CN104348189B - A kind of distributed power supply system - Google Patents

Abstract

The present invention relates to a kind of distributed power supply system. This system has intelligent monitor system, photovoltaic module that can prediction distribution formula power supply, the generated output of wind-powered electricity generation module, the situation of change of prediction load, and the ruuning situation of energy storage situation, the ruuning situation of diesel-driven generator and the power distribution network of Real-time Obtaining of battery module based on real-time detection, implement optimum control strategy.

Description

A kind of distributed power supply system

Affiliated technical field

The present invention relates to a kind of distributed power supply system.

Background technology

Along with economic development, electricity needs increases rapidly, and electrical network scale constantly expands, the drawback of ultra-large power systemAlso day by day show especially: operation stability is poor, be difficult to adapt to user to electric power safety and reliability requirement and diversified power demands.

Distributed power source (distributedgeneration, DG) is a kind of emerging electric power energy, comprises that photovoltaic sends outElectric system, wind generator system, generation system of micro turbine etc. Distributed generation technology has environmental protection, economic dispatch one isRow advantage, can meet the requirement of people's and economic environmental protection stable to electric power safety well, has caused extensive concern, and graduallyBe promoted and develop.

But these small-sized renewable energy power generation modes basically, exist that capacity is little, product efficiency is low, power supplyThe shortcomings such as poor reliability, cost of electricity-generating height, have greatly hindered the large-scale promotion application of such electricity generation system and have further sent outExhibition.

In addition, DG access power distribution network has brought new problem and challenge also to protection and the control of power distribution network. Photovoltaic and windElectric energy is a kind of intermittent energy source, inevitably exists generated output along with the strong and weak fluctuation of illumination and wind-force while being incorporated into the power networksThe shortcoming of strong and weak fluctuation. Therefore, the distributed power source access power distribution network of high permeability, operation, the scheduling that will inevitably give power distribution networkBring a lot of negative effects with management. In order to make power distribution network can adapt to the grid integration of distributed power source, in the urgent need to distributingFormula power supply operation is controlled and Access Control key technology is furtherd investigate, thereby reduce, distributed power source grid integration bringsAdverse effect is brought into play its positive booster action simultaneously.

Summary of the invention

For addressing the above problem, the invention provides a kind of distributed power supply system, this system has intelligent monitor system, canWith the photovoltaic module of prediction distribution formula power supply, the generated output of wind-powered electricity generation module, the situation of change of prediction load, and based on real-time inspectionThe ruuning situation of energy storage situation, the ruuning situation of diesel-driven generator and the power distribution network of Real-time Obtaining of the battery module of surveying, realExecute optimum control strategy.

To achieve these goals, the invention provides a kind of distributed power supply system, this distributed power supply system is by generating electricityAnd load system and monitoring system composition, described generating and load system comprise photovoltaic generating module, wind-powered electricity generation module, battery mouldPiece, diesel generator, load, dc bus, multiple AC/DC or DC/AC module, is characterized in that, this monitoring system comprises:

Photovoltaic generation monitoring module, for real-time monitoring photovoltaic generating module, and generated output to photovoltaic generating modulePredict;

Wind-powered electricity generation monitoring module, for monitoring in real time wind-powered electricity generation module, and predicts the generated output of wind-powered electricity generation module;

Battery monitor module, for monitoring in real time battery module;

Diesel-driven generator monitoring module, for monitoring in real time diesel-driven generator;

Load monitoring module, for the described load of monitoring in real time, and predicts the power situation of change of load;

Power distribution network contact module, for knowing in real time ruuning situation and relevant tune of power distribution network from power distribution network regulation and control centerDegree information;

The monitoring module that is incorporated into the power networks, for controlling, distributed power source connects or isolation power distribution network;

Middle control module, for determining the operation strategy of distributed power supply system, and to the each module in above-mentioned monitoring systemSend instruction, to carry out this operation strategy;

Bus module, for the liaison of the modules of this monitoring system;

Described photovoltaic generation monitoring module, wind-powered electricity generation monitoring module, load monitoring module include detecting unit and storage is singleUnit, can Real-time Obtaining distributed power source in photovoltaic generating module, the service data of wind-powered electricity generation module and the load variations feelings of loadCondition, and store data;

Described photovoltaic generation monitoring module, wind-powered electricity generation monitoring module, load monitoring module include arithmetic element, can distinguishAccording to the service data of existing photovoltaic generating module and wind-powered electricity generation module, in the distributed power source in following predetermined instant can be againRaw energy power output predicts, according to existing loaded load variations situation, the workload demand of load predicted;

Described battery monitor module, has secondary battery unit and discharges and recharges detection module, can detect in real time and obtain electric power storageThe energy storage of pond module charges and discharge capacity;

Described middle control module, has ALU, can be according to the schedule information of power distribution network, current batteries to store energyThe capacity that charges and discharge, following distributed power source in renewable sources of energy power output, diesel-driven generator power output and to futureThe variation of workload demand, as constraints, is set up distributed power source and is moved tactful object function;

Described power distribution network contact module can Real-time Obtaining power distribution network parameter comprise: bus numbering, title, load haveThe branch road of merit, reactive load, circuit number, headend node and endpoint node numbering, series resistance, series reactance;

Predict in the following way the power output of wind-powered electricity generation module at described wind-powered electricity generation monitoring module, described wind-powered electricity generation module comprisesWind-driven generator and SVG:

(201) in collection wind-powered electricity generation module, current all kinds of electric weight measured values are as the initial value of the predicted value of all kinds of electric weight, pre-Measured value comprises: the blower fan predicted value of gaining meritPredicted value that blower fan is idleBlower fan set end voltage predicted valueSVG is idlePredicted valueSVG set end voltage predicted valueWind-powered electricity generation module site (PCC) busbar voltage predicted value

(202) set up according to described predicted value the MPC optimizing control models being formed by optimization aim function and constraints,And solve the predicted value of the meritorious and idle output of wind-powered electricity generation module:

The object function of MPC optimizing control models is suc as formula shown in (1):

$\underset{Q_{WTG}^{set},V_{SVG}^{set}}{\min }(\underset{i=1}{\overset{N-1}{\Σ}}\underset{j=0}{\overset{M-1}{\Σ}}{\mathrm{\ρ}}^{t_{i,j}}{F}_1,\underset{i=1}{\overset{N-1}{\Σ}}\underset{j=0}{\overset{M-1}{\Σ}}{\mathrm{\ρ}}^{t_{i,j}}{F}_2)---\left(1\right)$

In formula (1)WithFor optimized variable,WithImplication is respectively the idle setting value of blower fan and SVG voltageSetting value; N is the number in time window Coverage Control cycle; M is the number containing future position under single control cycle; ρ is decay systemNumber, value ρ < 1; Time variable ti, j=(Mi+j) Δ t meaning is that current time plays j prediction in i control cyclePoint, Δ t is future position interval, Δ t is by wind-powered electricity generation modular power predicted time interval determination;

F1 is the variance level of wind-powered electricity generation module site busbar voltage and setting value, and F1 expression is suc as formula (2):

$F_1\left(t_{i,j}\right)={\&lsqb;V_{PCC}^{pre}\left(t_{i,j}\right)-V_{PCC}^{ref}\&rsqb;}^2---\left(2\right)$

In formula (2)Represent the reference value of PCC voltage, after extracting from main website control instruction, set;

F2 is the idle level of reserve of SVG, and F2 expression is suc as formula (3):

$F_2\left(t_{i,j}\right)={\&lsqb;Q_{SVG}^{pre}\left(t_{i,j}\right)-Q_{SVG}^{opr}\&rsqb;}^2---\left(3\right)$

In formula (3)For the idle best operating point of SVG;

The constraints of MPC optimizing control models, specifically comprises:

The blower fan prediction-constraint condition of gaining merit:

$P_{WTG}^{pre}\left(t_{i,j}\right)=\underset{k=1}{\overset{N_a}{\&Sigma;}}{\mathrm{\&phi;}}_k{P}_{WTG}^{pre}\left(t_{i,j-k}\right)+{\mathrm{\&epsiv;}}_{WTG}^{pre}\left(t_{i,j}\right)-\underset{k=1}{\overset{N_m}{\&Sigma;}}{\mathrm{\&theta;}}_k{\mathrm{\&epsiv;}}_{WTG}^{pre}\left(t_{i,j-k}\right)---\left(4\right)$

In formula (4)For the meritorious predicated error of blower fan; Na and Nm are respectively the exponent number of AR and MA model, and φ k and θ k are phaseClose weight, exponent number and weight are all determined according to the meritorious history value of blower fan; Ti, j-k (comprises for participating in calculated data in prediction) the corresponding moment, subscript k pushes away the k Δ t time before characterizing the prediction moment, works as ti, j-k≤0 o'clock, and meritorious predicted value should be gotCorresponding moment history value;

Prediction-constraint condition that blower fan is idle:

Blower fan is idle reaches setting value before controlling next time:

$Q_{WTG}^{pre}\left(t_{i,0}\right)=Q_{WTG}^{set}\left(t_{i-1,0}\right)---\left(5\right)$

Each future position in i control cycle, the change procedure of blower fan reactive power is with exponential function matching:

$Q_{WTG}^{pre}\left(t_{i,j}\right)=\frac{1-e^{-(t_{i,j}-t_{i,0})/T_s}}{1-e^{-M\mathrm{\&Delta;}t/T_s}}{Q}_{WTG}^{set}\left(t_{i,0}\right)+\frac{e^{-(t_{i,j}-t_{i,0})/T_s}-e^{-M\mathrm{\&Delta;}t/T_s}}{1-e^{-M\mathrm{\&Delta;}t/T_s}}{Q}_{WTG}^{pre}\left(t_{i,0}\right)---\left(6\right)$

In formula (6), Ts is the idle adjusting time constant of blower fan, can obtain according to the idle adjusting testing experiment of blower fan;

Prediction-constraint condition that SVG is idle:

Reference value that SVG is idleShown in (7):

$\begin{array}{c}{Q}_{SVG}^{ref}\left(t_{i,j}\right)=K_P\&lsqb;V_{SVG}^{pre}\left(t_{i,j}\right)-V_{SVG}^{set}\left(t_{i,0}\right)\&rsqb;+K_I\mathrm{\&Delta;}t\underset{k=0}{\overset{i\&times;M+j}{\&Sigma;}}\&lsqb;V_{SVG}^{pre}\left(t_{i,j-k}\right)-V_{SVG}^{set}\left(t_{i,-k}\right)\&rsqb;\\ +Q_{SVG}^{pre}\left(t_{0,0}\right)-K_P\&lsqb;V_{SVG}^{pre}\left(t_{0,0}\right)-V_{SVG}^{set}\left(t_{0,0}\right)\&rsqb;\end{array}---\left(7\right)$

In formula (7), KI and KP are respectively the coefficient of proportional component and integral element;

Predicted value that SVG is idle is suc as formula shown in (8):

$Q_{SVG}^{pre}\left(t_{i,j}\right)=Q_{SVG}^{ref}\left(t_{i,j-1}\right)+\&lsqb;Q_{SVG}^{pre}\left(t_{i,j-1}\right)-Q_{SVG}^{ref}\left(t_{i,j-1}\right)\&rsqb;e^{-(t_{i,j}-t_{i,j-1})/T_d}---\left(8\right)$

In formula (8), time constant Td is SVG power electronic equipment action delay;

Voltage prediction constraints:

$V^{pre}\left(t_{i,j}\right)-V^{pre}\left(t_{0,0}\right)=S\left[\begin{array}{c}{P}_{WTG}^{pre}\left(t_{i,j}\right)-P_{WTG}^{pre}\left(t_{0,0}\right)\\ Q_{WTG}^{pre}\left(t_{i,j}\right)-Q_{WTG}^{pre}\left(t_{0,0}\right)\\ Q_{SVG}^{pre}\left(t_{i,j}\right)-Q_{SVG}^{pre}\left(t_{0,0}\right)\end{array}\right]---\left(9\right)$

V in formula (9)^preFor the vector that blower fan machine end, SVG machine end and PCC busbar voltage predicted value form, S is sensitivityMatrix;

The constraints of system voltage, generator operation and SVG operation:

$\left\{\begin{array}{c}{V}^{\min }\&le;V^{pre}\left(t_{i,j}\right)\&le;V^{\max }\\ Q_{WTG}^{\min }\&le;Q_{WTG}^{pre}\left(t_{i,j}\right)\&le;Q_{WTG}^{\max }\\ Q_{SVG}^{\min }\&le;Q_{SVG}^{pre}\left(t_{i,j}\right)\&le;Q_{SVG}^{\max }\\ {\mathrm{\&Delta;Q}}_{WTG}^{\min }\&le;Q_{WTG}^{pre}\left(t_{i,0}\right)-Q_{WTG}^{pre}\left(t_{i-1,0}\right)\&le;{\mathrm{\&Delta;Q}}_{WTG}^{\max }\\ {\mathrm{\&Delta;Q}}_{SVG}^{\min }\&le;Q_{SVG}^{pre}\left(t_{i,0}\right)-Q_{SVG}^{pre}\left(t_{i-1,0}\right)\&le;{\mathrm{\&Delta;Q}}_{SVG}^{\max }\end{array}\right.---\left(10\right)$

V in formula (10)^maxAnd V^minBe respectively the upper limit by PCC, blower fan and SVG voltage prediction value construction system voltage vectorAnd lower limit, wherein PCC voltage limit value is provided by power distribution network control centre, and blower fan and SVG voltage limit value are according to device fabrication factoryThe normal range of operation that business provides is determined;WithBe respectively the idle operation bound of blower fan,WithBe respectively SVGIdle operation bound, the normal range of operation all providing according to device fabrication manufacturer is determined;WithBe respectively windClimbing bound that machine is idle,WithBe respectively the idle climbing bound of SVG, all need to tie through reactive speed experiment testFruit is determined.

Preferably, described wind-powered electricity generation module can be directly power distribution network power supply.

Preferably, described dc bus can be by DC/AC module to the load supplying in power distribution network and distributed system.

Preferably, diesel-driven generator is directly the load supplying in distributed system.

Distributed power supply system tool of the present invention has the following advantages: the power situation of change of (1) distributed power source can be by standardTrue prediction; (2) system is in system when load supplying, can take into account that power distribution network scheduling requires, distributed power source ruuning situationWith the workload demand of load, meet user in system and simultaneously, taken into account power supply reliability, improved power supply benefit simultaneously.

Brief description of the drawings

Fig. 1 shows the block diagram of a kind of distributed power source of the present invention;

Fig. 2 shows the flow chart of distributed power supply system operation of the present invention.

Detailed description of the invention

Fig. 1 shows a kind of distributed power supply system 10 of the present invention, and this distributed power supply system is by generating electricity and loadingSystem 200 and monitoring system 100 form, and described generating and load system 200 comprise photovoltaic generating module, wind-powered electricity generation module, electric power storagePond module, diesel oil wind turbine, load, dc bus, multiple AC/DC or DC/AC module, this monitoring system 100 comprises: photovoltaic is sent outElectricity monitoring module 104, for monitoring the photovoltaic generating module 201 of real-time generating and load system 200, and to photovoltaic generation mouldThe generated output of piece 201 is predicted; Wind-powered electricity generation monitoring module 105, for the wind-powered electricity generation of monitoring generating in real time and load system 200Module 202, and the generated output of wind-powered electricity generation module 202 is predicted; Diesel-driven generator monitoring module 109, for real-time monitoringDiesel-driven generator 207 in generating and load system 200; Battery monitor module 106, for monitoring generating in real time and load systemThe battery module 203 of system 200; Load monitoring module 108, for the load of monitoring generating in real time and load system 200204, and the power situation of change of load 204 is predicted; Power distribution network contact module 102, for adjusting from power distribution network 30 in real timeRuuning situation and the relevant schedule information of power distribution network 30 known at control center; Parallel control module 103, for generating and load systemSystem 200 connects or isolation power distribution network 30; Middle control module 107, for determining the power supply strategy of generating and load system 200, and toAbove-mentioned each module is sent instruction, to carry out this power supply strategy; Bus communication module 101, for each mould of this monitoring system 100The liaison of piece.

Bus communication module 101, for the communication between above-mentioned modules, described bus communication module 101 is by superfluousRemaining dual CAN bus is connected with other modules.

Photovoltaic generating module 201 comprises multiple photovoltaic arrays. Photovoltaic generating module 201 at least comprises voltage, current detectingEquipment and sunlight intensity checkout equipment and temperature testing equipment, described wind-powered electricity generation module 202 can directly supply for power distribution network 30Electricity. Described dc bus can be by DC/AC module to the load supplying in power distribution network and distributed system. Diesel-driven generator207 is directly the load supplying in distributed system.

Wind-powered electricity generation module comprises multiple wind-driven generators and SVG equipment. Wind-powered electricity generation monitoring module 105 at least comprises wind-driven generatorLevel pressure, electric current, frequency detection equipment, wind speed checkout equipment, and SVG voltage and current checkout equipment. Wind-driven generator defeatedGoing out power is determined by wind speed, wind direction and the unique characteristics of wind-driven generator site.

Photovoltaic generation based on real-time meteorological data, blower fan generating prediction, need predict temperature, illuminance, wind speed.Wherein, temperature prediction can adopt following methods:

Temperature data T1[24 in sample], T2[24] ... Tm[24], Δ T1[23], Δ T2[23], Δ T3[23], Δ T4[23], Δ T5[23]; Run to the temperature T [t] that the t moment monitors, Δ T[t-1]; The similarity of accounting temperature variation tendency; RightSimilarity is normalized; Temperature after the t moment is predicted, obtained T[24-t]. In addition, to illuminance, can by withThe similar method of temperature prediction is predicted, then utilizes the Mathematical Modeling of photovoltaic module to predict its power output.

Battery monitor module 106 at least comprises accumulator voltage, current detecting equipment and temperature testing equipment. WithIn the charge and discharge process of real-time monitoring battery module. Have more than needed/lack by regulating the charge/discharge of battery to store/supplementEnergy, the energy flow of battery can be described as,

The energy flow of battery can be described as,

[Soc^ref-Soc(k+1)]＝a[Soc^ref-Soc(k)]+ηEs(k)

Soc_min≤Soc(k)≤Soc_max

Wherein, Soc (k) is the capacity status of k moment battery, Soc^refIt is the reliability for ensureing energy-storage battery workAnd the setting value arranging, Es (k) represents electric weight mobile between energy storage device and other power equipment, the physics loss of energy storageCoefficient a ∈ (0,1), the efficiency for charge-discharge that η is battery, charge efficiency is designated as η_c, discharging efficiency is designated as η_d, and full between themBe enough to lower relation:

$\mathrm{\&eta;}=\left\{\begin{array}{cc}{\mathrm{\&eta;}}_c,& ifEs\left(k\right)>0\\ {\mathrm{\&eta;}}_d,& eise.\end{array}\right.$

The charge and discharge process of energy storage can be regarded as a dynamic process that simultaneously comprises continuous variable and discrete variable, thisIn adopt mixed logic dynamic model processing method, by introduce binary variable δ (k) represent the work of energy storage at current timeMake state,

Z(k)＝δ(k)Es(k)

Z (k) represents the electric weight of current time energy storage charge/discharge, and the dynamic characteristic of battery can be described as:

[Soc^ref-Soc(k+1)]＝a[Soc^ref-Soc(k)]+(η_c-η_d)Z(k)+η_dEs(k)

Meet following constraints: E₁δ(k)+E₂Z(k)≤E₃Es(k)+E₄

Wherein, coefficient matrix E₁，E₂，E₃And E₄Be in the time that logical proposition is converted to linear inequality binary variable andContinuous variable will be satisfied linear inequality constraint, can obtain by the derivation of mathematical formulae.

In the time that logical proposition is converted to linear inequality in binary variable and continuous variable process, the linearity that meetInequality constraints E₁δ(k)+E₂Z(k)≤E₃Es(k)+E₄, wherein coefficient matrix E₁，E₂，E₃And E₄Be respectively:

$\left\{\begin{array}{c}{E}_1={\left[\begin{array}{cccccc}{\mathrm{Soc}}^{ref}& -({\mathrm{Soc}}^{ref}+\mathrm{\&epsiv;})& {\mathrm{Soc}}^{ref}& {\mathrm{Soc}}^{ref}& -{\mathrm{Soc}}^{ref}& -{\mathrm{Soc}}^{ref}\end{array}\right]}^T\\ E_2={\left[\begin{array}{cccccc}0& 0& 1& -1& 1& -1\end{array}\right]}^T\\ E_3={\left[\begin{array}{cccccc}1& -1& 1& -1& 0& 0\end{array}\right]}^T\\ E_4={\left[\begin{array}{cccccc}{\mathrm{Soc}}^{ref}& -\mathrm{\&epsiv;}& {\mathrm{Soc}}^{ref}& {\mathrm{Soc}}^{ref}& 0& 0\end{array}\right]}^T\end{array}\right..$

Middle control module 107 at least comprises CPU element, data storage cell and display unit.

Power distribution network contact module 102 at least comprises Wireless Telecom Equipment. This Wireless Telecom Equipment can be wireline equipment orWireless device.

Parallel control module 103 at least comprises the inspection for detection of power distribution network and distributed power source voltage, electric current and frequencyMeasurement equipment, data acquisition unit and data processing unit. Data acquisition unit comprises collection pretreatment and A/D modular converter, adoptsJi Ba road telemetered signal amount, comprises grid side A phase voltage, electric current, the three-phase voltage of distributed electrical source, electric current. Remote measurement amount canA little less than changing strong ac signal (5A/100V) into inside without distortion by the high-precision current in terminal and voltage transformerThe signal of telecommunication enters A/D chip and carries out analog-to-digital conversion after filtering is processed, and the data signal after conversion is through data processing unit meterCalculate, obtain three-phase voltage current value and the power distribution network 30 side phase voltage current values of distributed power source 200 sides. This telemetered signal amount placeReason has adopted high-speed and high-density synchronized sampling, automatic frequency tracking technology to also have improved fft algorithm, so precision obtains fullyEnsure, can complete meritorious, idle and measurement and the processing of electric energy from first-harmonic to higher harmonic components of distributed power source 200 sides.

Referring to accompanying drawing 2, method of the present invention comprises the steps:

S1. photovoltaic generation in photovoltaic generation monitoring module 104, wind-powered electricity generation monitoring module 105 Real-time Obtaining distributed power sources 200The service data of module 201 and wind-powered electricity generation module 202, and store data, load monitoring module 108 Real-time Obtaining loads 204 negativeLotus situation of change;

S2. according to the service data of photovoltaic generating module 201 and wind-powered electricity generation module 202 in existing distributed power source 200, to notThe power output of coming in the distributed power source 200 in predetermined instant is predicted, according to the load in existing distributed power source 200204 load variations situation, predicts the workload demand of load 204;

S3. the energy storage that battery module 203 is obtained in the real-time detection of battery monitor module 106 charges and discharge capacity, power distribution network connectionNetwork module 102 detects the schedule information of power distribution network in real time;

S4. with the schedule information of power distribution network 30, the capacity that charges and discharge of current battery module 203 energy storage, following distributed electricalPower output in source and to the variation of following workload demand as constraints, set up distributed power source 200 power supply managementsObject function;

S5. above-mentioned power supply management object function is optimized, determines power supply strategy;

S6. carry out above-mentioned power supply strategy.

In step S3, power distribution network 30 parameters of obtaining comprise: bus numbering, title, load are meritorious, reactive load, lineThe branch road on road number, headend node and endpoint node numbering, series resistance, series reactance.

Above-mentioned object function is optimized in accordance with the following steps: utilize the PREDICTIVE CONTROL based on MIXED INTEGER quadratic programming belt restrainingOptimized algorithm, with the controlled power output of distributed power source, batteries to store energy charge and discharge electric weight and discharge and recharge the time, diesel-driven generator canControl power output and be performance variable to load demand, taking the optimum capacity of energy storage device as setting value, with distributed electricalThe energy supply and demand error in source is minimum as target, is meeting the controlled power output of distributed power source, the controlled output of diesel-driven generatorPower, batteries to store energy discharge and recharge with capacity physical constraint condition under, adjust distributed power source controlled power output, diesel oil is sent outThe controlled power output of motor, energy storage charge and discharge electric weight and discharge and recharge the time and regulating action to load demand, to reachThe target function value of distributed power source minimum.

In step S5, the object function of above-mentioned power supply management is optimized in accordance with the following steps: utilize based on MIXED INTEGER twoThe PREDICTIVE CONTROL optimized algorithm of inferior planning belt restraining, (comprises photovoltaic generation output work with the controlled power output of distributed power source 200Rate and wind power output power), the controlled power output of diesel-driven generator 207, battery module 203 energy storage charge and discharge electric weight and discharge and rechargeTime, be performance variable to load 204 workload demands, taking the optimum capacity of energy storage device (battery module 203) as setting value, withThe energy supply and demand error of distributed power source 200 is minimum as target, is meeting the controlled power output of distributed power source 200, electric power storage203 energy storage of pond module discharge and recharge with capacity physical constraint condition under, adjust distributed power source 200 controlled power output, diesel oilThe controlled power output of generator 207, energy storage charge and discharge electric weight and discharge and recharge the time and regulating action to load demand,To reach the target function value of distributed power source minimum.

In step S2, the power output in the distributed power source 200 in following predetermined instant is predicted, specifically adoptRealize by following mode, according to the power stage of intensity of illumination data and temperature data prediction photovoltaic generating module, according to wind speedThe power output of data prediction wind-powered electricity generation module.

In step S2, the regenerative resource power output in the distributed power source in following predetermined instant is carried out in advanceSurvey, comprise and predict the power stage of photovoltaic generating module and the power output of wind-powered electricity generation module.

Predict in the following way the power output of wind-powered electricity generation module at step S2, described wind-powered electricity generation module comprises wind-driven generatorAnd SVG:

(201) in collection wind-powered electricity generation module, current all kinds of electric weight measured values are as the initial value of the predicted value of all kinds of electric weight, pre-Measured value comprises: the blower fan predicted value of gaining meritPredicted value that blower fan is idleBlower fan set end voltage predicted valueSVG is idlePredicted valueSVG set end voltage predicted valueWind-powered electricity generation module site (PCC) busbar voltage predicted value

(202) set up according to described predicted value the MPC optimizing control models being formed by optimization aim function and constraints,And solve the predicted value of the meritorious and idle output of wind-powered electricity generation module:

The object function of MPC optimizing control models is suc as formula shown in (1):

$\underset{Q_{WTG}^{set},V_{SVG}^{set}}{\min }(\underset{i=1}{\overset{N-1}{\&Sigma;}}\underset{j=0}{\overset{M-1}{\&Sigma;}}{\mathrm{\&rho;}}^{t_{i,j}}{F}_1,\underset{i=1}{\overset{N-1}{\&Sigma;}}\underset{j=0}{\overset{M-1}{\&Sigma;}}{\mathrm{\&rho;}}^{t_{i,j}}{F}_2)---\left(1\right)$

In formula (1)WithFor optimized variable,WithImplication is respectively the idle setting value of blower fan and SVG voltageSetting value; N is the number in time window Coverage Control cycle; M is the number containing future position under single control cycle; ρ is decay systemNumber, value ρ < 1; Time variable ti, j=(Mi+j) Δ t meaning is that current time plays j prediction in i control cyclePoint, Δ t is future position interval, Δ t is by wind-powered electricity generation modular power predicted time interval determination;

F1 is the variance level of wind-powered electricity generation module site busbar voltage and setting value, and F1 expression is suc as formula (2):

$F_1\left(t_{i,j}\right)={\&lsqb;V_{PCC}^{pre}\left(t_{i,j}\right)-V_{PCC}^{ref}\&rsqb;}^2---\left(2\right)$

In formula (2)Represent the reference value of PCC voltage, after extracting from main website control instruction, set;

F2 is the idle level of reserve of SVG, and F2 expression is suc as formula (3):

$F_2\left(t_{i,j}\right)={\&lsqb;Q_{SVG}^{pre}\left(t_{i,j}\right)-Q_{SVG}^{opr}\&rsqb;}^2---\left(3\right)$

In formula (3)For the idle best operating point of SVG;

The constraints of MPC optimizing control models, specifically comprises:

The blower fan prediction-constraint condition of gaining merit:

$P_{WTG}^{pre}\left(t_{i,j}\right)=\underset{k=1}{\overset{N_a}{\&Sigma;}}{\mathrm{\&phi;}}_k{P}_{WTG}^{pre}\left(t_{i,j-k}\right)+{\mathrm{\&epsiv;}}_{WTG}^{pre}\left(t_{i,j}\right)-\underset{k=1}{\overset{N_m}{\&Sigma;}}{\mathrm{\&theta;}}_k{\mathrm{\&epsiv;}}_{WTG}^{pre}\left(t_{i,j-k}\right)---\left(4\right)$

In formula (4)For the meritorious predicated error of blower fan; Na and Nm are respectively the exponent number of AR and MA model, and φ k and θ k are phaseClose weight, exponent number and weight are all determined according to the meritorious history value of blower fan; Ti, j-k (comprises for participating in calculated data in prediction) the corresponding moment, subscript k pushes away the k Δ t time before characterizing the prediction moment, works as ti, j-k≤0 o'clock, and meritorious predicted value should be gotCorresponding moment history value;

Prediction-constraint condition that blower fan is idle:

Blower fan is idle reaches setting value before controlling next time:

$Q_{WTG}^{pre}\left(t_{i,0}\right)=Q_{WTG}^{set}\left(t_{i-1,0}\right)---\left(5\right)$

Each future position in i control cycle, the change procedure of blower fan reactive power is with exponential function matching:

$Q_{WTG}^{pre}\left(t_{i,j}\right)=\frac{1-e^{-(t_{i,j}-t_{i,0})/T_s}}{1-e^{-M\mathrm{\&Delta;}t/T_s}}{Q}_{WTG}^{set}\left(t_{i,0}\right)+\frac{e^{-(t_{i,j}-t_{i,0})/T_s}-e^{-M\mathrm{\&Delta;}t/T_s}}{1-e^{-M\mathrm{\&Delta;}t/T_s}}{Q}_{WTG}^{pre}\left(t_{i,0}\right)---\left(6\right)$

In formula (6), Ts is the idle adjusting time constant of blower fan, can obtain according to the idle adjusting testing experiment of blower fan.

Prediction-constraint condition that SVG is idle:

Reference value that SVG is idleShown in (7):

$\begin{array}{c}{Q}_{SVG}^{ref}\left(t_{i,j}\right)=K_P\&lsqb;V_{SVG}^{pre}\left(t_{i,j}\right)-V_{SVG}^{set}\left(t_{i,0}\right)\&rsqb;+K_I\mathrm{\&Delta;}t\underset{k=0}{\overset{i\&times;M+j}{\&Sigma;}}\&lsqb;V_{SVG}^{pre}\left(t_{i,j-k}\right)-V_{SVG}^{set}\left(t_{i,-k}\right)\&rsqb;\\ +Q_{SVG}^{pre}\left(t_{0,0}\right)-K_P\&lsqb;V_{SVG}^{pre}\left(t_{0,0}\right)-V_{SVG}^{set}\left(t_{0,0}\right)\&rsqb;\end{array}---\left(7\right)$

In formula (7), KI and KP are respectively the coefficient of proportional component and integral element;

Predicted value that SVG is idle is suc as formula shown in (8):

$Q_{SVG}^{pre}\left(t_{i,j}\right)=Q_{SVG}^{ref}\left(t_{i,j-1}\right)+\&lsqb;Q_{SVG}^{pre}\left(t_{i,j-1}\right)-Q_{SVG}^{ref}\left(t_{i,j-1}\right)\&rsqb;e^{-(t_{i,j}-t_{i,j-1})/T_d}---\left(8\right)$

In formula (8), time constant Td is SVG power electronic equipment action delay;

Voltage prediction constraints:

$V^{pre}\left(t_{i,j}\right)-V^{pre}\left(t_{0,0}\right)=S\left[\begin{array}{c}{P}_{WTG}^{pre}\left(t_{i,j}\right)-P_{WTG}^{pre}\left(t_{0,0}\right)\\ Q_{WTG}^{pre}\left(t_{i,j}\right)-Q_{WTG}^{pre}\left(t_{0,0}\right)\\ Q_{SVG}^{pre}\left(t_{i,j}\right)-Q_{SVG}^{pre}\left(t_{0,0}\right)\end{array}\right]---\left(9\right)$

V in formula (9)^preFor the vector that blower fan machine end, SVG machine end and PCC busbar voltage predicted value form, S is sensitivityMatrix;

The constraints of system voltage, generator operation and SVG operation:

$\left\{\begin{array}{c}{V}^{\min }\&le;V^{pre}\left(t_{i,j}\right)\&le;V^{\max }\\ Q_{WTG}^{\min }\&le;Q_{WTG}^{pre}\left(t_{i,j}\right)\&le;Q_{WTG}^{\max }\\ Q_{SVG}^{\min }\&le;Q_{SVG}^{pre}\left(t_{i,j}\right)\&le;Q_{SVG}^{\max }\\ {\mathrm{\&Delta;Q}}_{WTG}^{\min }\&le;Q_{WTG}^{pre}\left(t_{i,0}\right)-Q_{WTG}^{pre}\left(t_{i-1,0}\right)\&le;{\mathrm{\&Delta;Q}}_{WTG}^{\max }\\ {\mathrm{\&Delta;Q}}_{SVG}^{\min }\&le;Q_{SVG}^{pre}\left(t_{i,0}\right)-Q_{SVG}^{pre}\left(t_{i-1,0}\right)\&le;{\mathrm{\&Delta;Q}}_{SVG}^{\max }\end{array}\right.---\left(10\right)$

V in formula (10)^maxAnd V^minBe respectively the upper limit by PCC, blower fan and SVG voltage prediction value construction system voltage vectorAnd lower limit, wherein PCC voltage limit value is provided by power distribution network control centre, and blower fan and SVG voltage limit value are according to device fabrication factoryThe normal range of operation that business provides is determined;WithBe respectively the idle operation bound of blower fan,WithBe respectively SVGIdle operation bound, the normal range of operation all providing according to device fabrication manufacturer is determined;WithBe respectively blower fanIdle climbing bound,WithBe respectively the idle climbing bound of SVG, all need through reactive speed experimental resultsDetermine.

In S2, adopt Neural Network model predictive workload demand, concrete steps are as follows:

S211. gather 12 groups of active power and reactive power every day, continuous acquisition 8 days, has 96 groups of data P so altogetherAnd Q (k) (k), k=1,2 ..., 96.

S212. 96 groups of data P (k) and Q (k) are normalized, make N=1,2 ..., 96; First using 12 active-power Ps (k) of every day as one group of input vector R(m), 12 reactive power Qs (k) are as one group of input vector S (m), m=1, and 2 ..., 8, m represents the frequency of training of neutral net;Suppose in advance simultaneously the 9th day 12 active-power Ps ' (k) as the output vector R ' of predicted power, the 9th day 12 are idlePower Q ' is (k) as the output vector S ' of predicted power; The active power input vector of front like this 8 days is just R (1), R (2), R(3), R (4), R (5), R (6), R (7), R (8), the output vector of the 9th day prediction active power is R '; The reactive power of first 8 daysInput vector is just S (1), S (2), and S (3), S (4), S (5), S (6), S (7), S (8), the output of the 9th day prediction active power is vowedAmount is S '.

S213. the input layer using 8 groups of input vector R (m) and S (m) as neutral net, the transmission letter of hidden layer neuronNumber adopts S type tan tansig, and the neuronic transfer function of output layer adopts S type logarithmic function logsig, as Fig. 2 instituteShow, after 8 neural metwork trainings, just determined the weights of each connection weight in neutral net like this.

S214. for 8 active power input vector R (m), there is a in hidden layer neuron¹＝tansig(IW¹R+b¹)，Wherein a¹For hidden layer neuron output, IW¹For the weights of hidden layer neuron, b¹For the threshold value of hidden layer neuron; In outputLayer neuron has a²＝logsig(IW²a¹+b²), wherein a²For the output of output layer neuron, IW²For the neuronic power of output layerValue, b²For the neuronic threshold value of output layer.

S215. for 8 active power input vector S (m), there is c in hidden layer neuron¹＝tansig(IW¹S+b¹)，Wherein c¹For hidden layer neuron output, IW¹For the weights of hidden layer neuron, b¹For the threshold value of hidden layer neuron; In outputLayer neuron has c²＝logsig(IW²c1+b²), wherein c²For the output of output layer neuron, IW²For the neuronic power of output layerValue, b²For the neuronic threshold value of output layer.

S216. using the input vector R (8) of the 8th day and S (8) again as the input layer of neutral net, now neutral netThe output vector R ' of the predicted power of middle output and S ' are the power prediction normalized value of the 9th day, then calculate with renormalizationMethod,K=1,2 ..., 96, the vector value R (9) of output and R (9) are just12 active-power P ' (k) He 12 reactive power Qs ' (k) of the 9th day predicted power. So by that analogy, can repeatStep is above utilized data prediction to the power of ten days of second day to the 9th day, and the power of every day all can so belowWith out predicted.

In step S4, distributed power source general power Pg is constrained to:

Non-response scheduling period 1 time, P_g,min≤P_g(l)≤P_g,max，P_g,minFor distributed power source 200 can be from power distribution network30 peak powers that absorb, P_g,maxFor distributed power source 200 can be to the peak power of power distribution network 30 transmission powers;

Response scheduling period 2 times, P_g(2)＝P_set，P_setFor the interconnection power of response scheduling period 2 times requirements.

Above content is in conjunction with concrete preferred embodiment further description made for the present invention, can not assertSpecific embodiment of the invention is confined to these explanations. For general technical staff of the technical field of the invention,Do not depart under the prerequisite of the present invention design, make some being equal to substitute or obvious modification, and performance or purposes identical, all shouldBe considered as belonging to protection scope of the present invention.

Claims (4)

1. a distributed power supply system, this distributed power supply system is by generating electricity and load system and monitoring system form, described inGenerating and load system comprise photovoltaic generating module, wind-powered electricity generation module, battery module, diesel generator, load, direct current motherLine, multiple AC/DC or DC/AC module, is characterized in that, this monitoring system comprises:

Photovoltaic generation monitoring module, for monitoring in real time photovoltaic generating module, and carries out the generated output of photovoltaic generating modulePrediction;

Wind-powered electricity generation monitoring module, for monitoring in real time wind-powered electricity generation module, and predicts the generated output of wind-powered electricity generation module;

Battery monitor module, for monitoring in real time battery module;

Diesel-driven generator monitoring module, for monitoring in real time diesel-driven generator;

Load monitoring module, for the described load of monitoring in real time, and predicts the power situation of change of load;

Power distribution network contact module, for knowing in real time ruuning situation and the relevant scheduling letter of power distribution network from power distribution network regulation and control centerBreath;

The monitoring module that is incorporated into the power networks, for controlling, distributed power source connects or isolation power distribution network;

Middle control module, for determining the operation strategy of distributed power supply system, and each module in above-mentioned monitoring system is sentInstruction, to carry out this operation strategy;

Bus module, for the liaison of the modules of this monitoring system;

Described photovoltaic generation monitoring module, wind-powered electricity generation monitoring module, load monitoring module include detecting unit and memory cell, canWith the service data of photovoltaic generating module, wind-powered electricity generation module in Real-time Obtaining distributed power source and the load variations situation of load, andStorage data;

Described photovoltaic photoelectric monitoring module, wind-powered electricity generation monitoring module, load monitoring module include arithmetic element, basis respectivelyThe service data of existing photovoltaic generating module and wind-powered electricity generation module, to the renewable energy in the distributed power source in following predetermined instantSource power output predicts, according to existing loaded load variations situation, the workload demand of load predicted;

Described battery monitor module, has secondary battery unit and discharges and recharges detection module, can detect in real time and obtain battery mouldThe energy storage of piece charges and discharge capacity;

Described middle control module, has ALU, can filling according to the schedule information of power distribution network, current batteries to store energyPut the power output of renewable sources of energy power output in capacity, following distributed power source, diesel-driven generator and to loading futureThe variation of demand, as constraints, is set up distributed power source and is moved tactful object function;

Described power distribution network contact module can Real-time Obtaining power distribution network parameter comprise: bus numbering, title, load are meritorious, negativeLotus is idle, the branch road of circuit number, headend node and endpoint node numbering, series resistance, series reactance;

Predict in the following way the power output of wind-powered electricity generation module at described wind-powered electricity generation monitoring module, described wind-powered electricity generation module comprises wind-forceGenerator and SVG:

(201) gather in wind-powered electricity generation module current all kinds of electric weight measured values as the initial value of the predicted value of all kinds of electric weight, predicted valueComprise: the blower fan predicted value of gaining meritPredicted value that blower fan is idleBlower fan set end voltage predicted valuePrediction that SVG is idleValueSVG set end voltage predicted valueWind-powered electricity generation module site (PCC) busbar voltage predicted value

(202) set up according to described predicted value the MPC optimizing control models being formed by optimization aim function and constraints, and askSeparate the predicted value of the meritorious and idle output of wind-powered electricity generation module:

The object function of MPC optimizing control models is suc as formula shown in (1):

$\underset{Q_{WTG}^{set},V_{SVG}^{set}}{\min }(\underset{i=0}{\overset{N-1}{\&Sigma;}}\underset{j=0}{\overset{M-1}{\&Sigma;}}{\mathrm{\&rho;}}^{t_{i,j}}{F}_1,\underset{i=0}{\overset{N-1}{\&Sigma;}}\underset{j=0}{\overset{M-1}{\&Sigma;}}{\mathrm{\&rho;}}^{t_{i,j}}{F}_2)---\left(1\right)$

In formula (1)WithFor optimized variable,WithImplication is respectively the idle setting value of blower fan and SVG voltage is setValue; N is the number in time window Coverage Control cycle; M is the number containing future position under single control cycle; ρ is attenuation coefficient, getsValue ρ < 1; Time variable ti, j=(Mi+j) Δ t meaning is that current time plays j future position in i control cycle, ΔT is future position interval, and Δ t is by wind-powered electricity generation modular power predicted time interval determination;

F1 is the variance level of wind-powered electricity generation module site busbar voltage and setting value, and F1 expression is suc as formula (2):

$F_1\left(t_{i,j}\right)={\&lsqb;V_{PCC}^{pre}\left(t_{i,j}\right)-V_{PCC}^{ref}\&rsqb;}^2---\left(2\right)$

In formula (2)Represent the reference value of PCC voltage, after extracting from main website control instruction, set;

F2 is the idle level of reserve of SVG, and F2 expression is suc as formula (3):

$F_2\left(t_{i,j}\right)={\&lsqb;Q_{SVG}^{pre}\left(t_{i,j}\right)-Q_{SVG}^{opr}\&rsqb;}^2---\left(3\right)$

In formula (3)For the idle best operating point of SVG;

The constraints of MPC optimizing control models, specifically comprises:

The blower fan prediction-constraint condition of gaining merit:

$P_{WTG}^{pre}\left(t_{i,j}\right)=\underset{k=1}{\overset{N_a}{\mathrm{\&Sigma;}}}{\mathrm{\&phi;}}_k{P}_{WTG}^{pre}\left(t_{i,j-k}\right)+{\mathrm{\&epsiv;}}_{WTG}^{pre}\left(t_{i,j}\right)-\underset{k=1}{\overset{N_m}{\mathrm{\&Sigma;}}}{\mathrm{\&theta;}}_k{\mathrm{\&epsiv;}}_{WTG}^{pre}\left(t_{i,j-k}\right)---\left(4\right)$

In formula (4)For the meritorious predicated error of blower fan; Na and Nm are respectively the exponent number of AR and MA model, and φ k and θ k are relevant powerHeavy, exponent number and weight are all determined according to the meritorious history value of blower fan; Ti, j-k (comprises for participating in calculated data in prediction) the corresponding moment, subscript k pushes away the k Δ t time before characterizing the prediction moment, works as ti, j-k≤0 o'clock, and meritorious predicted value should be gotCorresponding moment history value;

Prediction-constraint condition that blower fan is idle:

Blower fan is idle reaches setting value before controlling next time:

$Q_{WTG}^{pre}\left(t_{i,0}\right)=Q_{WTG}^{set}\left(t_{i-1,0}\right)---\left(5\right)$

Each future position in i control cycle, the change procedure of blower fan reactive power is with exponential function matching:

$Q_{WTG}^{pre}\left(t_{i,j}\right)=\frac{1-e^{-(t_{i,j}-t_{i,0})/T_s},}{1-e^{-M\mathrm{\&Delta;}t/T_s},}{Q}_{WTG}^{set}\left(t_{i,0}\right)+\frac{e^{-(t_{i,j}-t_{i,0})/T_s}-e^{-M\mathrm{\&Delta;}t/T_s}}{1-e^{-M\mathrm{\&Delta;}t/T_s}}{Q}_{WTG}^{pre}\left(t_{i,0}\right)---\left(6\right)$

In formula (6), Ts is the idle adjusting time constant of blower fan, can obtain according to the idle adjusting testing experiment of blower fan;

Prediction-constraint condition that SVG is idle:

Reference value that SVG is idleShown in (7):

$\begin{array}{c}{Q}_{SVG}^{ref}\left(t_{i,j}\right)=K_P\&lsqb;V_{SVG}^{pre}\left(t_{i,j}\right)-V_{SVG}^{set}\left(t_{i,0}\right)\&rsqb;+K_I\mathrm{\&Delta;}t\underset{k=0}{\overset{i\&times;M+j}{\mathrm{\&Sigma;}}}\&lsqb;V_{SVG}^{pre}\left(t_{i,j-k}\right)-V_{SVG}^{set}\left(t_{i,-k}\right)\&rsqb;\\ +Q_{SVG}^{pre}\left(t_{0,0}\right)-K_P\&lsqb;V_{SVG}^{pre}\left(t_{0,0}\right)-V_{SVG}^{set}\left(t_{0,0}\right)\&rsqb;\end{array}---\left(7\right)$

In formula (7), KI and KP are respectively the coefficient of proportional component and integral element;

Predicted value that SVG is idle is suc as formula shown in (8):

$Q_{SVG}^{pre}\left(t_{i,j}\right)=Q_{SVG}^{ref}\left(t_{i,j-1}\right)+\&lsqb;Q_{SVG}^{pre}\left(t_{i,j-1}\right)-Q_{SVG}^{ref}\left(t_{i,j-1}\right)\&rsqb;e^{-(t_{i,j}-t_{i,j-1})/T_d}---\left(8\right)$

In formula (8), time constant Td is SVG power electronic equipment action delay;

Voltage prediction constraints:

$V^{pre}\left(t_{i,j}\right)-V^{pre}\left(t_{0,0}\right)=S\left[\begin{array}{c}{P}_{WTG}^{pre}\left(t_{i,j}\right)-P_{WTG}^{pre}\left(t_{0,0}\right)\\ Q_{WTG}^{pre}\left(t_{i,j}\right)-Q_{WTG}^{pre}\left(t_{0,0}\right)\\ Q_{SVG}^{pre}\left(t_{i,j}\right)-Q_{SVG}^{pre}\left(t_{0,0}\right)\end{array}\right]---\left(9\right)$

V in formula (9)^preFor the vector that blower fan machine end, SVG machine end and PCC busbar voltage predicted value form, S is sensitivity matrix;

The constraints of system voltage, generator operation and SVG operation:

$\left\{\begin{array}{c}{V}^{\min }\&le;V^{pre}\left(t_{i,j}\right)\&le;V^{\max }\\ Q_{WTG}^{\min }\&le;Q_{WTG}^{pre}\left(t_{i,j}\right)\&le;Q_{WTG}^{\max }\\ Q_{SVG}^{\min }\&le;Q_{SVG}^{pre}\left(t_{i,j}\right)\&le;Q_{SVG}^{\max }\\ {\mathrm{\&Delta;Q}}_{WTG}^{\min }\&le;Q_{WTG}^{pre}\left(t_{i,0}\right)-Q_{WTG}^{pre}\left(t_{i-1,0}\right)\&le;{\mathrm{\&Delta;Q}}_{WTG}^{\max }\\ {\mathrm{\&Delta;Q}}_{SVG}^{\min }\&le;Q_{SVG}^{pre}\left(t_{i,0}\right)-Q_{SVG}^{pre}\left(t_{i-1,0}\right)\&le;{\mathrm{\&Delta;Q}}_{SVG}^{\max }\end{array}\right.---\left(10\right)$

V in formula (10)^maxAnd V^minBe respectively by the upper limit of PCC, blower fan and SVG voltage prediction value construction system voltage vector and underLimit, wherein PCC voltage limit value is provided by power distribution network control centre, and blower fan and SVG voltage limit value are given according to device fabrication manufacturerThe normal range of operation going out is determined;WithBe respectively the idle operation bound of blower fan,WithBe respectively SVG withoutMerit operation bound, the normal range of operation all providing according to device fabrication manufacturer is determined;WithBe respectively blower fanIdle climbing bound,WithBe respectively the idle climbing bound of SVG, all need through reactive speed experimental resultsDetermine.

2. the system as claimed in claim 1, is characterized in that, described wind-powered electricity generation module can be directly power distribution network power supply.

3. the system as claimed in claim 1, is characterized in that, described dc bus can be by DC/AC module to power distribution networkWith the load supplying in distributed system.

4. the system as claimed in claim 1, is characterized in that, diesel-driven generator is directly that the load in distributed system suppliesElectricity.