CN104779630A - Capacity allocation method for hybrid energy storage system capable of restraining wind power output power fluctuation - Google Patents

Abstract

The invention discloses a capacity allocation method for a hybrid energy storage system capable of restraining wind power output power fluctuation. The wind power fluctuation component is extracted based on the national wind power integration standard by means of the improved moving average method, fluctuating power is allocated by means of the variable-filter time constant high-pass filtering method based on the charging states of energy storage equipment, in this way, over-charging and over-discharging of energy storage equipment can be avoided, and finally a capacity allocation model is established with the full-life circle economical efficiency optimization of energy storage equipment as an objective function. Compared with a traditional method according to which only one-time primary investment is considered, the method better accords with the actual condition of long-term operation of an energy storage system and has guiding significance for practical engineering construction.

Description

A kind of mixed energy storage system capacity collocation method stabilizing wind power output power fluctuation

Technical field

The invention belongs to micro-capacitance sensor technical field of energy storage, relate to a kind of mixed energy storage system capacity collocation method, be specifically related to a kind of mixed energy storage system capacity configuration optimizing method stabilizing wind power output power fluctuation.

Background technology

Have green, the wind generating technology of the feature such as clean, renewable is widely used and develops in worldwide, installed capacity increases year by year, is the effective way solving world today's energy crisis and problem of environmental pollution.But the power output of wind generator system is subject to the impact of the external factors such as natural conditions, there is intermittence and randomness, large-scale wind power is grid-connected causes adverse influence by the reliable and stable operation of bulk power grid and the quality of power supply, therefore needs the energy-storage system configuring certain capacity in order to stabilize wind power output power fluctuation.

The energy type energy-storage system of representative that traditional with storage battery is has higher specific capacity, is applicable to the discharge and recharge of long period yardstick, is generally used for meeting load power demands.Wind power output power fluctuation is except the wave component of long time scale, also exist a large amount of in short-term, the wave component of peak value, need to possess fast-response energy and the larger energy-storage system of charge-discharge electric power is stabilized, obviously traditional energy-storage system of accumulator cannot meet the demands, needs are that the power-type energy-storage system of representative coordinates with ultracapacitor, and common formation mixed energy storage system carries out stabilizing of regenerative resource output-power fluctuation.

Three technological difficulties of configuration mixed energy storage system capacity are: the extraction of (1) wind power output power wave component; (2) distribution of wind power output power wave component between dissimilar energy storage device; (3) scientific and reasonable capacity configuration model is set up.

For technological difficulties (1), conventional method is main mainly with low-pass filtering method, easily causes to stabilize that component is excessive thus that energy storage system capacity is configured is excessive, and economy is poor and effectively can not combine the relevant wind-electricity integration standard of country;

For technological difficulties (2), conventional method can not take into account the state-of-charge of energy storage device, may cause the super-charge super-discharge of energy storage device in the assignment procedure;

For technological difficulties (3), conventional method the foundation of optimization aim and constraints really timing there is certain deficiency.

Summary of the invention

In order to solve above-mentioned technical problem, the present invention proposes a kind of mixed energy storage system capacity configuration optimizing method stabilizing wind power output power fluctuation.

The technical solution adopted in the present invention is: a kind of mixed energy storage system capacity collocation method stabilizing wind power output power fluctuation, is characterized in that, comprise the following steps:

Step 1: process the real data that active power of wind power field exports, uses the moving average method of improvement and in conjunction with national wind-electricity integration standard, extracts in wind power output power and need energy-storage system to carry out the wave component stabilized;

Step 2: the state-of-charge taking into account each energy storage device, uses the high-pass filtering method of variable time constant filter between dissimilar energy storage device, carry out the distribution of wave component;

Step 3: set up the hybrid energy-storing selected topic capacity Optimal Allocation Model for the purpose of mixed energy storage system Life cycle economy optimum, use modified particle swarm optiziation, require as constraints with the technical characteristic of energy storage device and wind-electricity integration, carry out mixed energy storage system capacity and distribute rationally.

As preferably, the moving average method of the improvement described in step 1 in conjunction with national wind-electricity integration standard, extract in wind power output power and need energy-storage system to carry out the wave component stabilized, its specific implementation process comprises following sub-step:

Step 1.1: the moving average method that utilization improves calculates level and smooth grid-connected component and need energy-storage system to carry out power the wave component handled up;

Suppose that sliding window numerical value is N min, and suppose that N is that even number is described to facilitate, obtain grid-connected component and the min level undulate quantity of t wind-powered electricity generation thus, shown in (1.1), (1.2), (1.3):

P _ft＝(P _t-(N/2-1)+P _t-(N/2-2)+...+P _t+...+P _t+N/2)/N (1.1)；

P _mt＝P _t-P _ft(1.2)；

t＝N/2,N/2+1,...,M-N/2 (1.3)；

In formula: P _tit is the wind power of actual measurement in t minute; P _ftit is grid-connected component; P _mtit is min level wave component; M is measurement point sum;

Step 2: in conjunction with the requirement of national wind-electricity integration power fluctuation, the basis of moving average method is improved, adjusts for the part not meeting fluctuation rate of change, finally obtains the compensation power P of mixed energy storage system _hESS;

P _HESS＝P _mt+△P (1.4)；

In formula: Δ P is the power carrying out according to the fluctuation rate of change requirement of wind-powered electricity generation maximum power adjusting.

As preferably, N=15min.

As preferably, described national wind-electricity integration standard asks for an interview table 1;

The national wind-electricity integration standard of table 1

As preferably, the high-pass filtering method of the variable time constant filter of the utilization described in step 2 carries out the distribution of wave component between dissimilar energy storage device, and its specific implementation process comprises following sub-step:

Step 2.1: the compensation power P using high-frequency filter equalizer mixed energy storage system _hESSshown in (2.1):

$\left\{\begin{array}{c}{P}_{\mathrm{UC}}=P_{\mathrm{HESS}}*\frac{T_{\mathrm{UC}}s}{T_{\mathrm{UC}}s+1}\\ P_{\mathrm{BESS}}=P_{\mathrm{HESS}}-P_{\mathrm{UC}}\end{array}\right.---\left(2.1\right)$

Wherein, P _uCand P _bESSbe respectively ultracapacitor and storage battery stabilize power, T _uCit is the time constant filter of high pass filter.

Step 2.2: adjustment time constant filter, adopts fuzzy control theory to revise; The core content of described fuzzy control theory comprises:

Obfuscation: the input variable within the scope of domain is carried out Fuzzy processing, obtains fuzzy subset and membership function;

Formulate fuzzy rule: express the control decision of people's subjectivity for fuzzy control statement, form the domination set under different condition by many fuzzy control statements, this is the core of fuzzy control;

Fuzzy reasoning: according to fuzzy fuzzy quantity of exerting oneself;

De-fuzzy; Fuzzy quantity is changed into the precise control amount in domain;

Step 2.3: based on above-mentioned fuzzy control theory, when carrying out hybrid energy-storing and stabilizing power fluctuation, consider the state-of-charge of each energy storage device, when state-of-charge approaches bound, timely adjustment time constant filter, revise the charge-discharge electric power instruction of energy storage device, thus ensure that the state-of-charge of energy storage device is within zone of reasonableness all the time.

As preferably, the implementation procedure of the fuzzy control described in step 2 is, when energy storage device state-of-charge close to upper in limited time, if obtain charge power instruction, then regulate time constant filter suitably to reduce charge power, avoid overcharging; When energy storage device state-of-charge is prescribed a time limit close to lower, if obtain discharge power instruction, then regulate time constant filter to reduce discharge power, avoided putting.

As preferably, the implementation procedure of the fuzzy control described in step 2 is, with storage battery charge state SoC _bESSwith ultracapacitor state-of-charge SoC _uCas input, with the correction factor k of time constant filter for exporting;

First to SoC _bESSand SoC _uCbe normalized, obtaining its degree of membership is:

${\mathrm{\ξ}}_{\mathrm{bat}}=\frac{{\mathrm{SoC}}_{\mathrm{BESS}}-{\mathrm{SoC}}_{\mathrm{mid}}}{{\mathrm{SoC}}_{\mathrm{mid}}}---\left(2.2\right);$

${\mathrm{\ξ}}_{\mathrm{uc}}=\frac{{\mathrm{SoC}}_{\mathrm{UC}}-{\mathrm{SoC}}_{\mathrm{mid}}}{{\mathrm{SoC}}_{\mathrm{mid}}}---\left(2.3\right);$

SoC in formula _midfor the median of energy storage device state-of-charge;

By SoC _min≤ SoC≤SoC _maxknown ξ _batand ξ _uccontinuous domain be respectively [-a, a] and [-b, b], the concrete size of a with b is relevant with energy storage device technical characteristic, because the depth of discharge of ultracapacitor is greater than storage battery, therefore b>a and 0<b<1; When energy storage device degree of membership is a or b, represents that capacity is saturated, be full of electricity completely; When energy storage device degree of membership is-a or-b, represents that capacity is exhausted, discharge completely; ξ _batand ξ _ucfuzzy set be { NB (negative large), ZO (zero), PB (honest) }; The output variable of fuzzy control is the correction factor k of time constant filter, and its discrete domain is [-1 ,-0.5,0,0.5,1], and fuzzy set is { NB, NS (negative little), ZO, PS (just little), PB};

After the fuzzy set establishing input variable and output variable and membership function, now formulate fuzzy rule:

Work as P _hESS>0, when namely mixed energy storage system obtains the instruction of discharging, the following experience of fuzzy rule Main Basis is formulated:

(1) if the state-of-charge of storage battery and ultracapacitor is median, then time constant filter remains unchanged;

(2) if the state-of-charge of storage battery is less than normal and ultracapacitor state-of-charge is bigger than normal time, suitably tune up time constant filter T _uC, increase the discharge power of ultracapacitor, reduce the discharge power of storage battery;

(3) if the state-of-charge of storage battery is bigger than normal and ultracapacitor state-of-charge is less than normal time, suitably turn time constant filter T down _uC, reduce the discharge power of ultracapacitor, increase the discharge power of storage battery;

Work as P _hESS<0, when namely mixed energy storage system obtains the instruction of charging, the following experience of fuzzy rule Main Basis is formulated:

(1) if the state-of-charge of storage battery and ultracapacitor is median, then time constant filter remains unchanged;

(2) if the state-of-charge of storage battery is less than normal and ultracapacitor state-of-charge is bigger than normal time, suitably turn time constant filter T down _uC, reduce the charge power of ultracapacitor, increase the charge power of storage battery;

(3) if the state-of-charge of storage battery is bigger than normal and ultracapacitor state-of-charge is less than normal time, suitably tune up time constant filter T _uC, increase the charge power of ultracapacitor, reduce the charge power of storage battery;

Before state-of-charge judgement is carried out to storage battery and ultracapacitor, classification to be carried out to the state-of-charge of two kinds of energy storage devices, be divided into S _max, S _high, S _mid, S _lowand S _minfive classes; When state-of-charge mediates state, do not need to change time constant filter; When state-of-charge is on the low side, restriction electric discharge, increases charge power; When state-of-charge is higher, restriction charging, increases discharge power;

Because storage battery is different with the technical characteristic of ultracapacitor, the two five class state-of-charges also difference to some extent, is specifically categorized as following table 2;

Table 2: five class state-of-charges of storage battery and ultracapacitor

For obtaining output variable time constant filter correction factor k (-1≤k≤1), needing de-fuzzy to be exported accurately, adopting weighted mean method to carry out de-fuzzy process such as formula shown in (2.4);

$k=\frac{\underset{i}{\mathrm{\&Sigma;}}\underset{j}{\mathrm{\&Sigma;}}{f}_{1i}\left({\mathrm{SoC}}_{\mathrm{BESS}}\right)f_{2j}\left({\mathrm{SoC}}_{\mathrm{UC}}\right)k_{\mathrm{ij}}}{\mathrm{\&Sigma;\&Sigma;}{f}_{1i}\left({\mathrm{SoC}}_{\mathrm{BESS}}\right)f_{2j}\left({\mathrm{SoC}}_{\mathrm{UC}}\right)}---\left(2.4\right)$

In formula:

F _1i(SoC _bESS) be input variable SoC _bESSi-th be subordinate to angle value; f _2j(SoC _uC) be input variable SoC _uCjth be subordinate to angle value;

Obtaining revised time constant filter is:

T _UC ^*＝(1+k)T _UC(2.5)。

As preferably, the hybrid energy-storing selected topic capacity Optimal Allocation Model described in step 3, its target function and constraints are specially:

Described target function is such as formula shown in (3.1):

$\left\{\begin{array}{c}\min C=\frac{C_{\mathrm{iv}}+C_{\mathrm{om}}+C_{\mathrm{dc}}}{Y}\\ C_{\mathrm{iv}}=m*O_{\mathrm{bat}}+n*O_{\mathrm{uc}}\\ C_{\mathrm{om}}=Y(m*k_{\mathrm{bat}}+n*k_{\mathrm{uc}})\\ C_{\mathrm{dc}}=p_{\mathrm{bat}}(m*O_{\mathrm{bat}})+p_{\mathrm{uc}}(n*O_{\mathrm{uc}})\end{array}\right.---\left(3.1\right);$

In formula: C is the average annual expense of mixed energy storage system; Y is energy-storage system running time; C _ivit is the acquisition cost (investment cost) of energy storage device; M and n is the equipment number of storage battery and super capacitor respectively; O _batand O _ucthe unit price of storage battery and super capacitor respectively; C _dcrefer to energy storage device disposal replacement cost (disposal cost), p _batand p _ucthe replacing batch of storage battery and super capacitor respectively; C _omrefer to the operation expense (operation & maintenance cost) of energy storage device, k _batand k _ucthe maintenance unit price of storage battery and super capacitor respectively;

Described constraints, comprises the state-of-charge in charge and discharge process and charge and discharge power constraint, shown in (3.2):

$\left\{\begin{array}{c}{\mathrm{SoC}}_{\min }\&le;\mathrm{SoC}\&le;{\mathrm{SoC}}_{\max }\\ 0\&le;P_C\&le;P_{C.\max }\\ 0\&le;P_D\&le;P_{D.\max }\end{array}\right.---\left(3.2\right);$

In formula: SoC (state of charge) is the state-of-charge of energy storage device, and wherein the state-of-charge excursion of storage battery and super capacitor is respectively [0.2,1] and [0.1,1]; P _cand P _dthe charging and discharging power of energy storage device respectively.

As preferably, the improve PSO algorithm described in step 3, its specific implementation comprises following sub-step:

Step 3.1: the whole particle populations of initialization, setting iterations, material calculation, computational accuracy, target function;

Step 3.2: the fitness calculating each particle in an iterative process, obtains individual extreme value and the global extremum of each particle;

Step 3.3: the flying speed and the present position that are upgraded each particle by individual extreme value and global extremum;

Wherein inertia weight is introduced speed more new formula:

$v_{\mathrm{id}}^{k+1}={\mathrm{wv}}_{\mathrm{id}}^k+c_1{r}_1(p_{\mathrm{id}}-z_{\mathrm{id}}^k)+c_2{r}_2(p_{\mathrm{gd}}-z_{\mathrm{id}}^k)---\left(3.1\right);$

In formula, the flying speed of particle i in d dimension space after k+1 iteration; ω is inertia weight; c ₁and c ₂be Studying factors; r ₁and r ₂it is the random number of span (0,1); p _idit is the individual optimal value of particle; p _gdit is colony's optimal value of particle; be through the particle fitness value after k iteration;

When determining inertia weight value, first adopt adaptive weighting method to inertia weight coefficient assignment, its specific implementation formula is;

$\mathrm{\&omega;}=\left\{\begin{array}{cc}{\mathrm{\&omega;}}_{\min }+\frac{({\mathrm{\&omega;}}_{\max }-{\mathrm{\&omega;}}_{\min })*(f-f_{\min })}{f_{\mathrm{avg}}-f_{\min }},& f\&le;f_{\mathrm{avg}}\\ {\mathrm{\&mu;}}_{\max },& f>f_{\mathrm{avg}}\end{array}\right.---\left(3.2\right);$

In formula: ω _maxit is the maximum occurrences of inertia weight; ω _minit is the minimum value of inertia weight; F is the target function of problem to be solved, f _avgit is the average target value of all particle adaptive values in colony; f _minthe minimum target value of all particle adaptive values in colony, μ _maxthe maximum of inertia weight;

Step 3.4: when reaching iterations or meeting stopping criterion for iteration, finishing iteration process also exports optimal solution.

The present invention adopts the moving average method low pass wind power wave component improved in conjunction with national wind-electricity integration standard, the state-of-charge taking into account energy storage device adopts variable time constant filter method to distribute fluctuating power, the super-charge super-discharge of energy storage device can be avoided, finally set up capacity configuration model so that energy storage device Life cycle economy is optimum for target function, compared to traditional method only considering disposable initial outlay, more meet the actual conditions of energy-storage system long-time running, to Practical Project construction, there is directive significance.

Accompanying drawing explanation

Fig. 1: the hybrid energy-storing topology diagram being the embodiment of the present invention;

Fig. 2: the flow chart being the present embodiment;

Fig. 3: the wind-powered electricity generation real output being the embodiment of the present invention:

Fig. 4: be the process chart that improvement moving average method of the present invention extracts wind-electricity integration component and wave component;

Fig. 5: be the grid-connected component that obtains and wave component after wind-powered electricity generation real output and the process of the embodiment of the present invention,

Wherein (a) is wind power actual measured value and isolated lasting grid-connected component, and (b) is isolated hybrid energy-storing component;

Fig. 6: be fuzzy control flow chart of the present invention;

Fig. 7: the fluctuating power distribution condition being storage battery and super capacitor in the mixed energy storage system of the embodiment of the present invention; Fig. 8: the state-of-charge situation being dissimilar energy storage device after the utilization variable time constant filter method of the embodiment of the present invention;

Fig. 9: the particle cluster algorithm flow chart being the embodiment of the present invention;

Figure 10: be that the mixed energy storage system of the embodiment of the present invention stabilizes wind power output power ripple effect figure.

Embodiment

Understand for the ease of those of ordinary skill in the art and implement the present invention, below in conjunction with drawings and Examples, the present invention is described in further detail, should be appreciated that exemplifying embodiment described herein is only for instruction and explanation of the present invention, is not intended to limit the present invention.

Ask for an interview Fig. 1, be the hybrid energy-storing topology diagram of the present embodiment, mixed energy storage system is connected on the exit low-pressure side of wind energy turbine set through power inverter PCS, is typical centralized configuration.

Ask for an interview Fig. 2, the present embodiment, based on electric power verification system circuit, proposes a kind of mixed energy storage system capacity collocation method stabilizing wind power output power fluctuation, comprises the following steps:

Step 1: process the real data that active power of wind power field exports, uses the moving average method of improvement and in conjunction with national wind-electricity integration standard, extracts in wind power output power and need energy-storage system to carry out the wave component stabilized;

The wind energy turbine set installed capacity that the present embodiment is chosen is 20M, and real output data as shown in Figure 3; Figure 3 shows that certain installed capacity is the active power output valve (perunit value) of the wind energy turbine set actual measurement in 2014 of 20MW.

Ask for an interview Fig. 4, the particular content wherein improving moving average method is:

Select the window value of certain numerical value, then all numerical value in sliding window are carried out arithmetic average calculating, and using the numerical value of mean value as sliding window mid point, last constantly moving window repeats above process, until finally completing process to data obtains result.Use moving average method to process wind power output power data, wind power can be separated, obtain level and smooth grid-connected component and need energy-storage system to carry out power the wave component handled up.When using moving average method, suppose that sliding window numerical value is N min, and suppose that N is that even number is described to facilitate, obtain grid-connected component and the min level undulate quantity of t wind-powered electricity generation thus, shown in (1.1), (1.2), (1.3):

P _ft＝(P _t-(N/2-1)+P _t-(N/2-2)+...+P _t+...+P _t+N/2)/N (1.1)；

P _mt＝P _t-P _ft(1.2)；

t＝N/2,N/2+1,...,M-N/2 (1.3)；

In formula: P _tit is the wind power of actual measurement in t minute; P _ftit is grid-connected component; P _mtit is min level wave component; M is measurement point sum.

Moving average period N is an important parameter, but the selection of N has certain randomness.N numerical value is too little, then the wave component of wind power is too much superimposed upon on grid-connected component, causes power fluctuation to become large; N numerical value is too large, then the grid-connected component of wind power is too level and smooth, and wave component increases, and causes required stored energy capacitance greatly to increase.Research shows, is applicable to the common load possessing general characteristic during sliding window value 15min; In addition, for the load with greater impact characteristic, then select longer time span (30min).According to the load condition of example of calculation, choose N=15min.

Fixing moving average Period Length is selected to be conducive to calculating grid-connected component and wave component fast, but for the point that minority fluctuating range in wind power output power is very large, the grid-connected component likely obtained still does not meet the demands, therefore in conjunction with the requirement of national wind-electricity integration power fluctuation, the basis of moving average method is improved, the part not meeting fluctuation rate of change is adjusted, finally obtains the compensation power P of mixed energy storage system _hESS.

P _HESS＝P _mt+△P (1.4)；

In formula: Δ P is the power carrying out according to the fluctuation rate of change requirement of wind-powered electricity generation maximum power adjusting.

Its Chinese Home for the wind-electricity integration standard of different scales wind energy turbine set is:

Use the moving average method of improvement and in conjunction with national wind-electricity integration standard, extract in wind power output power and need energy-storage system to carry out the wave component stabilized, the grid-connected component obtained after final process and wave component are as shown in Figure 5.As seen from Figure 5, isolated lasting grid-connected component has good flatness, and wave component and hybrid energy-storing component then show larger fluctuation.

Step 2: the state-of-charge taking into account each energy storage device, uses the high-pass filtering method of variable time constant filter between dissimilar energy storage device, carry out the distribution of wave component;

The particular content of variable time constant filter method is:

The frequency height stabilizing power according to ultracapacitor and storage battery is different, and a kind of thinking uses the compensation power P of high-frequency filter equalizer mixed energy storage system _hESSshown in (1.5):

$\left\{\begin{array}{c}{P}_{\mathrm{UC}}=P_{\mathrm{HESS}}*\frac{T_{\mathrm{UC}}s}{T_{\mathrm{UC}}s+1}\\ P_{\mathrm{BESS}}=P_{\mathrm{HESS}}-P_{\mathrm{UC}}\end{array}\right.---\left(1.5\right)$

Wherein, P _uCand P _bESSbe respectively ultracapacitor and storage battery stabilize power, T _uCit is the time constant filter of high pass filter;

When adjusting time constant filter, fuzzy control theory is adopted to revise.The control strategy of people's subjectivity can be converted into computer language by fuzzy control, its basic procedure is processed into fuzzy quantity by after input signal obfuscation, obtained the output quantized by fuzzy inference system, thus control, wherein formulate the core that fuzzy rule is fuzzy control.Fuzzy control, compared with Traditional control, carrys out the decision-making of simulating human by rule base, be convenient to understand and implement.

Ask for an interview Fig. 6, the core content of fuzzy control comprises:

(1) obfuscation: the input variable within the scope of domain is carried out Fuzzy processing, obtains fuzzy subset and membership function;

(2) fuzzy rule is formulated: express the control decision of people's subjectivity for fuzzy control statement, form the domination set under different condition by many fuzzy control statements, this is the core of fuzzy control;

(3) fuzzy reasoning: according to fuzzy fuzzy quantity of exerting oneself;

(4) de-fuzzy; Fuzzy quantity is changed into the precise control amount in domain.

Based on above-mentioned fuzzy control principle, when carrying out hybrid energy-storing and stabilizing power fluctuation, consider the state-of-charge of each energy storage device, when state-of-charge approaches bound, timely adjustment time constant filter, revise the charge-discharge electric power instruction of energy storage device, thus ensure that the state-of-charge of energy storage device is within zone of reasonableness all the time.The target of fuzzy control is: when energy storage device state-of-charge is prescribed a time limit close to upper, if obtain charge power instruction, then regulate time constant filter suitably to reduce charge power, avoid overcharging; When energy storage device state-of-charge is prescribed a time limit close to lower, if obtain discharge power instruction, then regulate time constant filter to reduce discharge power, avoided putting.Improve energy storage device charging and discharging state by fuzzy control, reach the object increased the service life.Herein with storage battery charge state SoC _bESSwith ultracapacitor state-of-charge SoC _uCas input, with the correction factor k of time constant filter for exporting.

First to SoC _bESSand SoC _uCbe normalized, obtaining its degree of membership is:

${\mathrm{\&xi;}}_{\mathrm{bat}}=\frac{{\mathrm{SoC}}_{\mathrm{BESS}}-{\mathrm{SoC}}_{\mathrm{mid}}}{{\mathrm{SoC}}_{\mathrm{mid}}}---\left(1.6\right);$

${\mathrm{\&xi;}}_{\mathrm{uc}}=\frac{{\mathrm{SoC}}_{\mathrm{UC}}-{\mathrm{SoC}}_{\mathrm{mid}}}{{\mathrm{SoC}}_{\mathrm{mid}}}---\left(1.7\right);$

SoC in formula _midfor the median of energy storage device state-of-charge.

By SoC _min≤ SoC≤SoC _maxknown ξ _batand ξ _uccontinuous domain be respectively [-a, a] and [-b, b], the concrete size of a with b is relevant with energy storage device technical characteristic, because the depth of discharge of ultracapacitor is greater than storage battery, therefore b>a and 0<b<1.When energy storage device degree of membership is a or b, represents that capacity is saturated, be full of electricity completely; When energy storage device degree of membership is-a or-b, represents that capacity is exhausted, discharge completely.ξ _batand ξ _ucfuzzy set be { NB (negative large), ZO (zero), PB (honest) }.The output variable of fuzzy control is the correction factor k of time constant filter, and its discrete domain is [-1 ,-0.5,0,0.5,1], and fuzzy set is { NB, NS (negative little), ZO, PS (just little), PB}.

After the fuzzy set establishing input variable and output variable and membership function, now formulate fuzzy rule:

Work as P _hESS>0, when namely mixed energy storage system obtains the instruction of discharging, the following experience of fuzzy rule Main Basis is formulated:

(1) if the state-of-charge of storage battery and ultracapacitor is median, then time constant filter remains unchanged;

(2) if the state-of-charge of storage battery is less than normal and ultracapacitor state-of-charge is bigger than normal time, suitably tune up time constant filter T _uC, increase the discharge power of ultracapacitor, reduce the discharge power of storage battery;

(3) if the state-of-charge of storage battery is bigger than normal and ultracapacitor state-of-charge is less than normal time, suitably turn time constant filter T down _uC, reduce the discharge power of ultracapacitor, increase the discharge power of storage battery;

Work as P _hESS<0, when namely mixed energy storage system obtains the instruction of charging, the following experience of fuzzy rule Main Basis is formulated:

(1) if the state-of-charge of storage battery and ultracapacitor is median, then time constant filter remains unchanged;

(2) if the state-of-charge of storage battery is less than normal and ultracapacitor state-of-charge is bigger than normal time, suitably turn time constant filter T down _uC, reduce the charge power of ultracapacitor, increase the charge power of storage battery;

(3) if the state-of-charge of storage battery is bigger than normal and ultracapacitor state-of-charge is less than normal time, suitably tune up time constant filter T _uC, increase the charge power of ultracapacitor, reduce the charge power of storage battery;

Before state-of-charge judgement is carried out to storage battery and ultracapacitor, classification to be carried out to the state-of-charge of two kinds of energy storage devices, be divided into S _max, S _high, S _mid, S _lowand S _minfive classes.When state-of-charge mediates state, do not need to change time constant filter; When state-of-charge is on the low side, restriction electric discharge, increases charge power; When state-of-charge is higher, restriction charging, increases discharge power.Because storage battery is different with the technical characteristic of ultracapacitor, the two five class state-of-charges also difference to some extent, is specifically categorized as:

For obtaining output variable time constant filter correction factor k (-1≤k≤1), needing de-fuzzy to be exported accurately, adopting weighted mean method to carry out de-fuzzy process such as formula shown in (1.8) herein.

$k=\frac{\underset{i}{\mathrm{\&Sigma;}}\underset{j}{\mathrm{\&Sigma;}}{f}_{1i}\left({\mathrm{SoC}}_{\mathrm{BESS}}\right)f_{2j}\left({\mathrm{SoC}}_{\mathrm{UC}}\right)k_{\mathrm{ij}}}{\mathrm{\&Sigma;\&Sigma;}{f}_{1i}\left({\mathrm{SoC}}_{\mathrm{BESS}}\right)f_{2j}\left({\mathrm{SoC}}_{\mathrm{UC}}\right)}---\left(1.8\right);$

In formula:

F _1i(SoC _bESS) be input variable SoC _bESSi-th be subordinate to angle value;

F _2j(SoC _uC) be input variable SoC _uCjth be subordinate to angle value.

Obtaining revised time constant filter is:

T _UC ^*＝(1+k)T _UC(1.9)；

In mixed energy storage system after power division, storage battery and super capacitor stabilizes power and state-of-charge as shown in Figure 7,8.As shown in Figure 7, after power division, in the course of work of mixed energy storage system, ultracapacitor has stabilized the fluctuating power of high fdrequency component by discharge and recharge comparatively frequently, the fluctuating power component compared with low frequency then stabilized by storage battery, avoids discharge and recharge frequently.As shown in Figure 8, based on the high-pass filtering method of the variable time constant filter of fuzzy control, avoid the super-charge super-discharge of energy storage device, make its state-of-charge remain at a more rational numerical value, be conducive to subsequent time and fluctuating power is stabilized.

Step 3: set up the hybrid energy-storing selected topic capacity Optimal Allocation Model for the purpose of mixed energy storage system Life cycle economy optimum, use modified particle swarm optiziation, require as constraints with the technical characteristic of energy storage device and wind-electricity integration, carry out mixed energy storage system capacity and distribute rationally.

Wherein the target function of capacity Optimal Allocation Model and constraints are specially:

Minimum as optimization aim using the average annual expense of mixed energy storage system.Consider the long-time running of micro-grid system and energy-storage system needs to carry out necessary maintenance and replacing, obtain target function such as formula shown in (1.10):

$\left\{\begin{array}{c}\min C=\frac{C_{\mathrm{iv}}+C_{\mathrm{om}}+C_{\mathrm{dc}}}{Y}\\ C_{\mathrm{iv}}=m*O_{\mathrm{bat}}+n*O_{\mathrm{uc}}\\ C_{\mathrm{om}}=Y(m*k_{\mathrm{bat}}+n*k_{\mathrm{uc}})\\ C_{\mathrm{dc}}=p_{\mathrm{bat}}(m*O_{\mathrm{bat}})+p_{\mathrm{uc}}(n*O_{\mathrm{uc}})\end{array}\right.---\left(1.10\right);$

In formula: C is the average annual expense of mixed energy storage system; Y is energy-storage system running time; C _ivit is the acquisition cost (investment cost) of energy storage device; M and n is the equipment number of storage battery and super capacitor respectively; O _batand O _ucthe unit price of storage battery and super capacitor respectively; C _dcrefer to energy storage device disposal replacement cost (disposal cost), p _batand p _ucthe replacing batch of storage battery and super capacitor respectively; C _omrefer to the operation expense (operation & maintenance cost) of energy storage device, k _batand k _ucthe maintenance unit price of storage battery and super capacitor respectively.

The technical characteristic constraint of energy storage device, comprises the state-of-charge in charge and discharge process and charge and discharge power constraint, shown in (1.11):

$\left\{\begin{array}{c}{\mathrm{SoC}}_{\min }\&le;\mathrm{SoC}\&le;{\mathrm{SoC}}_{\max }\\ 0\&le;P_C\&le;P_{C.\max }\\ 0\&le;P_D\&le;P_{D.\max }\end{array}\right.---\left(1.11\right);$

In formula: SoC (state of charge) is the state-of-charge of energy storage device, and wherein the state-of-charge excursion of storage battery and super capacitor is respectively [0.2,1] and [0.1,1]; P _cand P _dthe charging and discharging power of energy storage device respectively.

For the capacity configuration model of the mixed energy storage system Life cycle economy optimum set up, using single battery number m and super capacitor monomer number n as the optimizing amount of solving, take into account wind-electricity integration standard and the constraint of energy storage device technical characteristic, particle cluster algorithm is adopted to solve, as shown in Figure 9, it specifically comprises the following steps process chart:

Step 3.1: the whole particle populations of initialization, setting iterations, material calculation, computational accuracy, target function;

Step 3.2: the fitness calculating each particle in an iterative process, obtains individual extreme value and the global extremum of each particle;

Step 3.3: the flying speed and the present position that are upgraded each particle by individual extreme value and global extremum;

Wherein inertia weight is introduced speed more new formula:

$v_{\mathrm{id}}^{k+1}={\mathrm{wv}}_{\mathrm{id}}^k+c_1{r}_1(p_{\mathrm{id}}-z_{\mathrm{id}}^k)+c_2{r}_2(p_{\mathrm{gd}}-z_{\mathrm{id}}^k)---\left(1.12\right);$

In formula, the flying speed of particle i in d dimension space after k+1 iteration; ω is inertia weight; c ₁and c ₂be Studying factors; r ₁and r ₂it is the random number of span (0,1); p _idit is the individual optimal value of particle; p _gdit is colony's optimal value of particle; be through the particle fitness value after k iteration;

When determining inertia weight value, first adopt adaptive weighting method to inertia weight coefficient assignment, its specific implementation formula is;

$\mathrm{\&omega;}=\left\{\begin{array}{cc}{\mathrm{\&omega;}}_{\min }+\frac{({\mathrm{\&omega;}}_{\max }-{\mathrm{\&omega;}}_{\min })*(f-f_{\min })}{f_{\mathrm{avg}}-f_{\min }},& f\&le;f_{\mathrm{avg}}\\ {\mathrm{\&mu;}}_{\max },& f>f_{\mathrm{avg}}\end{array}\right.---\left(1.13\right);$

In formula: ω _maxit is the maximum occurrences of inertia weight; ω _m ⁱ _nit is the minimum value of inertia weight; F is the target function of problem to be solved, f _avgit is the average target value of all particle adaptive values in colony; f _minthe minimum target value of all particle adaptive values in colony, μ _maxthe maximum of inertia weight;

Step 3.4: when reaching iterations or meeting stopping criterion for iteration, finishing iteration process also exports optimal solution.

Storage battery required in computational process and super capacitor parameter are:

Wherein, be C for rated capacity _bat(Ah) storage battery, when port voltage is U _bat, electric current is I _battime, the energy of energy-storage system of accumulator and power are such as formula shown in (1.12):

$\left\{\begin{array}{c}{E}_{\mathrm{bat}}=m{C}_{\mathrm{bat}}{U}_{\mathrm{bat}}\\ P_{\mathrm{bat}}={\mathrm{mU}}_{\mathrm{bat}}{I}_{\mathrm{bat}}\end{array}\right.---\left(1.14\right);$

Be C for capacitance _ucsuper capacitor, when terminal voltage is U _uc, electric current is I _uctime, the energy of super capacitor energy-storage system and power are such as formula shown in (1.13):

$\left\{\begin{array}{c}{E}_{\mathrm{uc}}=\frac{1}{2}n{C}_{\mathrm{uc}}{U_{\mathrm{uc}}}^2\\ P_{\mathrm{uc}}=n{U}_{\mathrm{uc}}{I}_{\mathrm{uc}}\end{array}\right.---\left(1.15\right);$

In computational process, the uniform units of energy is MWh, and the uniform units of power is MW.

Configuration result is:

Mixed energy storage system stabilizes the effect of wind power output power fluctuation as shown in Figure 10.As seen from Figure 10, after configuration mixed energy storage system, the output-power fluctuation of wind energy turbine set is well stabilized, and grid-connected power is smoother and meet Grid-connection standards.

Should be understood that, the part that this specification does not elaborate all belongs to prior art.

Should be understood that; the above-mentioned description for preferred embodiment is comparatively detailed; therefore the restriction to scope of patent protection of the present invention can not be thought; those of ordinary skill in the art is under enlightenment of the present invention; do not departing under the ambit that the claims in the present invention protect; can also make and replacing or distortion, all fall within protection scope of the present invention, request protection range of the present invention should be as the criterion with claims.

Claims (9)

1. stabilize a mixed energy storage system capacity collocation method for wind power output power fluctuation, it is characterized in that, comprise the following steps:

Step 1: process the real data that active power of wind power field exports, uses the moving average method of improvement and in conjunction with national wind-electricity integration standard, extracts in wind power output power and need energy-storage system to carry out the wave component stabilized;

Step 2: the state-of-charge taking into account each energy storage device, uses the high-pass filtering method of variable time constant filter between dissimilar energy storage device, carry out the distribution of wave component;

Step 3: set up the hybrid energy-storing selected topic capacity Optimal Allocation Model for the purpose of mixed energy storage system Life cycle economy optimum, use modified particle swarm optiziation, require as constraints with the technical characteristic of energy storage device and wind-electricity integration, carry out mixed energy storage system capacity and distribute rationally.

2. the mixed energy storage system capacity collocation method stabilizing wind power output power fluctuation according to claim 1, it is characterized in that: the moving average method of the improvement described in step 1 in conjunction with national wind-electricity integration standard, extracting in wind power output power needs energy-storage system to carry out the wave component stabilized, and its specific implementation process comprises following sub-step:

Step 1.1: the moving average method that utilization improves calculates level and smooth grid-connected component and need energy-storage system to carry out power the wave component handled up;

Suppose that sliding window numerical value is N min, and suppose that N is that even number is described to facilitate, obtain grid-connected component and the min level undulate quantity of t wind-powered electricity generation thus, shown in (1.1), (1.2), (1.3):

P _ft＝(P _t-(N/2-1)+P _t-(N/2-2)+...+P _t+...+P _t+N/2)/N (1.1)；

P _mt＝P _t-P _ft(1.2)；

t＝N/2,N/2+1,...,M-N/2 (1.3)；

In formula: P _tit is the wind power of actual measurement in t minute; P _ftit is grid-connected component; P _mtit is min level wave component; M is measurement point sum;

Step 2: in conjunction with the requirement of national wind-electricity integration power fluctuation, the basis of moving average method is improved, adjusts for the part not meeting fluctuation rate of change, finally obtains the compensation power P of mixed energy storage system _hESS;

P _HESS＝P _mt+△P (1.4)；

In formula: Δ P is the power carrying out according to the fluctuation rate of change requirement of wind-powered electricity generation maximum power adjusting.

3. the mixed energy storage system capacity collocation method stabilizing wind power output power fluctuation according to claim 2, is characterized in that: N=15min.

4. the mixed energy storage system capacity collocation method stabilizing wind power output power fluctuation according to claim 2, is characterized in that: described national wind-electricity integration standard asks for an interview table 1;

The national wind-electricity integration standard of table 1

5. the mixed energy storage system capacity collocation method stabilizing wind power output power fluctuation according to claim 1, it is characterized in that: the high-pass filtering method of the variable time constant filter of the utilization described in step 2 carries out the distribution of wave component between dissimilar energy storage device, its specific implementation process comprises following sub-step:

Step 2.1: the compensation power P using high-frequency filter equalizer mixed energy storage system _hESSshown in (2.1):

$\left\{\begin{array}{c}{P}_{\mathrm{UC}}=P_{\mathrm{HESS}}*\frac{T_{\mathrm{UC}}s}{T_{\mathrm{UC}}s+1}\\ P_{\mathrm{BESS}}=P_{\mathrm{HESS}}-P_{\mathrm{UC}}\end{array}\right.---\left(2.1\right)$

Wherein, P _uCand P _bESSbe respectively ultracapacitor and storage battery stabilize power, T _uCit is the time constant filter of high pass filter;

Step 2.2: adjustment time constant filter, adopts fuzzy control theory to revise; The core content of described fuzzy control theory comprises:

Obfuscation: the input variable within the scope of domain is carried out Fuzzy processing, obtains fuzzy subset and membership function;

Formulate fuzzy rule: express the control decision of people's subjectivity for fuzzy control statement, form the domination set under different condition by many fuzzy control statements, this is the core of fuzzy control;

Fuzzy reasoning: according to fuzzy fuzzy quantity of exerting oneself;

De-fuzzy; Fuzzy quantity is changed into the precise control amount in domain;

Step 2.3: based on above-mentioned fuzzy control theory, when carrying out hybrid energy-storing and stabilizing power fluctuation, consider the state-of-charge of each energy storage device, when state-of-charge approaches bound, timely adjustment time constant filter, revise the charge-discharge electric power instruction of energy storage device, thus ensure that the state-of-charge of energy storage device is within zone of reasonableness all the time.

6. the mixed energy storage system capacity collocation method stabilizing wind power output power fluctuation according to claim 5, it is characterized in that: the implementation procedure of the fuzzy control described in step 2 is, when energy storage device state-of-charge is prescribed a time limit close to upper, if obtain charge power instruction, then regulate time constant filter suitably to reduce charge power, avoid overcharging; When energy storage device state-of-charge is prescribed a time limit close to lower, if obtain discharge power instruction, then regulate time constant filter to reduce discharge power, avoided putting.

7. the mixed energy storage system capacity collocation method stabilizing wind power output power fluctuation according to claim 5, is characterized in that: the implementation procedure of the fuzzy control described in step 2 is, with storage battery charge state SoC _bESSwith ultracapacitor state-of-charge SoC _uCas input, with the correction factor k of time constant filter for exporting;

First to SoC _bESSand SoC _uCbe normalized, obtaining its degree of membership is:

${\mathrm{\&xi;}}_{\mathrm{bat}}=\frac{{\mathrm{SoC}}_{\mathrm{BESS}}-{\mathrm{SoC}}_{\mathrm{mid}}}{{\mathrm{SoC}}_{\mathrm{mid}}}---\left(2.2\right);$

${\mathrm{\&xi;}}_{\mathrm{uc}}=\frac{{\mathrm{SoC}}_{\mathrm{UC}}-{\mathrm{SoC}}_{\mathrm{mid}}}{{\mathrm{SoC}}_{\mathrm{mid}}}---\left(2.3\right);$

SoC in formula _midfor the median of energy storage device state-of-charge;

By SoC _min≤ SoC≤SoC _maxknown ξ _batand ξ _uccontinuous domain be respectively [-a, a] and [-b, b], the concrete size of a with b is relevant with energy storage device technical characteristic, because the depth of discharge of ultracapacitor is greater than storage battery, therefore b>a and 0<b<1; When energy storage device degree of membership is a or b, represents that capacity is saturated, be full of electricity completely; When energy storage device degree of membership is-a or-b, represents that capacity is exhausted, discharge completely; ξ _batand ξ _ucfuzzy set be { NB (negative large), ZO (zero), PB (honest) }; The output variable of fuzzy control is the correction factor k of time constant filter, and its discrete domain is [-1 ,-0.5,0,0.5,1], and fuzzy set is { NB, NS (negative little), ZO, PS (just little), PB};

After the fuzzy set establishing input variable and output variable and membership function, now formulate fuzzy rule:

Work as P _hESS>0, when namely mixed energy storage system obtains the instruction of discharging, the following experience of fuzzy rule Main Basis is formulated:

(1) if the state-of-charge of storage battery and ultracapacitor is median, then time constant filter remains unchanged;

(2) if the state-of-charge of storage battery is less than normal and ultracapacitor state-of-charge is bigger than normal time, suitably tune up time constant filter T _uC, increase the discharge power of ultracapacitor, reduce the discharge power of storage battery;

(3) if the state-of-charge of storage battery is bigger than normal and ultracapacitor state-of-charge is less than normal time, suitably turn time constant filter T down _uC, reduce the discharge power of ultracapacitor, increase the discharge power of storage battery;

Work as P _hESS<0, when namely mixed energy storage system obtains the instruction of charging, the following experience of fuzzy rule Main Basis is formulated:

(1) if the state-of-charge of storage battery and ultracapacitor is median, then time constant filter remains unchanged;

(2) if the state-of-charge of storage battery is less than normal and ultracapacitor state-of-charge is bigger than normal time, suitably turn time constant filter T down _uC, reduce the charge power of ultracapacitor, increase the charge power of storage battery;

(3) if the state-of-charge of storage battery is bigger than normal and ultracapacitor state-of-charge is less than normal time, suitably tune up time constant filter T _uC, increase the charge power of ultracapacitor, reduce the charge power of storage battery;

Before state-of-charge judgement is carried out to storage battery and ultracapacitor, classification to be carried out to the state-of-charge of two kinds of energy storage devices, be divided into S _max, S _high, S _mid, S _lowand S _minfive classes; When state-of-charge mediates state, do not need to change time constant filter; When state-of-charge is on the low side, restriction electric discharge, increases charge power; When state-of-charge is higher, restriction charging, increases discharge power;

Because storage battery is different with the technical characteristic of ultracapacitor, the two five class state-of-charges also difference to some extent, is specifically categorized as following table 2;

Table 2: five class state-of-charges of storage battery and ultracapacitor

For obtaining output variable time constant filter correction factor k (-1≤k≤1), needing de-fuzzy to be exported accurately, adopting weighted mean method to carry out de-fuzzy process such as formula shown in (2.4);

$k=\frac{\underset{i}{\mathrm{\&Sigma;}}\underset{j}{\mathrm{\&Sigma;}}{f}_{1i}\left({\mathrm{SoC}}_{\mathrm{BESS}}\right)f_{2j}\left({\mathrm{SoC}}_{\mathrm{UC}}\right)k_{\mathrm{ij}}}{\mathrm{\&Sigma;\&Sigma;}{f}_{1i}\left({\mathrm{SoC}}_{\mathrm{BESS}}\right)f_{2j}\left({\mathrm{SoC}}_{\mathrm{UC}}\right)}---\left(2.4\right)$

In formula:

F _1i(SoC _bESS) be input variable SoC _bESSi-th be subordinate to angle value; f _2j(SoC _uC) be input variable SoC _uCjth be subordinate to angle value;

Obtaining revised time constant filter is:

T _UC ^*＝(1+k)T _UC(2.5)。

8. the mixed energy storage system capacity collocation method stabilizing wind power output power fluctuation according to claim 1, is characterized in that: the hybrid energy-storing selected topic capacity Optimal Allocation Model described in step 3, and its target function and constraints are specially:

Described target function is such as formula shown in (3.1):

$\left\{\begin{array}{c}\min C=\frac{C_{\mathrm{iv}}+C_{\mathrm{om}}+C_{\mathrm{dc}}}{Y}\\ C_{\mathrm{iv}}=m*O_{\mathrm{bat}}+n*O_{\mathrm{uc}}\\ C_{\mathrm{om}}=Y(m*k_{\mathrm{bat}}+n*k_{\mathrm{uc}})\\ C_{\mathrm{dc}}=p_{\mathrm{bat}}(m*O_{\mathrm{bat}})+p_{\mathrm{uc}}(n*O_{\mathrm{uc}})\end{array}\right.---\left(3.1\right);$

In formula: C is the average annual expense of mixed energy storage system; Y is energy-storage system running time; C _ivit is the acquisition cost (investment cost) of energy storage device; M and n is the equipment number of storage battery and super capacitor respectively; O _batand O _ucthe unit price of storage battery and super capacitor respectively; C _dcrefer to energy storage device disposal replacement cost (disposal cost), p _batand p _ucthe replacing batch of storage battery and super capacitor respectively; C _omrefer to the operation expense (operation & maintenance cost) of energy storage device, k _batand k _ucthe maintenance unit price of storage battery and super capacitor respectively;

Described constraints, comprises the state-of-charge in charge and discharge process and charge and discharge power constraint, shown in (3.2):

$\left\{\begin{array}{c}{\mathrm{SoC}}_{\min }\&le;\mathrm{SoC}\&le;{\mathrm{SoC}}_{\max }\\ 0\&le;P_C\&le;P_{C.\max }\\ 0\&le;P_D\&le;P_{D.\max }\end{array}\right.---\left(3.2\right);$

In formula: SoC (state of charge) is the state-of-charge of energy storage device, and wherein the state-of-charge excursion of storage battery and super capacitor is respectively [0.2,1] and [0.1,1]; P _cand P _dthe charging and discharging power of energy storage device respectively.

9. the mixed energy storage system capacity collocation method stabilizing wind power output power fluctuation according to claim 8, it is characterized in that: the modified particle swarm optiziation described in step 3, its specific implementation comprises following sub-step:

Step 3.1: the whole particle populations of initialization, setting iterations, material calculation, computational accuracy, target function;

Step 3.2: the fitness calculating each particle in an iterative process, obtains individual extreme value and the global extremum of each particle;

Step 3.3: the flying speed and the present position that are upgraded each particle by individual extreme value and global extremum;

Wherein inertia weight is introduced speed more new formula:

$v_{\mathrm{id}}^{k+1}={\mathrm{wv}}_{\mathrm{id}}^k+c_1{r}_1(p_{\mathrm{id}}-z_{\mathrm{id}}^k)+c_2{r}_2(p_{\mathrm{gd}}-z_{\mathrm{id}}^k)---\left(3.1\right);$

In formula, the flying speed of particle i in d dimension space after k+1 iteration; ω is inertia weight; c ₁and c ₂be Studying factors; r ₁and r ₂it is the random number of span (0,1); p _idit is the individual optimal value of particle; p _gdit is colony's optimal value of particle; be through the particle fitness value after k iteration;

When determining inertia weight value, first adopt adaptive weighting method to inertia weight coefficient assignment, its specific implementation formula is;

$\mathrm{\&omega;}=\left\{\begin{array}{cc}{\mathrm{\&omega;}}_{\min }+\frac{({\mathrm{\&omega;}}_{\max }-{\mathrm{\&omega;}}_{\min })*(f-f_{\min })}{f_{\mathrm{avg}}-f_{\min }},& f\&le;f_{\mathrm{avg}}\\ {\mathrm{\&mu;}}_{\max },& f>f_{\mathrm{avg}}\end{array}\right.---\left(3.2\right);$

In formula: ω _maxit is the maximum occurrences of inertia weight; ω _minit is the minimum value of inertia weight; F is the target function of problem to be solved, f _avgit is the average target value of all particle adaptive values in colony; f _minthe minimum target value of all particle adaptive values in colony, μ _maxthe maximum of inertia weight;

Step 3.4: when reaching iterations or meeting stopping criterion for iteration, finishing iteration process also exports optimal solution.