and ultracapacitor) is all connected with DC bus by buck/boost power inverter, realizes the two-way flow of power.

At present, research is expanded to the control strategy of micro-grid energy storage system both at home and abroad.Chen Yizhe proposes the micro-grid energy storage system active control strategies based on short-term load forecasting, when considering the restriction of battery capacity, discharge and recharge number of times, according to the result of load prediction, the discharge and recharge of control storage battery initiatively, optimizes the charging and discharging curve of storage battery.Zhang Guoju etc. adopt ultracapacitor as the energy-storage units of micro-capacitance sensor, establish the small signal equivalent model that complementary PWM controls, and application double-closed-loop control and power feedforward link achieve the stable of DC bus-bar voltage.But single energy-storage units often well can not take into account the requirement of micro-capacitance sensor for energy and power.Dougal R A etc. proposes the concept of storage battery-super capacitor mixed energy storage, and demonstrates the superiority of mixed energy storage system in performance theoretically.Dai Yongxi etc., for pulse-type load, devise a kind of active and mix energy storage connected mode, optimize the discharge condition of storage battery.Wang Hongfu etc. adopt storage battery-ultracapacitor to mix energy-storage system, and storage battery and ultracapacitor stabilize the power fluctuation of low frequency and high frequency respectively according to respective characteristic, and level and smooth grid-connected honourable active power, improves the grid-connected quality of power supply.The existing research to hybrid energy-storing, mainly concentrates on the control strategy stabilizing grid-connected power, and to the electricity Real-time Balancing of independent operating micro-capacitance sensor and the research of voltage stabilization less.

Summary of the invention

The present invention be directed to the electricity Real-time Balancing of independent operating micro-capacitance sensor, the problem of voltage stabilization, propose a kind of micro-capacitance sensor mixed energy storage system power distribution method based on fuzzy control rule, it is a kind of power distribution strategies of the change time constant filter based on fuzzy control, the method is according to the state-of-charge of energy-storage units, the time constant of low pass filter is regulated, the charge-discharge electric power of dynamic adjustment energy-storage units, the state-of-charge of energy-storage units is made to maintain in certain scope, avoid energy-storage units overcharge, overdischarge, effectively maintain the stable of DC bus-bar voltage.

Technical scheme of the present invention is: a kind of micro-capacitance sensor mixed energy storage system power distribution method based on fuzzy control rule, according to the power demand of load and the power stage of distributed power source, calculates micro-capacitance sensor and mixes the performance number that energy-storage system needs to stabilize; According to the state-of-charge of energy-storage units, adopt fuzzy control rule to regulate the size of time constant filter, revise the compensation power value of storage battery and ultracapacitor, carry out the power optimization distribution that micro-capacitance sensor mixes energy-storage system.

Described fuzzy control rule regulates the size of time constant filter, specifically comprises:

1) synthetic load power P _hunregard superposing of low frequency and high frequency two kinds of components as, synthetic load power P _hunby only remaining low frequency part P after low pass filter _{bat_ref}, this part is absorbed by storage battery or compensates, i.e. P _{bat_ref}as the reference power of storage battery; Synthetic load power P _hunwith the actual power P of storage battery _batdo difference, acquired results P _{uc_ref}as the reference power of ultracapacitor;

2) to the state-of-charge SOC of storage battery _batwith the state-of-charge SOC of ultracapacitor _ucbe normalized, state-of-charge represents the percentage of current capacities and rated capacity, and the degree of membership of energy storage charge state is as follows:

${\mathrm{\ξ}}_{\mathrm{bat}\_\mathrm{soc}}=\frac{{\mathrm{SOC}}_{\mathrm{bat}}-{\mathrm{SOC}}_{\mathrm{bat}\_\mathrm{mid}}}{{\mathrm{SOC}}_{\mathrm{bat}\_\mathrm{mid}}}$

In formula: SOC _{bat_mid}for the intermediate value of storage battery charge state;

${\mathrm{\ξ}}_{\mathrm{uc}\_\mathrm{soc}}=\frac{{\mathrm{SOC}}_{\mathrm{uc}}-{\mathrm{SOC}}_{\mathrm{uc}\_\mathrm{mid}}}{{\mathrm{SOC}}_{\mathrm{uc}\_\mathrm{mid}}}$

In formula: SOC _{uc_mid}for the intermediate value of ultracapacitor state-of-charge;

By ξ _{bat_soc}and ξ _{uc_soc}as the input of fuzzy control, set the fuzzy set of input variable as { NB (negative large), ZE (zero), PB (honest) }, wherein ξ _{bat_soc}domain be [-a, a], ξ _{uc_soc}domain be [-b, b], wherein 0<a<1,0<b<1, a and b specifically neglect greatly the characteristic of energy-storage units and determine, and getting time constant correction △ λ is the output variable of fuzzy control, fuzzy set is { NB (negative large), NS (negative little), ZE (zero), PS (just little), PB (honest) }, corresponding fuzzy domain is {-1,-0.4,0,0.4,1};

3) fuzzy rule is formulated:

When synthetic load power P _hunduring <0:

A: when the state-of-charge of storage battery and ultracapacitor is median, maintain the time constant preset constant;

B: when the state-of-charge of storage battery is close to lower limit, and ultracapacitor state-of-charge turns time constant T down in limited time close to upper, increases the charge power of storage battery;

C: when the state-of-charge of storage battery is close to the upper limit, and ultracapacitor state-of-charge tunes up time constant T in limited time close to lower, increases the charge power of ultracapacitor;

When synthetic load power P _hunduring >0:

A: when the state-of-charge of storage battery and ultracapacitor is median, maintain the time constant preset constant;

B: when the state-of-charge of storage battery is close to lower limit, and ultracapacitor state-of-charge tunes up time constant T in limited time close to upper, reduces the discharge power of storage battery;

C: when the state-of-charge of storage battery is close to the upper limit, and ultracapacitor state-of-charge turns time constant T down in limited time close to lower, reduces the discharge power of ultracapacitor;

4) de-fuzzy obtains revising rear time constant:

Adopt weighted mean method to carry out deblurring calculating, obtain the correction factor △ λ of reference power ,-1≤△ λ≤1,

$\mathrm{\&Delta;\&lambda;}=\frac{\underset{i}{\mathrm{\&Sigma;}}\underset{j}{\mathrm{\&Sigma;}}{u}_{1i}\left({\mathrm{SOC}}_{\mathrm{bat}}\right)u_{2j}\left({\mathrm{SOC}}_{\mathrm{uc}}\right)\mathrm{\&Delta;}{\mathrm{\&lambda;}}_{\mathrm{ij}}}{\mathrm{\&Sigma;\&Sigma;}{u}_{\mathrm{ii}}\left({\mathrm{SOC}}_{\mathrm{bat}}\right)u_{2j}\left({\mathrm{SOC}}_{\mathrm{uc}}\right)},$

In formula: u _1i(SOC _bat) be input variable SOC _bati-th be subordinate to angle value; u _2j(SOC _uc) be input variable SOC _ucjth be subordinate to angle value, revised low pass filter time constant is:

T ^*＝T+△λT。

Beneficial effect of the present invention is: the micro-capacitance sensor mixed energy storage system power distribution method that the present invention is based on fuzzy control rule, the size of low-pass filtering time constant is distributed by fuzzy control rule Modulating Power, revise the compensation power value of storage battery and ultracapacitor, be conducive to the power-balance realizing micro-capacitance sensor mixed energy storage system.

Accompanying drawing explanation

Fig. 1 is microgrid power distribution map;

Fig. 2 is that power of the present invention tentatively distributes control block diagram;

Fig. 3 is input and output membership function figure of the present invention;

Fig. 4 is the state-of-charge figure of the present invention's energy-storage units when not adopting fuzzy control rule;

Fig. 5 is the state-of-charge figure of the present invention's energy-storage units when adopting fuzzy control rule.

Embodiment

The power division of the main synthetic load power of the control of hybrid energy-storing between energy-storage units in independent micro-grid.Propose the change time constant power distribution strategies based on fuzzy control rule, according to the state-of-charge of energy-storage units, formulate corresponding fuzzy control rule, the size of suitable adjustment low-pass filtering time constant, revise the compensation power value of storage battery and ultracapacitor.

Change time constant filter power distribution strategies based on fuzzy control rule:

The object of power division is the power-balance realizing system, according to the power demand of load and the power stage of distributed power source, calculates the performance number that mixed energy-storage system needs to stabilize.And then according to the discharge and recharge feature of storage battery, ultracapacitor and the state-of-charge of correspondence, calculate the value and power reference of storage battery and ultracapacitor respectively.

Because system adopts DC bus structure, must ensure system busbar voltage constant during system cloud gray model, the pass of busbar voltage and each power cell is:

$C_{\mathrm{dc}}{u}_{\mathrm{dc}}\frac{{\mathrm{du}}_{\mathrm{dc}}}{\mathrm{dt}}=P_{\mathrm{wt}}+P_{\mathrm{pv}}+P_{\mathrm{bat}}+P_{\mathrm{uc}}-P_{\mathrm{load}}---\left(1\right)$

In formula: C _dcfor dc-link capacitance; u _dcfor DC bus-bar voltage; P _wtfor wind power generation power; P _pvfor photovoltaic generation power; P _batfor storage battery power; P _ucfor ultracapacitor power; P _loadfor load power.

Work as P _bat, P _ucfor timing, expression storage battery, ultracapacitor send power, work as P _bat, P _ucfor time negative, represent storage battery, ultracapacitor absorbed power.

During system stable operation, DC bus-bar voltage is constant, power Real-time Balancing, and namely the difference power of distributed power source and load is absorbed by hybrid energy-storing unit completely or makes up, as shown in Equation 2.

P _wt+P _pv+P _bat+P _uc-P _load＝0 (2)

The present invention adopts the concept of synthetic load, namely the demand power of the power output of distributed power source (wind power generation unit, photovoltaic generation unit) and load is combined and considers.Because generator unit and load have certain stochastic volatility, synthetic load is made also to be the power of a random fluctuation.

P _load-(P _wt+P _pv)＝P _hun (3)

In formula: P _hunfor synthetic load power.

During micro-capacitance sensor steady operation, system power Real-time Balancing, the power that energy-storage system sends stabilizes synthetic load power completely, shown in (4).

P _bat+P _uc＝P _hun (4)

1, based on the power division of low-pass filtering:

Synthetic load power can regard the result of the wave component superposition of different frequency as, supposes to be decomposed into low-frequency fluctuation component and high-frequency fluctuation component, then stabilizes low-frequency fluctuation and high-frequency fluctuation respectively by different energy-storage units.In mixed energy storage system, storage battery energy density is high, is suitable for absorbing or sending low frequency power component; And ultracapacitor charge/discharge rates is fast, be suitable for absorbing or supplementary high frequency power component.

If the reference power of storage battery is P _{bat_ref}, ultracapacitor reference power be P _{uc_ref}.Fig. 2 power tentatively distributes in control block diagram, synthetic load power P _hunregard superposing of low frequency and high frequency two kinds of components as, synthetic load power P _hunby only remaining low frequency part P after low pass filter _{bat_ref}, this part is absorbed by storage battery or compensates, i.e. P _{bat_ref}as the reference power of storage battery; Synthetic load power P _hunwith the actual power P of storage battery _batdo difference, acquired results P _{uc_ref}as the reference power of ultracapacitor.

The reference power of storage battery is:

$P_{\mathrm{bat}\_\mathrm{ref}}=\frac{1}{1+\mathrm{TS}}{P}_{\mathrm{hun}}---\left(5\right)$

In formula: T is low-pass filtering time constant.

The terminal voltage U of usual storage battery _bchange very little, within a certain period of time U _bregard as invariable, then the reference current of storage battery is:

$i_{\mathrm{bat}\_\mathrm{ref}}=\frac{P_{\mathrm{bat}\_\mathrm{ref}}}{U_b}---\left(6\right)$

The reference power of ultracapacitor is:

P _{uc_ref}＝P _hun-P _bat (7)

From the above, time constant T is larger, and the power that storage battery is born is just milder, and the frequency range that ultracapacitor compensates is larger; Time constant T is less, and the power fluctuation that storage battery is born is larger, and the frequency range that ultracapacitor compensates is corresponding to be reduced.When adopting basic filtering algorithm, time constant T is definite value, does not consider the state-of-charge of energy-storage units, easily causes energy-storage units to overcharge, overdischarge.Therefore, the present invention, on the basis of basic filtering algorithm, adds charge state feedback link.When energy-storage units state-of-charge within the specific limits time, adopt the time constant preset, carry out power division, if when the state-of-charge of a certain energy-storage units is close to bound, based on fuzzy control regulation time constant, thus revise the reference power P of storage battery and ultracapacitor _{bat_ref}, P _{uc_ref}, the state-of-charge of energy-storage units is remained in prescribed limit.

2, the time constant based on fuzzy control rule adjusts:

For mixed energy-storage system, be difficult to arrange precise time constant to realize power division to low pass filter, for making the state-of-charge considering each energy-storage units during mixed energy-storage system power division, the present invention adopts fuzzy control strategy to regulate the size of time constant filter, and the power optimization carrying out mixed energy-storage system distributes.The present invention selects the state-of-charge SOC of energy-storage units _batand SOC _ucfor input variable, the correction factor △ λ of time constant is output variable.

1) input, the fuzzy subset exported and membership function is established:

First to the state-of-charge SOC of storage battery _batwith the state-of-charge SOC of ultracapacitor _ucbe normalized.State-of-charge represents the percentage of current capacities and rated capacity, and the degree of membership of energy storage charge state is as follows:

${\mathrm{\&xi;}}_{\mathrm{bat}\_\mathrm{soc}}=\frac{{\mathrm{SOC}}_{\mathrm{bat}}-{\mathrm{SOC}}_{\mathrm{bat}\_\mathrm{mid}}}{{\mathrm{SOC}}_{\mathrm{bat}\_\mathrm{mid}}}---\left(8\right)$

In formula: SOC _{bat_mid}for the intermediate value of storage battery charge state.

${\mathrm{\&xi;}}_{\mathrm{uc}\_\mathrm{soc}}=\frac{{\mathrm{SOC}}_{\mathrm{uc}}-{\mathrm{SOC}}_{\mathrm{uc}\_\mathrm{mid}}}{{\mathrm{SOC}}_{\mathrm{uc}\_\mathrm{mid}}}---\left(9\right)$

In formula: SOC _{uc_mid}for the intermediate value of ultracapacitor state-of-charge.

By ξ _{bat_soc}and ξ _{uc_soc}as the input of fuzzy control, set the fuzzy set of input variable as { NB (negative large), ZE (zero), PB (honest) }, wherein ξ _{bat_soc}domain be [-a, a], ξ _{uc_soc}domain be [-b, b], wherein 0<a<1,0<b<1, a and b specifically neglects greatly the characteristic of energy-storage units and determines, the discharge and recharge degree of depth of ultracapacitor is greater than storage battery in general, therefore b>a.Getting time constant correction △ λ is the output variable of fuzzy control, and fuzzy set is { NB (negative large), NS (negative little), ZE (zero), PS (just little), PB (honest) }, corresponding fuzzy domain is {-1 ,-0.4,0,0.4,1}.Fig. 3 is the membership function of input variable and output variable.

2) fuzzy rule is formulated

When synthetic load power P _hunduring <0, fuzzy rule is as shown in table 1.Now the formulation of fuzzy rule is based on following several experiences:

(1) when the state-of-charge of storage battery and ultracapacitor is median, the time constant preset is maintained constant;

(2) when the state-of-charge of storage battery is close to lower limit, and ultracapacitor state-of-charge suitably turns time constant T down in limited time close to upper, increases the charge power of storage battery;

(3) when the state-of-charge of storage battery is close to the upper limit, and ultracapacitor state-of-charge suitably tunes up time constant T in limited time close to lower, increases the charge power of ultracapacitor.

When synthetic load power P _hunduring >0, fuzzy rule is as shown in table 2.Now the formulation of fuzzy rule is based on following several experiences:

(1) when the state-of-charge of storage battery and ultracapacitor is median, the time constant preset is maintained constant;

(2) when the state-of-charge of storage battery is close to lower limit, and ultracapacitor state-of-charge suitably tunes up time constant T in limited time close to upper, reduces the discharge power of storage battery;

(3) when the state-of-charge of storage battery is close to the upper limit, and ultracapacitor state-of-charge suitably turns time constant T down in limited time close to lower, reduces the discharge power of ultracapacitor.

Table 1

Table 2

3) de-fuzzy

The present invention adopts weighted mean method to carry out deblurring calculating, obtains the correction factor △ λ (precise volume) of reference power ,-1≤△ λ≤1.

$\mathrm{\&Delta;\&lambda;}=\frac{\underset{i}{\mathrm{\&Sigma;}}\underset{j}{\mathrm{\&Sigma;}}{u}_{1i}\left({\mathrm{SOC}}_{\mathrm{bat}}\right)u_{2j}\left({\mathrm{SOC}}_{\mathrm{uc}}\right)\mathrm{\&Delta;}{\mathrm{\&lambda;}}_{\mathrm{ij}}}{\mathrm{\&Sigma;\&Sigma;}{u}_{\mathrm{ii}}\left({\mathrm{SOC}}_{\mathrm{bat}}\right)u_{2j}\left({\mathrm{SOC}}_{\mathrm{uc}}\right)}---\left(10\right)$

In formula: u _1i(SOC _bat) be input variable SOC _bati-th be subordinate to angle value; u _2j(SOC _uc) be input variable SOC _ucjth be subordinate to angle value.

Revised low pass filter time constant is:

T ^*＝T+△λT (11)

Fig. 4 and Fig. 5 is respectively the state-of-charge curve of the energy-storage units adopted before and after Fuzzy strategy, and when not using fuzzy control rule, ultracapacitor has many places to exceed the bound of state-of-charge; Under the constraint of fuzzy control rule, the excursion of ultracapacitor state-of-charge narrows to some extent, is controlled in rational scope.Because the capacity of storage battery is herein comparatively large, in scope that state-of-charge remains on substantially [0.4,0.6], capacity is sufficient, therefore prescribe a time limit on ultracapacitor state-of-charge will be crossed, tune up time constant T by fuzzy control rule, storage battery absorbs more power; Will cross in limited time lower at ultracapacitor state-of-charge, turn time constant T down by fuzzy control rule, storage battery sends more power, reduces the power that ultracapacitor should send, and avoids capacity of super capacitor exhausted.

Claims (2)

1. based on a micro-capacitance sensor mixed energy storage system power distribution method for fuzzy control rule, it is characterized in that, according to the power demand of load and the power stage of distributed power source, calculate micro-capacitance sensor and mix the performance number that energy-storage system needs to stabilize; According to the state-of-charge of energy-storage units, adopt fuzzy control rule to regulate the size of time constant filter, revise the compensation power value of storage battery and ultracapacitor, carry out the power optimization distribution that micro-capacitance sensor mixes energy-storage system.

2. according to claim 1 based on the micro-capacitance sensor mixed energy storage system power distribution method of fuzzy control rule, it is characterized in that, described fuzzy control rule regulates the size of time constant filter, specifically comprises:

1) synthetic load power P _hunregard superposing of low frequency and high frequency two kinds of components as, synthetic load power P _hunby only remaining low frequency part P after low pass filter _{bat_ref}, this part is absorbed by storage battery or compensates, i.e. P _{bat_ref}as the reference power of storage battery; Synthetic load power P _hunwith the actual power P of storage battery _batdo difference, acquired results P _{uc_ref}as the reference power of ultracapacitor;

2) to the state-of-charge SOC of storage battery _batwith the state-of-charge SOC of ultracapacitor _ucbe normalized, state-of-charge represents the percentage of current capacities and rated capacity, and the degree of membership of energy storage charge state is as follows:

In formula: SOC _{bat_mid}for the intermediate value of storage battery charge state;

In formula: SOC _{uc_mid}for the intermediate value of ultracapacitor state-of-charge;

By ξ _{bat_soc}and ξ _{uc_soc}as the input of fuzzy control, set the fuzzy set of input variable as { NB (negative large), ZE (zero), PB (honest) }, wherein ξ _{bat_soc}domain be [-a, a], ξ _{uc_soc}domain be [-b, b], wherein 0<a<1,0<b<1, a and b specifically neglect greatly the characteristic of energy-storage units and determine, and getting time constant correction △ λ is the output variable of fuzzy control, fuzzy set is { NB (negative large), NS (negative little), ZE (zero), PS (just little), PB (honest) }, corresponding fuzzy domain is {-1,-0.4,0,0.4,1};

3) fuzzy rule is formulated:

When synthetic load power P _hunduring <0:

A: when the state-of-charge of storage battery and ultracapacitor is median, maintain the time constant preset constant;

B: when the state-of-charge of storage battery is close to lower limit, and ultracapacitor state-of-charge turns time constant T down in limited time close to upper, increases the charge power of storage battery;

C: when the state-of-charge of storage battery is close to the upper limit, and ultracapacitor state-of-charge tunes up time constant T in limited time close to lower, increases the charge power of ultracapacitor;

When synthetic load power P _hunduring >0:

A: when the state-of-charge of storage battery and ultracapacitor is median, maintain the time constant preset constant;

B: when the state-of-charge of storage battery is close to lower limit, and ultracapacitor state-of-charge tunes up time constant T in limited time close to upper, reduces the discharge power of storage battery;

C: when the state-of-charge of storage battery is close to the upper limit, and ultracapacitor state-of-charge turns time constant T down in limited time close to lower, reduces the discharge power of ultracapacitor;

4) de-fuzzy obtains revising rear time constant:

Adopt weighted mean method to carry out deblurring calculating, obtain the correction factor △ λ of reference power ,-1≤△ λ≤1,

In formula: u _1i(SOC _bat) be input variable SOC _bati-th be subordinate to angle value; u _2j(SOC _uc) be input variable SOC _ucjth be subordinate to angle value, revised low pass filter time constant is:

T ^*＝T+△λT 。