CN104851053A - Wind-photovoltaic-energy-storage-contained method for power supply reliability evaluation method of distribution network - Google Patents

Abstract

The invention relates to a wind-photovoltaic-energy-storage-contained method for power supply reliability evaluation of a distribution network. The method comprises the following steps: time sequence simulation is carried out on a fan and a photovoltaic unit to obtain a time sequence power output model and an output time sequence of energy steerage is calculated; sampling and simulation of a time sequence operation state of an overall system are carried out by using a sequential Monte Carlo method, antithetic sampling is carried out to improve an algorithm precision, and calculation is carried out to obtain load power failure time reduced by wind-photovoltaic-energy storage system island operation and a frequency; and statistics of power supply reliability indexes of the overall system is carried out according to all load point indexes. Compared with the prior art, the method has high evaluation precision; the economy of the energy storage system is evaluation; and a reference basis is provided for an optimal allocated capacity for energy storage in an active distribution network in terms of reliability.

Description

A kind of distribution network reliability appraisal procedure containing wind-light storage

Technical field

The present invention relates to a kind of Reliability Estimation Method, especially relate to a kind of distribution network reliability appraisal procedure containing wind-light storage.

Background technology

The shortage of resource, the consumption of the energy and the demand of environmental protection facilitate the development of distributed new.Along with the development of active distribution network (active distribution network, ADN) technology ^[1-2], a large amount of distributed power sources (distributed generator, DG) accesses power distribution network, and distributed power source is exerted oneself as photovoltaic, wind-powered electricity generation etc. and had stochastic volatility, will produce a series of impact to distribution network reliability.Energy storage device relies on its fast power to regulate and the feature had concurrently for accumulation of energy power is exerted oneself for level and smooth DG has vital role, and the islet operation of the power distribution network made has feasibility, effectively improve the ability of power distribution network reliable power supply, is ingredient important in active distribution network.

Research for conventional electrical distribution net reliability assessment is more ripe, and appraisal procedure is analytical method mainly ^[3]and simulation ^[4].Analytical method, mainly through enumerating the fault state of system, utilizes reliability model of unit statistical system reliability index.Document [3] " appoint pretty; Zhang Yongjun; Ren Zhen; etc. the medium voltage distribution network reliability assessment based on feeder line piecemeal equivalence improves FMEA method [J]. Electric Power Automation Equipment; 2007,27 (12): 53-57. " and propose a kind of distribution network reliability block algorithm based on fault mode consequences analysis method and carry out reliability assessment.And simulation produces random number based on component reliability parameter and carries out sampling and then simulate running status that system may occur and calculate reliability index by the method for probability statistics.Document [4] " fourth is bright; Zhang Jing; Li Shenghu. based on the evaluating reliability of distribution network model [J] of sequential Monte Carlo emulation. and electric power network technique; 2004; 28 (3): 38-42. " consider circuit capacity constraint, trend distribution and load time-varying model, utilize sequential Monte-Carlo simulation to assess distribution network reliability.

Due to the access of DG and energy storage device, active distribution network is made to have the ability of islet operation, when feeder line in power distribution network breaks down may there is the decoupled mode containing DG in distribution system, this makes the Reliability Evaluation model of active distribution network and method produce great change compared to traditional distribution system, corresponding Calculation of Reliability process also becomes complicated thereupon, to the research that this Chinese scholars is correlated with.Wherein, part research have ignored the randomness that DG exerts oneself, and DG is reduced to controlled firm power source, and have ignored the effect of energy storage device.Document [5] " thunder shakes; Wei Gang; Cai Yang; etc. containing electricity distribution network model and the Calculation of Reliability [J] of distributed power source Area Node. Automation of Electric Systems; 2011; 35 (1): 39-43. " quote the network partitioning method of DG Area Node, consider that multiple DG access scheme is assessed power supply reliability.Document [6] " Liu Chuanquan; Zhang Yan. take into account the distribution network reliability [J] of distributed power source. Automation of Electric Systems; 2007; 31 (22): 46-49. " overall power distribution net decoupled mode, utilize a kind of minimal path method of improvement to the distribution network reliability analytical calculation containing distributed power source.Document [7] " CELLI G; GHIANI E; SOMA G.G; et al.Active distributionnetwork reliability assessment with a pseudo sequential Monte Carlo method [C] //PowerTech; June 19-23; 2011, IEEE Trondheim:1-8. " in conjunction with the Control management system in active distribution network, adopt the reliability of pseudo-sequential Monte Carlo simulation to active distribution network to assess in the area of two kinds of different management modes.And the effect considering energy storage device that another part research is concise and to the point, document [8] " beam benefits; Cheng Lin; Liu Sige. based on Monte Carlo simulation containing microgrid evaluating reliability of distribution network [J]. electric power network technique; 2011; 35 (10): 76-81. " establish wind-light storage model and adopt sequential Monte Carlo simulation to carry out reliability assessment to the power distribution network containing microgrid, but do not consider the sustainable time of stored energy capacitance on the impact of reliability and wind-light storage islet operation.This section of list of references is as follows:

[1] Zhao Bo, Wang Caisheng, Zhou Jinhui, etc. status in quo and prospect development [J] of active distribution network. Automation of Electric Systems, 2014,38 (18): 125-134.

[2] model tomorrow, Zhang Zuping, Su Aoxue, etc. the initiatively research [J] of distribution system possible technique. Proceedings of the CSEE, 2013,33 (22): 12-18.

[3] appoint pretty, Zhang Yongjun, Ren Zhen, etc. based on feeder line piecemeal equivalence medium voltage distribution network reliability assessment improve FMEA method [J]. Electric Power Automation Equipment, 2007,27 (12): 53-57.

[4] fourth is bright, Zhang Jing, Li Shenghu. based on the evaluating reliability of distribution network model [J] of sequential Monte Carlo emulation. and electric power network technique, 2004,28 (3): 38-42.

[5] thunder shakes, Wei Gang, Cai Yang, etc. containing electricity distribution network model and the Calculation of Reliability [J] of distributed power source Area Node. Automation of Electric Systems, 2011,35 (1): 39-43.

[6] Liu Chuanquan, Zhang Yan. take into account the distribution network reliability [J] of distributed power source. Automation of Electric Systems, 2007,31 (22): 46-49.

[7]CELLI G,GHIANI E,SOMA G.G,et al.Active distribution networkreliability assessment with a pseudo sequential Monte Carlo method[C]//PowerTech,June 19-23,2011,IEEE Trondheim:1-8.

[8] beam benefits, Cheng Lin, Liu Sige. based on Monte Carlo simulation containing microgrid evaluating reliability of distribution network [J]. electric power network technique, 2011,35 (10): 76-81.

Summary of the invention

Object of the present invention is exactly provide a kind of distribution network reliability appraisal procedure containing wind-light storage to overcome defect that above-mentioned prior art exists.

Object of the present invention can be achieved through the following technical solutions:

Containing the distribution network reliability appraisal procedure of wind-light storage, it is characterized in that, comprise the following steps: first time stimulatiom is carried out to blower fan, photovoltaic and obtain sequential and to exert oneself model, and calculate the time series of exerting oneself of energy storage; Then adopt the sequential running status of sequential Monte Carlo sampling emulation whole system, and improve arithmetic accuracy by antithesis sampling, calculate the load power off time and frequency that obtain and reduced by wind-light storage system islet operation; Whole system power supply reliability index is added up finally by each load point index.

The method is specially:

1) read in test macro data, choose test time limit N _year, emulation starts;

2) according to the sequential power stage curve of wind speed, intensity of illumination historical statistical data acquisition wind-power electricity generation, photovoltaic generation, and each node load curve is inputted;

3) exert oneself according to energy storage discharge and recharge strategy obtain the time series of exerting oneself of energy storage device in conjunction with wind, light;

4) the lasting failure free time T of all elements is obtained _f, and fault correction time T _r, thus obtain the operation-failure sequence of whole system, accumulative total simulation time;

5) analyze all event of failures, the impact that analysis of failure load is powered, determine to cause power failure the scope of load and position because of fault, the outer load point failure-frequency of statistics isolated island and power off time;

6) island operation state is judged, if isolated island internal component fault or energy storage fault, then cannot islet operation, the failure-frequency of cumulative statistics isolated island internal loading point and power off time; Otherwise perform step 7);

7) load in isolated island to be exerted oneself P in conjunction with blower fan during simulated time t _wtgt (), photovoltaic are exerted oneself P _sorload T normal working hours in situation of the exerting oneself calculating islet operation situation of (t) and energy storage device _b; If T _b> T _rillustrate that isolated island internal loading normally runs in the system failure period, not cumulative island internal loading power off time, and judge whether seamless switching, if yes, directly perform step 8), if NO, after accumulating once frequency of power cut, perform step 8); If T _b< T _rthen add up T between the power failure of isolated island internal loading _r-T _b, and perform step 8 after accumulating once frequency of power cut);

8) judge whether total simulation time is less than N _yearif be less than, proceed 4)-7) step, otherwise, add up the reliability index of each load point, and perform step 9);

9) carry out antithesis sampling, repeat step 3)-8), add up each load point reliability index, and obtain the statistical value of final load point reliability index;

10) obtained the power supply reliability index of whole system by load point statistical indicator, be analyzed with conventional electrical distribution net reliability; Change capacity of energy storing device, analyze stored energy capacitance on the impact of power supply reliability and obtain energy storage device build cost-benefit coefficient δ curve.

Described is specially according to the sequential power stage curve of wind speed, intensity of illumination historical statistical data acquisition wind-power electricity generation, photovoltaic generation:

1) the power stage model that can be obtained aerogenerator by the probability distribution of wind speed as shown in the formula:

In formula, P _wtgt () is exerted oneself for blower fan t; v _tfor t wind speed; v _in, v _n, v _outbe respectively the incision wind speed of blower fan, wind rating and cut-out wind speed; P _nfor the rated power of blower fan; a ₁, a ₂, a ₃for the coefficient of polynomial fitting of blower fan power curve non-linear partial;

2) output power of photovoltaic generation depends on intensity of illumination, and intensity of illumination obeys β distribution, and photovoltaic is exerted oneself as follows with the relational model of intensity of illumination:

$P_{\mathrm{sor}\left(t\right)}=\left\{\begin{array}{cc}{P}_{\mathrm{sn}}({I_t}^2/\left(I_{\mathrm{sn}}{I}_c\right))& 0\&le;I_t<I_c\\ P_{\mathrm{sn}}(I_t/I_{\mathrm{sn}})& I_c\&le;I_t<I_{\mathrm{sn}}\\ P_{\mathrm{sn}}& I_t\&GreaterEqual;I_{\mathrm{sn}}\end{array}\right.$

In formula, P _sort () is exerted oneself in real time for photovoltaic; P _snfor photovoltaic array rated power; I _t, I _c, I _snbe respectively t hour real-time lighting intensity, a certain setting light intensity and specified intensity of illumination.

The time series of exerting oneself of described energy storage device specifically obtains as follows:

$\left\{\begin{array}{cc}{Q}_{\mathrm{in}}={\&Integral;}_0^{T_b}[P_{\mathrm{wtg}}\left(t\right)+P_{\mathrm{sor}}\left(t\right)-P_L\left(t\right)]\mathrm{dt}& P_{\mathrm{wtg}}\left(t\right)+P_{\mathrm{sor}}\left(t\right)>P_L\left(t\right)\\ Q_{\mathrm{out}}={\&Integral;}_0^{T_b}[P_L\left(t\right)-(P_{\mathrm{wtg}}\left(t\right)+P_{\mathrm{sor}}\left(t\right))]\mathrm{dt}& P_{\mathrm{wtg}}\left(t\right)+P_{\mathrm{sor}}\left(t\right)<P_L\left(t\right)\end{array}\right.$

Energy storage device charge-discharge electric power need be less than maximum charge-discharge electric power permissible value, and the inner dump energy of energy storage device need meet certain capacity restriction simultaneously:

$\left\{\begin{array}{c}{P}_c\left(t\right)\&le;P_{\mathrm{in}-\max }\\ -P_c\left(t\right)\&le;P_{\mathrm{out}-\max }\\ Q_{\mathrm{re}}+Q_{\mathrm{in}}-Q_{\mathrm{out}}\&GreaterEqual;Q_{\min }\end{array}\right.$

In formula, T _bfor isolated island parallel-adder settle-out time; Q _inand Q _outfor the discharge and recharge of energy storage device in the islet operation duration; P _lt () is t payload; P _ct () is t charge-discharge electric power; P _in-maxand P _out-maxfor the maximum charge and discharge power of energy storage; Q _refor energy storage initial time electricity; Q _minit is the minimum value that the inner dump energy of energy storage need meet; Energy storage device grid-connected floating charge when if system is normally run, then Q _refor capacity of energy storing device.

The lasting failure free time T of all elements of described acquisition _f, and fault correction time T _rbe specially:

T _f＝-ln x/λ

T _r＝-ln y/μ

Thus simulating the operation of element-fault-time sequence, x, y meet the equally distributed random number between (0,1), and λ is element failure rate, and μ is element repair rate.

The all event of failures of described analysis are specially:

1) DG accesses electrical network by on-load switch, when first being disconnected by isolating switch on its upstream feeder line after DG breaks down, then isolated by DG by DG grid-connected place on-load switch, the load in isolating switch downstream has a power failure once, but due to the power supply recovering rapidly load, disregard power off time;

2) DG accesses electrical network by isolating switch, and now DG itself breaks down then out of service by this switch, disregards its impact on power distribution network normal power supply;

3) DG is accessed by isolating switch, and isolated island site are grid-connected by high-speed switch, can realize seamless switching, now not add up the frequency of power cut of isolated island internal load.

Described cost-benefit coefficient δ is calculated as follows:

The cost that energy storage device needs are considered is: acquisition cost, operating cost, maintenance cost, can be expressed from the next:

M _s＝M _cQ _c

M _k＝(k ₁+k ₂)M _s

In formula, M _s, M _kbe respectively acquisition cost and operation expense; k ₁, k ₂for operation, maintenance factor; Q _c, M _cbe respectively capacity and the unit capacity cost of investment of energy storage device;

From reliability perspectives, the income that energy storage device brings be due to reduce the system blackout time to user saves by the economic loss brought of power failure, represent with the income that the reduction Δ ENS of the not enough index of the total electricity of system brings, define the economy that following cost-benefit coefficient δ weighs energy storage device:

$\mathrm{\&delta;}=\frac{\mathrm{\&Delta;ENS}\&times;M_r}{M_s+M_k}$

In formula, M _rfor the loss of outage of unit power load, the income that this formula Middle molecule brings for accumulator system, denominator is investment and the operating cost of energy storage device, and therefore, the economic benefit of larger this accumulator system of expression of δ value is better.

Compared with prior art, the present invention has taken into full account the randomness and timing that in active distribution network, wind-powered electricity generation and photovoltaic generation are exerted oneself, and during islet operation energy storage discharge and recharge strategy on the impact of islet operation time, establish the model that active distribution network wind-light storage system is exerted oneself, and adopt antithesis Monte Carlo Analogue Method to assess system power supply reliability, improve the precision of traditional Monte carlo algorithm; In fault treating procedure, based on the different modes of distributed power source access electrical network, and whether isolated island has seamless switching ability, have employed different fault handling patterns, and reliability assessment result is tallied with the actual situation more; Finally, the investment construction cost of energy storage device is considered from cost-benefit angle, the economic return brought by the lifting of power supply reliability of evaluates calculation, based on the cost benefit coefficient of the energy storage device that the present invention defines, have evaluated the economy of accumulator system, provide reference frame from the best configuration capacity of reliability perspectives to energy storage active distribution network.

Accompanying drawing explanation

Fig. 1 is element outage model schematic diagram;

Fig. 2 is particular flow sheet of the present invention;

Fig. 3 is the IEEE RBTS BUS6 system schematic improved;

Fig. 4 is energy storage device cost-benefit ratio curve map.

Embodiment

Below in conjunction with the drawings and specific embodiments, the present invention is described in detail.

1. honourable power stage model

1.1 wind driven generator output power models

At present in research both at home and abroad, the probability distribution adopting Two-parameter Weibull Distribution to carry out matching wind speed more, probability density function is as follows:

$f\left(v\right)=\left(\frac{k}{c}\right){\left(\frac{v}{c}\right)}^{k-1}\exp {(-\frac{v}{c})}^k---\left(1\right)$

In formula, k and c is the two-parameter of Weibull distribution, and v is wind speed.

The power stage model that can be obtained aerogenerator by the probability distribution of wind speed as shown in the formula:

In formula, P _wtgt () is exerted oneself for blower fan t; v _tfor t wind speed; v _in, v _n, v _outbe respectively the incision wind speed of blower fan, wind rating and cut-out wind speed; P _nfor the rated power of blower fan.A ₁, a ₂, a ₃for the coefficient of polynomial fitting of power curve non-linear partial;

The power module of 1.2 photovoltaic generations

The output power of photovoltaic generation depends primarily on intensity of illumination, and intensity of illumination generally obeys β distribution, and photovoltaic is exerted oneself as follows with the relational model of intensity of illumination:

$P_{\mathrm{sor}\left(t\right)}=\left\{\begin{array}{cc}{P}_{\mathrm{sn}}({I_t}^2/\left(I_{\mathrm{sn}}{I}_c\right))& 0\&le;I_t<I_c\\ P_{\mathrm{sn}}(I_t/I_{\mathrm{sn}})& I_c\&le;I_t<I_{\mathrm{sn}}\\ P_{\mathrm{sn}}& I_t\&GreaterEqual;I_{\mathrm{sn}}\end{array}\right.---\left(3\right)$

In formula, P _sort () is exerted oneself in real time for photovoltaic; P _snfor photovoltaic array rated power; I _t, I _c, I _snbe respectively t hour real-time lighting intensity, a certain certain light intensity and specified intensity of illumination, unit is kW/m ².

1.3 blower fans and photovoltaic generation time series analysis

In existing research, many employings probability model carries out matching to wind speed and intensity of illumination and obtains data and calculate exerting oneself of wind-powered electricity generation photovoltaic, these methods can obtain wind speed in certain statistical time range and intensity of illumination data, but can not reflect the true timing condition of wind speed and intensity of illumination exactly.In order to obtain blower fan better, the sequential of photovoltaic generation exerts oneself model, reflect the timing of exerting oneself of wind-light storage system better, the present invention chooses the actual wind speed of assessment area and the statistics of intensity of illumination and the sequential obtaining blower fan and photovoltaic generation is thus exerted oneself.

The discharge and recharge strategy of 2 energy storage devices and mathematical model thereof

The discharge and recharge strategy of 2.1 energy storage devices

Accumulator system is the important component part of active distribution network, can the exerting oneself of effectively level and smooth wind-power electricity generation and photovoltaic, improve the utilization factor of wind energy, sun power, and the islet operation that the access of energy storage is power distribution network provides feasibility, effectively can improve the reliable power supply ability of system.When distribution network systems breaks down, system containing DG and energy storage device can continue to provide electric energy to the user in isolated island by islet operation, but the constraints such as the undulatory property of exerting oneself due to blower fan and photovoltaic generation, the finite capacity of accumulator system, the time that isolated island can maintain operation is also limited, if in the finite time of islet operation, faulty equipment is repaired, and isolated island is incorporated into the power networks again, and isolated island internal loading can obtain continued power; Otherwise if faulty equipment is greater than the maintained working time of isolated island repair time, then isolated island internal loading will lose power supply within a period of time.The present invention, when evaluates calculation power off time, has fully taken into account the sustainable time of isolated island based on said process.

Therefore, form isolated island based on randomness power supply and energy storage to power to power failure load restoration, the discharge and recharge strategy that isolated island can maintain time of operation and energy storage is closely related, the present invention takes into account blower fan and photovoltaic is exerted oneself, consider that following energy storage discharge and recharge strategy obtains the model of exerting oneself of energy storage: when the output power sum of blower fan and photovoltaic is greater than this moment load, energy storage device charges; When being less than load, energy storage device discharges.This kind of method of operation can make full use of wind-force and solar energy resources, obtains maximum blower fan photovoltaics and exerts oneself.

The mathematical model that 2.2 energy storage devices are exerted oneself

Following energy storage charging and recharging model is obtained by above-mentioned energy storage discharge and recharge strategy:

$\left\{\begin{array}{cc}{Q}_{\mathrm{in}}={\&Integral;}_0^{T_b}[P_{\mathrm{wtg}}\left(t\right)+P_{\mathrm{sor}}\left(t\right)-P_L\left(t\right)]\mathrm{dt}& P_{\mathrm{wtg}}\left(t\right)+P_{\mathrm{sor}}\left(t\right)>P_L\left(t\right)\\ Q_{\mathrm{out}}={\&Integral;}_0^{T_b}[P_L\left(t\right)-(P_{\mathrm{wtg}}\left(t\right)+P_{\mathrm{sor}}\left(t\right))]\mathrm{dt}& P_{\mathrm{wtg}}\left(t\right)+P_{\mathrm{sor}}\left(t\right)<P_L\left(t\right)\end{array}\right.---\left(4\right)$

Energy storage device charge-discharge electric power need be less than maximum charge-discharge electric power permissible value, and the inner dump energy of energy storage device need meet certain capacity restriction simultaneously:

$\left\{\begin{array}{c}{P}_c\left(t\right)\&le;P_{\mathrm{in}-\max }\\ -P_c\left(t\right)\&le;P_{\mathrm{out}-\max }\\ Q_{\mathrm{re}}+Q_{\mathrm{in}}-Q_{\mathrm{out}}\&GreaterEqual;Q_{\min }\end{array}\right.---\left(5\right)$

In formula, T _bfor isolated island parallel-adder settle-out time; Q _inand Q _outfor the discharge and recharge of energy storage device in the islet operation duration; P _lt () is t payload; P _ct () is t charge-discharge electric power; P _in-maxand P _out-maxfor the maximum charge and discharge power of energy storage; Q _refor energy storage initial time electricity; Q _minit is the minimum value that the inner dump energy of energy storage need meet.Energy storage device grid-connected floating charge when if system is normally run, then Q _refor capacity of energy storing device.

Based on above-mentioned model, when Reliability Evaluation, take into full account the supported working time of isolated island, i.e. T _b, the assessment result of distribution network reliability is tallied with the actual situation more, and this does not accomplish in the research of forefathers.

The cost-benefit of 3 energy storage devices

In active distribution network, energy storage device cooperation wind-force, photovoltaic generation possess islet operation ability, customer outage hours can be shortened, improve power supply reliability, but the Construction and operation of energy storage device needs to carry out great amount of investment, therefore inevitably need the economy problems considering that energy storage is built.

Energy storage device needs the cost of consideration mainly: acquisition cost, operating cost, maintenance cost etc., can be expressed from the next:

M _s＝M _cQ _c(6)

M _k＝(k ₁+k ₂)M _s(7)

In formula, M _s, M _kbe respectively acquisition cost and operation expense; k ₁, k ₂for operation, maintenance factor; Q _c, M _cbe respectively capacity and the unit capacity cost of investment of energy storage device; Above-mentioned each cost is the year value at cost after according to conversion tenure of use of energy storage device.

From reliability perspectives, the income that energy storage device brings is mainly due to the economic loss brought by power failure that the minimizing system blackout time saves to user, and the income that the present invention brings with the reduction Δ ENS of the not enough index of the total electricity of system represents.Invention defines the economy that following cost-benefit coefficient δ weighs energy storage device:

$\mathrm{\&delta;}=\frac{\mathrm{\&Delta;ENS}\&times;M_r}{M_s+M_k}---\left(8\right)$

In formula, M _rfor the loss of outage of unit power load, the income that this formula Middle molecule brings for accumulator system, denominator is investment and the operating cost of energy storage device, and therefore, the economic benefit of larger this accumulator system of expression of δ value is better.

4 containing the active distribution network reliability assessment of wind-light storage system

4.1 antithesis Monte Carlo Analogue Method

Monte Carlo Analogue Method is the important method of evaluating reliability of distribution network, is mainly divided into sequential and non-sequential Monte Carlo method.Wherein, sequential Monte Carlo method is the temporal model generation system status switch randomly according to element, then according to time ordered pair system state sequence carry out sampling analysis.Exert oneself in the active distribution network of the DG with undulatory property there is wind-powered electricity generation, photovoltaic generation etc., sequential Monte Carlo simulation can carry out sequential sampling to elements all in network and DG, more adequately simulates the running status of network.The coefficient of variation β of sample is the foundation judging this arithmetic accuracy and convergence:

$\mathrm{\&beta;}=\frac{\sqrt{V\left(F\right)}}{\tilde{E}\left(F\right)\&CenterDot;\sqrt{n}}---\left(9\right)$

In formula, for the estimated value of trial function sample average, V (F) is the variance of trial function sample, and the less then arithmetic accuracy of β is more accurate, and n is frequency in sampling.

There is the shortcoming that calculated amount is large, precision is low in tradition sequential Monte Carlo simulation, the present invention adopts antithesis Monte Carlo Analogue Method to improve the precision of Calculation of Reliability, document [10] obtains probability distribution and the border of section tidal current based on the method and boundary flow optimized algorithm, then predict the blocking probability of transmission cross-section, demonstrate the validity of method.The thought of the method is: Monte Carlo method when sampling to former random quantity, simultaneously to former stochastic variable negative correlation and another random quantity with identical expectation value sample, and determine the estimated value of variable by the linear combination of two random quantitys.

As stochastic variable X _iwhen meeting common distribution as exponential distribution, can by satisfied (0,1) upper another stochastic variable Y equally distributed _iobtained by mathematics change.By Y _iproduce one group of sample value, utilize 1-Y simultaneously _iproduce another group sample value, then negative correlation that what these two groups of sample values were tried to achieve be contemplated to be.Corresponding X is obtained through mathematics change respectively according to two groups of sample values _i ¹, X _i ², and try to achieve mathematical expectation E respectively ₁, E ₂, and the variance V of correspondence ₁(E ₁), V ₂(E ₂), then obtain total expectation according to linear transformation and be respectively with variance:

$E\left(X\right)=\frac{1}{2}(E_1+E_2)---\left(10\right)$

$V\left(X\right)=\frac{1}{4}[V_1\left(E_1\right)+V_2\left(E_2\right)+2\mathrm{cov}(V_1\left(E_1\right),V_2\left(E_2\right))]---\left(11\right)$

Due to E ₁, E ₂calculated by the stochastic variable of a pair negative correlation to obtain, then covariance cov (V ₁(E ₁), V ₂(E ₂)) < 0, therefore variance V (X) will reduce, and from formula (9), coefficient of variation β also will reduce, and the precision of Monte Carlo Analogue Method will get a promotion.

Fault handling pattern in 4.2 algorithms

The present invention to the tupe of power distribution network internal fault consider very specific and comprehensive, more accurate assessment result can be obtained, specific as follows:

In active distribution network, the fast recovery of power supply after active management and fault is carried out to DG, energy storage, various switchgear, need based on perfect electrical power distribution automatization system (Distribution Automation System, DAS), in power distribution network, all switches all achieve three distant (remote measurement, remote signalling, remote control) function, remote-controlled or the automatic detection of isolated island realizes and switches from net, under this assumed condition, the present invention is based on following fault handling pattern and reliability is analyzed.

When in electrical network, circuit, on-off element or transformer break down, the breaker actuation that fault element upstream is nearest, excision fault element, then the block switch that fault element upstream is nearest is opened, and after isolated fault element, non-faulting region, upstream restores electricity, therefore, the affected power failure load in upstream adds up a frequency of power cut, and due to the existence of automation equipment in power distribution network, does not consider the time of localization of fault and switching device grid switching operation.Therefore, upstream region power off time is shorter, disregards power off time.The outer load of an isolated island then accumulative frequency of power cut in the downstream that fault effects arrives, and according to the maintenance of equipment time cumulative statistics fault outage time.Whether the above-mentioned processing mode being isolated island external load, then seamless switching and islet operation duration can add up its frequency of power cut and power off time according to isolated island for isolated island internal load.When power supply in isolated island cannot meet all loads in isolated island, then according to significance level, sorted by load, optimum choice sub-load is powered.

As DG or energy storage device fault, generally be connected with electrical network by on-load switch or isolating switch due to when DG or energy storage access power distribution network, the difference of switchtype, fault handling pattern is also different, make on distribution network reliability, to there is different impacts during DG fault, the present invention has carried out analyzing respectively to three kinds of different patterns: 1) DG accesses electrical network by on-load switch, when first being disconnected by isolating switch on its upstream feeder line after DG breaks down, by DG grid-connected place on-load switch, DG is isolated again, the load in isolating switch downstream has a power failure once, but due to the power supply recovering rapidly load, disregard power off time, 2) DG accesses electrical network by isolating switch, and now DG itself breaks down then out of service by this switch, can disregard its impact on power distribution network normal power supply, 3) DG is accessed by isolating switch, and isolated island site (PCC point) is grid-connected by high-speed switch, can seamless switching be realized, now not add up the frequency of power cut of isolated island internal load.The present invention has carried out simulation analysis respectively to these three kinds of different modes.

The outage model of 4.3 elements

The present invention comprises DG, energy storage device to elements all in power distribution network and all adopts two state models, and as shown in Figure 1, element state is " operation-fault-operation ", and wherein λ is element failure rate, and μ is element repair rate.Think that element continues failure free time T _fwith fault correction time T _rall meet exponential distribution, its probability density function is:

$\left\{\begin{array}{c}f\left(t\right)={\mathrm{\&lambda;e}}^{-\mathrm{\&lambda;t}}\\ g\left(t\right)={\mathrm{\&mu;e}}^{-\mathrm{\&lambda;t}}\end{array}\right.---\left(12\right)$

In formula, f (t), g (t) are respectively the probability that element t breaks down and the probability repaired.Then T _f, T _rsampling can be carried out by formula (13) (14) based on uniform random number to obtain:

T _f＝-ln x/λ (13)

T _r＝-ln y/μ (14)

Thus simulating the operation of element-fault-time sequence, x, y meet the equally distributed random number between (0,1), and quantity is N, i.e. all numbers of elements.

4.4 reliability estimation method flow processs

Reliability estimation method main thought of the present invention is: carry out time stimulatiom to blower fan, photovoltaic and obtain sequential and to exert oneself model, and take into account the time series of exerting oneself of energy storage, adopt the sequential running status of sequential Monte Carlo sampling emulation whole system, and improve arithmetic accuracy by antithesis sampling, calculate the load power off time and frequency that obtain and reduced by wind-light storage system islet operation, and by each load point indicator-specific statistics whole system power supply reliability index.As shown in Figure 2, specific algorithm flow process is as follows:

1) read in test macro data, choose test time limit N _year, emulation starts.

2) utilize formula (2)-(3) to obtain the sequential power stage curve of wind-power electricity generation, photovoltaic generation according to wind speed, intensity of illumination historical statistical data, and input each node load curve.

3) according to energy storage discharge and recharge strategy, formula (4) is utilized to exert oneself obtain the time series of exerting oneself of energy storage device in conjunction with wind, light.

4) N number of satisfied (0,1) upper equally distributed random number x is obtained all elements T by formula (13) is generated respectively _f, in all elements, minimum normal working hours is as system up-time, then produces a random number y, obtains T by formula (14) _ras system failure time, thus obtain the operation-failure sequence of whole system, accumulative total simulation time.

5) analyze all event of failures, based on the fault handling pattern above described in 4.2 joints, fault is analyzed the impact that load is powered, determine to cause power failure the scope of load and position because of fault, the outer load point failure-frequency of statistics isolated island and power off time.Judge island operation state, if isolated island internal component fault or energy storage fault, then cannot islet operation, the failure-frequency of cumulative statistics isolated island internal loading point and power off time; Otherwise continue next step.

6) load in isolated island to be exerted oneself P in conjunction with blower fan during simulated time t, photovoltaic _wtg(t), P _sorload T normal working hours in situation of the exerting oneself calculating islet operation situation of (t) and energy storage device _b.If T _b> T _rillustrate that isolated island internal loading normally runs in the system failure period, cumulative island internal loading power off time, on the contrary the T that then adds up _r-T _btime and the number of stoppages.Work as T _b> T _rand isolated island is when can realize seamless switching, do not add up the frequency of power cut of isolated island internal loading.

7) judge whether total simulation time is less than N _yearif be less than, proceed 4)-6) process, otherwise, add up the reliability index of each load point.

8) obtain other N number of satisfied (0,1) equally distributed random number 1-x by x, carry out antithesis sampling, repetitive process 3)-7), add up each load point reliability index; Obtain the statistical value of final load point reliability index based on formula (10) in conjunction with preceding step statistical value.

9) obtained the power supply reliability index of whole system by load point statistical indicator, be analyzed with conventional electrical distribution net reliability; Change capacity of energy storing device, analyze stored energy capacitance on the impact of power supply reliability and obtain energy storage device build cost-benefit coefficient δ curve.

Application example

Below in conjunction with case, the present invention is described in detail.

The present invention utilizes the statistical value of the annual wind speed in somewhere, Shanghai and intensity of illumination, utilizes matlab software programming simulated program that the IEEE RBTS BUS6 system that Fig. 3 improves is carried out to reliability assessment and analyzed result.In blower fan, photovoltaic and energy storage device connecting system F5 feeder line, forming isolated island during upstream failure is that load LP14-LP18 powers.System amounts to 30 circuits, 23 distribution transformings, 23 load point, and line length and the data such as load index and number of users are shown in document [12], and line failure rate is 0.06 time/(akm ^-1); Blower fan rated power is 2MW, v _in, v _n, v _outbe respectively 2.5m/s, 8m/s, 20m/s; Photovoltaic battery panel rated power is 1MW, I _c, I _snbe respectively 0.15kW/m2,0.8kW/m2; The capacity of energy storage device is 2MWh, and maximum charge-discharge electric power is 0.5MW, inner residue minimum capacity Q _minfor 0.2MWh.In system, the dependability parameter of each element is as shown in table 1:

Table 1

Analysis of simulation result

The system power supply reliability index that the present invention calculates has: system System average interruption frequency index (SAIFI), system System average interruption duration index (SAIDI), user on average have a power failure lasting index (CAIDI), on average to power Availability Index (ASAI) and the not enough index (ENS) of the total electricity of system.Simulation time is 400 years, four kinds of situations are divided to carry out simulation analysis to example shown in Fig. 3, system power supply reliability index is obtained as shown in table 2 through simulation calculation, wherein in pattern one, DG accesses electrical network by on-load switch, in pattern two, DG accesses electrical network by isolating switch, pattern three kinds of DG also access electrical network by isolating switch, and simultaneously the PCC point place switch of isolated island is fast chopper, can realize and from net seamless switching.Last is classified as not containing the power supply reliability index of this power distribution network of emulation gained during wind-light storage.

Table 2

The cost-benefit analysis of energy storage device

According to formula (6)-(8), being assumed to 600,000 yuan of a/MWh, is 1,000,000 yuan/MWh, operation maintenance coefficient, obtains because installed capacity changes the cost-benefit curve brought, as shown in Figure 4:

As can be seen from Figure 4, along with the growth of capacity of energy storing device, the power off time of system reduces, and cost-benefit ratio δ increases, but increasing degree is more and more slow.Increase after reaching peak value, the lifting due to capacity of energy storing device cannot reduce the power off time of system, and now investment continues growth cannot bring more income, and curve starts to decline.δ > 1 shows that income is by unnecessary investment, and namely energy storage device brings economic return while bringing power supply reliability to promote under this corresponding capacity.

Curve obtains maximum cost-benefit ratio when reaching peak value, and the stored energy capacitance now configured is about 2.85MWh, and the optimum reaching system reliability and economy is coordinated, and this capacity is this system optimal stored energy capacitance.

Claims (7)

1. containing the distribution network reliability appraisal procedure of wind-light storage, it is characterized in that, comprise the following steps: first time stimulatiom is carried out to blower fan, photovoltaic and obtain sequential and to exert oneself model, and calculate the time series of exerting oneself of energy storage; Then adopt the sequential running status of sequential Monte Carlo sampling emulation whole system, and improve arithmetic accuracy by antithesis sampling, calculate the load power off time and frequency that obtain and reduced by wind-light storage system islet operation; Whole system power supply reliability index is added up finally by each load point index.

2. a kind of distribution network reliability appraisal procedure containing wind-light storage according to claim 1, it is characterized in that, the method is specially:

1) read in test macro data, choose test time limit N _year, emulation starts;

2) according to the sequential power stage curve of wind speed, intensity of illumination historical statistical data acquisition wind-power electricity generation, photovoltaic generation, and each node load curve is inputted;

3) exert oneself according to energy storage discharge and recharge strategy obtain the time series of exerting oneself of energy storage device in conjunction with wind, light;

4) the lasting failure free time T of all elements is obtained _f, and fault correction time T _r, thus obtain the operation-failure sequence of whole system, accumulative total simulation time;

5) analyze all event of failures, the impact that analysis of failure load is powered, determine to cause power failure the scope of load and position because of fault, the outer load point failure-frequency of statistics isolated island and power off time;

6) island operation state is judged, if isolated island internal component fault or energy storage fault, then cannot islet operation, the failure-frequency of cumulative statistics isolated island internal loading point and power off time; Otherwise perform step 7);

7) load in isolated island to be exerted oneself P in conjunction with blower fan during simulated time t _wtgt (), photovoltaic are exerted oneself P _sorload T normal working hours in situation of the exerting oneself calculating islet operation situation of (t) and energy storage device _b; If T _b> T _rillustrate that isolated island internal loading normally runs in the system failure period, not cumulative island internal loading power off time, and judge whether seamless switching, if yes, directly perform step 8), if NO, after accumulating once frequency of power cut, perform step 8); If T _b< T _rthen add up T between the power failure of isolated island internal loading _r-T _b, and perform step 8 after accumulating once frequency of power cut);

8) judge whether total simulation time is less than N _yearif be less than, proceed 4)-7) step, otherwise, add up the reliability index of each load point, and perform step 9);

9) carry out antithesis sampling, repeat step 3)-8), add up each load point reliability index, and obtain the statistical value of final load point reliability index;

10) obtained the power supply reliability index of whole system by load point statistical indicator, be analyzed with conventional electrical distribution net reliability; Change capacity of energy storing device, analyze stored energy capacitance on the impact of power supply reliability and obtain energy storage device build cost-benefit coefficient δ curve.

3. a kind of distribution network reliability appraisal procedure containing wind-light storage according to claim 2, is characterized in that, described is specially according to the sequential power stage curve of wind speed, intensity of illumination historical statistical data acquisition wind-power electricity generation, photovoltaic generation:

1) the power stage model that can be obtained aerogenerator by the probability distribution of wind speed as shown in the formula:

In formula, P _wtgt () is exerted oneself for blower fan t; v _tfor t wind speed; v _in, v _n, v _outbe respectively the incision wind speed of blower fan, wind rating and cut-out wind speed; P _nfor the rated power of blower fan; a ₁, a ₂, a ₃for the coefficient of polynomial fitting of blower fan power curve non-linear partial;

2) output power of photovoltaic generation depends on intensity of illumination, and intensity of illumination obeys β distribution, and photovoltaic is exerted oneself as follows with the relational model of intensity of illumination:

$P_{\mathrm{sor}}\left(t\right)=\left\{\begin{array}{cc}{P}_{\mathrm{sn}}({I_t}^2/\left(I_{\mathrm{sn}}{I}_c\right))& 0\&le;I_t<I_c\\ P_{\mathrm{sn}}(I_t/I_{\mathrm{sn}})& I_c\&le;I_t<I_{\mathrm{sn}}\\ P_{\mathrm{sn}}& I_t\&GreaterEqual;I_{\mathrm{sn}}\end{array}\right.$

In formula, P _sort () is exerted oneself in real time for photovoltaic; P _snfor photovoltaic array rated power; I _t, I _c, I _snbe respectively t hour real-time lighting intensity, a certain setting light intensity and specified intensity of illumination.

4. a kind of distribution network reliability appraisal procedure containing wind-light storage according to claim 3, it is characterized in that, the time series of exerting oneself of described energy storage device specifically obtains as follows:

$\left\{\begin{array}{cc}{Q}_{\mathrm{in}}={\&Integral;}_0^{T_b}[P_{\mathrm{wtg}}\left(t\right)+P_{\mathrm{sor}}\left(t\right)-P_L\left(t\right)]\mathrm{dt}& P_{\mathrm{wtg}}\left(t\right)+P_{\mathrm{sor}}\left(t\right)>P_L\left(t\right)\\ Q_{\mathrm{out}}={\&Integral;}_0^{T_b}[P_L\left(t\right)-(P_{\mathrm{wtg}}\left(t\right)+p_{\mathrm{sor}}\left(t\right))]\mathrm{dt}& P_{\mathrm{wtg}}\left(t\right)+P_{\mathrm{sor}}\left(t\right)<P_L\left(t\right)\end{array}\right.$

Energy storage device charge-discharge electric power need be less than maximum charge-discharge electric power permissible value, and the inner dump energy of energy storage device need meet certain capacity restriction simultaneously:

$\left\{\begin{array}{c}{P}_c\left(t\right)\&le;P_{\mathrm{in}-\max }\\ -P_c\left(t\right)\&le;P_{\mathrm{out}-\max }\\ Q_{\mathrm{re}}+Q_{\mathrm{in}}-Q_{\mathrm{out}}\&GreaterEqual;Q_{\min }\end{array}\right.$

In formula, T _bfor isolated island parallel-adder settle-out time; Q _inand Q _outfor the discharge and recharge of energy storage device in the islet operation duration; P _lt () is t payload; P _ct () is t charge-discharge electric power; P _in-maxand P _out-maxfor the maximum charge and discharge power of energy storage; Q _refor energy storage initial time electricity; Q _minit is the minimum value that the inner dump energy of energy storage need meet; Energy storage device grid-connected floating charge when if system is normally run, then Q _refor capacity of energy storing device.

5. a kind of distribution network reliability appraisal procedure containing wind-light storage according to claim 4, is characterized in that, the lasting failure free time T of all elements of described acquisition _f, and fault correction time T _rbe specially:

T _f＝-lnx/λ

T _r＝-lny/μ

Thus simulating the operation of element-fault-time sequence, x, y meet the equally distributed random number between (0,1), and λ is element failure rate, and μ is element repair rate.

6. a kind of distribution network reliability appraisal procedure containing wind-light storage according to claim 5, it is characterized in that, all event of failures of described analysis are specially:

1) DG accesses electrical network by on-load switch, when first being disconnected by isolating switch on its upstream feeder line after DG breaks down, then isolated by DG by DG grid-connected place on-load switch, the load in isolating switch downstream has a power failure once, but due to the power supply recovering rapidly load, disregard power off time;

2) DG accesses electrical network by isolating switch, and now DG itself breaks down then out of service by this switch, disregards its impact on power distribution network normal power supply;

3) DG is accessed by isolating switch, and isolated island site are grid-connected by high-speed switch, can realize seamless switching, now not add up the frequency of power cut of isolated island internal load.

7. a kind of distribution network reliability appraisal procedure containing wind-light storage according to claim 6, it is characterized in that, described cost-benefit coefficient δ is calculated as follows:

The cost that energy storage device needs are considered is: acquisition cost, operating cost, maintenance cost, can be expressed from the next:

M _s＝M _cQ _c

M _k＝(k ₁+k ₂)M _s

In formula, M _s, M _kbe respectively acquisition cost and operation expense; k ₁, k ₂for operation, maintenance factor; Q _c, M _cbe respectively capacity and the unit capacity cost of investment of energy storage device;

From reliability perspectives, the income that energy storage device brings be due to reduce the system blackout time to user saves by the economic loss brought of power failure, represent with the income that the reduction Δ ENS of the not enough index of the total electricity of system brings, define the economy that following cost-benefit coefficient δ weighs energy storage device:

$\mathrm{\&delta;}=\frac{\mathrm{\&Delta;ENS}\&times;M_r}{M_s+M_k}$

In formula, M _rfor the loss of outage of unit power load, the income that this formula Middle molecule brings for accumulator system, denominator is investment and the operating cost of energy storage device, and therefore, the economic benefit of larger this accumulator system of expression of δ value is better.