CN106228466A - Power retailing company Purchasing combination Design Method and system - Google Patents

Abstract

The present invention provides a kind of power retailing company Purchasing combination Design Method and system, gather power retailing company load and electricity price data, gather and simulate power retailing company sale of electricity influence factor, build the Purchasing combination optimisation strategy model of power retailing company, and be standardized processing, use the Purchasing combination optimisation strategy model after analytical method solving standardization, acquisition optimal decision vector, optimal decision vector is denormalized, obtain true decision vector, and calculate power retailing company profit under optimal decision, according to true decision vector and the profit under optimal decision, generate power retailing company Purchasing combination scheme.During whole, consider power retailing company load, electricity price data and sell an influence factor, and building Purchasing combination optimisation strategy model, calculating power retailing corporate profit, power retailing company Purchasing combination scheme can be generated according to market situation and trend.

Description

Power retailing company Purchasing combination Design Method and system

Technical field

The present invention relates to electric power network technical field, particularly relate to power retailing company Purchasing combination Design Method With system.

Background technology

Along with the development of science and technology, electricity has become as the indispensable material of people's productive life, people's need to electricity The amount of asking is the most increasing.Huge power demand, not only to whole power supply unit (such as electric station, transformer station and circuit Maintenance is patrolled and examined) new requirement is proposed, also there is a new difficult problem to need to overcome and solve for sale of electricity unit.

At present, power sales opening is the important content of power market reform.Come from the international practice of Power Market Construction Seeing, power sales is open contributes to reducing electric cost, Optimization of Energy Structure, and then improves the safety and economy fortune of power system OK.

For power retailing company, Purchasing combination different in conract market, ahead market and Real-time markets Impact can be brought to corporate profit, use which kind of Purchasing combination Cai Nengshi corporate profit to maximize, it is achieved optimum power purchase has become It it is a crucial problem.And there is no a kind of power retailing company Purchasing combination scheme at present to instruct power retailing company optimum Power purchase.

Summary of the invention

Based on this, it is necessary to instruct power retailing company closing for there is no power retailing company Purchasing combination scheme at present The about problem of optimum power purchase in market, ahead market and Real-time markets, it is provided that a kind of power retailing company Purchasing combination scheme Method for designing and system, to design power retailing company Purchasing combination scheme, instruct power retailing company in conract market, day Optimum power purchase in front market and Real-time markets.

A kind of power retailing company Purchasing combination Design Method, including step:

Gather power retailing company load and electricity price data；

Gathering and simulate power retailing company sale of electricity influence factor, sale of electricity influence factor includes load transfer and market part Volume；

According to power retailing company load and electricity price data and power retailing company sale of electricity influence factor, build electric power zero Sell the Purchasing combination optimisation strategy model of company, and be standardized Purchasing combination optimisation strategy model processing；

Use the Purchasing combination optimisation strategy model after analytical method solving standardization, it is thus achieved that optimal decision to Amount；

Optimal decision vector is denormalized, obtains true decision vector, and calculate power retailing company in optimal decision Under profit；

According to true decision vector and the profit under optimal decision, generate power retailing company Purchasing combination scheme.

A kind of power retailing company Purchasing combination Scheme Design System, including:

First acquisition module, is used for gathering power retailing company load and electricity price data；

Second acquisition module, is used for gathering and simulate power retailing company sale of electricity influence factor, and sale of electricity influence factor include Load transfer and the market share；

Model construction module, for according to power retailing company load and electricity price data and power retailing company sale of electricity shadow The factor of sound, builds the Purchasing combination optimisation strategy model of power retailing company, and marks Purchasing combination optimisation strategy model Quasi-ization processes；

Parsing module, for using the Purchasing combination optimisation strategy model after analytical method solving standardization, obtains Obtain optimal decision vector；

Computing module, for being denormalized optimal decision vector, obtains true decision vector, and it is public to calculate power retailing Department's profit under optimal decision；

Instruct module, for according to true decision vector and the profit under optimal decision, generating power retailing company power purchase Assembled scheme.

Power retailing company of the present invention Purchasing combination Design Method and system, gather power retailing company load and electricity Valence mumber evidence, gathers and simulates power retailing company sale of electricity influence factor, builds the Purchasing combination optimisation strategy of power retailing company Model, and be standardized processing, use the Purchasing combination optimisation strategy model after analytical method solving standardization, obtain Optimal decision vector, optimal decision vector is denormalized, obtains true decision vector, and calculate power retailing company Profit under excellent decision-making, according to true decision vector and the profit under optimal decision, generates Purchasing combination side of power retailing company Case.During whole, consider power retailing company load, electricity price data and sell an influence factor, and building power purchase group Close optimisation strategy model, calculate power retailing corporate profit, power retailing company can be generated purchase according to market situation and trend Electricity assembled scheme, to instruct optimum power purchase in conract market, ahead market and Real-time markets to generate profit maximization.

Accompanying drawing explanation

Fig. 1 is the schematic flow sheet of power retailing company of the present invention first embodiment of Purchasing combination Design Method；

Fig. 2 is the schematic flow sheet of power retailing company of the present invention second embodiment of Purchasing combination Design Method；

Fig. 3 is load degree of bias δ_LAt interval [-θ_L/ 2, θ_LProfit distribution schematic diagram when changing in/2]；

Fig. 4 is Spot Price degree of bias δ_PAt interval [-θ_P/ 2, θ_PProfit distribution schematic diagram when changing in/2]；

Fig. 5 is that the profit of power retailing company is by load θ_LWith electricity price θ_PThe impact of interval change；

Fig. 6 is the structural representation of power retailing company of the present invention first embodiment of Purchasing combination Scheme Design System；

Fig. 7 is the structural representation of power retailing company of the present invention second embodiment of Purchasing combination Scheme Design System.

Detailed description of the invention

Power retailing company the most respectively can from conract market, ahead market and Real-time markets power purchase, electric power zero The company that sells can power purchase by the hour in ahead market and Real-time markets.On the other hand, the contract power purchase one of power retailing company As be long-term, can according to the contract energy decomposition method decided through consultation in advance power energy allocation to each period, it will be recognized that specific to certain Day is definite value.The power purchase different from these three kinds of Market Selection affects power retailing corporate profit than regular meeting, it addition, power retailing is public Department also can be in the face of different Trading risk.How rational allocation power purchase in conract market, ahead market and Real-time markets Ratio, becomes and instructs power retailing company optimum power purchase committed step, power retailing company of the present invention Purchasing combination conceptual design Method and system is starting point based on foregoing, proposes one and can generate rationally and effectively instruct power retailing company closing The about method for designing of optimum power purchase scheme and system in market, ahead market and Real-time markets.

As it is shown in figure 1, a kind of power retailing company Purchasing combination Design Method, including step:

S100: gather power retailing company load and electricity price data.

Power retailing company load and electricity price data can gather acquisition with the data of electrically-based retail records.Record number According to including historical record data and real-time recorded data, on the one hand associate between analytical data from history long term data, On the other hand data real-time variable feature is considered.

Non-essential, step S100 specifically includes consideration tou power price, gathers power retailing company load and electricity price data, And utilize strengthening interval number to characterize power retailing company load and electricity price data.

Assume to be divided into Z tou power price time section every day, period z time a length of Tz (z=1,2 ..., Z), sale of electricity valency is γ_Z($/MWh).If the basic period is 1h, then Σ Tz=24 (h).If the electricity price of period t is ρ (t), then available following point of ρ (t) Section function representation:

$\mathrm{\&rho;}\left(t\right)=\left\{\begin{array}{c}{\mathrm{\&gamma;}}_1,0<t\&le;T_1\\ {\mathrm{\&gamma;}}_2,T_1<t\&le;T_1+T_2\\ .\\ .\\ .\\ {\mathrm{\&gamma;}}_z,\underset{l=1}{\overset{z-1}{\&Sigma;}}{T}_l<t\&le;\underset{i=1}{\overset{z}{\&Sigma;}}{T}_l\\ .\\ .\\ .\\ {\mathrm{\&gamma;}}_Z,\underset{l=1}{\overset{Z-1}{\&Sigma;}}{T}_l<t\&le;24\end{array}\right.---\left(1\right)$

May make up electricity price vector

S200: gather and simulate power retailing company sale of electricity influence factor, sale of electricity influence factor includes load transfer and city Market share.

Power retailing company sells an influence factor and mainly includes meeting transfer and the market share.If certain institute of power retailing company Accounting for the market share isIts value keeps stable within one period, can be defined as follows:

In formula: l (t) is period t original loads demand defined above, Q (t) is bearing that this power retailing company is carried Lotus summation.Then can use the market share of following formula simulation Duo Jia electric power retail company:

In formula: β₁And β₂It is respectively this power retailing company loyalty customer proportion and other power retailing company is loyal Client's proportion；ω_z(z=1,2 ..., Z) it is the coefficient of association relevant to electricity price, for characterizing because price factor is caused Client flowing；Ferr (a) is error function；ο_zAnd σ_zIt is respectively average price and the standard deviation of this period market guidance.Formula (3) mould Similar power retailing company, customer type (loyalty), Electricity price fluctuation etc. are intended to the power retailing houses market part studied The impact of volume.Loyalty customer is more weak with associating of other influence factor, and non-loyalty customer is selected retailer to have weight by price Affect.Under tou power price pattern, user can be by response tou power price power cost saving cost, and major way is that load turns Move, the load of price higher period will transfer to the relatively low period.Keep in longer period in view of tou power price parameter Constant, before and after the transfer of definable load power consumption situation of change per hour:

$Q^{\&prime;}\left(t\right)=Q\left(t\right)\&lsqb;1+\underset{k=1,k\&NotEqual;t}{\overset{24}{\&Sigma;}}\mathrm{\&alpha;}(k,t)\frac{\mathrm{\&rho;}\left(k\right)-\mathrm{\&rho;}\left(t\right)}{\mathrm{\&rho;}\left(t\right)}\&rsqb;---\left(4\right)$

In formula: Q (t) and Q ' (t) is respectively the prediction power consumption before period t load shifts and the correction after load transfer Power consumption；α (k, t) is elastic scalar index from k period to the t period that shift from of load, and 0≤α (k, t)≤1.

S300: according to power retailing company load and electricity price data and power retailing company sale of electricity influence factor, build The Purchasing combination optimisation strategy model of power retailing company, and be standardized Purchasing combination optimisation strategy model processing.

Gather ready data based on step S100 and step S200, build the Purchasing combination optimization of power retailing company Policy model, and be standardized Purchasing combination optimisation strategy model processing.Rigorous mathematical formulae will be used and tie below Close step S300 in Fig. 2 detailed description embodiment wherein and implement process.

As in figure 2 it is shown, wherein one be in example, step S300 includes:

S310: according to power retailing company load and electricity price data and power retailing company sale of electricity influence factor, calculate Power retailing company dynamoelectric benefit next day.

Constructed power retailing corporate decision Optimized model includes the load and electricity price represented by interval number, electric power Retail company is power purchase in conract market, ahead market and Real-time markets, and power purchase ratio is respectively WithAccording to L^±=[l^±(1),...,l^±(t),...l^±(24)]^TWith the electricity price information in each market, the actual load amount of power retailing company Q′^±(t) (actual load amount Q '^±Correction power consumption accumulated value after the transfer of (t) load) can be based on L^±With tou power price data P_sT () is tried to achieve.Based on factors such as aforesaid tou power price, load transfer, the market shares, power retailing company can be asked for as the following formula The dynamoelectric benefit of next day:

$I^{\&PlusMinus;}=\underset{t=1}{\overset{24}{\&Sigma;}}{I}^{\&PlusMinus;}\left(t\right)=\underset{t=1}{\overset{24}{\&Sigma;}}\&lsqb;P_s\left(t\right)-M_c^{\&PlusMinus;}\left(t\right)P_c\left(t\right)-M_{da}^{\&PlusMinus;}\left(t\right)P_{da}\left(t\right)-M_{rt}^{\&PlusMinus;}\left(t\right)P_{rt}^{\&PlusMinus;}\left(t\right)\&rsqb;Q^{\&prime;\&PlusMinus;}\left(t\right)---\left(5\right)$

S320: be target to the maximum with day dynamoelectric benefit summation, builds the Purchasing combination optimisation strategy mould of power retailing company Type.

It is target to the maximum purchasing dynamoelectric benefit summation day, sets up the Purchasing combination optimisation strategy model of power retailing company.

S330: use function expression characterize respectively power retailing company any time in conract market, ahead market with And the power purchase ratio of Real-time markets, power retailing company any time in conract market, the purchasing of ahead market and Real-time markets Electricity ratio sum is 1.

Power retailing company in conract market, the power purchase ratio sum of ahead market, Real-time markets be 1, it may be assumed that

$M_c^{\&PlusMinus;}\left(t\right)+M_{da}^{\&PlusMinus;}\left(t\right)+M_{rt}^{\&PlusMinus;}\left(t\right)=1,\&ForAll;t---\left(6\right)$

Power retailing company can power purchase by the hour in ahead market and Real-time markets.On the other hand, power retailing is public Department contract power purchase be usually long-term, can according to the contract energy decomposition method decided through consultation in advance power energy allocation to each period, It is believed that specific to being one day definite value.If Q_cT () is the contract purchase of electricity of certain period in decomposing, i.e. have

$Q^{\&prime;\&PlusMinus;}\left(t\right)M_c^{\&PlusMinus;}\left(t\right)={Q^{\&PlusMinus;}}_c\left(t\right)={\mathrm{\&lambda;}}^{\&PlusMinus;},\&ForAll;t---\left(7\right)$

Q_cmin≤λ^±=[λ^-,λ⁺]≤Q_cmax (8)

Additionally, the fluctuation range in the power purchase ratio of ahead market and Real-time markets can be described as

$M_{damin}\&le;M_{da}^{\&PlusMinus;}\left(t\right)\&le;M_{damax}---\left(9\right)$

$M_{rtmin}\&le;M_{rt}^{\&PlusMinus;}\left(t\right)\&le;M_{rtmax}---\left(10\right)$

S340: remove redundancy decision vector in function expression respectively, and adjust feasible zone difference, it is thus achieved that the order of magnitude is close And the standardization decision variable of linear independence.

Followed by standardization.First remove redundancy decision vector, adjust feasible zone difference, it is thus achieved that order of magnitude phase Closely, the standardization decision variable of linear independence.

$X^{\&prime;\&PlusMinus;}={\&lsqb;{\mathrm{\&lambda;}}^{\&prime;\&PlusMinus;},M_{da}^{\&prime;\&PlusMinus;}\left(1\right),...,M_{da}^{\&prime;\&PlusMinus;}\left(t\right),...,M_{da}^{\&prime;\&PlusMinus;}\left(24\right)\&rsqb;}^T---\left(11\right)$

$M_{rt}^{\&PlusMinus;}\left(t\right)=1-M_{da}^{\&PlusMinus;}\left(t\right)-{\mathrm{\&lambda;}}^{\&PlusMinus;}/Q^{\&prime;\&PlusMinus;}\left(t\right)---\left(12\right)$

$M_{da}^{\&prime;\&PlusMinus;}\left(t\right)=M_{da}^{\&PlusMinus;}\left(t\right)-M_{damin}---\left(13\right)$

${\mathrm{\&lambda;}}^{\&prime;\&PlusMinus;}\left(t\right)=\frac{{\mathrm{\&lambda;}}^{\&PlusMinus;}\left(t\right)-Q_{c\min }}{Q_{c\max }-Q_{c\min }}---\left(14\right)$

S350: standardization decision variable is substituted into Purchasing combination optimisation strategy model, it is thus achieved that purchasing after standardization Electricity Combinatorial Optimization Policy model.

Standardization decision vector substitution Purchasing combination optimisation strategy model is obtained standardized Purchasing combination optimisation strategy Model:

$I_0^{\&PlusMinus;}=\underset{t=1}{\overset{24}{\&Sigma;}}\left(P_s\right(t)-P_{rt}^{\&PlusMinus;}(t\left)\right)Q^{\&prime;\&PlusMinus;}\left(t\right)+Q_{c\min }\underset{t=1}{\overset{24}{\&Sigma;}}\left(P_{rt}^{\&PlusMinus;}\right(t)-P_c(t\left)\right)+\underset{t=1}{\overset{24}{\&Sigma;}}{Q}^{\&prime;\&PlusMinus;}\left(t\right)\left(P_{rt}^{\&PlusMinus;}\right(t)-P_{da}(t\left)\right)M_{da\min }---\left(16\right)$

And

0≤λ′^±≤1 (18)

$0\&le;M_{da}^{\&prime;\&PlusMinus;}\left(t\right)\&le;M_{damax}-M_{damin},\&ForAll;t---\left(19\right)$

${\mathrm{\&lambda;}}^{\&prime;\&PlusMinus;}+Q^{\&prime;}\left(t\right)M_{da}^{\&prime;\&PlusMinus;}\left(t\right)\&le;Q^{\&prime;}\left(t\right)(1-M_{rtmin}-M_{damin})-Q_{cmin},\&ForAll;t---\left(20\right)$

$-{\mathrm{\&lambda;}}^{\&prime;\&PlusMinus;}-Q^{\&prime;}\left(t\right)M_{da}^{\&PlusMinus;}\left(t\right)\&le;(M_{rtmax}+M_{da\min }-1)Q^{\&prime;}\left(t\right)+Q_{cmin},\&ForAll;t---\left(21\right)$

Comprehensively can obtain constrained parameters matrix A formula (20) and B formula (21).So far the Purchasing combination optimization of power retailing company is obtained The criteria optimization model of decision-making, the most i.e. can use analytic method to solve it.

S400: use the Purchasing combination optimisation strategy model after analytical method solving standardization, it is thus achieved that optimum is certainly Plan vector.

Below the load and electricity price data that use A area are described in detail the power purchase group of constructed power retailing company Close optimisation strategy model and the side of solving.

In the A area load of 2013 and electricity price data, arbitrarily choose the data of 4 days, respectively January 1, May 12 Day, August 20 and the December load data of 29 days, ahead market electricity price data, Real-time markets electricity price data.Due to power retailing Business is when determining power purchase strategy, and following load and Spot Price data are unknown, and front having addressed can use y^±=[y^-,y⁺] =[y_real-θ/2,y_real+ θ/2] such range format describes, and the excursion of interval number is θ.If power retailing company is every User is valuated by three periods of natural gift: 1) 00:00 point at midnight presses low ebb electricity price charging to 7:00 in morning；2) 7:00 in morning point is extremely Electricity price charging when evening, 22:00 pressed peak；3) 22:00 in evening point-midnight 00:00 presses low ebb electricity price charging.Give this successively The electricity price of three periods is: γ 1=35 $/MWh, γ 2=50 $/MWh, γ 3=35 $/MWh；Three periods time a length of: T1= 7h, T2=15h, T3=2h.For this tou power price mechanism, power load can be adjusted or shift by user, to not Impact produces or saves electric power expenditure on the premise of life.For peak load period (i.e. 7:00 in morning point to 22:00 in evening), electric power Demand will shift to price low-valley interval, if transferring ratio is 0.08；Low-valley interval (i.e. 00:00 at midnight point to 7:00 point in morning, And 22:00 point in evening-midnight 00:00 point) receive the electricity needs being transferred out by the peak load period, if price low-valley interval is inhaled The inflow ratio of peak load received is 0.15.Owing to the peak load period is 15 hours, two low-valley intervals 9 hours altogether, and peak load The load radix of period is bigger, and the load transferring ratio of such peak load period is less than the inflow ratio of the received peak load of low-valley interval Rate.So, the elastic scalar index α (t of load transfer₁,t₂) can use following method to determine:

$\mathrm{\&alpha;}(t_1,t_2)=\left\{\begin{array}{c}0.08,t_1\&Element;{\mathrm{\&Lambda;}}_v,t_2\&Element;{\mathrm{\&Lambda;}}_p;\\ 0.15,t_1\&Element;{\mathrm{\&Lambda;}}_p,t_2\&Element;{\mathrm{\&Lambda;}}_v;\\ 0,t_1,t_2\&Element;{\mathrm{\&Lambda;}}_g,g=p,v.\end{array}\right.---\left(22\right)$

In formula: rate period and low ebb rate period when subscript p and v represent peak respectively；Λ_vAnd Λ_pIt is respectively low ebb electricity price Rate period set when period set and peak.

S500: be denormalized optimal decision vector, obtains true decision vector, and calculates power retailing company at optimum Profit under decision-making.

The optimal decision vector obtained based on step S400, and according to above-mentioned formula (5), calculates power retailing company Profit under excellent decision-making.Continue to launch explanation as a example by examples detailed above.

In instantiation, given β₁=0.1, β₂=0.2, ω 1=0.7, ω₂=1, ω₃=0.7, o1=34, ο₂= 49, ο₃=37, σ₁=2, σ₂=3, σ₃=2.Then can try to achieve city shared by power retailing company according to tou power price data and formula (3) Market shareGiven θ_L=10, θ_P=2, Q_cmin=3000MWh, Q_cmax=8000MWh, M_damax=0.8M_damin=0.1, M_rtmax=0.8, M_rtmin=0.1, contract power purchase valency Pc=30 $/MWh, carried out Purchasing combination optimization to selected 4 days respectively, obtain Corresponding optimum results, and try to achieve Purchasing combination optimum results corresponding power retailing corporate profit, as shown in table 1.With 2013 It is denormalized as a example by January 1, in, available concrete Purchasing combination decision scheme shown in table 2.

The Expect Profits that table 1 power retailing company optimum Purchasing combination strategy is corresponding

Table 2 power retailing company optimum Purchasing combination scheme

S600: according to true decision vector and the profit under optimal decision, generate power retailing company Purchasing combination scheme.

Profit under the true decision vector calculated based on step before and optimal decision, generates power retailing public Department's Purchasing combination scheme, to instruct power retailing company power purchase.Specifically, the power retailing company Purchasing combination scheme of generation Described in have power retailing company power purchase ratio in conract market, ahead market and Real-time markets such that it is able to realize Optimum power purchase maximizes with power retailing corporate profit.

Power retailing company of the present invention Purchasing combination Design Method, gathers power retailing company load and electricity price number According to, gather and simulate power retailing company sale of electricity influence factor, build the Purchasing combination optimisation strategy model of power retailing company, And be standardized processing, use the Purchasing combination optimisation strategy model after analytical method solving standardization, it is thus achieved that Excellent decision vector, is denormalized optimal decision vector, obtains true decision vector, and calculates power retailing company and determine at optimum Profit under plan, according to true decision vector and the profit under optimal decision, generates power retailing company Purchasing combination scheme.Whole During individual, consider power retailing company load, electricity price data and sell an influence factor, and building Purchasing combination optimization Policy model, calculates power retailing corporate profit, can generate power retailing company Purchasing combination according to market situation and trend Scheme, to instruct optimum power purchase in conract market, ahead market and Real-time markets to generate profit maximization.

As in figure 2 it is shown, wherein an embodiment also includes after step S600:

S700: analyze power retailing company load and the electricity price impact on profit.

Below as a example by 1 day January in 2013, the impact that power retailing company is gained by analystal section data characteristic.First The impact on profit of the prediction degree of bias of analysis load and Spot Price interval amount.If degree of bias δ ∈ [-θ/2, θ/2], i.e. uncertain region Between be y^±=[y^-,y⁺]=[y_real-θ/2+δ,y_real+θ/2+δ]；The δ<0 when predicting left avertence, otherwise then δ>0.When load degree of bias δ L and Spot Price degree of bias δ P is respectively at interval [-θ_L/ 2, θ_L/ 2] and [-θ_P/ 2, θ_P/ 2], during change, corresponding profit interval changes The most as shown in Figures 3 and 4.Can be seen that from Fig. 3 and 4, it was predicted that the degree of bias is linear approximate relationship on the impact that profit is distributed.δ L exists Interval [-θ_L/ 2, θ_L/ 2] in, during change, the fluctuation range of profit is not so good as δ P at interval [-θ_P/ 2, θ_P/ 2] in, fluctuation during change shows Writing, this is owing to both relative skewnesses vary in size.The order of magnitude of load is 10⁵, the order of magnitude of electricity price is then 10²-10³ Between, so the former relative skewness is 10^-4, the latter is then 10^-2, so the latter's slope is bigger.Profit fluctuation range relative to For total profit the most greatly, can accept.Visible, the uncertainty describing load and Spot Price by interval amount is simpler Single and real.

Then the analystal section weight range change impact on profit.Work as θ_LAnd θ_PRespectively in [0,100] and [0,4] is interval During change, with θ_LFor x-axis, θ_PFor y-axis, power retailing corporate profit is z-axis, and the variation relation of three is as shown in Figure 5.Interval model Enclosing the least, uncertainty degree is the lowest；Vice versa.From fig. 5, it can be seen that the interval range of load and Spot Price is the biggest, electricity Power retail company make a profit faced by risk the biggest.As can be seen from Table 1, the profit area that the interval optimized algorithm of strengthening obtains is used Between smaller, i.e. profit distribution ratio comparatively dense, and can obtain the expected value of day profit according to " suitably interval ", this can be electric power Retail company determines that Purchasing combination decision-making provides technical support.

As shown in Figure 6, a kind of power retailing company Purchasing combination Scheme Design System, including:

First acquisition module 100, is used for gathering power retailing company load and electricity price data.

Second acquisition module 200, is used for gathering and simulating power retailing company sale of electricity influence factor, and sale of electricity influence factor wraps Include load transfer and the market share.

Model construction module 300, for selling with electricity price data and power retailing company according to power retailing company load Electricity influence factor, builds the Purchasing combination optimisation strategy model of power retailing company, and enters Purchasing combination optimisation strategy model Column criterionization processes.

Parsing module 400, for using the Purchasing combination optimisation strategy model after analytical method solving standardization, Acquisition optimal decision vector.

Computing module 500, for being denormalized optimal decision vector, obtains true decision vector, and calculates electric power zero Sell company's profit under optimal decision.

Instruct module 600, for according to true decision vector and the profit under optimal decision, generating power retailing company and purchase Electricity assembled scheme.

Power retailing company of the present invention Purchasing combination Scheme Design System, it is public that the first acquisition module 100 gathers power retailing Department's load and electricity price data, the second acquisition module 200 gathers and simulates power retailing company sale of electricity influence factor, model construction mould Block 300 builds the Purchasing combination optimisation strategy model of power retailing company, and is standardized processing, and parsing module 400 uses Purchasing combination optimisation strategy model after analytical method solving standardization, it is thus achieved that optimal decision vector, computing module 500 Optimal decision vector is denormalized, obtains true decision vector, and calculate power retailing company profit under optimal decision, Instruct module 600 according to true decision vector and the profit under optimal decision, generate power retailing company Purchasing combination scheme.Whole During individual, consider power retailing company load, electricity price data and sell an influence factor, and building Purchasing combination optimization Policy model, calculates power retailing corporate profit, can generate power retailing company Purchasing combination according to market situation and trend Scheme, to instruct optimum power purchase in conract market, ahead market and Real-time markets, it is achieved profit maximization.

Wherein in an embodiment, the first acquisition module 100 considers tou power price, gather power retailing company load with Electricity price data, and utilize strengthening interval number to characterize power retailing company load and electricity price data.

Such as Fig. 7, wherein in an embodiment, model construction module 300 includes:

Profit computing unit 310, for selling with electricity price data and power retailing company according to power retailing company load Electricity influence factor, calculates power retailing company dynamoelectric benefit next day.

Model construction unit 320, for being target to the maximum with day dynamoelectric benefit summation, builds the power purchase of power retailing company Combinatorial Optimization Policy model.

Function characterization unit 330, is used for using function expression to characterize power retailing company any time respectively at contract The power purchase ratio of market, ahead market and Real-time markets, power retailing company any time in conract market, ahead market with And the power purchase ratio sum of Real-time markets is 1.

First processing unit 340, for removing redundancy decision vector in function expression respectively, and it is poor to adjust feasible zone Different, it is thus achieved that the order of magnitude is close and the standardization decision variable of linear independence.

Second processing unit 350, for substituting into Purchasing combination optimisation strategy model by standardization decision variable, it is thus achieved that standard Purchasing combination optimisation strategy model after change process.

Wherein in an embodiment, the computing formula of power retailing company dynamoelectric benefit next day is:

$I^{\&PlusMinus;}=\underset{t=1}{\overset{24}{\&Sigma;}}{I}^{\&PlusMinus;}\left(t\right)=\underset{t=1}{\overset{24}{\&Sigma;}}\&lsqb;P_s\left(t\right)-M_c^{\&PlusMinus;}\left(t\right)P_c\left(t\right)-M_{da}^{\&PlusMinus;}\left(t\right)P_{da}\left(t\right)-M_{rt}^{\&PlusMinus;}\left(t\right)P_{rt}^{\&PlusMinus;}\left(t\right)\&rsqb;Q^{\&prime;\&PlusMinus;}\left(t\right)$

In formula, I^±For power retailing company dynamoelectric benefit next day, t is each hour, P_sT () is t electricity price, For t power retailing company in conract market power purchase ratio,For t power retailing company in conract market power purchase Ratio,For t power retailing company in Day-ahead Electricity Purchase ratio,Exist for t power retailing company Real-time markets power purchase ratio, Q '^±T () is power retailing company load.

As it is shown in fig. 7, wherein in an embodiment, power retailing company Purchasing combination Scheme Design System also includes:

Analyze module 700, for analyzing power retailing company load and the electricity price impact on profit.

Embodiment described above only have expressed the several embodiments of the present invention, and it describes more concrete and detailed, but also Can not therefore be construed as limiting the scope of the patent.It should be pointed out that, come for those of ordinary skill in the art Saying, without departing from the inventive concept of the premise, it is also possible to make some deformation and improvement, these broadly fall into the protection of the present invention Scope.Therefore, the protection domain of patent of the present invention should be as the criterion with claims.

Claims (10)

1. a power retailing company Purchasing combination Design Method, it is characterised in that include step:

Gather power retailing company load and electricity price data；

Gathering and simulate power retailing company sale of electricity influence factor, described sale of electricity influence factor includes load transfer and market part Volume；

According to described power retailing company load and electricity price data and described power retailing company sale of electricity influence factor, build institute State the Purchasing combination optimisation strategy model of power retailing company, and described Purchasing combination optimisation strategy model is standardized place Reason；

Use the Purchasing combination optimisation strategy model after analytical method solving standardization, it is thus achieved that optimal decision vector；

Described optimal decision vector is denormalized, obtains true decision vector, and calculate described power retailing company at optimum Profit under decision-making；

According to described true decision vector and the profit under described optimal decision, generate power retailing company Purchasing combination scheme.

Power retailing company the most according to claim 1 Purchasing combination Design Method, it is characterised in that described collection Power retailing company load includes with the step of electricity price data:

Consider tou power price, gather power retailing company load and electricity price data, and utilize strengthening interval number to characterize described electric power Retail company's load and electricity price data.

Power retailing company the most according to claim 1 Purchasing combination Design Method, it is characterised in that described basis Described power retailing company load and electricity price data and described power retailing company sale of electricity influence factor, build described electric power zero Sell the Purchasing combination optimisation strategy model of company, and described Purchasing combination optimisation strategy model is standardized the step processed Including:

According to described power retailing company load and electricity price data and described power retailing company sale of electricity influence factor, calculate electricity Power retail company dynamoelectric benefit next day；

It is target to the maximum with day dynamoelectric benefit summation, builds the Purchasing combination optimisation strategy model of described power retailing company；

Use function expression characterize respectively described power retailing company any time in conract market, ahead market and in real time The power purchase ratio in market, described power retailing company any time in conract market, ahead market and the power purchase of Real-time markets Ratio sum is 1；

Remove redundancy decision vector in described function expression respectively, and adjust feasible zone difference, it is thus achieved that the order of magnitude is close and line The standardization decision variable that property is unrelated；

Described standardization decision variable is substituted into described Purchasing combination optimisation strategy model, it is thus achieved that the power purchase after standardization Combinatorial Optimization Policy model.

Power retailing company the most according to claim 3 Purchasing combination Design Method, it is characterised in that described electric power The computing formula of retail company's dynamoelectric benefit next day is:

$I^{\&PlusMinus;}=\underset{t=1}{\overset{24}{\&Sigma;}}{I}^{\&PlusMinus;}\left(t\right)=\underset{t=1}{\overset{24}{\&Sigma;}}\&lsqb;P_s\left(t\right)-M_c^{\&PlusMinus;}\left(t\right)P_c\left(t\right)-M_{da}^{\&PlusMinus;}\left(t\right)P_{da}\left(t\right)-M_{rt}^{\&PlusMinus;}\left(t\right)P_{rt}^{\&PlusMinus;}\left(t\right)\&rsqb;Q^{\&prime;\&PlusMinus;}\left(t\right)$

In formula, I^±For described power retailing company dynamoelectric benefit next day, t is each hour, P_sT () is t electricity price, For power retailing company described in t in conract market power purchase ratio,For power retailing company described in t at contract Market power purchase ratio,For power retailing company described in t in Day-ahead Electricity Purchase ratio,Described in t Power retailing company is at Real-time markets power purchase ratio, Q '^±T () is described power retailing company actual load.

Power retailing company the most according to claim 1 Purchasing combination Design Method, it is characterised in that described basis Described true decision vector and the profit under described optimal decision, after generating the step of power retailing company Purchasing combination scheme Also include:

Analyze described power retailing company load and the electricity price impact on profit.

6. a power retailing company Purchasing combination Scheme Design System, it is characterised in that including:

First acquisition module, is used for gathering power retailing company load and electricity price data；

Second acquisition module, is used for gathering and simulating power retailing company sale of electricity influence factor, and described sale of electricity influence factor includes Load transfer and the market share；

Model construction module, for selling with electricity price data and described power retailing company according to described power retailing company load Electricity influence factor, builds the Purchasing combination optimisation strategy model of described power retailing company, and described Purchasing combination is optimized plan Slightly model is standardized processing；

Parsing module, for using the Purchasing combination optimisation strategy model after analytical method solving standardization, it is thus achieved that Excellent decision vector；

Computing module, for being denormalized described optimal decision vector, obtains true decision vector, and calculates described electric power zero Sell company's profit under optimal decision；

Instruct module, for according to described true decision vector and the profit under described optimal decision, generating power retailing company Purchasing combination scheme.

Power retailing company the most according to claim 6 Purchasing combination Scheme Design System, it is characterised in that described first Acquisition module considers tou power price, gathers power retailing company load and electricity price data, and it is described to utilize strengthening interval number to characterize Power retailing company load and electricity price data.

Power retailing company the most according to claim 6 Purchasing combination Scheme Design System, it is characterised in that described model Structure module includes:

Profit computing unit, for selling with electricity price data and described power retailing company according to described power retailing company load Electricity influence factor, calculates power retailing company dynamoelectric benefit next day；

Model construction unit, for being target to the maximum with day dynamoelectric benefit summation, builds the power purchase group of described power retailing company Close optimisation strategy model；

Function characterization unit, is used for using function expression to characterize described power retailing company any time respectively in contract city , ahead market and the power purchase ratio of Real-time markets, described power retailing company any time in conract market, ahead market And the power purchase ratio sum of Real-time markets is 1；

First processing unit, for removing redundancy decision vector in described function expression respectively, and adjusts feasible zone difference, obtains The order of magnitude is close and the standardization decision variable of linear independence；

Second processing unit, for substituting into described Purchasing combination optimisation strategy model by described standardization decision variable, it is thus achieved that mark Purchasing combination optimisation strategy model after quasi-ization process.

Power retailing company the most according to claim 8 Purchasing combination Scheme Design System, it is characterised in that described electric power The computing formula of retail company's dynamoelectric benefit next day is:

$I^{\&PlusMinus;}=\underset{t=1}{\overset{24}{\&Sigma;}}{I}^{\&PlusMinus;}\left(t\right)=\underset{t=1}{\overset{24}{\&Sigma;}}\&lsqb;P_s\left(t\right)-M_c^{\&PlusMinus;}\left(t\right)P_c\left(t\right)-M_{da}^{\&PlusMinus;}\left(t\right)P_{da}\left(t\right)-M_{rt}^{\&PlusMinus;}\left(t\right)P_{rt}^{\&PlusMinus;}\left(t\right)\&rsqb;Q^{\&prime;\&PlusMinus;}\left(t\right)$

In formula, I^±For described power retailing company dynamoelectric benefit next day, t is each hour, P_sT () is t electricity price, For power retailing company described in t in conract market power purchase ratio,For power retailing company described in t at contract Market power purchase ratio,For power retailing company described in t in Day-ahead Electricity Purchase ratio,Described in t Power retailing company is at Real-time markets power purchase ratio, Q '^±T () is described power retailing company actual load.

Power retailing company the most according to claim 6 Purchasing combination Scheme Design System, it is characterised in that also include:

Analyze module, for analyzing described power retailing company load and the electricity price impact on profit.