CN105160451A - Electric-automobile-contained micro electric network multi-target optimization scheduling method - Google Patents

Abstract

The invention relates to an electric-automobile-contained micro electric network multi-target optimization scheduling method. The method is characterized by comprising steps that, 1), a mode of access of an electric automobile to a micro electrical network is determined, discharging and charging load distribution characteristic superposition of a single electric automobile under different access modes is carried out to obtain discharging and charging load distribution characteristics of the electric automobile; (2), the electric automobile is taken as a micro electric network scheduling object which is added for electric network optimization scheduling, and an micro electric network scheduling model in consideration of large-scale electric automobile access is established according to the discharging and charging load distribution characteristics of the electric automobile; and 3), a particle swarm optimization algorithm based on the automatic recombination mechanism is employed to solve the micro electric network scheduling model in consideration of large-scale electric automobile access, economical efficiency of micro electric network scheduling under various scheduling strategies is compared and analyzed, and thereby the optimum scheduling strategy is obtained. Compared with the prior art, the method further has advantages of comprehensive consideration and effective and feasible performance.

Description

A kind of micro-capacitance sensor Multiobjective Optimal Operation method containing electric automobile

Technical field

The present invention relates to micro-capacitance sensor scheduling field, especially relate to a kind of micro-capacitance sensor Multiobjective Optimal Operation method containing electric automobile.

Background technology

Micro-capacitance sensor, as the management of a kind of new distribution type electric power network and provisioning technique, can provide convenient for renewable energy system accesses power distribution network, and realize the management of Demand-side Energy Efficient and major network electric power energy efficiency utilization.

The Optimal Scheduling of micro-capacitance sensor is similar to the Optimal Scheduling of traditional bulk power grid, but has its singularity and complicacy.Optimal Scheduling, main target is that the cost in order to realize operation of power networks process is minimum.Problem along with environmental pollution causes the concern of people day by day, and environmental pollution cost is put in the problem of optimizing scheduling by a lot of researchist now.And in the Optimized Operation of micro-capacitance sensor, reduce operating cost and the discharge of decreasing pollution thing, have the energy-saving and emission-reduction of whole electrical network and also have great meaning.

In recent years, along with government's energy-conserving and environment-protective and high and new technology relevant policies reinforcement and implement, use the number of users of electric automobile (ElectricVehicle, EV) constantly to increase, the power storage amount of these electric automobiles is considerable simultaneously.But electric automobile access electrical network is flexibly and disperses, and not by the restriction of room and time, this feature will increase the instability of electrical network, and affects the quality of power supply of electrical network.Be similar to distributed power source, if electric automobile is accessed micro-capacitance sensor, can avoid or reduce electric automobile and directly access impact on electrical network.

At present, access problem on a large scale for micro-capacitance sensor Multiobjective Optimal Operation and electric automobile, scholar both domestic and external has carried out a series of research work, and achieves the achievement of some theory and practice aspects.Chen Dawei and Zhu Guiping establishes the Optimal Operation Model of the micro-capacitance sensor taking into account environmental factor, but is just multiplied by fixing weights summation respectively to two objective function the lowest coursing costs and pollution plot network minimal, in fact remains single object optimization scheduling.Yang Qi etc. mainly have studied four kinds and relate to grid-connected and hardware configuration that is islet operation micro-capacitance sensor economic load dispatching system, and analyze the effect of energy-storage units.S.W.Hadley etc. have studied EVs and return the statistics rule that moment and day travel distance for the last time, establish the statistical model of EVs charge requirement, and analyze EVs and charge at random on the impact of network load.Han Haiying etc. consider electric automobile by period charge and discharge process, and establish the micro-capacitance sensor Optimal Operation Model containing the EVs that can network on a large scale, obtain a bill Unit Combination and exert oneself.But these processing modes are relatively simple, and a lot of aspect requires further study discussion.

Summary of the invention

Object of the present invention is exactly provide a kind of to overcome defect that above-mentioned prior art exists to consider comprehensive, the effective and feasible micro-capacitance sensor Multiobjective Optimal Operation method containing electric automobile.

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

Containing a micro-capacitance sensor Multiobjective Optimal Operation method for electric automobile, it is characterized in that, the method comprises the following steps:

1) determine electric automobile access micro-capacitance sensor pattern, by the separate unit electric automobile under different access module put fill load distribution performance superposition obtain putting of electric automobile fill load distribution performance;

2) electric automobile is joined in micro-capacitance sensor Optimized Operation as micro-capacitance sensor scheduler object, and fill according to putting of electric automobile the micro-capacitance sensor Optimal Operation Model that load distribution performance sets up the extensive electric automobile access of consideration;

3) employing considers based on the particle swarm optimization algorithm of automatic recombination mechanism the micro-capacitance sensor Optimal Operation Model that extensive electric automobile accesses, and the micro-capacitance sensor scheduling economy under the multiple different scheduling strategy of comparative analysis, thus obtain optimal scheduling strategy.

Described step 1) in pattern comprise the V0G pattern of unidirectional unordered charging and the V2G pattern of two-way orderly discharge and recharge.

Described step 2) in consider that the optimization object function of the micro-capacitance sensor Optimal Operation Model of extensive electric automobile access is:

Micro-capacitance sensor management and running cost Obj ₁minimum:

The disposal cost Obj discharged pollutants ₂minimum:

Micro-capacitance sensor scheduling integrated cost is minimum:

minObj ₃＝m ₁Obj ₁+m ₂Obj ₂

Wherein, C _gfor the fuel cost of distributed power source, C _oMfor operation expense, C _dPfor the depreciable cost of generator unit, C _gridfor micro-capacitance sensor and bulk power grid electric energy switching cost, C _eVfor micro-capacitance sensor and electric car electric energy switching cost, C _lfor load synthesis cost of compensation, Δ T _tfor, T is, j is distributed generation unit numbering in micro-capacitance sensor, and t is for running the period, and k is discharged pollutant type, C _kfor processing every kilogram of expense discharged pollutants, γ _jk(P _jt) for being generator unit j output P in micro-capacitance sensor _jtthe weight of the pollutant k produced during electric energy, γ _gridk(P _gridt) be power distribution network output P _gridtthe weight of the pollutant k produced during electric energy, m ₁, m ₂for the weight of operating cost and discharge costs;

Constraint condition is:

Power-balance retrains:

Cold/heating power balance constraint:

W _Load＝W _MT+W _FC

Distributed generation unit active-power P _jtbound retrains:

The exchange power P of micro-capacitance sensor and major network _gridlimit value retrains:

The power P of energy-storage units _sBtconstraint and charged SOC _sBtconstraint:

P _SBmin≤P _SBt≤P _SBmax

SOC _SBmin≤SOC _SBt≤SOC _SBmax

SOC _end＝SOC ₀

The power P of electric automobile _evtconstraint and charged SOC _eVtconstraint:

P _EVmin≤P _EVt≤P _EVmax

SOC _EVmin≤SOC _EVt≤SOC _EVmax

Wherein, P _jtfor the power that generator unit j in t period micro-capacitance sensor sends, P _gridtfor the power that t period power distribution network transmits to micro-capacitance sensor, P _batterytfor the power that t period accumulator sends, P _loadtfor the workload demand in t period micro-capacitance sensor, W _loadfor whole micro-grid system cold/thermal load demands, W _mTfor miniature gas turbine waste heat flue gas provide cold/thermal power, W _fCfor fuel cell power generation produce heat provide cold/thermal power, for the minimum output power of distributed generation unit j, for the peak power output of distributed generation unit j, for the minimum power that common point circuit between micro-capacitance sensor and power distribution network can transmit, for the peak power that common point circuit between micro-capacitance sensor and power distribution network can transmit, P _sBmin, P _sBmaxbe respectively minimum power and the peak power of accumulator cell charging and discharging, SOC _sBmin, SOC _sBmaxminimum value and the maximal value of storage battery charge state respectively, SOC ₀and SOC _endbe respectively the state-of-charge of initial time 0 and end time 24 accumulator in the dispatching cycle, P _eVmin, P _eVmaxbe respectively minimum power and the peak power of electric automobile discharge and recharge, SOC _min, SOC _maxminimum value and the maximal value of batteries of electric automobile state-of-charge respectively.

Under the V0G pattern of unidirectional unordered charging, the electrical demand filling electric automobile in load distribution performance of putting of electric automobile is:

E _EV＝η _EV·d

Wherein, η _eVfor the electrical demand coefficient of electric automobile mileage, the distance travelled d obeys logarithm normal distribution of every platform electric automobile, its probability density function is:

Duration of charging f (x) meets normal distribution:

Wherein, μ _dand μ _sfor expectation value, σ _dand σ _sfor standard deviation.。

Under the V2G pattern of two-way orderly discharge and recharge, electric automobile put the discharge period T filled in load distribution performance _disc1for:

Charge power in conjunction with electric automobile can obtain duration of charge, thus obtains lasting duration of charging T _disc2for:

Single electric automobile required charging load was consumed gross energy, i.e. P among one day _eVfor:

Wherein, T _all-discfor putting to the total duration that discharges needed for state-of-charge lower limit when electric automobile is full of electricity, P _discfor electric automobile discharge power, SOC _maxand SOC _minbe respectively the bound of storage battery charge state, D is electric automobile daily travel, W ₁₀₀for electric automobile hundred kilometers of power consumption, T _{end_disc}for electric discharge finish time, T _{start_disc}for networking discharging time, P _cfor charging electric vehicle power.

Described step 2) in multiple different scheduling strategies comprise micro-grid connection operation reserve and micro-capacitance sensor islet operation strategy.

Described step 3) specifically comprise the following steps:

31) according to the access module of electric automobile, the model parameter of each distributed generation unit, each objective function parameters and each constraint condition parameter in setting micro-capacitance sensor, and introduce uncontrollable measurable distributed power source and exert oneself and cold electric load parameter;

32) controlled unit miniature gas turbine of exerting oneself, fuel cell, diesel-driven generator, accumulator and major network interchange power are tieed up particles as five, and set particle cluster algorithm parameter, comprise population, solution space dimension, maximum iteration time, particle maximal rate and restructuring index r;

33) calculate the fitness of each particle, and record the target function value of the current individual extreme value of each particle and correspondence, and then obtain all extreme values and corresponding target function value, and select individual optimal value and global optimum;

34) the current number of times of iteration adds one, upgrades population and carries out position and speed;

35) judged result whether meet Premature Convergence standard and restructuring number of times whether reach the value preset, if result meets Premature Convergence standard and restructuring number of times does not reach the value preset, then recombinate population and restructuring index r=r+1, and return step 33), otherwise carry out step 36);

36) judge whether to meet the condition of convergence, if so, then obtain global optimum or reach maximum iteration time, finishing iteration process, if not, then returning step 33), continue iterative operation.

Compared with prior art, the present invention has the following advantages:

Consider the discharge and recharge electrical characteristics of electric automobile and the use habit of car owner, formulate electric automobile and adopt unidirectional unordered V0G and two-way orderly V2G pattern access micro-capacitance sensor respectively, thus establish the micro-capacitance sensor Optimized Operation mathematical model that a consideration electric automobile accesses in a large number, use operation expense minimum simultaneously, the highest and minimum three optimization aim of comprehensive cost of environmental benefit, under six kinds of scheduling strategies preset, the different electric automobile access way of comparative analysis is on the impact of micro-capacitance sensor economical operation, thus checking electric automobile with the access of V2G pattern and V0G pattern access build validity and the feasibility of micro-capacitance sensor Optimal Operation Model.

Accompanying drawing explanation

Fig. 1 is micro-capacitance sensor Optimized Operation analysis process figure under the random charge mode of electric automobile.

Fig. 2 is the charging electric vehicle carry calculation process flow diagram based on Monte Carlo Analogue Method.

Fig. 3 is electric automobile load curve map under unidirectional unordered V0G pattern.

Fig. 4 is electric automobile load curve map under two-way orderly V2G pattern.

Fig. 5 is the lower accumulator cell charging and discharging strategy that is incorporated into the power networks.

Fig. 6 is the predicted power curve map of PV, WT.

Fig. 7 is that the optimization of tactful 1 optimization aim three times each generator units is exerted oneself situation map.

Fig. 8 is that the optimization of tactful 3 optimization aim three times each generator units is exerted oneself situation map.

Fig. 9 is that the optimization of tactful 4 optimization aim three times each generator units is exerted oneself situation map.

Embodiment

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

Embodiment:

The present invention is directed to after electric automobile adds micro-capacitance sensor, on the impact of Optimized Operation.The present invention contains distributed power source and comprises photovoltaic (Photovoltaic for this reason, PV), wind-powered electricity generation (WindTurbine, WT), fuel cell (FuelCell, FC), miniature gas turbine (MicroTurbine, MT), diesel motor (DieselGenerator, DSG) and energy-storage units comprise accumulator (Battery, Bat), also contemplate the access of electric automobile.

The Optimal Scheduling of micro-capacitance sensor is an economical operation optimization problem, and what consider a problem lays particular stress on difference, and objective function is not identical yet.The present invention establishes and considers operating cost and environmental benefit, and both integrated costs of taking into account are as the micro-capacitance sensor multiple goal economic load dispatching model of optimization aim.

1, optimization aim

(1) micro-capacitance sensor management and running cost minimization

The present invention is mainly for the economic load dispatching cost in micro-capacitance sensor operational process, and therefore each distributed power source initial investment construction cost is not counted in scheduling cost category, and when only considering that distributed power source scheduling is exerted oneself, its operation expense and fuel cost etc.; In addition, electric automobile belongs to car owner's private property, and the purchase of electric automobile and upkeep cost are calculated voluntarily by car owner and bear, and are not counted in micro-capacitance sensor operating cost, but this part expense is an influence factor when formulating the electricity price to electric automobile power purchase.So the optimization aim of micro-capacitance sensor operating cost comprises the fuel cost C of distributed power source _g, operation expense C _oM, generator unit depreciable cost C _dP, micro-capacitance sensor and bulk power grid electric energy switching cost C _grid, micro-capacitance sensor and electric car electric energy switching cost C _eVand load synthesis cost of compensation C _l, expression is:

(2) disposal cost discharged pollutants is minimum

CO can be produced when the miniature gas turbine MT in micro-capacitance sensor, fuel cell FC and diesel-driven generator DG operation and bulk power grid unit generation ₂, SO ₂, NO _xdeng pollutant, thus pollutant emission processing cost can be produced.The objective function of micro-capacitance sensor emission treatment cost minimization can be expressed as:

In formula, j is distributed generation unit numbering in micro-capacitance sensor, 1 ~ N; T runs the period, 1 ~ T; K is discharged pollutant type (CO ₂, SO ₂, NO _xdeng); C _kit is process every kilogram of expense discharged pollutants (unit/kg); γ _jk(P _jt) be that in micro-capacitance sensor, generator unit j exports P _jtthe weight (kg/kW) of the pollutant k produced during electric energy; γ _gridk(P _gridt) power distribution network output P _gridtthe weight (kg/kW) of the pollutant k produced during electric energy.

(3) micro-capacitance sensor scheduling integrated cost is minimum

minObj ₃＝m ₁Obj ₁+m ₂Obj ₂(3)

In formula, Obj ₃for micro-capacitance sensor integrated cost; Obj ₁for operating cost; Obj ₂for the processing cost that discharges pollutants; m ₁, m ₂being respectively the weight of operating cost and discharge costs, for balancing the impact of the energy and environment, This document assumes that operating cost is identical with the weight of discharge costs, getting m here ₁=m ₂=1.

2, constraint condition

(1) power-balance constraint

P _jtthe power (kW) that in t period micro-capacitance sensor, generator unit j sends; P _gridtbe the power (kW) that t period power distribution network transmits to micro-capacitance sensor, represent power during negative value by micro-capacitance sensor reverse transfer to power distribution network; P _batterytbe the power (kW) that t period accumulator sends, during negative value, represent accumulator absorbed power; P _loadtit is the workload demand (kW) in t period micro-capacitance sensor.

(2) cold/heating power balance constraint

Cold/cogeneration unit need meet cooling and heating load constraint:

W _Load＝W _MT+W _FC(5)

In formula: W _loadfor whole micro-grid system cold/thermal load demands, kW; W _mTfor miniature gas turbine waste heat flue gas provide cold/thermal power, kW; W _fCfor fuel cell power generation produce heat provide cold/thermal power, kW.

(3) generator unit active power bound constraint

In formula, for the minimum output power of distributed generation unit j; for the peak power output of distributed generation unit j.

(4) the Power Exchange limit value of micro-capacitance sensor and major network retrains

In formula, the minimum power and peak power that between micro-capacitance sensor and power distribution network, common point circuit can transmit respectively.

(5) power constraint of energy-storage units and charged constraint

The charge-discharge electric power that battery system should meet and state-of-charge (SOC) are constrained to:

P _SBmin≤P _SBt≤P _SBmax(8)

SOC _SBmin≤SOC _SBt≤SOC _SBmax(9)

In formula, P _sBmin, P _sBmaxbe respectively minimum power and the peak power of accumulator cell charging and discharging; SOC _sBtthe state-of-charge of accumulator in the t period; SOC _sBmin, SOC _sBmaxminimum value and the maximal value of storage battery charge state respectively.

In addition, because the Optimized Operation of micro-capacitance sensor to accumulator presents cycle dynamics, This document assumes that the SOC of accumulator was consistent whole story dispatching cycle of one day, namely constraint condition is met:

SOC _end＝SOC ₀(10)

Wherein, SOC ₀and SOC _endbe respectively the state-of-charge of initial time 0 and end time 24 accumulator in the dispatching cycle.

(6) power constraint of electric automobile and charged constraint

For electric automobile, the charge-discharge electric power that also should meet and state-of-charge are constrained to:

P _EVmin≤P _EVt≤P _EVmax(11)

SOC _EVmin≤SOC _EVt≤SOC _EVmax(12)

In formula, P _eVmin, P _eVmaxbe respectively minimum power and the peak power of electric automobile discharge and recharge; SOC _eVtthe state-of-charge of batteries of electric automobile in the t period; SOC _min, SOC _maxminimum value and the maximal value of batteries of electric automobile state-of-charge respectively.

3, electric automobile model

(1) electric automobile unordered charging time power characteristic

It is relevant that electric automobile puts the uncertain factors such as charge power and electric automobile during traveling mileage, duration of charging, therefore needs to emulate by platform electric automobile by generating the random occurrence obeying statistical law.All electric automobile powertrace superpositions just can be obtained total charge power curve, and its flow process as shown in Figure 1.

To every platform electric automobile, the probability of nearly 14% can not be gone on a journey, if trip, its daily travel d is similar to obeys logarithm normal distribution, and its probability density function is:

In formula: distribution parameter μ _d=3.2, σ _d=0.88, be distributed as electric automobile daily travel average and standard variance.Corresponding charging electric vehicle electrical demand is E _eV:

E _EV＝η _EV·d(14)

Under the unordered access module of VOG, due to most of car owner go home after at once will to charging electric vehicle, so hypothesis finally returns the moment in one day start the moment of charging exactly, then the moment that starts to charge meets normal distribution:

In formula: μ _s=17.6; σ _s=3.4.

(2) power characteristic during the orderly discharge and recharge of electric automobile

The orderly discharge and recharge of electric automobile, refer to when electric automobile accesses in a large number, through policies such as electricity price guiding, utilize the mode of tou power price to regulate and control electric automobile and under the prerequisite first meeting electric automobile user use habit, electric automobile discharge and recharge is dispatched in order.In electricity price electric discharge peak period time grid-connected, electricity price low-valley interval charges; During isolated island, in the electric discharge of load peak period, charge in the load valley period.

Electric automobile daily consumption energy can be obtained by daily travel S, thus obtain state-of-charge when batteries of electric automobile networks:

In formula, C is the total volume of batteries of electric automobile.

Discharge period is:

In formula: T _all-discfor putting to the total duration that discharges needed for state-of-charge lower limit when electric automobile is full of electricity; T _discfor the actual discharge duration; P _discfor electric automobile discharge power; S _{oC, max}and S _{oC, min}be respectively the bound of storage battery charge state.

Networking discharging time T _{start_disc}by last return moment t ₀with the energy requirement of micro-capacitance sensor through judging to obtain.

Electric discharge finish time T _{end_disc}by discharging time and discharge period jointly determine, the upper limit is this day finish time 24:00.

The electric discharge period is T _{start_disc}~ T _{end_disc}, during this period, electric automobile discharges according to discharge power, is added up by the electric discharge load of N electric automobile, namely obtains the total discharge power discharged in the period.

Due to the restriction of discharge time, part electric automobile is not discharged completely, so start initial state-of-charge when charging, different because of the difference of discharged condition, this initial state-of-charge is uniquely determined by above-mentioned discharge scenario.Single electric automobile required charging load was consumed gross energy among one day, namely

Charge power in conjunction with electric automobile can obtain duration of charge, thus obtains the power load distributing that charges.

(3) Monte Carlo Analogue Method solves

Monte Carlo Method (MonteCarloMethod) is a kind of statistical method estimating mathematical function based on random sampling and stochastic simulation.Its basic solution throughway is: for problem to be solved, according to the statistical law of physical phenomenon itself, or the probability model depending on stochastic variable that arteface one is applicable, make the statistic of some stochastic variable be with the solution of Solve problems.Monte carlo method is according to following 2 theories:

1. the law of large numbers: in function f (x) field of definition [a, b], extract N number of several xi randomly with non-uniform probability distribution, the arithmetic mean of functional value sum converges on the expectation value of function.After extracting abundant random sample, the Monte Carlo estimated value of integration will converge on the correct result of this integration, and namely stochastic variable statistic is:

2. central limit theorem: the stochastic variable that a large amount of faint factor adds up obeys single normal distribution.The error ε of monte carlo method depends on standard deviation sigma and number of samples N, and is directly proportional to standard deviation sigma, is inversely proportional to, that is: with number of samples N side square root

Monte Carlo sampling mean approximation is a kind of effective ways solving random optimization, be also called Method of Stochastic, also statistical test method is claimed, it is for verifying that the constraint condition of Probability Forms provides effective approach, it is mainly used in and solves mathematics, the problem of the aspect such as engineer applied and production management, the basic thought of Monte Carlo sampling mean approximation method is: first set up a probability model, its certain parameter is made to equal the solution of problem, then according to the distribution of hypothesis, concrete value (this process is named again sampling) is selected to stochastic variable, thus construct a deterministic model, calculate result, again by the result that multiple sampling is tested, obtain the statistical property of parameter, finally calculate the approximate value of solution.Based on Monte Carlo simulation charging electric vehicle carry calculation flow process as shown in Figure 2.

In figure, N is electric automobile simulation quantity, and n is the electric automobile that present day analog calculates.The probability distribution that system input information comprises electric automobile total scale, various charging behavior occurs, possible charge period and the probability distribution of initiation of charge time, from initial SOC probability distribution corresponding to many duration constraints, dissimilar charging behavior.To separate unit charging electric vehicle carry calculation, first to determine the charging behavior of this car, if this car has multiple charging behavior, one that systematically discusses meets U (0,1) equally distributed random number, according to the probability distribution that difference charging behavior occurs, determines the charging behavior of vehicle.

(4) electric automobile discharge and recharge case

For the micro-capacitance sensor that a residential block is formed, carried model is verified.There are 400 family residents this residential block, existing given following hypothesis:

1. on average every 2 family families have an electric automobile, and Ji Gai community has 200 electric automobiles;

2. select BYD E6 vehicle herein, parameter is as follows: capacity Q=60kWh; Charge power P _dh=10kW; Power consumption S _1kWh=4.762km/ (kWh); Electric discharge efficiency eta=85%;

3. average each electric motor car stroke every day 34.76km, whole electric automobile day charge volume be 1460kWh.

4. in order to encourage car owner participate in electricity price guide under the orderly discharge and recharge plan of electric automobile, drafting micro-capacitance sensor to the price of electric automobile power purchase is in real time to the price of conventional load sale of electricity, such electric automobile often discharge 1kWh car owner can benefit about 0.6 yuan (the paddy period electricity price wherein during EVs charging is 0.37 yuan/kWh, EVs electric discharge time peak period electricity price be 1.03 yuan/kWh).Along with the maturation in market and the perfect of technology, micro-capacitance sensor supvr suitably can turn down the price to electric automobile power purchase, to regain the cost of repacking charge and discharge device.

The present invention simulates the orderly discharge and recharge part throttle characteristics of different scales electric automobile access electrical network, because the topmost function of electric automobile is still as the vehicles, consider being accustomed to car of user, in orderly discharge and recharge, the 07:00-17:00 period, the discharge and recharge carry calculation not participating in dispatching orderly discharge and recharge is different from unordered discharge and recharge Fig. 3, and the daily load curve of its emulation respectively as shown in Figure 4.

4, micro-capacitance sensor running optimizatin scheduling strategy:

The micro-capacitance sensor scheduler object that the present invention considers comprises photovoltaic power generation equipment (Photovoltaic, PV), wind power generating set (WindTurbine, WT), miniature gas turbine (MicroTurbine, MT), diesel motor (DieselGenerator, DG), fuel cell (FuelCell, FC), accumulator (StorageBattery, SB), the major network of electric automobile (electricvehicles, EVs) and exchange electric energy.Adopt tou power price pattern to formulate micro-capacitance sensor Optimized Operation strategy, according to external electrical network load condition, whole day 24h is divided into paddy period (00:00-07:00 and 23:00-24:00), at ordinary times section (07:00-10:00,15:00-18:00 and 21:00-23:00) and peak period (10:00-15:00 and 18:00-21:00).Micro-capacitance sensor scheduling was a thread period with 1 hour, first power load and the cooling and heating load in current scheduling moment is predicted, and the situation of exerting oneself of photovoltaic power generation equipment and wind power generating set, and monitor the state-of-charge of accumulator, with in each scheduling slot, micro-capacitance sensor operating cost is minimum, maximum and the integrated cost of environmental benefit is minimum is optimization aim, according to different scheduling strategies, obtain the optimum results under different scheduling strategy, and determine the meritorious state of exerting oneself of controllable type generator unit in micro-capacitance sensor, the charge-discharge electric power curve of accumulator and the active power situation exchanged with major network.

(1) micro-capacitance sensor Optimized Operation basic scheduling strategy

1. the scheduling strategy of photovoltaic and wind-powered electricity generation

Because sun power and wind-force belong to clean energy resource, not to environment, therefore preferentially use the electric energy that photovoltaic power generation equipment and wind power generating set send, and arrange accumulator to stablize their output-power fluctuation, make them actual exert oneself more to meet dope force curve.

2. accumulator cell charging and discharging scheduling strategy

The present invention mainly considers the accumulator effect micro-capacitance sensor dispatching system from three aspects:

The first, stablizing the fluctuation that blower fan and photovoltaic are exerted oneself, guaranteeing that they can exert oneself by doping force curve;

The second, when micro-grid connection is run, arrange accumulator in the charging of paddy period, in section disconnection at ordinary times and in the electric discharge of peak period, as shown in Figure 5.Consider that the state-of-charge of accumulator is all periodic cycle in every day, namely the discharge and recharge of a day will get back to the state-of-charge of beginning in this day after terminating.Meanwhile, consider the impact of discharge and recharge on its serviceable life of accumulator, get that its charged lower limit is 20%, the upper limit is 100% here;

3rd, when micro-capacitance sensor islet operation, when photovoltaic and exerting oneself of wind-powered electricity generation meet power load need, and when having surplus, accumulator can store unnecessary electric energy; When photovoltaic and wind power output can not meet power load, accumulator can discharge not enough with supplementary power.Here take into account the periodicity of accumulator cell charging and discharging equally, and get that its charged lower limit is 20%, the upper limit is 90%.

(2) micro-capacitance sensor Optimized Operation strategy

Because electric automobile has the characteristic of storage of electrical energy, after a large amount of access micro-capacitance sensor, mobile energy storage device can be regarded as, conventional energy storage device or standby generator sets can be replaced to a certain extent by reasonable arrangement, thus improve the utilization factor of electric automobile and reduce the investment of building micro-capacitance sensor.Research emphasis of the present invention draws the micro-capacitance sensor scheduling strategy with feasibility, and reasonable arrangement electric automobile access micro-capacitance sensor realizes larger performance driving economy.Therefore, the micro-capacitance sensor Optimized Operation strategy of six consideration electric automobile accesses has been formulated herein.

Strategy 1: micro-grid connection is run, distributed generation unit PV, WT, MT, DG, FC and major network participate in Optimized Operation jointly, the photovoltaic power generation equipment PV and wind power generating set WT that gives priority in arranging for exerts oneself, fuel cell FC, to run under " with electricity fixed heat " pattern, miniature gas turbine MT runs under the pattern of " electricity determining by heat ", accumulator SB is in the electric discharge of peak period, the charging of paddy period, and it is not enough that diesel-driven generator DG fills up residue, jointly meets electricity consumption and cooling and heating load.Can two-way exchange electric energy between micro-capacitance sensor and major network.

Strategy 2: micro-capacitance sensor islet operation, the photovoltaic power generation equipment PV and wind power generating set WT that gives priority in arranging for exerts oneself, and fuel cell FC runs under " with the fixed heat of electricity " pattern, and miniature gas turbine MT runs under the pattern of " electricity determining by heat ".When sending electric energy and exceeding power load, accumulator SB stores unnecessary electric energy; When sending electric energy and can not meeting power load, diesel-driven generator DG and accumulator SB output power supplementary power deficiency;

Strategy 3: micro-grid connection is run, electric automobile adopts the V0G pattern of unidirectional unordered charging.The charge condition of consideration electric automobile EV, regards pure power load as by electric automobile.Power load (conventional load+electric automobile) provides electric energy by distributed generation unit PV, WT, MT, DG, FC and accumulator SB and collaborative the exerting oneself of major network.

Strategy 4: micro-grid connection is run, electric automobile adopts the V2G pattern of two-way orderly charging.Consider the flash-over characteristic of electric automobile EV, concentrate charging to store electric energy at a low price at major network paddy period electric automobile EVs and accumulator SB, now load (conventional load+charging electric vehicle load) provides electric energy by distributed generation unit PV, WT, MT, DG, FC and major network; Do not carry out discharge and recharge at section electric automobile EV at ordinary times as the vehicles, conventional load provides electric energy by distributed generation unit PV, WT, MT, DG, FC and accumulator SB and major network; And start to discharge for conventional load provides electric energy according to certain probability at peak period electric automobile, provide electric energy by distributed generation unit PV, WT, MT, DG, FC and collaborative the exerting oneself of accumulator SB, if there is unnecessary electric energy, feed back in major network.

Strategy 5: micro-capacitance sensor islet operation, electric automobile adopts the V0G pattern of unidirectional unordered charging.The charge condition of consideration electric automobile EV, regards pure power load as by electric automobile.Power load (conventional load+electric automobile) provides electric energy by distributed generation unit PV, WT, MT, DG, FC and collaborative the exerting oneself of accumulator SB.When distributed generation unit PV, WT, MT, DG, FC and accumulator SB maximum output can not meet load electric energy demand, make the active power balance between supply and demand of whole micro-capacitance sensor by temporarily excising interruptible load.

Strategy 6: micro-capacitance sensor islet operation, electric automobile adopts the V2G pattern of two-way orderly charging.Consider the flash-over characteristic of electric automobile EV, during load valley, electric automobile EV concentrates the electric energy that charging storage regenerative resource sends, and now load (conventional load+charging electric vehicle load) is exerted oneself by distributed generation unit PV, WT, MT, DG, FC and provided electric energy; When load is mild, electric automobile EV does not carry out discharge and recharge as the vehicles, and conventional load provides electric energy by distributed generation unit PV, WT, MT, DG, FC and collaborative the exerting oneself of accumulator SB; And electric automobile starts electric discharge according to certain probability when load peak, conventional load provides electric energy by distributed generation unit PV, WT, MT, DG, FC and collaborative the exerting oneself of accumulator SB simultaneously, when distributed generation unit PV, WT, MT, DG, FC and accumulator SB maximum output can not meet load electric energy demand, make the active power balance between supply and demand of whole micro-capacitance sensor by temporarily excising interruptible load.

4, sample calculation analysis

(1) optimum configurations

The search time of present case is one day of summer, formulates the operational plan of 00:00-24:00 period on the same day according to 1 hour time interval.Fig. 6 give photovoltaic, wind energy 24 time discontinuity surface generating prediction curve; Each distributed power source related data is as shown in table 1.Table 2 gives micro-capacitance sensor real-time electrical load requirement, and table 3 gives micro-capacitance sensor real-time cooling workload demand; Table 4 gives the segmentation electricity price of power distribution network; Table 5 gives each generator unit pollutant discharge coefficient.

Each distributed electrical source dates in table 1 micro-capacitance sensor

The real-time electrical load requirement of table 2 micro-capacitance sensor (kW)

Table 3 micro-capacitance sensor real-time cooling workload demand (kW)

Table 4 micro-capacitance sensor and power distribution network electricity price scheme

Note: the peak period is: 10:00 ~ 15:00,18:00 ~ 21:00; Section is at ordinary times: 7:00 ~ 10:00,15:00 ~ 18:00,21:00 ~ 23:00; The paddy period is: 23:00 ~ 24:00,0:00 ~ 7:00.

The each generator unit pollutant discharge coefficient of table 5

(2) interpretation of result and discussion

Under different scheduling strategy and different target function, Optimized Operation total expenses is as shown in table 6.

Table 6 Optimized Operation result total expenses compares

According to the data in table 6, make analysis:

1) scheduling strategy 1,3,4 relatively in, be both the state of being incorporated into the power networks, have dropped 8.2%, 7.9% and 8.0% respectively before adopting optimization aim 1,2,3 times operation total expensess of V2G access module to compare EVs access, that compares employing V0G pattern have dropped 12.8%, 12.0% and 12.4% especially respectively.Under V0G pattern is described, electric automobile increases power load merely, only increases the operating cost of micro-capacitance sensor; And under V2G pattern, electric automobile absorbs the unnecessary electric energy of distributed power source as mobile energy storage device, and the electricity price taking full advantage of major network peak interval of time is poor, charging is concentrated during low electricity price, electric discharge peak clipping during high electricity price, decrease exert oneself burden and the dependence to major network of distributed power source, as Figure 7-9.

2) scheduling strategy 2,5,6 relatively in, be both island operation state, be rise 7.6%, 6.2% and 7.5% respectively before adopting optimization aim 1,2,3 times operation total expensess of V0G pattern to compare EVs access, and rise 9.8%, 12.4% and 12.9% more respectively before adopting optimization aim 1,2,3 times operation total expensess of V2G access module to compare EVs access.Illustrate under islet operation, the access of electric automobile both increases the operating cost of micro-capacitance sensor, this is because distributed power source is exerted oneself, cost is higher than the cost from major network power purchase, simultaneously because a large amount of electric automobile concentrates the load of charging excessive, it is not enough to make up to the effect of conventional electric load peak load shifting the expense that distributed power source additionally runs.Therefore, under the islet operation of short time, need to work out more reasonably charging electric vehicle plan.

3) scheduling strategy 3,5 relatively in, electric automobile is both under V0G access module, total expenses during islet operation under optimization aim 1,2,3 all exceeds 20.6%, 27.3% and 30.5% respectively than when being incorporated into the power networks, this is because during islet operation the cost of electricity-generating of distributed power source than high from the cost of major network power purchase.

4) at scheduling strategy 4, 6 relatively in, electric automobile is both under V2G access module, optimization aim 1 during islet operation, 2, total expenses under 3 also all exceeds 41.0% than when being incorporated into the power networks respectively, 53.2% and 56.5%, this is because during islet operation the cost of electricity-generating of distributed power source than high from the cost of major network power purchase, but the ratio exceeded under V0G pattern in comparing analysis (3), also to have more a lot, this is because EVs concentrates the increase load of charging excessive, call diesel-driven generator in a large number to exert oneself, improve the total expenses of islet operation, especially taking into account with the total expenses under the optimization aim 2 and 3 of environmental benefit, amplification reaches 53.2% and 56.5% respectively.

Claims (7)

1., containing a micro-capacitance sensor Multiobjective Optimal Operation method for electric automobile, it is characterized in that, the method comprises the following steps:

1) determine electric automobile access micro-capacitance sensor pattern, by the separate unit electric automobile under different access module put fill load distribution performance superposition obtain putting of electric automobile fill load distribution performance;

2) electric automobile is joined in micro-capacitance sensor Optimized Operation as micro-capacitance sensor scheduler object, and fill according to putting of electric automobile the micro-capacitance sensor Optimal Operation Model that load distribution performance sets up the extensive electric automobile access of consideration;

3) employing considers based on the particle swarm optimization algorithm of automatic recombination mechanism the micro-capacitance sensor Optimal Operation Model that extensive electric automobile accesses, and the micro-capacitance sensor scheduling economy under the multiple different scheduling strategy of comparative analysis, thus obtain optimal scheduling strategy.

2. a kind of micro-capacitance sensor Multiobjective Optimal Operation method containing electric automobile according to claim 1, it is characterized in that, the pattern in described step 1) comprises the V0G pattern of unidirectional unordered charging and the V2G pattern of two-way orderly discharge and recharge.

3. a kind of micro-capacitance sensor Multiobjective Optimal Operation method containing electric automobile according to claim 1, is characterized in that, described step 2) in consider that the optimization object function of the micro-capacitance sensor Optimal Operation Model of extensive electric automobile access is:

Micro-capacitance sensor management and running cost Obj ₁minimum:

${\mathrm{minObj}}_1=\underset{t=1}{\overset{T}{\Σ}}{\mathrm{\ΔT}}_t\[C_G+C_{OM}+C_{DP}+C_{Grid}+C_{EV}+C_L\]$

The disposal cost Obj discharged pollutants ₂minimum:

${\mathrm{minObj}}_2=\underset{t=1}{\overset{T}{\Σ}}{\mathrm{\ΔT}}_t\[\underset{j=1}{\overset{N}{\Σ}}{C}_k{\mathrm{\γ}}_{jk}\left(P_{jt}\right)+C_k{\mathrm{\γ}}_{gridk}\left(P_{gridt}\right)\]$

Micro-capacitance sensor scheduling integrated cost is minimum:

minObj ₃＝m ₁Obj ₁+m ₂Obj ₂

Wherein, C _gfor the fuel cost of distributed power source, C _oMfor operation expense, C _dPfor the depreciable cost of generator unit, C _gridfor micro-capacitance sensor and bulk power grid electric energy switching cost, C _eVfor micro-capacitance sensor and electric car electric energy switching cost, C _lfor load synthesis cost of compensation, Δ T _tfor the unit time period, T is dispatching cycle, and j is distributed generation unit numbering in micro-capacitance sensor, and t is for running the period, and k is discharged pollutant type, C _kfor processing every kilogram of expense discharged pollutants, γ _jk(P _jt) for being generator unit j output P in micro-capacitance sensor _jtthe weight of the pollutant k produced during electric energy, γ _gridk(P _gridt) be power distribution network output P _gridtthe weight of the pollutant k produced during electric energy, m ₁, m ₂for the weight of operating cost and discharge costs;

Constraint condition is:

Power-balance retrains:

$\underset{j=1}{\overset{N}{\Σ}}{P}_{jt}+P_{gridt}+P_{batteryt}=P_{loadt}$

Cold/heating power balance constraint:

W _Load＝W _MT+W _FC

Distributed generation unit active-power P _jtbound retrains:

$P_j^{min}\≤P_{jt}\≤P_j^{max}$

The exchange power P of micro-capacitance sensor and major network _gridlimit value retrains:

$P_{grid}^{\min }\≤P_{grid}\≤P_{grid}^{\max }$

The power P of energy-storage units _sBtconstraint and charged SOC _sBtconstraint:

P _SBmin≤P _SBt≤P _SBmax

SOC _SBmin≤SOC _SBt≤SOC _SBmax

SOC _end＝SOC ₀

The power P of electric automobile _evtconstraint and charged SOC _eVtconstraint:

P _EVmin≤P _EVt≤P _EVmax

SOC _EVmin≤SOC _EVt≤SOC _EVmax

Wherein, P _jtfor the power that generator unit j in t period micro-capacitance sensor sends, P _gridtfor the power that t period power distribution network transmits to micro-capacitance sensor, P _batterytfor the power that t period accumulator sends, P _loadtfor the workload demand in t period micro-capacitance sensor, W _loadfor whole micro-grid system cold/thermal load demands, W _mTfor miniature gas turbine waste heat flue gas provide cold/thermal power, W _fCfor fuel cell power generation produce heat provide cold/thermal power, for the minimum output power of distributed generation unit j, for the peak power output of distributed generation unit j, for the minimum power that common point circuit between micro-capacitance sensor and power distribution network can transmit, for the peak power that common point circuit between micro-capacitance sensor and power distribution network can transmit, P _sBmin, P _sBmaxbe respectively minimum power and the peak power of accumulator cell charging and discharging, SOC _sBmin, SOC _sBmaxminimum value and the maximal value of storage battery charge state respectively, SOC ₀and SOC _endbe respectively the state-of-charge of initial time 0 and end time 24 accumulator in the dispatching cycle, P _eVmin, P _eVmaxbe respectively minimum power and the peak power of electric automobile discharge and recharge, SOC _min, SOC _maxminimum value and the maximal value of batteries of electric automobile state-of-charge respectively.

4. a kind of micro-capacitance sensor Multiobjective Optimal Operation method containing electric automobile according to claim 2, it is characterized in that, under the V0G pattern of unidirectional unordered charging, the electrical demand filling electric automobile in load distribution performance of putting of electric automobile is:

E _EV＝η _EV·d

Wherein, η _eVfor the electrical demand coefficient of electric automobile mileage, the distance travelled d obeys logarithm normal distribution of every platform electric automobile, its probability density function is:

$f(d,{\mathrm{\μ}}_d,{\mathrm{\σ}}_d)=\frac{1}{{\mathrm{d\σ}}_d\sqrt{2\mathrm{\π}}}{e}^{-\frac{{(\ln d-\mathrm{\μ})}^2}{2{{\mathrm{\σ}}_d}^2}}$

Duration of charging f _sx () meets normal distribution:

$f_s\left(x\right)=\left\{\begin{array}{cc}\frac{1}{{\mathrm{\&sigma;}}_s\sqrt{2\mathrm{\&pi;}}}{e}^{-\frac{(x-{\mathrm{\&mu;}}_s)}{2{{\mathrm{\&sigma;}}_s}^2}},& ({\mathrm{\&mu;}}_s-12)<x<24\\ \frac{1}{{\mathrm{\&sigma;}}_s\sqrt{2\mathrm{\&pi;}}}{e}^{-\frac{(x+24-{\mathrm{\&mu;}}_s)}{2{{\mathrm{\&sigma;}}_s}^2}},& 0<x<({\mathrm{\&mu;}}_s-12)\end{array}\right.$

Wherein, μ _dand μ _sfor expectation value, σ _dand σ _sfor standard deviation.

5. a kind of micro-capacitance sensor Multiobjective Optimal Operation method containing electric automobile according to claim 2, is characterized in that, under the V2G pattern of two-way orderly discharge and recharge, electric automobile put the discharge period T filled in load distribution performance _disc1for:

$T_{disc1}=T_{all-disc}-\frac{{\mathrm{DW}}_{100}}{100P_{disc}}$

$T_{all-disc}=\frac{({\mathrm{SOC}}_{max}-{\mathrm{SOC}}_{min})}{P_{disc}}$

Charge power in conjunction with electric automobile can obtain duration of charge, thus obtains lasting duration of charging T _disc2for:

$T_{disc2}=\frac{P_{EV}}{P_c}$

Single electric automobile required charging load was consumed gross energy, i.e. P among one day _eVfor:

$P_{EV}=\frac{{\mathrm{DW}}_{100}}{100}+P_{disc}(T_{end\_disc}-T_{start\_disc})$

Wherein, T _all-discfor putting to the total duration that discharges needed for state-of-charge lower limit when electric automobile is full of electricity, P _discfor electric automobile discharge power, SOC _maxand SOC _minbe respectively the bound of storage battery charge state, D is electric automobile daily travel, W ₁₀₀for electric automobile hundred kilometers of power consumption, T _{end_disc}for electric discharge finish time, T _{start_disc}for networking discharging time, P _cfor charging electric vehicle power.

6. a kind of micro-capacitance sensor Multiobjective Optimal Operation method containing electric automobile according to claim 1, is characterized in that, described step 2) in multiple different scheduling strategies comprise micro-grid connection operation reserve and micro-capacitance sensor islet operation strategy.

7. a kind of micro-capacitance sensor Multiobjective Optimal Operation method containing electric automobile according to claim 1, it is characterized in that, described step 3) specifically comprises the following steps:

31) according to the access module of electric automobile, the model parameter of each distributed generation unit, each objective function parameters and each constraint condition parameter in setting micro-capacitance sensor, and introduce uncontrollable measurable distributed power source and exert oneself and cold electric load parameter;

32) controlled unit miniature gas turbine of exerting oneself, fuel cell, diesel-driven generator, accumulator and major network interchange power are tieed up particles as five, and set particle cluster algorithm parameter, comprise population, solution space dimension, maximum iteration time, particle maximal rate and restructuring index r;

33) calculate the fitness of each particle, and record the target function value of the current individual extreme value of each particle and correspondence, and then obtain all extreme values and corresponding target function value, and select individual optimal value and global optimum;

34) the current number of times of iteration adds one, upgrades population and carries out position and speed;

35) judged result whether meet Premature Convergence standard and restructuring number of times whether reach the value preset, if result meets Premature Convergence standard and restructuring number of times does not reach the value preset, then recombinate population and restructuring index r=r+1, and return step 33), otherwise carry out step 36);

36) judge whether to meet the condition of convergence, if so, then obtain global optimum or reach maximum iteration time, finishing iteration process, if not, then returning step 33), continue iterative operation.