CN106127389A - A kind of virtual plant combined heat and power scheduling Robust Optimization Model - Google Patents

Abstract

The invention provides a kind of virtual plant combined heat and power scheduling Robust Optimization Model, this model aggregation unit includes distributed power generation unit, Wind turbines, photovoltaic unit, cogeneration of heat and power (CHP) unit, boiler, electricity energy storage, hot energy storage, electric load and thermic load, and considers that CHP unit participates in SRM sight.The uncertain problem faced for virtual plant (VPP) and the risk thus brought, use robust optimization (RO) to process EM electricity price, SRM electricity price, wind power output, photovoltaic are exerted oneself, electric load and the uncertainty of thermic load, and set up risk quantification index, the robustness of balance RO model and economy.The model that the present invention provides well solves when simultaneously participating in EM and SRM, and before VPP day, combined heat and power Scheduling Optimization Model sets up problem, improves the motility of decision-making, thus adds the profit of VPP.Further, the introducing of RO model effectively reduces system risk, thus chooses suitable robust coefficient for policymaker and provide effective reference.

Description

A kind of virtual plant combined heat and power scheduling Robust Optimization Model

Technical field

The invention belongs to electric power system power source scheduling field, excellent particularly to a kind of virtual plant combined heat and power scheduling robust Change model.

Background technology

In recent years, cogeneration of heat and power (combined heat and power, CHP) unit is flourishing with the advantage of energy-conserving and environment-protective How development, but its generating and the feature that is closely connected of heat supply reduce its motility, therefore, effectively manage it and become Problem demanding prompt solution.Virtual plant (virtualpower plant, VPP) technology is to solve this problem to provide new think of Road.It is polymerized CHP unit with VPP form, controls it by EMS and run, CHP unit or even VPP can be realized overall Optimizing scheduling.

VPP faces electricity price in scheduling process, regenerative resource is exerted oneself and the multiple uncertain factor such as load fluctuation, gives certainly Plan and system safety operation bring the biggest difficulty.Therefore, use rational scheduling mode, quantify as far as possible or weaken uncertain because of The element impact on scheduling strategy, it is achieved profit maximization becomes the focus of academia research.Robust optimizes (robust Optimization, RO) process probabilistic method as a kind of, have without knowing uncertain parameter probability distribution, meter Calculate the advantages such as quick, ability of avoiding risk is good.RO passes through robustness and the economy of robust coefficient adjustment system, and robust coefficient is more Greatly, system robustness is the strongest, and the risk faced is the least.Further, existing research majority considers that robust coefficient to expected profit and is determined The impact of plan scheme, seldom relates to the quantitative analysis of risk, causes the conservative of decision scheme unavoidably or advances rashly, meanwhile, and the most non-face Face RO and process SRM electricity price and the probabilistic problem of thermic load.

Summary of the invention

Goal of the invention: provide a kind of virtual plant combined heat and power scheduling Robust Optimization Model, solve to simultaneously participate in energy When market (EM) and spinning reserve market (SRM), before VPP day, combined heat and power Scheduling Optimization Model sets up problem.Use RO process EM electricity price, SRM electricity price, wind power output, photovoltaic are exerted oneself, electric load and the uncertainty of thermic load, reduce the risk that VPP faces, And set up risk quantification index, reduce the blindness of unascertained decision.

Technical scheme: the present invention provides a kind of virtual plant combined heat and power scheduling Robust Optimization Model, comprises the following steps:

Step 1: set up under EM and SRM combined heat and power Scheduling Optimization Model before VPP day；

Step 2: use RO to process EM electricity price, SRM electricity price, wind power output, photovoltaic are exerted oneself, electric load and thermic load the most true Qualitative, set up RO model；

Step 3: set up risk quantification index, weighs the risk of systems face.

Further, step 1 comprises the following steps:

For VPP network operator, its target is that net profit maximizes, and therefore the object function of Optimized model is as follows:

$max\underset{t=1}{\overset{T}{\Σ}}(C_t^m-C_t^{dg}-C_t^{chp}-C_t^{boil}-C_t^s-C_t^{dr})$

Object function comprises six parts, and respectively VPP participates in electricity market profit gainedDG costCHP machine Composition is originallyBoiler costEnergy storage costDR costEvery part expression is as follows:

$C_t^m={\mathrm{\λ}}_t^{em}{P}_t^{em}+{\mathrm{\λ}}_t^{srm}{R}_t^{srm}$

In formula:It is respectively t period EM electricity price and SRM electricity price；Be respectively t period VPP at EM and The competitive bidding amount of SRM, for decision variable.

DG cost includes operating cost, start-up and shut-down costs and Environmental costs:

$C_t^{dg}=\underset{i=1}{\overset{n_i}{\mathrm{\Σ}}}\left(\begin{array}{c}\left(k_i{P}_{i,t}^{dg}+k_i^f{\mathrm{\μ}}_{i,t}^o\right)+\left({\mathrm{\λ}}_i^{su}{\mathrm{\μ}}_{i,t}^{su}+{\mathrm{\λ}}_i^{sd}{\mathrm{\μ}}_{i,t}^{sd}\right)+\\ P_{i,t}^{dg}\underset{e=1}{\overset{n_e}{\mathrm{\Σ}}}{Q}_{i,t}^{dg}\left(V_e+Y_e\right)\end{array}\right)$

In formula: n_iFor DG unit number；For t period i-th cell DG output, for decision variable；k_i、 Fuel cost, fixed cost, start-up cost and stopping cost for i-th cell DG；Boolean variable Table respectively Show whether t period i-th cell DG works, starts, stops, be to put 1, otherwise set to 0；n_eFor the pollutant sum considered；For I-th cell DG e item pollutant discharge amount；V_e、Y_eIt is respectively e item pollutant environmental value, the fine order of magnitude.

CHP unit cost includes that fuel cost, start-up and shut-down costs and Environmental costs, boiler cost include fuel cost and environment Cost, the CHP unit considered herein and boiler are all with natural gas as fuel, such as following formula:

$C_t^{chp}=\underset{l=1}{\overset{n_l}{\mathrm{\Σ}}}\left(\begin{array}{c}\frac{860{\mathrm{\λ}}_t^{ng}{P}_{l,t}^{echp}}{H^{ng}{\mathrm{\η}}_l^{chp}}+({\mathrm{\λ}}_l^{su}{\mathrm{\μ}}_{l,t}^{su}+{\mathrm{\λ}}_l^{sd}{\mathrm{\μ}}_{l,t}^{sd})+\\ \frac{P_{l,t}^{echp}}{{\mathrm{\η}}_l^{chp}}\underset{e=1}{\overset{n_e}{\Σ}}{Q}_{l,e}^{chp}(V_e+Y_e)\end{array}\right)$

$C_t^{boil}=\underset{k=1}{\overset{n_k}{\Σ}}(\frac{860{\mathrm{\λ}}_t^{ng}{P}_{k,t}^{boil}}{H^{ng}{\mathrm{\η}}_k^{boil}}+\frac{P_{k,t}^{boil}}{{\mathrm{\η}}_k^{boil}}\underset{e=1}{\overset{n_e}{\Σ}}{Q}_{k,e}^{boil}(V_e+Y_e\left)\right)$

In formula: n_l、n_kIt is respectively CHP unit and boiler unit number；For t period Gas Prices；H^ngFor natural gas heat Value；860 are converted into the Conversion of measurement unit ratio of kcal for kWh；For t period l unit CHP unit output power and Kth unit boiler heat supplying power, for decision variable；It is l unit CHP unit and kth unit boiler efficiency, needs It is noted thatIt is only the power supplying efficiency of CHP unit, not whole efficiency； It is that l unit CHP starts and stops Cost；Boolean variableRepresent whether t period l unit CHP starts, stops, and is to put 1, otherwise sets to 0 respectively；It is l unit CHP unit and kth unit boiler e item pollutant discharge amount, V_e、Y_eIt is respectively e item to pollute Substance environment is worth, the fine order of magnitude；

Energy storage cost includes ES cost and TS cost, and it linearly closes with charge and discharge power and storage, heat release power approximation System:

$C_t^s=\underset{x=1}{\overset{n_x}{\Σ}}\left(a_x^{es}\right(P_{x,t}^{esc}+P_{x,t}^{esd})+b_x^{es})+\underset{z=1}{\overset{n_z}{\Σ}}\left(a_z^{ts}\right(P_{z,t}^{tsc}+P_{z,t}^{tsd})+b_z^{ts})$

In formula: n_x、n_zIt is respectively ES and TS unit number；For t period xth unit ES charge and discharge power, for certainly Plan variable；For t period z unit TS storage, heat release power, for decision variable； For ES and TS Cost coefficient.

DR cost statement is shown as when VPP interrupts customer charge, need to pay certain compensation.In view of different interruption degree pair The influence degree of user is different, will interrupt making up price and load rejection horizontal hook, and interruption level is the highest, and making up price is more Height, is specifically expressed as follows:

$C_t^{dr}=\underset{m=1}{\overset{n_m}{\Σ}}\left({\mathrm{\λ}}_m^{curt}{P}_{m,t}^{elcurt}\right)$

In formula: n_mFor interrupting horizontal progression；It is that m level interrupts level compensating price；Interrupt for t period m level Horizontal break load, for decision variable.

It is as follows that VPP operationally needs to meet constraints:

1) DG constraint.

$P_{i,t}^{dg}\≥P_i^{\min }{\mathrm{\μ}}_{i,t}^o$

$P_{i,t}^{dg}+R_{i,t}^{dg}\≤P_i^{\max }{\mathrm{\μ}}_{i,t}^o$

$R_{i,t}^{dg}\≤r_i^u{t}^r$

$-r_i^d\≤P_{i,t}^{dg}-P_{i,t-1}^{dg}\≤r_i^u$

${\mathrm{\μ}}_{i,t}^o-{\mathrm{\μ}}_{i,t-1}^o\≤{\mathrm{\μ}}_{i,t}^{su}$

${\mathrm{\μ}}_{i,t-1}^o-{\mathrm{\μ}}_{i,t}^o\≤{\mathrm{\μ}}_{i,t}^{sd}$

In formula:It is respectively i-th cell DG maximum, minimum output power；Standby for t period i-th cell DG With capacity, for decision variable；r_i ^u、r_i ^dBe respectively i-th cell DG upwards, climbing rate downwards；t^rFor the active service time.

2) CHP Unit commitment.

$P_{l,t}^{tchp}=k_l^{chp}{P}_{l,t}^{echp}$

$P_{l,t}^{echp}\≥P_l^{\min }{\mathrm{\μ}}_{l,t}^o$

$P_{l,t}^{echp}+R_{l,t}^{echp}\≤P_l^{\max }{\mathrm{\μ}}_{l,t}^o$

$R_{l,t}^{echp}\≤r_l^u{t}^r$

$-r_l^d\≤P_{l,t}^{echp}-P_{l,t-1}^{echp}\≤r_l^u$

${\mathrm{\μ}}_{l,t}^o-{\mathrm{\μ}}_{l,t-1}^o\≤{\mathrm{\μ}}_{l,t}^{su}$

${\mathrm{\μ}}_{l,t-1}^o-{\mathrm{\μ}}_{l,t}^o\≤{\mathrm{\μ}}_{l,t}^{sd}$

In formula:For t period l unit CHP unit heating power；It is l unit CHP unit hotspot stress, with CHP machine unit characteristic is relevant；It is respectively l unit CHP unit output power maximum, minimum；For t period l Unit CHP unit reserve capacity, for decision variable；r_l ^u、r_l ^dBe respectively l unit CHP unit upwards, climbing rate downwards；Boolean VariableRepresent whether t period l unit CHP unit runs, and is to put 1, otherwise sets to 0.

3) boiler constraint.

$0\≤P_{k,t}^{boil}\≤P_k^{max}$

In formula:For kth unit boiler maximum heating power.

4) ES and TS constraint.

$0\≤P_{x,t}^{esc}\≤P_x^{cmax}$

$0\≤P_{x,t}^{esd}\≤P_x^{dmax}$

$S_x^{\min }\≤S_{x,t}^{es}\≤S_x^{\max }$

$S_{x,t}^{es}=S_{x,t-1}^{es}+{\mathrm{\η}}_x^{esc}{P}_{x,t}^{esc}-\frac{P_{x,t}^{esd}}{{\mathrm{\η}}_x^{esd}}$

$S_{x,0}^{es}=S_x^{esi}$

$S_{x,24}^{es}=S_x^{esf}$

$0\≤P_{z,t}^{tsc}\≤P_z^{cmax}$

$0\≤P_{z,t}^{tsd}\≤P_z^{dmax}$

$S_z^{\min }\≤S_{z,t}^{ts}\≤S_z^{\max }$

$S_{z,t}^{ts}=S_{z,t-1}^{ts}+{\mathrm{\η}}_z^{tsc}{P}_{z,t}^{tsc}-\frac{P_{z,t}^{tsd}}{{\mathrm{\η}}_z^{tsd}}$

$S_{z,0}^{ts}=S_z^{tsi}$

$S_{z,24}^{ts}=S_z^{tsf}$

In formula:It is respectively xth unit ES maximum charge and discharge power；Store up for t period xth unit ES Electricity；It is respectively xth unit ES reserve of electricity upper and lower limit；It is respectively xth unit ES charge and discharge effect Rate；It is respectively xth unit ES beginning, end reserve of electricity；It is respectively the storage of z unit TS maximum, heat release merit Rate；For t period z unit TS quantity of heat storage；It is respectively z unit TS quantity of heat storage upper and lower limit；Point It is not z unit TS storage, exothermal efficiency；It is respectively z unit TS beginning, end quantity of heat storage.

5) DR constraint.

$0\≤P_{m,t}^{elcurt}\≤k_m^{curt}{P}_t^{el}$

$P_t^{elcurt}=\underset{m=1}{\overset{n_m}{\Σ}}{P}_{m,t}^{elcurt}$

$P_t^{elcurt}+R_t^{el}\≤\underset{m=1}{\overset{n_m}{\Σ}}\left(k_m^{curt}{P}_t^{el}\right)$

In formula:It is that m level interrupts horizontal break coefficient；P_t ^elFor t period electric load；P_t ^elcurtInterrupt negative for the t period Lotus；For t period reserve capacity for load variation in power, for decision variable.

6) electricity, heating power balance constraint.

$\underset{w=1}{\overset{n_w}{\mathrm{\Σ}}}{P}_{w,t}^{wp}+\underset{s=1}{\overset{n_s}{\mathrm{\Σ}}}{P}_{s,t}^{pv}+\underset{i=1}{\overset{n_i}{\mathrm{\Σ}}}{P}_{i,t}^{dg}+\underset{l=1}{\overset{n_l}{\mathrm{\Σ}}}{P}_{l,t}^{echp}+\underset{x=1}{\overset{n_x}{\mathrm{\Σ}}}{P}_{x,t}^{esd}=P_t^{em}+P_t^{el}-P_t^{elcurt}+\underset{x=1}{\overset{n_x}{\mathrm{\Σ}}}{P}_{x,t}^{esc}$

$\underset{l=1}{\overset{n_l}{\Σ}}{P}_{l,t}^{tchp}+\underset{k=1}{\overset{n_k}{\Σ}}{P}_{k,t}^{boil}+\underset{z=1}{\overset{n_z}{\Σ}}{P}_{z,t}^{tsd}\≥P_t^{tl}+\underset{z=1}{\overset{n_z}{\Σ}}{P}_{z,t}^{tsc}$

In formula: n_w、n_sIt is respectively Wind turbines, photovoltaic unit number；It is respectively t period Wind turbines w, photovoltaic Unit s exerts oneself；P_t ^tlFor t period thermic load.

7) competitive bidding spare capacity retrains.

$R_t^{srm}=\underset{i}{\overset{n_i}{\mathrm{\Σ}}}{R}_{i,t}^{dg}+\underset{l}{\overset{n_l}{\mathrm{\Σ}}}{R}_{l,t}^{echp}+R_t^{el}$

8) VPP system reserve constraint.

In VPP, DG, CHP unit, interruptible load all can provide system reserve, but do not include standby to SRM competitive bidding Capacity:

$\begin{array}{c}\underset{i=1}{\overset{n_i}{\mathrm{\Σ}}}({P_i}^{\max }-P_{i,t}^{\mathrm{dg}}-R_{i,t}^{\mathrm{dg}})+\underset{l=1}{\overset{n_l}{\mathrm{\Σ}}}({P_l}^{\max }-P_{l,t}^{\mathrm{echp}}-R_{l,t}^{\mathrm{echp}})+\\ (\underset{m=1}{\overset{n_m}{\mathrm{\Σ}}}\left(k_m^{\mathrm{curt}}{P_t}^{\mathrm{el}}\right)-{P_t}^{\mathrm{elcurt}}-R_t^{\mathrm{el}})\≥R_t^a\end{array}$

In formula:For spare capacity needed for t period VPP.

Further, step 2 comprises the following steps:

EM electricity price in above-mentioned model, SRM electricity price, wind power output, photovoltaic are exerted oneself, electric load and thermic load all exist uncertain Property, use RO to process such uncertain problem, with a herein_ijRepresent above-mentioned uncertain parameter, it is considered to a_ijForm is as follows:

$a_{ij}\∈\[{\tilde{a}}_{ij}-{\hat{a}}_{ij},{\tilde{a}}_{ij}+{\hat{a}}_{ij}\]$

Wherein:

${\tilde{a}}_{ij}=\frac{1}{2}({\stackrel{\‾}{a}}_{ij}+{\underset{\‾}{a}}_{ij})$

${\hat{a}}_{ij}=\frac{1}{2}({\stackrel{\‾}{a}}_{ij}-{\underset{\‾}{a}}_{ij})$

In formula: a _ijIt is respectively the upper and lower limit of uncertain parameter,I.e. think that uncertain parameter is at it Fluctuation in upper and lower limit interval range.

Above-mentioned uncertain parameter form considers the situation that uncertain parameter is the worst, and the decision-making thus done has the strongest Conservative, but lose economy.Introduce robust coefficient Г, Γ ∈ [0, | J |] for this, wherein, J is the collection of all uncertain parameter Close, now, uncertain parameter a_ijInterval beWhen Γ=0, do not consider the most true of uncertain parameter Qualitative, this model is consistent with deterministic optimization model, and system robustness is poor.Along with the continuous increase of Γ, system robustness by Gradually improving, economy constantly declines.As Γ=| J |, it is the most conservative form.By regulation robust coefficient Г, the most available The optimal solution of different conservative, takes into account robustness and the economy of decision scheme.

Use RO to process EM electricity price, SRM electricity price, wind power output, photovoltaic are exerted oneself, electric load and the uncertainty of thermic load, Gained object function and constraints are as follows:

$max\underset{t=1}{\overset{T}{\mathrm{\Σ}}}\left(\frac{1}{2}\right({\stackrel{\‾}{\mathrm{\λ}}}_t^{em}+{\underset{\‾}{\mathrm{\λ}}}_t^{em})P_t^{em}+\frac{1}{2}({\stackrel{\‾}{\mathrm{\λ}}}_t^{srm}+{\underset{\‾}{\mathrm{\λ}}}_t^{srm})R_t^{srm}-C_t^{dg}-C_t^{chp}-C_t^{boil}-C_t^s-C_t^{dr})-{\mathrm{\Γ}}^m{\mathrm{\υ}}^m-\underset{t=1}{\overset{T}{\Σ}}(q_t^{em}+q_t^{srm})$

${\mathrm{\υ}}^m+q_t^{em}\≥\frac{1}{2}\left({\stackrel{\‾}{\mathrm{\λ}}}_t^{em}-{\underset{\‾}{\mathrm{\λ}}}_t^{em}\right)y_t^{em}$

${\mathrm{\υ}}^m+q_t^{srm}\≥\frac{1}{2}\left({\stackrel{\‾}{\mathrm{\λ}}}_t^{srm}-{\underset{\‾}{\mathrm{\λ}}}_t^{srm}\right)y_t^{srm}$

$-y_t^{em}\≤P_t^{em}\≤y_t^{em}$

$-y_t^{srm}\≤P_t^{srm}\≤y_t^{srm}$

$P_{w,t}^{wp}+{\mathrm{\Γ}}_{w,t}^{wp}{\mathrm{\υ}}_{w,t}^{wp}+q_{w,t}^{wp}\≤\frac{1}{2}\left({\stackrel{\‾}{P}}_{w,t}^{wp}-{\underset{\‾}{P}}_{w,t}^{wp}\right)$

$P_{s,t}^{\mathrm{pv}}+{\mathrm{\Γ}}_{s,t}^{\mathrm{pv}}{\mathrm{\υ}}_{s,t}^{\mathrm{pv}}+q_{s,t}^{\mathrm{pv}}\≤\frac{1}{2}\left({\stackrel{\‾}{P}}_{s,t}^{\mathrm{pv}}-{\underset{\‾}{P}}_{s,t}^{\mathrm{pv}}\right)$

$P_t^{el}-{\mathrm{\Γ}}_t^{el}{\mathrm{\υ}}_t^{el}-q_t^{el}\≥\frac{1}{2}\left({\stackrel{\‾}{P}}_t^{el}-{\underset{\‾}{P}}_t^{el}\right)$

$P_t^{tl}-{\mathrm{\Γ}}_t^{tl}{\mathrm{\υ}}_t^{tl}-q_t^{tl}\≥\frac{1}{2}\left({\stackrel{\‾}{P}}_t^{tl}+{\underset{\‾}{P}}_t^{tl}\right)$

${\mathrm{\υ}}_{w,t}^{wp}+q_{w,t}^{wp}\≥\frac{1}{2}\left({\stackrel{\‾}{P}}_{w,t}^{wp}-{\underset{\‾}{P}}_{w,t}^{wp}\right)y_{w,t}^{wp}$

${\mathrm{\υ}}_{s,t}^{pv}+q_{s,t}^{pv}\≥\frac{1}{2}({\stackrel{\‾}{P}}_{s,t}^{pv}-{\underset{\‾}{P}}_{s,t}^{pv})y_{s,t}^{pv}$

${\mathrm{\υ}}_t^{\mathrm{el}}+q_t^{\mathrm{el}}\≥\frac{1}{2}({\stackrel{\‾}{P}}_t^{\mathrm{el}}-{\underset{\‾}{P}}_t^{\mathrm{el}})y_t^{\mathrm{el}}$

${\mathrm{\υ}}_t^{tl}+q_t^{tl}\≥\frac{1}{2}({\stackrel{\‾}{P}}_t^{tl}-{\underset{\‾}{P}}_t^{tl})y_t^{tl}$

$y_{w,t}^{wp},y_{s,t}^{pv},y_t^{el},y_t^{el}\≥1$

${\mathrm{\υ}}^m,q_t^{em},y_t^{em},q_t^{srm},y_t^{srm},{\mathrm{\υ}}_{w,t}^{wp},q_{w,t}^{wp},y_{w,t}^{wp},$

${\mathrm{\υ}}_{s,t}^{pv},q_{s,t}^{pv},y_{s,t}^{pv},{\mathrm{\υ}}_t^{el},q_t^{el},y_t^{el},{\mathrm{\υ}}_t^{tl},q_t^{tl},y_t^{tl}\≥0$

In formula: P _t ^el、 P _b ^tlRespectively Exert oneself for EM electricity price, SRM electricity price, wind power output, photovoltaic, electric load and thermic load upper and lower limit；Γ^m、 Be respectively electricity price, wind power output, photovoltaic are exerted oneself, electric load and thermic load robust coefficient, υ^m、 Auxiliary for introduce Help variable.

Further, step 3 comprises the following steps:

As robust coefficient Γ ＜ | J |, all uncertain parameter fluctuation situation cannot be included in uncertain parameter interval, unavoidably Can bring certain risk to system, robust coefficient is the least, and risk is the biggest.Therefore, set up rational quantification of targets risk level, Robustness and the economy of VPP system just can be better balanced.The metric of system risk is general and loses loading, lose load Persistent period etc. are linked up with, and loading, the risk cost of its correspondence are lost in main considerationExpression formula is as follows:

$C_t^{ens}={\mathrm{\λ}}_t^{ens}{P}_t^{ens}$

In formula:Lose load fine for the t period, when VPP cannot feed system internal loading, need to force excise customer charge Time, heavy fine to be paid for this, thereforeNumerical value the biggest.Loading is lost, when supply electricity in VPP for the t period When amount is more than demand electricity,On the contrary, if VPP delivery is insufficient for load and electric power market demand, then

$P_t^{ens}=P_t^{el}-P_t^{elcurt}+P_t^{em}+\underset{x=1}{\overset{n_x}{\mathrm{\Σ}}}{P}_{x,t}^{esc}-\underset{w=1}{\overset{n_w}{\mathrm{\Σ}}}{P}_{w,t}^{wp}-\underset{s=1}{\overset{n_s}{\mathrm{\Σ}}}{P}_{s,t}^{pv}-\underset{i=1}{\overset{n_i}{\mathrm{\Σ}}}{P}_{i,t}^{dg}-\underset{l=1}{\overset{n_l}{\mathrm{\Σ}}}{P}_{l,t}^{echp}-\underset{x=1}{\overset{n_x}{\mathrm{\Σ}}}{P}_{x,t}^{esd}$

Include risk cost in object function, be the profit after VPP meter and risk.

For calculating VPP risk cost, Monte-carlo Simulation EM electricity price, SRM electricity price, wind power output, photovoltaic is used to go out Power, electric load and thermic load situation.The scene produced due to each Monte Carlo simulation is different, and corresponding mistake loading is the most not Identical, choose arbitrary scene institute gain and loss loading the most unreasonable to characterize system mistake loading.Therefore, expected value is used hereinRepresenting that t period VPP loses loading, gained expression formula is as follows:

$E\left(P_t^{ens}\right)=\underset{d=1}{\overset{n_d}{\Σ}}\left(\frac{1}{n_d}{P}_{d,t}^{ens}\right)$

In formula: n_dFor scene number；Loading is lost for t period d scene.

Beneficial effect: the invention have the advantages that and technique effect:

(1) the invention provides a kind of virtual plant combined heat and power Scheduling Optimization Model, solve polymerization cogeneration of heat and power machine The virtual plant of group participates in modeling problem in the case of energy and spinning reserve market at the same time；

(2) use robust optimization to process energy market electricity price, spinning reserve market electricity price, wind power output, photovoltaic are exerted oneself, electricity Load and the uncertainty of thermic load, effectively reduce the risk that virtual plant faces, and increase the profit of virtual plant.

(3) set up risk quantification index, the robustness of balance RO model and economy, thus provide effectively ginseng for policymaker Examine.

Accompanying drawing explanation

Fig. 1 is EM electricity price schematic diagram；

Fig. 2 is SRM electricity price schematic diagram；

Fig. 3 is wind power output schematic diagram；

Fig. 4 is that photovoltaic is exerted oneself schematic diagram；

Fig. 5 is electric load schematic diagram data；

Fig. 6 is thermic load schematic diagram data；

What Fig. 7 was robust coefficient on VPP profit and risk cost affects schematic diagram.

Detailed description of the invention

Below in conjunction with the accompanying drawings and specific embodiment, it is further elucidated with the present invention, it should be understood that these embodiments are merely to illustrate The present invention rather than limit the scope of the present invention, after having read the present invention, each to the present invention of those skilled in the art The amendment planting the equivalent form of value all falls within the application claims limited range.

A kind of virtual plant combined heat and power scheduling Robust Optimization Model, comprises the following steps:

Step 1: set up under EM and SRM combined heat and power Scheduling Optimization Model before VPP day；

Step 2: use RO to process EM electricity price, SRM electricity price, wind power output, photovoltaic are exerted oneself, electric load and thermic load the most true Qualitative, set up RO model；

Step 3: set up risk quantification index, weighs the risk of systems face.

Further, step 1 comprises the following steps:

For VPP network operator, its target is that net profit maximizes, and therefore the object function of Optimized model is as follows:

$max\underset{t=1}{\overset{T}{\Σ}}(C_t^m-C_t^{dg}-C_t^{chp}-C_t^{boil}-C_t^s-C_t^{dr})$

Object function comprises six parts, and respectively VPP participates in electricity market profit gainedDG costCHP machine Composition is originallyBoiler costEnergy storage costDR costEvery part expression is as follows:

$C_t^m={\mathrm{\λ}}_t^{em}{P}_t^{em}+{\mathrm{\λ}}_t^{srm}{R}_t^{srm}$

In formula:It is respectively t period EM electricity price and SRM electricity price；It is respectively t period VPP at EM With the competitive bidding amount of SRM, for decision variable.

DG cost includes operating cost, start-up and shut-down costs and Environmental costs:

$C_t^{dg}=\underset{i=1}{\overset{n_i}{\mathrm{\Σ}}}\left(\begin{array}{c}\left(k_i{P}_{i,t}^{dg}+k_i^f{\mathrm{\μ}}_{i,t}^o\right)+\left({\mathrm{\λ}}_i^{su}{\mathrm{\μ}}_{i,t}^{su}+{\mathrm{\λ}}_i^{sd}{\mathrm{\μ}}_{i,t}^{sd}\right)+\\ P_{i,t}^{dg}\underset{e=1}{\overset{n_e}{\mathrm{\Σ}}}{Q}_{i,t}^{dg}\left(V_e+Y_e\right)\end{array}\right)$

In formula: n_iFor DG unit number；For t period i-th cell DG output, for decision variable；k_i、 Fuel cost, fixed cost, start-up cost and stopping cost for i-th cell DG；Boolean variable Table respectively Show whether t period i-th cell DG works, starts, stops, be to put 1, otherwise set to 0；n_eFor the pollutant sum considered；For I-th cell DG e item pollutant discharge amount；V_e、Y_eIt is respectively e item pollutant environmental value, the fine order of magnitude.

CHP unit cost includes that fuel cost, start-up and shut-down costs and Environmental costs, boiler cost include fuel cost and environment Cost, the CHP unit considered herein and boiler are all with natural gas as fuel, such as following formula:

$C_t^{chp}=\underset{l=1}{\overset{n_l}{\mathrm{\Σ}}}\left(\begin{array}{c}\frac{860{\mathrm{\λ}}_t^{ng}{P}_{l,t}^{echp}}{H^{ng}{\mathrm{\η}}_l^{chp}}+({\mathrm{\λ}}_l^{su}{\mathrm{\μ}}_{l,t}^{su}+{\mathrm{\λ}}_l^{sd}{\mathrm{\μ}}_{l,t}^{sd})+\\ \frac{P_{l,t}^{echp}}{{\mathrm{\η}}_l^{chp}}\underset{e=1}{\overset{n_e}{\Σ}}{Q}_{l,e}^{chp}(V_e+Y_e)\end{array}\right)$

$C_t^{boil}=\underset{k=1}{\overset{n_k}{\Σ}}(\frac{860{\mathrm{\λ}}_t^{ng}{P}_{k,t}^{boil}}{H^{ng}{\mathrm{\η}}_k^{boil}}+\frac{P_{k,t}^{boil}}{{\mathrm{\η}}_k^{boil}}\underset{e=1}{\overset{n_e}{\Σ}}{Q}_{k,e}^{boil}(V_e+Y_e\left)\right)$

In formula: n_l、n_kIt is respectively CHP unit and boiler unit number；For t period Gas Prices；H^ngFor natural gas heat Value；860 are converted into the Conversion of measurement unit ratio of kcal for kWh；For t period l unit CHP unit output power and Kth unit boiler heat supplying power, for decision variable；It is l unit CHP unit and kth unit boiler efficiency, needs It is noted thatIt is only the power supplying efficiency of CHP unit, not whole efficiency； It is that l unit CHP starts and stops Cost；Boolean variableRepresent whether t period l unit CHP starts, stops, and is to put 1, otherwise sets to 0 respectively；It is l unit CHP unit and kth unit boiler e item pollutant discharge amount, V_e、Y_eIt is respectively e item to pollute Substance environment is worth, the fine order of magnitude；

Energy storage cost includes ES cost and TS cost, and it linearly closes with charge and discharge power and storage, heat release power approximation System:

$C_t^s=\underset{x=1}{\overset{n_x}{\Σ}}\left(a_x^{es}\right(P_{x,t}^{esc}+P_{x,t}^{esd})+b_x^{es})+\underset{z=1}{\overset{n_z}{\Σ}}\left(a_z^{ts}\right(P_{z,t}^{tsc}+P_{z,t}^{tsd})+b_z^{ts})$

In formula: n_x、n_zIt is respectively ES and TS unit number；For t period xth unit ES charge and discharge power, for certainly Plan variable；For t period z unit TS storage, heat release power, for decision variable； For ES and TS Cost coefficient.

DR cost statement is shown as when VPP interrupts customer charge, need to pay certain compensation.In view of different interruption degree pair The influence degree of user is different, will interrupt making up price and load rejection horizontal hook, and interruption level is the highest, and making up price is more Height, is specifically expressed as follows:

$C_t^{dr}=\underset{m=1}{\overset{n_m}{\Σ}}\left({\mathrm{\λ}}_m^{curt}{P}_{m,t}^{elcurt}\right)$

In formula: n_mFor interrupting horizontal progression；It is that m level interrupts level compensating price；Interrupt for t period m level Horizontal break load, for decision variable.

It is as follows that VPP operationally needs to meet constraints:

1) DG constraint.

$P_{i,t}^{dg}\≥P_i^{\min }{\mathrm{\μ}}_{i,t}^o$

$P_{i,t}^{dg}+R_{i,t}^{dg}\≤P_i^{\max }{\mathrm{\μ}}_{i,t}^o$

$R_{i,t}^{dg}\≤r_i^u{t}^r$

$-r_i^d\≤P_{i,t}^{dg}-P_{i,t-1}^{dg}\≤r_i^u$

${\mathrm{\μ}}_{i,t}^o-{\mathrm{\μ}}_{i,t-1}^o\≤{\mathrm{\μ}}_{i,t}^{su}$

${\mathrm{\μ}}_{i,t-1}^o-{\mathrm{\μ}}_{i,t}^o\≤{\mathrm{\μ}}_{i,t}^{sd}$

In formula:It is respectively i-th cell DG maximum, minimum output power；Standby for t period i-th cell DG With capacity, for decision variable；r_i ^u、r_i ^dBe respectively i-th cell DG upwards, climbing rate downwards；t^rFor the active service time.

2) CHP Unit commitment.

$P_{l,t}^{tchp}=k_l^{chp}{P}_{l,t}^{echp}$

$P_{l,t}^{echp}\≥P_l^{\min }{\mathrm{\μ}}_{l,t}^o$

$P_{l,t}^{echp}+R_{l,t}^{echp}\≤P_l^{\max }{\mathrm{\μ}}_{l,t}^o$

$R_{l,t}^{echp}\≤r_l^u{t}^r$

$-r_l^d\≤P_{l,t}^{echp}-P_{l,t-1}^{echp}\≤r_l^u$

${\mathrm{\μ}}_{l,t}^o-{\mathrm{\μ}}_{l,t-1}^o\≤{\mathrm{\μ}}_{l,t}^{su}$

${\mathrm{\μ}}_{l,t-1}^o-{\mathrm{\μ}}_{l,t}^o\≤{\mathrm{\μ}}_{l,t}^{sd}$

In formula:For t period l unit CHP unit heating power；It is l unit CHP unit hotspot stress, with CHP machine unit characteristic is relevant；P_l ^max、P_l ^minIt is respectively l unit CHP unit output power maximum, minimum；For t period l Unit CHP unit reserve capacity, for decision variable；Be respectively l unit CHP unit upwards, climbing rate downwards；Boolean VariableRepresent whether t period l unit CHP unit runs, and is to put 1, otherwise sets to 0.

3) boiler constraint.

$0\≤P_{k,t}^{boil}\≤P_k^{max}$

In formula:For kth unit boiler maximum heating power.

4) ES and TS constraint.

$0\≤P_{x,t}^{esc}\≤P_x^{cmax}$

$0\≤P_{x,t}^{\mathrm{esd}}\≤P_x^{\mathrm{dm}ax}$

$S_x^{\min }\≤S_{x,t}^{es}\≤S_x^{\max }$

$S_{x,t}^{es}=S_{x,t-1}^{es}+{\mathrm{\η}}_x^{esc}{P}_{x,t}^{esc}-\frac{P_{x,t}^{esd}}{{\mathrm{\η}}_x^{esd}}$

$S_{x,0}^{es}=S_x^{esi}$

$S_{x,24}^{es}=S_x^{esf}$

$0\≤P_{z,t}^{tsc}\≤P_z^{cmax}$

$0\≤P_{z,t}^{tsd}\≤P_z^{dmax}$

$S_z^{\min }\≤S_{z,t}^{ts}\≤S_z^{\max }$

$S_{z,t}^{ts}=S_{z,t-1}^{ts}+{\mathrm{\η}}_z^{tsc}{P}_{z,t}^{tsc}-\frac{P_{z,t}^{tsd}}{{\mathrm{\η}}_z^{tsd}}$

$S_{z,0}^{ts}=S_z^{tsi}$

$S_{z,24}^{ts}=S_z^{tsf}$

In formula:It is respectively xth unit ES maximum charge and discharge power；For t period xth unit ES storage electricity Amount；It is respectively xth unit ES reserve of electricity upper and lower limit；It is respectively xth unit ES charge and discharge efficiency；It is respectively xth unit ES beginning, end reserve of electricity；It is respectively the storage of z unit TS maximum, heat release power；For t period z unit TS quantity of heat storage；It is respectively z unit TS quantity of heat storage upper and lower limit；Respectively It is z unit TS storage, exothermal efficiency；It is respectively z unit TS beginning, end quantity of heat storage.

5) DR constraint.

$0\≤P_{m,t}^{elcurt}\≤k_m^{curt}{P}_t^{el}$

$P_t^{elcurt}=\underset{m=1}{\overset{n_m}{\Σ}}{P}_{m,t}^{elcurt}$

$P_t^{elcurt}+R_t^{el}\≤\underset{m=1}{\overset{n_m}{\Σ}}\left(k_m^{curt}{P}_t^{el}\right)$

In formula:It is that m level interrupts horizontal break coefficient；P_t ^elFor t period electric load；P_t ^elcurtInterrupt negative for the t period Lotus；For t period reserve capacity for load variation in power, for decision variable.

6) electricity, heating power balance constraint.

$\underset{w=1}{\overset{n_w}{\mathrm{\Σ}}}{P}_{w,t}^{wp}+\underset{s=1}{\overset{n_s}{\mathrm{\Σ}}}{P}_{s,t}^{pv}+\underset{i=1}{\overset{n_i}{\mathrm{\Σ}}}{P}_{i,t}^{dg}+\underset{l=1}{\overset{n_l}{\mathrm{\Σ}}}{P}_{l,t}^{echp}+\underset{x=1}{\overset{n_x}{\mathrm{\Σ}}}{P}_{x,t}^{esd}=P_t^{em}+P_t^{el}-P_t^{elcurt}+\underset{x=1}{\overset{n_x}{\mathrm{\Σ}}}{P}_{x,t}^{esc}$

$\underset{l=1}{\overset{n_l}{\Σ}}{P}_{l,t}^{tchp}+\underset{k=1}{\overset{n_k}{\Σ}}{P}_{k,t}^{boil}+\underset{z=1}{\overset{n_z}{\Σ}}{P}_{z,t}^{tsd}\≥P_t^{tl}+\underset{z=1}{\overset{n_z}{\Σ}}{P}_{z,t}^{tsc}$

In formula: n_w、n_sIt is respectively Wind turbines, photovoltaic unit number；It is respectively t period Wind turbines w, photovoltaic Unit s exerts oneself；P_t ^tlFor t period thermic load.

7) competitive bidding spare capacity retrains.

$R_t^{srm}=\underset{i}{\overset{n_i}{\mathrm{\Σ}}}{R}_{i,t}^{dg}+\underset{l}{\overset{n_l}{\mathrm{\Σ}}}{R}_{l,t}^{echp}+R_t^{el}$

8) VPP system reserve constraint.

In VPP, DG, CHP unit, interruptible load all can provide system reserve, but do not include standby to SRM competitive bidding Capacity:

$\begin{array}{c}\underset{i=1}{\overset{n_i}{\mathrm{\Σ}}}({P_i}^{\max }-P_{i,t}^{\mathrm{dg}}-R_{i,t}^{\mathrm{dg}})+\underset{l=1}{\overset{n_l}{\mathrm{\Σ}}}({P_l}^{\max }-P_{l,t}^{\mathrm{echp}}-R_{l,t}^{\mathrm{echp}})+\\ (\underset{m=1}{\overset{n_m}{\mathrm{\Σ}}}\left(k_m^{\mathrm{curt}}{P_t}^{\mathrm{el}}\right)-{P_t}^{\mathrm{elcurt}}-R_t^{\mathrm{el}})\≥R_t^a\end{array}$

In formula:For spare capacity needed for t period VPP.

Further, step 2 comprises the following steps:

EM electricity price in above-mentioned model, SRM electricity price, wind power output, photovoltaic are exerted oneself, electric load and thermic load all exist uncertain Property, use RO to process such uncertain problem, with a herein_ijRepresent above-mentioned uncertain parameter, it is considered to a_ijForm is as follows:

$a_{ij}\∈\[{\tilde{a}}_{ij}-{\hat{a}}_{ij},{\tilde{a}}_{ij}+{\hat{a}}_{ij}\]$

Wherein:

${\tilde{a}}_{ij}=\frac{1}{2}({\stackrel{\‾}{a}}_{ij}+{\underset{\‾}{a}}_{ij})$

${\hat{a}}_{ij}=\frac{1}{2}({\stackrel{\‾}{a}}_{ij}-{\underset{\‾}{a}}_{ij})$

In formula: a _ijIt is respectively the upper and lower limit of uncertain parameter,I.e. think that uncertain parameter is at it Fluctuation in upper and lower limit interval range.

Above-mentioned uncertain parameter form considers the situation that uncertain parameter is the worst, and the decision-making thus done has the strongest Conservative, but lose economy.Introduce robust coefficient Г, Γ ∈ [0, | J |] for this, wherein, J is the collection of all uncertain parameter Close, now, uncertain parameter a_ijInterval beWhen Γ=0, do not consider the most true of uncertain parameter Qualitative, this model is consistent with deterministic optimization model, and system robustness is poor.Along with the continuous increase of Γ, system robustness by Gradually improving, economy constantly declines.As Γ=| J |, it is the most conservative form.By regulation robust coefficient Г, the most available The optimal solution of different conservative, takes into account robustness and the economy of decision scheme.

Use RO to process EM electricity price, SRM electricity price, wind power output, photovoltaic are exerted oneself, electric load and the uncertainty of thermic load, Gained object function and constraints are as follows:

$\max \underset{t=1}{\overset{T}{\mathrm{\Σ}}}\left(\frac{1}{2}\right({\stackrel{\‾}{\mathrm{\λ}}}_t^{em}+{\underset{\‾}{\mathrm{\λ}}}_t^{em})P_t^{em}+\frac{1}{2}({\stackrel{\‾}{\mathrm{\λ}}}_t^{srm}+{\underset{\‾}{\mathrm{\λ}}}_t^{srm})R_t^{srm}-C_t^{dg}-C_t^{chp}-C_t^{boil}-C_t^s-C_t^{dr})-{\mathrm{\Γ}}^m{\mathrm{\υ}}^m-\underset{t=1}{\overset{T}{\Σ}}(q_t^{em}+q_t^{srm})$

${\mathrm{\υ}}^m+q_t^{em}\≥\frac{1}{2}\left({\stackrel{\‾}{\mathrm{\λ}}}_t^{em}-{\underset{\‾}{\mathrm{\λ}}}_t^{em}\right)y_t^{em}$

${\mathrm{\υ}}^m+q_t^{srm}\≥\frac{1}{2}\left({\stackrel{\‾}{\mathrm{\λ}}}_t^{srm}-{\underset{\‾}{\mathrm{\λ}}}_t^{srm}\right)y_t^{srm}$

$-y_t^{em}\≤P_t^{em}\≤y_t^{em}$

$-y_t^{srm}\≤P_t^{srm}\≤y_t^{srm}$

$P_{w,t}^{wp}+{\mathrm{\Γ}}_{w,t}^{wp}{\mathrm{\υ}}_{w,t}^{wp}+q_{w,t}^{wp}\≤\frac{1}{2}\left({\stackrel{\‾}{P}}_{w,t}^{wp}+{\underset{\‾}{P}}_{w,t}^{wp}\right)$

$P_{s,t}^{pv}+{\mathrm{\Γ}}_{s,t}^{pv}{\mathrm{\υ}}_{s,t}^{pv}+q_{s,t}^{pv}\≤\frac{1}{2}\left({\stackrel{\‾}{P}}_{s,t}^{pv}+{\underset{\‾}{P}}_{s,t}^{pv}\right)$

$P_t^{el}-{\mathrm{\Γ}}_t^{el}{\mathrm{\υ}}_t^{el}-q_t^{el}\≥\frac{1}{2}\left({\stackrel{\‾}{P}}_t^{el}+{\underset{\‾}{P}}_t^{el}\right)$

$P_t^{tl}-{\mathrm{\Γ}}_t^{tl}{\mathrm{\υ}}_t^{tl}-q_t^{tl}\≥\frac{1}{2}\left({\stackrel{\‾}{P}}_t^{tl}+{\underset{\‾}{P}}_t^{tl}\right)$

${\mathrm{\υ}}_{w,t}^{wp}+q_{w,t}^{wp}\≥\frac{1}{2}\left({\stackrel{\‾}{P}}_{w,t}^{wp}-{\underset{\‾}{P}}_{w,t}^{wp}\right)y_{w,t}^{wp}$

${\mathrm{\υ}}_{s,t}^{pv}+q_{s,t}^{pv}\≥\frac{1}{2}\left({\stackrel{\‾}{P}}_{s,t}^{pv}-{\underset{\‾}{P}}_{s,t}^{pv}\right)y_{s,t}^{pv}$

${\mathrm{\υ}}_t^{el}+q_t^{el}\≥\frac{1}{2}\left({\stackrel{\‾}{P}}_t^{el}-{\underset{\‾}{P}}_t^{el}\right)y_t^{el}$

${\mathrm{\υ}}_t^{tl}+q_t^{tl}\≥\frac{1}{2}\left({\stackrel{\‾}{P}}_t^{tl}-{\underset{\‾}{P}}_t^{tl}\right)y_t^{tl}$

$y_{w,t}^{wp},y_{s,t}^{pv},y_t^{el},y_t^{tl}\≥1$

${\mathrm{\υ}}^m,q_t^{em},y_t^{em},q_t^{srm},y_t^{srm},{\mathrm{\υ}}_{w,t}^{wp},q_{w,t}^{wp},y_{w,t}^{wp},$

${\mathrm{\υ}}_{s,t}^{pv},q_{s,t}^{pv},y_{s,t}^{pv},{\mathrm{\υ}}_t^{el},q_t^{el},y_t^{el},{\mathrm{\υ}}_t^{tl},q_t^{tl},y_t^{tl}\≥0$

In formula: P _t ^el、 P _t ^tlRespectively Exert oneself for EM electricity price, SRM electricity price, wind power output, photovoltaic, electric load and thermic load upper and lower limit；Γ^m、 Be respectively electricity price, wind power output, photovoltaic are exerted oneself, electric load and thermic load robust coefficient, υ^m、 Auxiliary for introduce Help variable.

Further, step 3 comprises the following steps:

As robust coefficient Γ ＜ | J |, all uncertain parameter fluctuation situation cannot be included in uncertain parameter interval, unavoidably Can bring certain risk to system, robust coefficient is the least, and risk is the biggest.Therefore, set up rational quantification of targets risk level, Robustness and the economy of VPP system just can be better balanced.The metric of system risk is general and loses loading, lose load Persistent period etc. are linked up with, and loading, the risk cost of its correspondence are lost in main considerationExpression formula is as follows:

$C_t^{ens}={\mathrm{\λ}}_t^{ens}{P}_t^{ens}$

In formula:Lose load fine for the t period, when VPP cannot feed system internal loading, need to force excise customer charge Time, heavy fine to be paid for this, thereforeNumerical value the biggest.Loading is lost, when supply electricity in VPP for the t period When amount is more than demand electricity,On the contrary, if VPP delivery is insufficient for load and electric power market demand, then

$P_t^{ens}=P_t^{el}-P_t^{elcurt}+P_t^{em}+\underset{x=1}{\overset{n_x}{\mathrm{\Σ}}}{P}_{x,t}^{esc}-\underset{w=1}{\overset{n_w}{\mathrm{\Σ}}}{P}_{w,t}^{wp}-\underset{s=1}{\overset{n_s}{\mathrm{\Σ}}}{P}_{s,t}^{pv}-\underset{i=1}{\overset{n_i}{\mathrm{\Σ}}}{P}_{i,t}^{dg}-\underset{l=1}{\overset{n_l}{\mathrm{\Σ}}}{P}_{l,t}^{echp}-\underset{x=1}{\overset{n_x}{\mathrm{\Σ}}}{P}_{x,t}^{esd}$

Include risk cost in object function, be the profit after VPP meter and risk.

For calculating VPP risk cost, Monte-carlo Simulation EM electricity price, SRM electricity price, wind power output, photovoltaic is used to go out Power, electric load and thermic load situation.The scene produced due to each Monte Carlo simulation is different, and corresponding mistake loading is the most not Identical, choose arbitrary scene institute gain and loss loading the most unreasonable to characterize system mistake loading.Therefore, expected value is used hereinRepresenting that t period VPP loses loading, gained expression formula is as follows:

$E\left(P_t^{ens}\right)=\underset{d=1}{\overset{n_d}{\Σ}}\left(\frac{1}{n_d}{P}_{d,t}^{ens}\right)$

In formula: n_dFor scene number；Loading is lost for t period d scene.

The present invention is introduced below as a example by a VPP:

This VPP includes 4 DG (including 2 miniature gas turbines, 2 fuel cells), 3 Wind turbines, 2 photovoltaic machines Electric load and thermic load in group, 5 CHP units, 1 boiler, 5 ES unit, 5 TS unit, regions.VPP dispatching cycle is 1 My god, it was divided into for 24 periods.EM electricity price, SRM electricity price, wind power output, photovoltaic are exerted oneself, electric load and thermic load upper and lower limit such as Fig. 1-6 institute Show.The SRM active service time is 10min, and losing load fine is 1000 $/MW.

1) VPP behavior is to profit effect

The SRM behavior shadow to profit is participated in for weighing CHP unit in VPP participates in EM and SRM, polymerization CHP unit and VPP Ringing, as a example by deterministic optimization model, build following 5 kinds of different schemes, scheme is arranged and acquired results is as shown in table 1.

The VPP constructing plan that 15 kinds of table is different

From table 1 scheme 1,2,5 contrast it can be seen that when VPP is only involved in EM or SRM profit respectively less than simultaneously participate in EM and SRM sight, wherein, is only involved in profit during SRM lower.For scheme 3, it should be noted that for comparing cogeneration and heat Electricity point is for impact on VPP profit, respectively with the DG identical with CHP unit peak power output and boiler instead CHP unit Powering and for thermal output, result shows, realizing cogeneration with VPP form polymerization CHP unit can increase VPP profit.With scheme 5 Comparing, scheme 4 does not considers that CHP unit participates in SRM sight, reduces the motility of decision-making, thus profit reduces.In sum, In VPP participates in EM and SRM, polymerization CHP unit, VPP, the participation SRM behavior of CHP unit all can be effectively increased VPP profit.

What in VPP, probabilistic essence was that it comprises all kinds of probabilistic collects, and wherein comprises probabilistic folded Add or offset, using RO to process single class uncertainty acquired results the most unreasonable.Therefore, consider herein using RO process In the case of above-mentioned all uncertainties, when disregarding risk after VPP profit, meter and risk VPP profit and risk cost with robust The situation of change of coefficient, acquired results is as shown in Figure 7.It can be seen that VPP profit and risk cost are with robust system when disregarding risk Number increase be gradually lowered, and count and risk after VPP profit with the increase of robust coefficient present first rise after downward trend, and Maximum is reached when robust coefficient is 30%.This is owing to the increase of robust coefficient improves the conservative of decision-making, thus drops Low economy, when therefore disregarding risk, VPP profit reduces.But, the increase of robust coefficient also improves the robustness of system, Robust coefficient is the biggest, and system robustness is the strongest, loses load risk the lowest, shows as risk cost and reduces, VPP profit after meter and risk Profit increases.But, when robust coefficient is excessive, the reduction of VPP risk cost is not enough to make up the loss that conservative strategy is brought, After meter and risk, VPP profit still can reduce.When robust coefficient reaches 100%, uncertain parameter interval is included all uncertain Parameter fluctuation situation, now system risk cost is 0, but its decision-making is overly conservative, therefore meter and risk after profit minimum.This Outward, in figure, robust coefficient is 0 i.e. deterministic optimization model result, and when illustrating to consider VPP risk, RO model can improve VPP profit.

Above simulation results show institute of the present invention structure model validation and practicality, illustrate that VPP participates in EM and SRM, poly- In closing CHP unit and VPP, the participation SRM behavior of CHP unit all can be effectively increased VPP profit.Further, RO model improves system Robustness, reduces system risk, thus adds VPP profit, and risk cost features the risk that VPP faces well, Reduce the blindness of unascertained decision, thus choose suitable robust coefficient for policymaker and effective reference is provided.

Claims (4)

1. a virtual plant combined heat and power scheduling Robust Optimization Model, it is characterised in that: comprise the following steps:

Step 1: set up under EM and SRM combined heat and power Scheduling Optimization Model before VPP day；

Step 2: use RO to process EM electricity price, SRM electricity price, wind power output, photovoltaic are exerted oneself, electric load and thermic load uncertain Property, set up RO model；

Step 3: set up risk quantification index, weighs the risk of systems face.

Virtual plant combined heat and power the most according to claim 1 scheduling Robust Optimization Model, it is characterised in that: described step 1 comprises the following steps:

For VPP network operator, its target is that net profit maximizes, and therefore the object function of Optimized model is as follows:

$max\underset{t=1}{\overset{T}{\Σ}}(C_t^m-C_t^{dg}-C_t^{chp}-C_t^{boil}-C_t^s-C_t^{dr})$

Object function comprises six parts, and respectively VPP participates in electricity market profit gainedDG costCHP machine forms ThisBoiler costEnergy storage costDR costEvery part expression is as follows:

$C_t^m={\mathrm{\λ}}_t^{em}{P}_t^{em}+{\mathrm{\λ}}_t^{srm}{R}_t^{srm}$

In formula:It is respectively t period EM electricity price and SRM electricity price；It is respectively t period VPP at EM and SRM Competitive bidding amount, for decision variable；

DG cost includes operating cost, start-up and shut-down costs and Environmental costs:

$C_t^{dg}=\underset{i=1}{\overset{n_i}{\mathrm{\Σ}}}\left(\begin{array}{c}\left(k_i{P}_{i,t}^{dg}+k_i^f{\mathrm{\μ}}_{i,t}^o\right)+\left({\mathrm{\λ}}_i^{su}{\mathrm{\μ}}_{i,t}^{su}+{\mathrm{\λ}}_i^{sd}{\mathrm{\μ}}_{i,t}^{sd}\right)+\\ P_{i,t}^{dg}\underset{e=1}{\overset{n_e}{\mathrm{\Σ}}}{Q}_{i,t}^{dg}\left(V_e+Y_e\right)\end{array}\right)$

In formula: n_iFor DG unit number；For t period i-th cell DG output, for decision variable；k_i、 It is The fuel cost of i cells D G, fixed cost, start-up cost and stopping cost；Boolean variable When representing t respectively Whether section i-th cell DG works, starts, stops, and is to put 1, otherwise sets to 0；n_eFor the pollutant sum considered；It it is the i-th list Unit's DG e item pollutant discharge amount；V_e、Y_eIt is respectively e item pollutant environmental value, the fine order of magnitude；

CHP unit cost includes that fuel cost, start-up and shut-down costs and Environmental costs, boiler cost include that fuel cost becomes with environment This, the CHP unit considered herein and boiler are all with natural gas as fuel, such as following formula:

$C_t^{chp}=\underset{l=1}{\overset{n_l}{\mathrm{\Σ}}}\left(\begin{array}{c}\frac{860{\mathrm{\λ}}_t^{ng}{P}_{l,t}^{echp}}{H^{ng}{\mathrm{\η}}_l^{chp}}+({\mathrm{\λ}}_l^{su}{\mathrm{\μ}}_{l,t}^{su}+{\mathrm{\λ}}_l^{sd}{\mathrm{\μ}}_{l,t}^{sd})+\\ \frac{P_{l,t}^{echp}}{{\mathrm{\η}}_l^{chp}}\underset{e=1}{\overset{n_e}{\Σ}}{Q}_{l,e}^{chp}(V_e+Y_e)\end{array}\right)$

$C_t^{boil}=\underset{k=1}{\overset{n_k}{\Σ}}(\frac{860{\mathrm{\λ}}_t^{ng}{P}_{k,t}^{boil}}{H^{ng}{\mathrm{\η}}_k^{boil}}+\frac{P_{k,t}^{boil}}{{\mathrm{\η}}_k^{boil}}\underset{e=1}{\overset{n_e}{\Σ}}{Q}_{k,e}^{boil}(V_e+Y_e\left)\right)$

In formula: n_l、n_kIt is respectively CHP unit and boiler unit number；For t period Gas Prices；H^ngFor heating value of natural gas； 860 are converted into the Conversion of measurement unit ratio of kcal for kWh；For t period l unit CHP unit output power and kth Unit boiler heat supplying power, for decision variable；It is l unit CHP unit and kth unit boiler efficiency, needs It is bright,It is only the power supplying efficiency of CHP unit, not whole efficiency； It is that l unit CHP starts and stops into This；Boolean variableRepresent whether t period l unit CHP starts, stops, and is to put 1, otherwise sets to 0 respectively；It is l unit CHP unit and kth unit boiler e item pollutant discharge amount, V_e、Y_eIt is respectively e item to pollute Substance environment is worth, the fine order of magnitude；

Energy storage cost includes ES cost and TS cost, and it approximates linear with charge and discharge power and storage, heat release power:

$C_t^s=\underset{x=1}{\overset{n_x}{\Σ}}\left(a_x^{es}\right(P_{x,t}^{esc}+P_{x,t}^{esd})+b_x^{es})+\underset{z=1}{\overset{n_z}{\Σ}}\left(a_z^{ts}\right(P_{z,t}^{tsc}+P_{z,t}^{tsd})+b_z^{ts})$

In formula: n_x、n_zIt is respectively ES and TS unit number；For t period xth unit ES charge and discharge power, become for decision-making Amount；For t period z unit TS storage, heat release power, for decision variable； For becoming of ES and TS This coefficient；

DR cost statement is shown as when VPP interrupts customer charge, need to pay certain compensation, it is contemplated that different interruption degree are to user Influence degree different, making up price and load rejection horizontal hook will be interrupted, it is the highest to interrupt level, and making up price is the highest, tool Body is expressed as follows:

$C_t^{dr}=\underset{m=1}{\overset{n_m}{\Σ}}\left({\mathrm{\λ}}_m^{curt}{P}_{m,t}^{elcurt}\right)$

In formula: n_mFor interrupting horizontal progression；It is that m level interrupts level compensating price；Level is interrupted for t period m level Interruptible load, for decision variable；

It is as follows that VPP operationally needs to meet constraints:

1) DG constraint:

$P_{i,t}^{dg}\≥P_i^{\min }{\mathrm{\μ}}_{i,t}^o$

$R_{i,t}^{dg}+R_{i,t}^{dg}\≤P_i^{\max }{\mathrm{\μ}}_{i,t}^o$

$R_{i,t}^{dg}\≤r_i^u{t}^r$

$-r_i^d\≤P_{i,t}^{dg}-P_{i,t-1}^{dg}\≤r_i^u$

${\mathrm{\μ}}_{i,t}^o-{\mathrm{\μ}}_{i,t-1}^o\≤{\mathrm{\μ}}_{i,t}^{su}$

${\mathrm{\μ}}_{i,t-1}^o-{\mathrm{\μ}}_{i,t}^o\≤{\mathrm{\μ}}_{i,t}^{sd}$

In formula:It is respectively i-th cell DG maximum, minimum output power；For the t standby appearance of period i-th cell DG Amount, for decision variable；Be respectively i-th cell DG upwards, climbing rate downwards；t^rFor the active service time；

2) CHP Unit commitment:

$P_{l,t}^{tchp}=k_l^{chp}{P}_{l,t}^{echp}$

$P_{l,t}^{echp}\≥P_l^{\min }{\mathrm{\μ}}_{l,t}^o$

$P_{l,t}^{echp}+R_{l,t}^{echp}\≤P_l^{\max }{\mathrm{\μ}}_{l,t}^o$

$R_{l,t}^{echp}\≤r_l^u{t}^r$

$-r_l^d\≤P_{l,t}^{echp}-P_{l,t-1}^{echp}\≤r_l^u$

${\mathrm{\μ}}_{l,t}^o-{\mathrm{\μ}}_{l,t-1}^o\≤{\mathrm{\μ}}_{l,t}^{su}$

${\mathrm{\μ}}_{l,t-1}^o-{\mathrm{\μ}}_{l,t}^o\≤{\mathrm{\μ}}_{l,t}^{sd}$

In formula:For t period l unit CHP unit heating power；It is l unit CHP unit hotspot stress, with CHP unit Characteristic is relevant；It is respectively l unit CHP unit output power maximum, minimum；For t period l unit CHP Unit reserve capacity, for decision variable；Be respectively l unit CHP unit upwards, climbing rate downwards；Boolean variable Represent whether t period l unit CHP unit runs, and is to put 1, otherwise sets to 0；

3) boiler constraint:

$0\≤P_{k,t}^{boil}\≤P_k^{max}$

In formula:For kth unit boiler maximum heating power.

4) ES and TS constraint:

$0\≤P_{x,t}^{esc}\≤P_x^{cmax}$

$0\≤P_{x,t}^{esd}\≤P_x^{dmax}$

$S_x^{\min }\≤S_{x,t}^{es}\≤S_x^{\max }$

$S_{x,t}^{es}=S_{x,t-1}^{es}+{\mathrm{\η}}_x^{esc}{P}_{x,t}^{esc}-\frac{P_{x,t}^{esd}}{{\mathrm{\η}}_x^{esd}}$

$S_{x,0}^{es}=S_x^{esi}$

$S_{x,24}^{es}=S_x^{esf}$

$0\≤P_{z,t}^{tsc}\≤P_z^{cmax}$

$0\≤P_{z,t}^{tsd}\≤P_z^{dmax}$

$S_z^{\min }\≤S_{z,t}^{ts}\≤S_z^{\max }$

$S_{z,t}^{ts}=S_{z,t-1}^{ts}+{\mathrm{\η}}_z^{tsc}{P}_{z,t}^{tsc}-\frac{P_{z,t}^{tsd}}{{\mathrm{\η}}_z^{tsd}}$

$S_{z,0}^{ts}=S_z^{tsi}$

$S_{z,24}^{ts}=S_z^{tsf}$

In formula:It is respectively xth unit ES maximum charge and discharge power；For t period xth unit ES reserve of electricity；It is respectively xth unit ES reserve of electricity upper and lower limit；It is respectively xth unit ES charge and discharge efficiency；It is respectively xth unit ES beginning, end reserve of electricity；It is respectively the storage of z unit TS maximum, heat release power；For t period z unit TS quantity of heat storage；It is respectively z unit TS quantity of heat storage upper and lower limit；Respectively It is z unit TS storage, exothermal efficiency；It is respectively z unit TS beginning, end quantity of heat storage；

5) DR constraint:

$0\≤P_{m,t}^{elcurt}\≤k_{mt}^{curt}{P}_t^{el}$

$P_t^{elcurt}=\underset{m=1}{\overset{n_m}{\Σ}}{P}_{m,t}^{elcurt}$

$P_t^{elcurt}+R_t^{el}\≤\underset{m=1}{\overset{n_m}{\Σ}}\left(k_{mt}^{curt}{P}_t^{el}\right)$

In formula:It is that m level interrupts horizontal break coefficient；For t period electric load；For t period interruptible load；For T period reserve capacity for load variation in power, for decision variable；

6) electricity, heating power balance constraint:

$\underset{w=1}{\overset{n_w}{\mathrm{\Σ}}}{P}_{w,t}^{wp}+\underset{s=1}{\overset{n_s}{\mathrm{\Σ}}}{P}_{s,t}^{pv}+\underset{i=1}{\overset{n_i}{\mathrm{\Σ}}}{P}_{i,t}^{dg}+\underset{l=1}{\overset{n_l}{\mathrm{\Σ}}}{P}_{l,t}^{echp}+\underset{x=1}{\overset{n_x}{\mathrm{\Σ}}}{P}_{x,t}^{esd}=P_t^{em}+P_t^{el}-P_t^{elcurt}+\underset{x=1}{\overset{n_x}{\mathrm{\Σ}}}{P}_{x,t}^{esc}$

$\underset{l=1}{\overset{n_l}{\Σ}}{P}_{l,t}^{tchp}+\underset{k=1}{\overset{n_k}{\Σ}}{P}_{k,t}^{boil}+\underset{z=1}{\overset{n_z}{\Σ}}{P}_{z,t}^{tsd}\≥P_t^{tl}+\underset{z=1}{\overset{n_z}{\Σ}}{P}_{z,t}^{tsc}$

In formula: n_w、n_sIt is respectively Wind turbines, photovoltaic unit number；It is respectively t period Wind turbines w, photovoltaic unit s Exert oneself；For t period thermic load；

7) competitive bidding spare capacity retrains:

$R_t^{srm}=\underset{i}{\overset{n_i}{\mathrm{\Σ}}}{R}_{i,t}^{dg}+\underset{l}{\overset{n_l}{\mathrm{\Σ}}}{R}_{l,t}^{echp}+R_t^{el}$

8) VPP system reserve constraint:

In VPP, DG, CHP unit, interruptible load all can provide system reserve, but do not include to the spare capacity of SRM competitive bidding:

$\begin{array}{c}\underset{i=1}{\overset{n_i}{\mathrm{\Σ}}}(P_i^{\max }-P_{i,t}^{dg}-R_{i,t}^{dg})+\underset{l=1}{\overset{n_l}{\mathrm{\Σ}}}(P_l^{\max }-P_{l,t}^{echp}-R_{l,t}^{echp})+\\ \left(\underset{m=1}{\overset{n_m}{\mathrm{\Σ}}}\right(k_m^{curt}{P}_t^{el})-P_t^{elcurt}-R_t^{el})\≥R_t^a\end{array}$

In formula:For spare capacity needed for t period VPP.

VPP combined heat and power the most according to claim 1 scheduling Robust Optimization Model, it is characterised in that: described step 2 includes Following steps:

EM electricity price in above-mentioned model, SRM electricity price, wind power output, photovoltaic are exerted oneself, electric load and thermic load all exist uncertainty, RO is used to process such uncertain problem, with a herein_ijRepresent above-mentioned uncertain parameter, it is considered to a_ijForm is as follows:

$a_{ij}\∈\[{\tilde{a}}_{ij}-{\hat{a}}_{ij},{\tilde{a}}_{ij}+{\hat{a}}_{ij}\]$

Wherein:

${\hat{a}}_{ij}=\frac{1}{2}({\stackrel{\‾}{a}}_{ij}-{\underset{\‾}{a}}_{ij})$

${\hat{a}}_{ij}=\frac{1}{2}({\stackrel{\‾}{a}}_{ij}-{\underset{\‾}{a}}_{ij})$

In formula: a _ijIt is respectively the upper and lower limit of uncertain parameter,I.e. think uncertain parameter thereon, under Fluctuation in limit interval range；

Above-mentioned uncertain parameter form considers the situation that uncertain parameter is the worst, and the decision-making thus done has the strongest guarding Property, but lose economy；Introduce robust coefficient Г, Γ ∈ [0, | J] for this, wherein, J is the set of all uncertain parameter, this Time, uncertain parameter a_ijInterval beWhen Γ=0, do not consider the uncertainty of uncertain parameter, This model is consistent with deterministic optimization model, and system robustness is poor；Along with the continuous increase of Γ, system robustness gradually carries Height, economy constantly declines.As Γ=| J |, it is the most conservative form；By regulation robust coefficient Г, i.e. available difference The optimal solution of conservative, takes into account robustness and the economy of decision scheme；

Use RO to process EM electricity price, SRM electricity price, wind power output, photovoltaic are exerted oneself, electric load and the uncertainty of thermic load, gained Object function and constraints are as follows:

$max\underset{t=1}{\overset{T}{\mathrm{\Σ}}}\left(\frac{1}{2}\right({\stackrel{\‾}{\mathrm{\λ}}}_t^{em}+{\underset{\‾}{\mathrm{\λ}}}_t^{em})P_t^{em}+\frac{1}{2}({\stackrel{\‾}{\mathrm{\λ}}}_t^{srm}+{\underset{\‾}{\mathrm{\λ}}}_t^{srm})R_t^{srm}-C_t^{dg}-C_t^{chp}-C_t^{boil}-C_t^s-C_t^{dr})-{\mathrm{\Γ}}^m{\mathrm{\υ}}^m-\underset{t=1}{\overset{T}{\Σ}}(q_t^{em}+q_t^{srm})$

${\mathrm{\υ}}^m+q_t^{em}\≥\frac{1}{2}\left({\stackrel{\‾}{\mathrm{\λ}}}_t^{em}-{\underset{\‾}{\mathrm{\λ}}}_t^{em}\right)y_t^{em}$

${\mathrm{\υ}}^m+q_t^{srm}\≥\frac{1}{2}\left({\stackrel{\‾}{\mathrm{\λ}}}_t^{srm}-{\underset{\‾}{\mathrm{\λ}}}_t^{srm}\right)y_t^{srm}$

$-y_t^{em}\≤P_t^{em}\≤y_t^{em}$

$-y_t^{srm}\≤P_t^{srm}\≤y_t^{srm}$

$P_{w,t}^{wp}+{\mathrm{\Γ}}_{w,t}^{wp}{\mathrm{\υ}}_{w,t}^{wp}+q_{w,t}^{wp}\≤\frac{1}{2}\left({\stackrel{\‾}{P}}_{w,t}^{wp}-{\underset{\‾}{P}}_{w,t}^{wp}\right)$

$P_{s,t}^{pv}+{\mathrm{\Γ}}_{s,t}^{pv}{\mathrm{\υ}}_{s,t}^{pv}+q_{s,t}^{pv}\≤\frac{1}{2}\left({\stackrel{\‾}{P}}_{s,t}^{pv}+{\underset{\‾}{P}}_{s,t}^{pv}\right)$

$P_t^{el}-{\mathrm{\Γ}}_t^{el}{\mathrm{\υ}}_t^{el}-q_t^{el}\≥\frac{1}{2}\left({\stackrel{\‾}{P}}_t^{el}-{\underset{\‾}{P}}_t^{el}\right)$

$P_t^{el}-{\mathrm{\Γ}}_t^{el}{\mathrm{\υ}}_t^{el}-q_t^{el}\≥\frac{1}{2}\left({\stackrel{\‾}{P}}_t^{el}-{\underset{\‾}{P}}_t^{el}\right)$

${\mathrm{\υ}}_{w,t}^{wp}+q_{w,t}^{wp}\≥\frac{1}{2}\left({\stackrel{\‾}{P}}_{w,t}^{wp}-{\underset{\‾}{P}}_{w,t}^{wp}\right)y_{w,t}^{wp}$

${\mathrm{\υ}}_{s,t}^{pv}+q_{s,t}^{pv}\≥\frac{1}{2}({\stackrel{\‾}{P}}_{s,t}^{pv}-{\underset{\‾}{P}}_{s,t}^{pv})y_{s,t}^{pv}$

${\mathrm{\υ}}_t^{el}+q_t^{el}\≥\frac{1}{2}({\stackrel{\‾}{P}}_t^{el}-{\underset{\‾}{P}}_t^{el})y_t^{el}$

${\mathrm{\υ}}_t^{tl}+q_t^{tl}\≥\frac{1}{2}({\stackrel{\‾}{P}}_t^{tl}-{\underset{\‾}{P}}_t^{tl})y_t^{tl}$

$y_{w,t}^{wp},y_{s,t}^{pv},y_t^{el},y_t^{tl}\≥1$

${\mathrm{\υ}}^m,q_t^{em},y_t^{em},q_t^{srm},y_t^{srm},{\mathrm{\υ}}_{w,t}^{wp},q_{w,t}^{wp},y_{w,t}^{wp},$

${\mathrm{\υ}}_{s,t}^{pv},q_{s,t}^{pv},y_{s,t}^{pv},{\mathrm{\υ}}_t^{el},q_t^{el},y_t^{el},{\mathrm{\υ}}_t^{tl},q_t^{tl},y_t^{tl}\≥0$

In formula:It is respectively EM Bound that the bound of electricity price, the bound of SRM electricity price, the bound of wind power output, photovoltaic are exerted oneself, electric load Bound and the bound of thermic load；Γ^m、Respectively electricity price, wind power output, photovoltaic goes out Power, electric load and thermic load robust coefficient, υ^m、 For the auxiliary variable introduced.

VPP combined heat and power the most according to claim 1 scheduling Robust Optimization Model, it is characterised in that: described step 3 includes Following steps:

As robust coefficient Γ ＜ | J |, all uncertain parameter fluctuation situation cannot be included in uncertain parameter interval, can give unavoidably System brings certain risk, and robust coefficient is the least, and risk is the biggest；Therefore, rational quantification of targets risk level, ability are set up Robustness and the economy of VPP system are better balanced；The metric of system risk is general and loses loading, lose based model for load duration Time etc. are linked up with, and loading, the risk cost of its correspondence are lost in main considerationExpression formula is as follows:

$C_t^{ens}={\mathrm{\λ}}_t^{ens}{P}_t^{ens}$

In formula:Lose load fine for the t period, when VPP cannot feed system internal loading, need to force excise customer charge time, Heavy fine is paid, therefore for thisNumerical value the biggest；Loading is lost for the t period, big when supplying electricity in VPP When demand electricity,On the contrary, if VPP delivery is insufficient for load and electric power market demand, then

$P_t^{ens}=P_t^{el}-P_t^{elcurt}+P_t^{em}+\underset{x=1}{\overset{n_x}{\mathrm{\Σ}}}{P}_{x,t}^{esc}-\underset{w=1}{\overset{n_w}{\mathrm{\Σ}}}{P}_{w,t}^{wp}-\underset{s=1}{\overset{n_s}{\mathrm{\Σ}}}{P}_{s,t}^{pv}-\underset{i=1}{\overset{n_i}{\mathrm{\Σ}}}{P}_{i,t}^{dg}-\underset{l=1}{\overset{n_l}{\mathrm{\Σ}}}{P}_{l,t}^{echp}-\underset{x=1}{\overset{n_x}{\mathrm{\Σ}}}{P}_{x,t}^{esd}$

Include risk cost in object function, be the profit after VPP meter and risk；

For calculating VPP risk cost, employing Monte-carlo Simulation EM electricity price, SRM electricity price, wind power output, photovoltaic are exerted oneself, electricity Load and thermic load situation；The scene produced due to each Monte Carlo simulation is different, and corresponding mistake loading also and differs, Choose arbitrary scene institute gain and loss loading the most unreasonable to characterize system mistake loading；Therefore, expected value is used hereinRepresenting that t period VPP loses loading, gained expression formula is as follows:

$E\left(P_t^{ens}\right)=\underset{d=1}{\overset{n_d}{\Σ}}\left(\frac{1}{n_d}{P}_{d,t}^{ens}\right)$

In formula: n_dFor scene number；Loading is lost for t period d scene.