CN117713240A - Optimal scheduling method for power distribution network-cloud energy storage system considering wind-solar uncertainty - Google Patents
Optimal scheduling method for power distribution network-cloud energy storage system considering wind-solar uncertainty Download PDFInfo
- Publication number
- CN117713240A CN117713240A CN202311724876.6A CN202311724876A CN117713240A CN 117713240 A CN117713240 A CN 117713240A CN 202311724876 A CN202311724876 A CN 202311724876A CN 117713240 A CN117713240 A CN 117713240A
- Authority
- CN
- China
- Prior art keywords
- energy storage
- power
- cloud energy
- distribution network
- constraint
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000004146 energy storage Methods 0.000 title claims abstract description 278
- 238000000034 method Methods 0.000 title claims abstract description 57
- 238000005457 optimization Methods 0.000 claims abstract description 33
- 238000000342 Monte Carlo simulation Methods 0.000 claims abstract description 9
- 230000001360 synchronised effect Effects 0.000 claims abstract description 9
- 238000006467 substitution reaction Methods 0.000 claims abstract description 7
- 230000005540 biological transmission Effects 0.000 claims description 42
- 238000007599 discharging Methods 0.000 claims description 33
- 230000005611 electricity Effects 0.000 claims description 27
- 230000004044 response Effects 0.000 claims description 27
- 239000000446 fuel Substances 0.000 claims description 12
- 239000008186 active pharmaceutical agent Substances 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 9
- 230000009194 climbing Effects 0.000 claims description 9
- 238000010977 unit operation Methods 0.000 claims description 8
- 230000009467 reduction Effects 0.000 claims description 7
- 238000010248 power generation Methods 0.000 claims description 6
- 230000001174 ascending effect Effects 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 239000002699 waste material Substances 0.000 claims description 2
- 230000003993 interaction Effects 0.000 abstract description 2
- 238000004458 analytical method Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010835 comparative analysis Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
Landscapes
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention relates to the technical field of multi-energy interconnection system scheduling, and discloses a power distribution network-cloud energy storage system optimal scheduling method considering wind-light uncertainty. The method comprises the steps of carrying out collaborative modeling on a power distribution network, a cloud energy storage system and cloud energy storage users, establishing a deterministic model and an uncertainty model of double-layer optimization scheduling of the power distribution network-cloud energy storage system, which take wind and light uncertainty into account, generating a wind and light generator set output uncertainty output scene by using a Monte Carlo simulation method and a synchronous back substitution method, and solving by using a Gurobi solver. According to the method, the cloud energy storage operation mode is considered in the optimal scheduling of the power distribution network, so that the scattered energy storage resources of the user side can be fully utilized; when the distribution network makes a day-ahead dispatch plan, only information interaction is needed with the main body of the cloud energy storage operator, so that the management difficulty of the distribution network can be greatly reduced.
Description
Technical Field
The invention relates to the technical field of multi-energy interconnection system scheduling, in particular to a power distribution network-cloud energy storage system optimal scheduling method considering wind-light uncertainty.
Background
With the increase of the proportion of new energy access distribution network and the improvement of energy storage technology, the cloud energy storage mode is a new mode for energy storage management at the user side in the future. A large amount of scattered energy storage resources with high idle rate and large management difficulty exist on the user side of the power distribution network, if the scattered energy storage resources can be reasonably utilized, the consumption of new energy can be promoted, the running cost of the power distribution network is reduced, and the utilization rate of the idle scattered energy storage resources can be improved. In the prior art, energy storage of a certain scale is generally equivalent to integral management through a cluster effect, besides the centralized energy storage, the available potential of the scattered energy storage resources of the electric automobile, the user self-built energy storage, the demand response load and other user sides is very large, if the energy storage resources can be reasonably utilized, the new energy consumption can be promoted, the running cost of a power distribution network can be reduced, and the utilization rate of the idle scattered energy storage resources can be improved. Therefore, the cloud energy storage mode is applied to power distribution network optimization scheduling, and the double-layer optimization scheduling method of the power distribution network-cloud energy storage system for researching and taking wind and light uncertainty into consideration has important significance.
Disclosure of Invention
Aiming at the problems, the invention aims to provide an optimal scheduling method of a power distribution network-cloud energy storage system for accounting for wind-solar uncertainty, which is characterized in that a gas turbine set, a wind-solar generator set, a transformer substation, a cloud energy storage system, electric automobile users participating in a cloud energy storage mode, uninterruptible power supply users and demand response load users in the power distribution network are subjected to collaborative modeling, in the process of optimal scheduling, the power distribution network only needs to interact with the cloud energy storage system, the power distribution network manages charging and discharging of the cloud energy storage system, a cloud energy storage operator manages charging and discharging of the cloud energy storage users, and a Gurobi solver is adopted for solving. The technical proposal is as follows:
A power distribution network-cloud energy storage system optimization scheduling method considering wind and light uncertainty comprises the following steps:
step 1: determining each component of a power distribution network-cloud energy storage system, namely a power distribution network, a cloud energy storage system and a cloud energy storage user, wherein the cloud energy storage user comprises an electric automobile user, an uninterruptible power supply user and a demand response load user;
step 2: establishing a deterministic model of double-layer optimized scheduling of a power distribution network-cloud energy storage system considering wind-solar uncertainty;
the deterministic model comprises an upper model and a lower model, and the constraint of the upper model comprises cloud energy storage system constraint, power distribution network unit constraint and power balance constraint; the constraints of the lower model comprise cloud energy storage user constraints and power distribution network tide constraints, and the cloud energy storage user constraints comprise electric vehicle constraints, uninterruptible power supply constraints and demand response load constraints;
step 3: establishing an uncertainty model of double-layer optimization scheduling of a power distribution network-cloud energy storage system considering wind-solar uncertainty;
when wind and light uncertainty is considered, a typical wind and light output scene is taken into an upper model, and a scheduling strategy considering the wind and light output uncertainty is obtained by combining the uncertainty model, so that the power distribution network can still stably operate when wind and light output fluctuates;
Step 4: on the premise of meeting the safety constraint of each component of the system, a double-layer optimization scheduling model of the power distribution network-cloud energy storage system is established, wherein the wind-light uncertainty of the double-layer optimization scheduling model is considered; the objective function of the upper model is that the running cost of the power distribution network is the lowest, and the optimal output plan of each power supply in the power distribution network and the charge and discharge plan of the cloud energy storage system are obtained by solving; the objective function of the lower model is the maximum profit of the cloud energy storage operator; based on a cloud energy storage system charge-discharge plan obtained by solving the upper model, and in combination with the operation constraint of a power distribution network, the charge-discharge plans of each electric automobile user, uninterruptible power supply user and demand response load user are formulated;
step 5: generating an uncertain output scene of the wind-light generator set by using a Monte Carlo simulation method, then carrying out scene reduction by using a synchronous substitution method, and finally generating typical wind and light output scenes;
step 6: and (3) inputting operation parameters of the power distribution network-cloud energy storage system, and solving a double-layer optimization scheduling model of the power distribution network-cloud energy storage system, which takes wind and light uncertainty into consideration, by adopting a commercial solver to obtain an optimization scheduling strategy of the double-layer optimization scheduling model.
Further, in the step 1, the power distribution network includes a substation unit, a gas unit, a wind-light generator unit, a cloud energy storage system, a cloud energy storage user, an electric load, a bus node and a power transmission line.
Further, the cloud energy storage system constraint, the power distribution network unit constraint and the power balance constraint in the constraint of the upper layer model in the step 2 are specifically:
1) Cloud energy storage system constraints
The constraints of the cloud energy storage system comprise the charge and discharge power and capacity constraint of the cloud energy storage system, and a cloud energy storage operator determines the upper limit and the lower limit of the charge and discharge power of the cloud energy storage system in each period and the upper limit and the lower limit of the net charge power in the day by estimating the capacity limit of the cloud energy storage user which is managed by the cloud energy storage operator and is acceptable to schedule in each period;
wherein: t is a time index, and T is a scheduling total period; e is an index of the cloud energy storage system, and E is a set of the cloud energy storage systems in the power distribution network;and->Discharging power and charging power of cloud energy storage system e for t period, +.>And->The upper limit of discharge power and the upper limit of charge power of the cloud energy storage system e in the t period are respectively +.>And->The lower limit of the discharge power and the lower limit of the charge power of the cloud energy storage system e in the t period are respectively set; />For the net charge of the cloud energy storage system e, +.>And->The upper limit and the lower limit of the net charge of the cloud energy storage system e are respectively set; o (O) e,t For the capacity of the t-period cloud energy storage system e, < >>And->The upper limit and the lower limit of the capacity of the cloud energy storage system e are respectively set;
2) Power distribution network unit constraint
The power distribution network unit constraint comprises a gas unit operation constraint, a wind-light unit output constraint and a transformer substation transmission power constraint;
a) Gas turbine unit operation constraint
The gas unit operation constraint comprises the upper and lower output limit constraint of the gas unit, the minimum start-up time constraint, the minimum stop time constraint, the start-up and stop fuel consumption constraint and the climbing constraint; the following formula is shown:
K g,t ≥k g ·(x g,t -x g,t-1 ),K g,t ≥0,g∈G,t∈T
D g,t ≥d g ·(x g,t-1 -x g,t ),D g,t ≥0,g∈G,t∈T
wherein: g is an index of the gas units, and G is a set of the gas units in the power distribution network; x is x g,t When the working state of the gas unit g is the t period, the gas unit g is in a starting state when the working state is 1, and is in a stopping state when the working state is 0; p (P) g,t And Q g,t The active power and the reactive power output by the gas turbine unit g in the t period are respectively,and->The maximum limit value and the minimum limit value of the active output of the gas unit g are respectively; />And->The maximum limit value and the minimum limit value of the reactive output of the gas unit g are respectively; />And->The start-up time counter and the stop time counter of the gas unit g in the t period are respectively; />And->The minimum start-up time and the minimum stop time of the gas unit g are respectively; k (K) g,t And D g,t Fuel consumption, k for starting up and stopping the gas unit g at t time intervals g And d g Fuel consumption for starting up and stopping the gas unit g respectively; / >And->The ascending slope rate and the descending slope rate of the gas unit g are respectively;
b) Wind-solar unit output constraint
The output constraint of the wind power unit and the photovoltaic unit comprises an active output constraint and a reactive output constraint;
wherein: s and W are respectively a photovoltaic generator set and a wind generating set in the power distribution network,and->Respectively the predicted output values of the photovoltaic generator set s and the wind generating set w in the period t, P s,t And P w,t The actual output values of the photovoltaic generator set s and the wind generating set w at the period t are respectively Q s,t And Q w,t Reactive power output of the photovoltaic generator set s and the wind generator set w in the t period is +.>And->The power factor angles of the photovoltaic generator set s and the wind generating set w are respectively;
c) Substation transmission power constraint
The substation transmission power constraint comprises an active power limit constraint and a reactive power limit constraint;
wherein: z is a substation set in the power distribution network,for t-period distribution network, purchasing power from z-direction upper power grid of transformer substation, < >>Reactive power provided to the distribution network by the substation z for the period t upper grid, +.>And->Active power transmitted by the upper power grid through the z-direction power distribution network of the transformer substationUpper and lower power limit,/->And->The upper limit and the lower limit of reactive power transmitted by the upper power grid through the z-direction power distribution network of the transformer substation are respectively set;
3) Power balance constraint
The power balance constraints include active power balance constraints, reactive power balance constraints:
wherein D is a node set where the load is located; p (P) t loss The estimated loss power of the power distribution network at the t period in the upper model is obtained; c is reactive power compensation equipment set, Q c,t Reactive power emitted by the reactive compensation equipment c in the t period; p (P) d,t And Q d,t Active power demand and reactive power demand for the t-period load d.
Further, cloud energy storage user constraint and power distribution network tide constraint in the constraint of the lower model in the step 2 are specifically;
1) Cloud energy storage user constraints
The cloud energy storage user constraint comprises an electric automobile constraint, a demand response load constraint and an uninterruptible power supply constraint;
a) Constraint of electric automobile
The electric vehicle constraint comprises an electric vehicle charge-discharge capacity constraint, an electric vehicle charge-discharge power constraint, an electric vehicle minimum charge-discharge time constraint and an electric vehicle charging station charge-discharge total power limit constraint;
wherein: v is an electric automobile set;and->The charging power and the discharging power of the electric automobile v in the t period are respectively; o (O) v,t Is the capacity eta of the electric automobile v in the period t v The charging and discharging efficiency of the electric automobile is improved; />And->The upper limit and the lower limit of the capacity of the electric automobile v are respectively; / >The method is characterized in that the method is the initial electric quantity of the electric automobile v when the electric automobile v is connected to the network, < >>The expected electric quantity when the electric automobile v leaves the charging station; />Net charge for electric vehicle v; />And->Respectively representing the upper limit and the lower limit of v discharge power of the electric automobile, < >>The discharge state of the electric automobile v in the t period is 1, and the discharge state is 0, and the discharge state is not; />Andrespectively representing the upper limit and the lower limit of v charging power of the electric automobile, < >>The charging state of the electric automobile v in the t period is 1, and the charging state is not charged when the charging state is 0; />Off-grid time counter for t-period electric automobile after charging>Minimum off-grid time length after the electric automobile is charged; />Off-grid time counter for t-period electric automobile after being powered off>The minimum off-grid time is the minimum off-grid time after the electric automobile is discharged; />On-line charging time counter for t-period electric automobile>The method comprises the steps of (1) enabling an electric automobile to be in a network for a minimum charging time; />For t period electric automobile online discharge duration counter, < >>Minimum discharge for electric automobile on networkDuration of time; a (Y) is a set of devices in the electric vehicle charging station Y, Y is a set of electric vehicle charging stations,/->And->Maximum charging power and discharging power of the electric vehicle charging station y respectively;
b) Demand response load constraints
The demand response load comprises an interruptible load, and the constraint condition is an interruptible load limit constraint;
wherein:an interrupt load amount of the interruptible load d for the t period; />The participation state of the interruptible load d in the t period is 1, and the interrupt state is represented when the participation state is 0, and the interrupt state is not represented when the participation state is 0; alpha d The ratio of the interruptible load capacity to the total load;
c) Uninterruptible power supply restraint
The uninterrupted power supply constraint comprises uninterrupted power supply capacity constraint and uninterrupted power supply charge-discharge power constraint;
O u,T ＝O u,0
wherein: u is an uninterruptible power supply set;and->The charging power and the latter discharging power of the uninterruptible power supply u in the t period are respectively; o (O) u,t For the capacity of the uninterrupted power supply u in the period of t, eta u The charging and discharging efficiency of the uninterrupted power supply is improved; />And->The upper limit and the lower limit of the u capacity of the uninterrupted power supply are set; o (O) u,T And O u,0 The capacities of the uninterruptible power supply u at the end time and the initial time in the scheduling period are respectively set; />And->Respectively represent the upper limit and the lower limit of the discharge power of the uninterruptible power supply u, < >>The discharge state of the uninterruptible power supply u in the t period is 1, and the discharge state is 0, so that the discharge state is not discharged; />And->Respectively representing the upper limit and the lower limit of the charging power of the uninterruptible power supply u, < >>The charging state of the uninterruptible power supply u in the t period is 1, and the charging state is 0, namely the charging state is not charged;
The sum of the charge and discharge power of each cloud energy storage user in the lower model is equal to the charge and discharge power of each cloud energy storage system obtained by solving the upper model;
wherein: a (e) is a device set of the cloud energy storage system e;
2) Power distribution network tide constraint
Linearizing power flow constraint of a power distribution network by adopting a second order cone relaxation method, wherein the operation constraint of the power distribution network comprises node power balance constraint of the power distribution network, line voltage drop limit constraint of the power distribution network, second order cone constraint and line operation constraint; wherein, the network loss difference value of the upper layer model and the lower layer modelEqual to the estimated net loss value P of the upper model t loss And the actual net loss value of the lower model +.>Is a difference in (2);
wherein: a (i) is a collection of distribution network devices directly connected to node i; l and N bus Respectively collecting power transmission lines and nodes of the power distribution network; s (l) and r (l) are a transmitting end node and a receiving end node of the power transmission line l; p (P) l,t And Q l,t Respectively t time periodActive power flow and reactive power flow of the electric line l;and->The active network loss and the reactive network loss of the power transmission line l in the t period are calculated; />The network loss difference value between the upper layer model and the lower layer model; omega l,t Is the current square value of the power transmission line l in the t period, and ψ j,t Sum phi i,t The voltage square values of the node j and the node i in the t period are respectively; r is (r) l And x l The resistance value and the reactance value of the power transmission line l are respectively; />And->Square values representing upper and lower limits of current allowed to flow through transmission line l, respectively, +.>And->The square values of the upper and lower voltage limits that node i can withstand are shown, respectively.
Further, in the step 3, the wind-light output uncertainty model is considered as follows:
wherein: superscript sce indicates the value of the variable under scene sce, and γ represents the set of typical wind and light output scenes;the emergency climbing rate of the gas unit; />Is an adjustable range of substation transmission power in an uncertainty scene.
Further, in the step 4 power distribution network-cloud energy storage system double-layer optimization scheduling model
(1) In the upper model, the running cost of the power distribution network in the objective function comprises the use cost of the cloud energy storage system, the electricity purchasing cost of the upper power grid, the electricity generating cost of the gas turbine set and the electricity discarding penalty of the renewable energy, and the expression of the objective function is as follows:
minF 1 ＝F M +F E +F G +F Voll
wherein: f (F) 1 F for the running cost of the distribution network M F, purchasing electricity cost for upper-level power grid of power distribution network E F for the use cost of the cloud energy storage system G F is the power generation cost of the gas unit Voll The cost is punished for renewable energy source waste electricity;
wherein:
wherein:the electricity purchase price corresponding to the t period;
When the power distribution network needs to charge/discharge the cloud energy storage system, the cloud energy storage operator pays corresponding service cost, and the cloud energy storage operator only needs to pay net charge cost to the power distribution network for the net charge part of the cloud energy storage system;
wherein:service unit price for charging/discharging of cloud energy storage system e, < >>A net charging unit price for the cloud energy storage system e;
wherein: m is M g (. Cndot.) is a heat rate curve,fuel unit price for period t;
wherein:and->The unit price of electricity discarding punishment of the photovoltaic generator set and the wind turbine generator set is respectively;
(2) In the lower model, the main body is a cloud energy storage system, and the profit of a cloud energy storage operator in the objective function comprises the following parts:
max(F 2 -F P )＝F E -(F U +F V +F I )-F P
wherein: f (F) 2 The total profit obtained for the cloud energy storage operator is equal to the cloud energy storage system use cost paid by the power distribution network to the cloud energy storage operator minus the cloud energy storage user scheduling cost paid by the cloud energy storage operator to the cloud energy storage user, F U Scheduling costs for uninterruptible Power supply users, F V Scheduling cost for electric automobile user, F I Scheduling costs for interruptible load users, F P Punishment cost is conducted on the network loss difference value between the upper layer model and the lower layer model;
the cloud energy storage operator pays the discharging/charging service charge to the electric automobile user, and the electric automobile user only needs to pay the net charging charge for the net charging part of the electric automobile;
Wherein:the unit price of charge/discharge service of the electric automobile; />The net charging unit price of the electric automobile;
the demand response load only considers the interruptible load of a user, and the cloud energy storage system induces the user to cut down electricity consumption in a compensation mode to be used as a part of the discharge power of the cloud energy storage system;
wherein:giving a compensation unit price capable of interrupting load to a cloud energy storage system operator;
the cloud energy storage operator pays the discharge/charge service charge to the uninterruptible power supply user;
wherein:the unit price is used for charging/discharging of the uninterrupted power supply;
when solving the output plans of all units in the power distribution network and the charge and discharge plans of the cloud energy storage system, the upper model does not calculate the power flow of the power distribution network, so that the network loss in the active power balance constraint is a predicted value; when a lower model makes a charge and discharge plan of each cloud energy storage user, carrying out power flow calculation of the power distribution network to obtain an actual network loss value; the network loss difference value is the difference value between the actual network loss and the estimated network loss;
wherein: c p Penalty unit price for the net loss difference.
Further, the process of step 5 is specifically as follows:
based on basic prediction scene of wind power and photovoltaic power generation, wind power output prediction error is assumedAnd photovoltaic output prediction error- >Meets the standard normal distribution, i.e.)>Generating an uncertainty output scene by a Monte Carlo simulation method, wherein in the scene h, the output of wind power and photovoltaic is respectively as follows:
wherein:and->The predicted force values of the wind driven generator w and the photovoltaic generator s in the t period under the scene sce are respectively; />And->Respectively predicting force values of the wind driven generator w and the photovoltaic generator s in a t period under a basic prediction scene;and->The output prediction error values of the wind driven generator w and the photovoltaic generator s in the t period under the scene sce are respectively.
The scene is reduced by a synchronous generation method, a plurality of typical scenes are screened out for model solving, so that the aim of improving the solving efficiency is fulfilled, and the scene reduction steps are as follows:
step a: setting SS as an initial scene set; DS is a scene set to be deleted, and the initial state is an empty set; the Euclidean distance between each scene in the initial scene set SS is calculated:
wherein: DT (DT) r,k For the (r) th scene omega r And the kth scene Ω k Euclidean distance between, P t r And P t k The method comprises the steps that a force value is predicted for a wind-solar unit in a t period under a scene r and a scene k respectively; omega shape u Is a set of all scenes;
step b: calculating scene Ω r Probability distance PD from all other scenes k (r) the smaller the value, the higher the similarity of the scene with other scenes;
wherein: ρ r Is scene omega r Is a probability of occurrence of (1); r=1, 2,..n, N is the total number of scenes;
step c: calculating the probability distance of each scene, if one scene omega exists b Probability distance PD from all other scenes k (b)＝minPD k (r); then select scene Ω b As a scene to be deleted;
step d: calculating each scene and scene omega in the initial scene set SS b If DT is the Euclidean distance of b,d ＝minDT b,k K noteqb, b = 1,2, N; then delete scene Ω b Scene omega b The probability of occurrence is added to scene Ω d Applying; updating scene sets and occurrence probabilities of various scenes:
SS＝SS-{Ω b }；DS＝DS+{Ω b }
ρ d ＝ρ d +ρ b ；ρ b ＝0
step e: repeating the steps b-d until the number of scenes meets the target set value.
Further, the operation parameters of the power distribution network-cloud energy storage system include the output cost, climbing capacity, minimum starting and stopping time, electric load size, power limit transmitted by a power transmission line, current limit of the power transmission line, voltage limit of a node, charge and discharge service cost of the cloud energy storage system, capacity limit of the cloud energy storage system, charge and discharge power limit of the cloud energy storage system, charge and discharge service cost of an electric vehicle user, capacity limit of the electric vehicle user, charge and discharge power limit of the electric vehicle user, charge and discharge service cost of an uninterruptible power supply user, capacity limit of the uninterruptible power supply user, charge and discharge power limit of the uninterruptible power supply user, charge and discharge compensation cost of demand response load and capacity limit of demand response load.
The beneficial effects of the invention are as follows:
aiming at the problems of high idle rate and high management difficulty of scattered energy storage resources at a user side of a power distribution network, the invention provides the power distribution network-cloud energy storage system optimal scheduling method considering wind-solar uncertainty, and the cloud energy storage operation mode is considered in the power distribution network optimal scheduling, so that the scattered energy storage resources at the user side can be fully utilized. When the distribution network makes a day-ahead dispatch plan, only information interaction is needed with the main body of the cloud energy storage operator, so that the management difficulty of the distribution network can be greatly reduced; according to the power distribution network-cloud energy storage system optimization scheduling method considering wind-light uncertainty, the uncertainty of wind-light output is processed by incorporating an uncertainty scene, and stable operation of the system is ensured when wind-light output fluctuates by adjusting charge and discharge of the cloud energy storage system.
Drawings
FIG. 1 is a topology of a 33 node power distribution network system; GT-gas units; c-reactive compensation device; WT-wind turbine generator; a PV-photovoltaic unit; u-uninterruptible power supply users; EV-electric car users; d-interruptible load user; TS-substation.
Fig. 2 (a) shows the wind and light uncertainty output scenario—the photovoltaic unit output of different scenarios.
Fig. 2 (b) shows a wind-solar uncertainty output scenario—different scenarios of wind turbine set output.
Fig. 3 (a) shows the individual unit output-optimizing scheduling strategy of example 1.
Fig. 3 (b) shows the output of each unit-the charge and discharge power of the energy storage user according to example 1.
Fig. 4 is a graph of the cloud energy storage system charge and discharge power under various scenarios of the set output of example 2.
Detailed Description
For a detailed description of the disclosed embodiments, the invention is further described below with reference to the accompanying drawings and specific examples.
The invention discloses a double-layer optimization scheduling method of a power distribution network-cloud energy storage system, which takes wind and light uncertainty into account, and comprises the following steps of:
step 1: the various components of the distribution network-cloud energy storage system are determined.
Example 1 shows the determination of the individual components of a distribution network-cloud energy storage system: the power distribution network comprises a transformer substation unit, a gas unit, a wind-light generator unit, a cloud energy storage system, a cloud energy storage user, an electric load, bus nodes and a power transmission line, wherein the cloud energy storage user comprises an electric automobile user, an uninterruptible power supply user and a demand response load user.
Step 2: and establishing a deterministic model of double-layer optimized scheduling of the power distribution network-cloud energy storage system considering wind-solar uncertainty.
Embodiment 2 presents a deterministic model for building a power distribution network-cloud energy storage system double-layer optimization schedule that accounts for wind-solar uncertainty:
(1) Upper model
The constraints of the upper model comprise cloud energy storage system constraints, power distribution network unit constraints and power balance constraints. The power distribution network unit constraint comprises a gas unit operation constraint, a wind-light unit output constraint and a transformer substation transmission power constraint;
the constraints of the cloud energy storage system comprise the charge and discharge power and capacity constraint of the cloud energy storage system, and a cloud energy storage operator needs to estimate the acceptable scheduled cloud energy storage user capacity limit value managed by the cloud energy storage operator in each period, so as to determine the upper limit and the lower limit of the charge and discharge power of the cloud energy storage system in each period and the upper limit and the lower limit of the net charge power in the current day.
Wherein: t is a time index, T is a scheduling total period, E is a set of cloud energy storage systems in the distribution network, discharging and charging power of cloud energy storage system e for t period, +.>The upper limit of the discharging power and the charging power of the cloud energy storage system e in the t period is respectively +.>The lower limit of the discharging power and the charging power of the cloud energy storage system e in the t period is respectively +.>For the net charge of the cloud energy storage system e, +.>The upper limit and the lower limit of the net charge of the cloud energy storage system e are respectively. O (O) e,t For the capacity of the t-period cloud energy storage system e, < >>The upper limit and the lower limit of the capacity of the cloud energy storage system e are respectively set.
The gas unit operation constraint comprises the upper and lower output limit constraint of the gas unit, the minimum start-up time constraint, the minimum stop time constraint, the start-up and stop fuel consumption constraint and the climbing constraint. The following formula is shown:
Wherein: g is a set of gas units in the power distribution network, and x is g,t When the working state of the gas unit g is the t period, the gas unit g is in a starting state when the working state is 1, and is in a stopping state when the working state is 0; p (P) g,t 、Q g,t The active power and the reactive power output by the gas turbine unit g in the t period are respectively,the maximum and minimum limit values of the active output of the gas unit g are respectively; />The maximum and minimum reactive output limit values of the gas unit g are respectively; />The start-up time counter and the stop time counter are respectively used for the gas unit g;the minimum start-up time and the minimum stop time of the gas unit g are respectively; k (K) g,t 、D g,t Fuel consumption for starting up and stopping the gas turbine unit g at t time interval, k g 、d g Fuel consumption for starting up and stopping the gas unit g respectively; />The ascending slope rate and the descending slope rate of the gas unit g are respectively.
The output constraint of wind power and photovoltaic units comprises an active output constraint and a reactive output constraint.
Wherein: s, W are respectively a photovoltaic generator set and a wind generator set in the power distribution network,the predicted output values of the photovoltaic generator set s and the wind generating set w at the period t and P are respectively s,t 、P w,t The actual output values of the photovoltaic generator set s and the wind generator set w at the period t are respectively Q s,t 、Q w,t The reactive power output of the photovoltaic generator set s and the wind generating set w in the period t is calculated, The power factor angles of the photovoltaic generator set s and the wind generator set w are respectively.
The substation transmission power constraints include active power limit constraints and reactive power limit constraints.
In the middle of: z is a substation set in the power distribution network,for t-period distribution network, purchasing power from z-direction upper power grid of transformer substation, < >>Reactive power provided to the distribution network by the substation z for the period t upper grid, +.> The upper limit and the lower limit of active power transmitted by the upper power grid through the z-direction power distribution network of the transformer substation are respectively +.>The upper limit and the lower limit of reactive power transmitted by the upper power grid through the z-direction power distribution network of the transformer substation are respectively set.
The output of each unit of the power distribution network, the charge and discharge power, the load and the network loss power of the cloud energy storage system are required to be balanced in each scheduling period, and the power balance constraint comprises an active power balance constraint and a reactive power balance constraint.
Wherein D is a node set where the load is located;the estimated loss power of the power distribution network at the t period in the upper model is obtained; c is reactive power compensation equipment set, Q c,t Reactive power emitted by the reactive compensation equipment c in the t period; p (P) d,t 、Q d,t For t-period loadd active and reactive power requirements.
(2) Lower layer model
The constraints of the lower model comprise cloud energy storage user constraints and power distribution network tide constraints, and the cloud energy storage user constraints comprise electric automobile constraints, uninterruptible power supply constraints and demand response load constraints. The related constraints of the electric vehicle mainly comprise the constraint of the charge and discharge capacity of the electric vehicle, the constraint of the charge and discharge power of the electric vehicle, the constraint of the minimum charge and discharge time of the electric vehicle and the constraint of the limit value of the total charge and discharge power of the charging station of the electric vehicle.
Wherein: v is an electric automobile set;the charging and discharging power of the electric automobile v in the period t respectively; o (O) v,t Is the capacity eta of the electric automobile v in the period t v The charging and discharging efficiency of the electric automobile is improved; />The upper limit and the lower limit of the capacity of the electric automobile v are respectively; />The method is characterized in that the method is the initial electric quantity of the electric automobile v when the electric automobile v is connected to the network, < >>The expected electric quantity when the electric automobile v leaves the charging station; />Net charge for electric vehicle v; />Respectively representing upper and lower limits of v discharge power of the electric automobile>The discharge state of the electric automobile v in the t period is 1, and the discharge state is 0, and the discharge state is not; />Respectively representing upper and lower limits of v charging power of the electric automobile>The charging state of the electric automobile v in the t period is 1, and the charging state is not charged when the charging state is 0; />Off-grid duration counter after electric automobile is charged>Minimum off-grid time length after the electric automobile is charged; />Counter for off-grid time length after electric automobile is discharged>The minimum off-grid time is the minimum off-grid time after the electric automobile is discharged;on-line charging time counter for electric automobile, < >>The method comprises the steps of (1) enabling an electric automobile to be in a network for a minimum charging time; />Is an electric steamOn-line discharge time counter for vehicle>The on-grid minimum discharge time length of the electric automobile is set; a (Y) is a set of devices in the electric vehicle charging station Y, Y is a set of electric vehicle charging stations,/- >The maximum charge and discharge power of the electric vehicle charging station y is respectively.
The demand response load includes an interruptible load, with the constraint being an interruptible load limit constraint.
Wherein:an interrupt load amount of the interruptible load d for the t period; />The participation state of the interruptible load d in the t period is 1, and the interrupt state is represented when the participation state is 0, and the interrupt state is not represented when the participation state is 0; alpha d In order to be a proportion of the interruptible load capacity to the total load.
The uninterruptible power supply related constraints include uninterruptible power supply capacity constraints and uninterruptible power supply charge-discharge power constraints.
O u,T ＝O u,0
Wherein: u is an uninterruptible power supply set;the charging and discharging power of the uninterruptible power supply u in the period t respectively; o (O) u,t For the capacity of the uninterrupted power supply u in the period of t, eta u The charging and discharging efficiency of the uninterrupted power supply is improved; /> The upper limit and the lower limit of the u capacity of the uninterrupted power supply are adopted; o (O) u,T And O u,0 The capacities of the uninterruptible power supply u at the end time and the initial time in the scheduling period are respectively set;respectively represent the upper limit and the lower limit of the u discharge power of the uninterrupted power supply>The discharge state of the uninterruptible power supply u in the t period is 1, and the discharge state is 0, so that the discharge state is not discharged; />Respectively represent the upper limit and the lower limit of the charging power of the uninterruptible power supply u>The charging state of the uninterruptible power supply u is t, and when the charging state is 1, the charging state is represented, When 0, no charge is indicated.
The sum of the charge and discharge power of each cloud energy storage user in the lower model is equal to the charge and discharge power of each cloud energy storage system obtained by solving the upper model.
/>
Wherein: a (e) is a device set of the cloud energy storage system e.
And linearizing the power flow constraint of the power distribution network by adopting a second order cone relaxation method, wherein the operation constraint of the power distribution network comprises power balance constraint of nodes of the power distribution network, voltage drop limit constraint of lines of the power distribution network, second order cone constraint and line operation constraint. Wherein, the network loss difference value of the upper layer model and the lower layer modelEqual to the estimated net loss value of the upper model +.>And the actual net loss value of the lower model +.>Is a difference in (c).
Wherein: a (i) is a collection of distribution network devices directly connected to node i; l, N bus Respectively collecting power transmission lines and nodes of the power distribution network; s (l) and r (l) are a transmitting end node and a receiving end node of the power transmission line l; p (P) l,t 、Q l,t Active and reactive power flows of the power transmission line l at the t period are respectively shown;the active power loss and the reactive power loss of the transmission line l in the t period are calculated; />The network loss difference value between the upper layer model and the lower layer model; omega l,t Is the current square value of the power transmission line l in the t period, and ψ i,t The square value of the voltage of the node i in the t period; r is (r) l And x l The resistance and reactance values of the transmission line l are respectively. / >Square values of upper and lower limits of current allowed to flow through transmission line/are respectively indicated, < >> The square values of the upper and lower limits of the voltage that node i can withstand are shown, respectively.
Step 3: and establishing an uncertainty model of double-layer optimization scheduling of the power distribution network-cloud energy storage system considering wind and light uncertainty.
When the wind-light uncertainty is considered, a typical wind-light output scene is taken into an upper model, and a scheduling strategy considering the wind-light output uncertainty is obtained by solving the uncertainty model, so that the power distribution network can still stably operate when the wind-light output fluctuates.
Embodiment 3 presents an uncertainty model for building a power distribution network-cloud energy storage system double-layer optimization schedule that accounts for wind-solar uncertainty:
and generating an uncertainty output scene by adopting a Monte Carlo simulation method, and then carrying out scene reduction by adopting a synchronous back substitution method to obtain a typical wind and light generator set output scene set gamma, wherein the scene sequence number is sce.
Wherein: superscript sce indicates the value of the variable under scene sce, and γ represents the set of typical wind and light output scenes;the emergency climbing rate of the gas unit; />Is an adjustable range of substation transmission power in an uncertainty scene.
Step 4: on the premise of meeting the safety constraint of each component of the system, a double-layer optimization scheduling model of the power distribution network-cloud energy storage system taking wind and light uncertainty into account is established, the objective function of an upper layer model is the lowest running cost of the power distribution network, the optimal output plan of each power supply in the power distribution network and the charge and discharge plan of the cloud energy storage system are obtained through solving, the objective function of a lower layer model is the maximum profit of a cloud energy storage operator, the charge and discharge plan of the cloud energy storage system obtained through solving the upper layer model is based, and the charge and discharge plans of each electric automobile user, uninterrupted power source user and demand response load user are formulated by combining the running constraint of the power distribution network.
Example 4 gives the objective function of a power distribution network-cloud energy storage system double-layer optimization scheduling model taking into account wind-solar uncertainty. In a double-layer optimization scheduling model of a power distribution network-cloud energy storage system, an objective function of an upper layer model is the minimum running cost of the power distribution network, and comprises the use cost of the cloud energy storage system, the electricity purchasing cost of an upper-level power grid, the electricity generating cost of a gas turbine set and the electricity discarding penalty of renewable energy sources, wherein the expression of the objective function is as follows:
minF 1 ＝F M +F E +F G +F Voll
wherein: f (F) 1 F for the running cost of the distribution network M F, purchasing electricity cost for upper-level power grid of power distribution network E F for the use cost of the cloud energy storage system G F is the power generation cost of the gas unit Voll And the cost is punished for renewable energy.
Wherein:
wherein:and the electricity purchase price corresponding to the t period is set.
When the power distribution network needs to be charged/discharged by the cloud energy storage system, corresponding service fees need to be paid to the cloud energy storage operators, and the cloud energy storage operators only need to pay net charge fees to the power distribution network for the net charge amount part of the cloud energy storage system.
Wherein:service unit price for charging/discharging of cloud energy storage system e, < >>The net charge unit price for the cloud energy storage system e.
Wherein: m is M g (. Cndot.) is a heat rate curve,is the fuel unit price of the t period.
Wherein:the unit price is punished for the abandoned electricity of the photovoltaic generator set and the wind turbine generator set respectively.
In the lower model, the main body is a cloud energy storage system, the objective function is that the profit of a cloud energy storage operator is maximum, and the profit of the cloud energy storage operator comprises the following parts:
max(F 2 -F P )＝F E -(F U +F V +F I )-F P
wherein: f (F) 2 The total profit obtained for the cloud energy storage operator is equal to the cloud energy storage system use cost paid by the power distribution network to the cloud energy storage operator minus the cloud energy storage user scheduling cost paid by the cloud energy storage operator to the cloud energy storage user, F U Scheduling costs for uninterruptible Power supply users, F V Scheduling cost for electric automobile user, F I Scheduling costs for interruptible load users, F P Penalty costs for the net loss difference between the upper and lower models.
The cloud energy storage operator pays the electric vehicle user for discharging/charging service charge, and the electric vehicle user only needs to pay the net charging charge for the electric vehicle net charging part.
Wherein:the unit price of charge/discharge service of the electric automobile; />The unit price of net charge of the electric automobile.
The demand response load only considers the user interruptible load, and the cloud energy storage system adopts a compensation mode to induce the user to cut down electricity consumption, so as to be used as a part of the discharge power of the cloud energy storage system.
Wherein:and giving a compensation unit price capable of interrupting the load to the cloud energy storage system operator.
The cloud energy storage operator pays the uninterruptible power supply user for the discharge/charge service.
Wherein:the unit price is used for charging/discharging of the uninterrupted power supply.
When solving the output plans of all units in the power distribution network and the charge and discharge plans of the cloud energy storage system, the upper model does not calculate the power flow of the power distribution network, so that the network loss in the active power balance constraint is a predicted value. And when the lower model makes a charge and discharge plan of each cloud energy storage user, carrying out power flow calculation of the power distribution network, and obtaining an actual network loss value. The network loss difference value is the difference value between the actual network loss and the estimated network loss.
Wherein: c p Penalty unit price for the net loss difference.
Step 5: and generating an output uncertainty output scene of the wind-light generator set by using a Monte Carlo simulation method and a synchronous back-substitution method.
In a double-layer optimization scheduling model of the power distribution network-cloud energy storage system considering wind-light uncertainty, the output plans of each unit obtained by solving are all predicted output based on wind light, however, in practice, wind-light output is always accompanied with fluctuation, so that the uncertainty is required to be considered; firstly, generating a large number of uncertain output scenes of the wind-light generator set by a Monte Carlo simulation method, then, performing scene reduction by a synchronous substitution method, and finally, generating typical wind and light output scenes.
Embodiment 5 provides a method for generating an output uncertainty output scene of a wind-light generator set:
based on basic prediction scene of wind power and photovoltaic power generation, wind power output prediction error is assumedAnd photovoltaic output prediction error->Meets the standard normal distribution, i.e.)>Generating an uncertainty output scene by a Monte Carlo simulation method, wherein in the scene sce, the output of wind power and photovoltaic power are respectively as follows:
wherein:and->The predicted force values of the wind driven generator w and the photovoltaic generator s in the t period under the scene sce are respectively; / >And->Respectively predicting force values of the wind driven generator w and the photovoltaic generator s in a t period under a basic prediction scene;and->The output prediction error values of the wind driven generator w and the photovoltaic generator s in the t period under the scene sce are respectively.
Because the number of scenes is too large to cause the solving speed to be slow, the scenes are reduced by a synchronous generation method, and a plurality of typical scenes are screened out for model solving so as to achieve the purpose of improving the solving efficiency, and the scene reduction comprises the following steps:
(1) Setting SS as an initial scene set; DS is a scene set to be deleted, and the initial state is an empty set; omega shape h (h=1, 2,., N) represents the h-th output scenario, scenario Ω, in the initial scenario set SS h The corresponding occurrence probability is ρ h . Calculating Euclidean distance DT between scenes in an initial scene set SS r,k ：
Wherein: DT (DT) r,k For the (r) th scene omega r And the kth scene Ω k Euclidean distance between, P t r And P t k Wind-solar unit for t time periods under scene r and scene k respectivelyPredicting a force value; omega shape u Is a set of all scenes.
(2)PD k (r) is scene Ω r The probability distance from all other scenes is smaller, which means that the similarity between the scene and other scenes is higher. Calculating probability distance of each scene, if PD k (b)＝minPD k (r), (r=1, 2,., N), then the scene Ω is selected b As a scene to be deleted. Wherein the probability distance PD k (r)：
(3) Calculating each scene and scene omega in scene set SS b If DT is the Euclidean distance of b,d ＝minDT b,k (k. Noteq. B, b=1, 2, once again, n.). Then delete scene Ω b Scene omega b The probability of occurrence is added to scene Ω d And (3) upper part. Updating scene sets and occurrence probabilities of various scenes:
SS＝SS-{Ω b }；DS＝DS+{Ω b }
ρ d ＝ρ d +ρ b ；ρ b ＝0
(4) Repeating (2) - (3) until the number of scenes meets the target set value.
Step 6: and (3) inputting operation parameters of the power distribution network-cloud energy storage system, and solving a double-layer optimization scheduling model of the power distribution network-cloud energy storage system, which takes wind and light uncertainty into consideration, by adopting a commercial solver to obtain an optimization scheduling strategy of the double-layer optimization scheduling model.
In the step 6, the operation parameters of the power distribution network-cloud energy storage system comprise the output cost, climbing capacity, minimum starting and stopping time, electric load size, power limit transmitted by a power transmission line, current limit of the power transmission line, voltage limit of a node, charge and discharge service cost of the cloud energy storage system, capacity limit of the cloud energy storage system, charge and discharge power limit of the cloud energy storage system, charge and discharge service cost of an electric vehicle user, capacity limit of the electric vehicle user, charge and discharge power limit of the electric vehicle user, charge and discharge service cost of an uninterruptible power supply user, capacity limit of the uninterruptible power supply user, charge and discharge power limit of the uninterruptible power supply user, demand response load and discharge compensation cost, demand response load capacity limit and the like.
The effects of the present invention will be described in detail by means of specific examples.
(1) The description of the examples is given.
As shown in fig. 1, the power distribution network system includes 33 nodes and 32 power transmission lines; solid lines are power transmission circuits, and solid points are nodes; numbers beside the solid points are the numbers of the nodes; the connection condition of each unit is shown in the figure, and the double-layer optimization scheduling of the power distribution network-cloud energy storage system, which is researched by the calculation example and takes 24 hours as a research period, and the interval time is 1 hour.
(2) Example results analysis.
In order to study the influence of different operation schemes of the power distribution network-cloud energy storage system which takes the wind-solar uncertainty into consideration on the economy and the safety of the system, wind-solar output uncertainty scenes shown in fig. 2 (a) and 2 (b) are extracted firstly; the following two examples were set for comparative analysis: calculation example 1: optimizing and scheduling a power distribution network containing a cloud energy storage system; calculation example 2: on the basis of the calculation example 1, the uncertainty of wind-light output is considered. The method comprises the steps of carrying out a first treatment on the surface of the
The output of each generator set in example 1 is shown in fig. 3 (a) and 3 (b). In example 1, a cloud energy storage system participates in optimal scheduling of a power distribution network. The cloud energy storage system is mainly focused on electricity consumption valley period, namely 01: 00-05: 00 and 24: 00-period charging, namely 12 in the electricity consumption peak period: 00-14: 00 and 18: 00-20: and discharging in a 00 period. The distribution network load comprises a base load and a cloud energy storage system charging load, and the peak-valley difference of a load curve is 2.7p.u. In the upper model, the total running cost of the power distribution network is 93401.2 yuan; the electricity purchasing cost of the upper-level power grid is 46542.6 yuan; the use cost of the gas unit is 45382.1 yuan; the cloud energy storage operator is paid 3187.5 yuan for charge/discharge service and 1711.0 yuan for net charge. In the lower model, a cloud energy storage operator pays 399.1 yuan of charge/discharge service charge to an electric automobile user and charges 2037.0 yuan of net charge; the charge/discharge service cost of the UPS is 611.6 yuan; the load interruption compensation cost is 562.2 yuan; finally, the total income of the cloud energy storage operators is 1940.6 yuan. In addition, for the electric automobile user, a certain benefit can be obtained by participating in the cloud energy storage operation mode, and the net charging cost is greatly reduced.
The output of each genset in example 2 is shown in fig. 4. In the example 2, the charging schedule of the cloud energy storage system in the uncertainty scene has larger fluctuation compared with the basic scene, but the overall change trend is unchanged, because in the uncertainty scene, the charging power of each period of the cloud energy storage system is still limited by the number of electric vehicles connected in the period, the schedulable UPS capacity and the interruptible load capacity. Under the influence of wind-light output fluctuation, the charge and discharge of the cloud energy storage system are greatly changed. In scenario 3, 21: 00-23: and the output of the wind turbine generator is increased within the 00 period, and the cloud energy storage system absorbs redundant wind power by increasing the charge quantity. In scenario 5, 02: and the output of the wind turbine generator at the moment 00 is reduced, at the moment, the cloud energy storage system reduces the charging power and discharges to meet the load requirement of the power distribution network, the charging power of the cloud energy storage system at the moment is reduced to 0.2p.u. from 0.6p.u. of the basic scene, and the discharging power is increased to 0.5p.u. from 0p.u. of the basic scene. Therefore, on the premise that the cloud energy storage system meets the charging requirement of a cloud energy storage user, the fluctuation of wind and light output can be dealt with by adjusting the charging and discharging power of each period, and the flexibility and the stability of the system operation are improved.
Through the analysis of the combination of the calculation example 1 and the calculation example 2, the cloud energy storage system can guide the user side scattered energy storage resources to charge in the electricity utilization valley period and discharge in the peak period, so that the running cost of the power distribution network and the net charge cost of the cloud energy storage users are reduced, the peak-valley difference of loads is reduced, and the mutual benefits and win-win of the power distribution network, the cloud energy storage operators and the cloud energy storage users are realized. The double-layer optimization operation model of the power distribution network-cloud energy storage system, which is provided by the invention, is used for processing the uncertainty of wind and light output by incorporating an uncertainty scene, and ensuring the stable operation of the system when the wind and light output fluctuates by adjusting the charge and discharge of the cloud energy storage system.
The foregoing description is only specific embodiments of the present invention, but not limited to the scope of the invention, and all equivalent changes or substitutions made by the specification and drawings of the present invention, directly or indirectly, should be included in the scope of the present invention.
Claims (8)
1. The optimal scheduling method of the power distribution network-cloud energy storage system considering wind-solar uncertainty is characterized by comprising the following steps of:
step 1: determining each component of a power distribution network-cloud energy storage system, namely a power distribution network, a cloud energy storage system and a cloud energy storage user, wherein the cloud energy storage user comprises an electric automobile user, an uninterruptible power supply user and a demand response load user;
Step 2: establishing a deterministic model of double-layer optimized scheduling of a power distribution network-cloud energy storage system considering wind-solar uncertainty;
the deterministic model comprises an upper model and a lower model, and the constraint of the upper model comprises cloud energy storage system constraint, power distribution network unit constraint and power balance constraint; the constraints of the lower model comprise cloud energy storage user constraints and power distribution network tide constraints, and the cloud energy storage user constraints comprise electric vehicle constraints, uninterruptible power supply constraints and demand response load constraints;
step 3: establishing an uncertainty model of double-layer optimization scheduling of a power distribution network-cloud energy storage system considering wind-solar uncertainty;
when wind and light uncertainty is considered, a typical wind and light output scene is taken into an upper model, and a scheduling strategy considering the wind and light output uncertainty is obtained by combining the uncertainty model, so that the power distribution network can still stably operate when wind and light output fluctuates;
step 4: on the premise of meeting the safety constraint of each component of the system, a double-layer optimization scheduling model of the power distribution network-cloud energy storage system is established, wherein the wind-light uncertainty of the double-layer optimization scheduling model is considered; the objective function of the upper model is that the running cost of the power distribution network is the lowest, and the optimal output plan of each power supply in the power distribution network and the charge and discharge plan of the cloud energy storage system are obtained by solving; the objective function of the lower model is the maximum profit of the cloud energy storage operator; based on a cloud energy storage system charge-discharge plan obtained by solving the upper model, and in combination with the operation constraint of a power distribution network, the charge-discharge plans of each electric automobile user, uninterruptible power supply user and demand response load user are formulated;
Step 5: generating an uncertain output scene of the wind-light generator set by using a Monte Carlo simulation method, then carrying out scene reduction by using a synchronous substitution method, and finally generating typical wind and light output scenes;
step 6: and (3) inputting operation parameters of the power distribution network-cloud energy storage system, and solving a double-layer optimization scheduling model of the power distribution network-cloud energy storage system, which takes wind and light uncertainty into consideration, by adopting a commercial solver to obtain an optimization scheduling strategy of the double-layer optimization scheduling model.
2. The optimal scheduling method for the power distribution network-cloud energy storage system according to claim 1, wherein in the step 1, the power distribution network comprises a transformer substation unit, a gas unit, a wind-light generator unit, a cloud energy storage system, a cloud energy storage user, an electric load, a bus node and a power transmission line.
3. The optimization scheduling method for the power distribution network-cloud energy storage system considering wind-solar uncertainty according to claim 1, wherein cloud energy storage system constraint, power distribution network unit constraint and power balance constraint in the constraint of the upper layer model in the step 2 are specifically:
1) Cloud energy storage system constraints
The constraints of the cloud energy storage system comprise the charge and discharge power and capacity constraint of the cloud energy storage system, and a cloud energy storage operator determines the upper limit and the lower limit of the charge and discharge power of the cloud energy storage system in each period and the upper limit and the lower limit of the net charge power in the day by estimating the capacity limit of the cloud energy storage user which is managed by the cloud energy storage operator and is acceptable to schedule in each period;
Wherein: t is a time index, and T is a scheduling total period; e is an index of the cloud energy storage system, and E is a set of the cloud energy storage systems in the power distribution network;and->Discharging power and charging power of cloud energy storage system e for t period, +.>And->The upper limit of discharge power and the upper limit of charge power of the cloud energy storage system e in the t period are respectively +.>And->The lower limit of the discharge power and the lower limit of the charge power of the cloud energy storage system e in the t period are respectively set; />For the net charge of the cloud energy storage system e, +.>And->The upper limit and the lower limit of the net charge of the cloud energy storage system e are respectively set; o (O) e,t For the capacity of the t-period cloud energy storage system e, < >>And->The upper limit and the lower limit of the capacity of the cloud energy storage system e are respectively set;
2) Power distribution network unit constraint
The power distribution network unit constraint comprises a gas unit operation constraint, a wind-light unit output constraint and a transformer substation transmission power constraint;
a) Gas turbine unit operation constraint
The gas unit operation constraint comprises the upper and lower output limit constraint of the gas unit, the minimum start-up time constraint, the minimum stop time constraint, the start-up and stop fuel consumption constraint and the climbing constraint; the following formula is shown:
K g,t ≥k g ·(x g,t -x g,t-1 ),K g,t ≥0,g∈G,t∈T
D g,t ≥d g ·(x g,t-1 -x g,t ),D g,t ≥0,g∈G,t∈T
wherein: g is an index of the gas units, and G is a set of the gas units in the power distribution network; x is x g,t When the working state of the gas unit g is the t period, the gas unit g is in a starting state when the working state is 1, and is in a stopping state when the working state is 0; p (P) g,t And Q g,t The active power and the reactive power output by the gas turbine unit g in the t period are respectively,and->The maximum limit value and the minimum limit value of the active output of the gas unit g are respectively; />And->The maximum limit value and the minimum limit value of the reactive output of the gas unit g are respectively; />And->The start-up time counter and the stop time counter of the gas unit g in the t period are respectively; />And->The minimum start-up time and the minimum stop time of the gas unit g are respectively; k (K) g,t And D g,t Fuel consumption, k for starting up and stopping the gas unit g at t time intervals g And d g Fuel consumption for starting up and stopping the gas unit g respectively; />And->The ascending slope rate and the descending slope rate of the gas unit g are respectively;
b) Wind-solar unit output constraint
The output constraint of the wind power unit and the photovoltaic unit comprises an active output constraint and a reactive output constraint;
wherein: s and W are respectively a photovoltaic generator set and a wind generating set in the power distribution network,and->Respectively the predicted output values of the photovoltaic generator set s and the wind generating set w in the period t, P s,t And P w,t The actual output values of the photovoltaic generator set s and the wind generating set w at the period t are respectively Q s,t And Q w,t Reactive power output of the photovoltaic generator set s and the wind generator set w in the t period is +.>And->The power factor angles of the photovoltaic generator set s and the wind generating set w are respectively;
c) Substation transmission power constraint
The substation transmission power constraint comprises an active power limit constraint and a reactive power limit constraint;
wherein: z is a transformer substation set in a power distribution networkThe combination of the two components is carried out,the power distribution network purchases electric power to the upper power network through the z-direction power substation at the t period,reactive power provided to the distribution network by the substation z for the period t upper grid, +.>And->The upper limit and the lower limit of active power transmitted by the upper power grid through the z-direction power distribution network of the transformer substation are respectively +.>And->The upper limit and the lower limit of reactive power transmitted by the upper power grid through the z-direction power distribution network of the transformer substation are respectively set;
3) Power balance constraint
The power balance constraints include active power balance constraints, reactive power balance constraints:
wherein D is a node set where the load is located; p (P) t loss The estimated loss power of the power distribution network at the t period in the upper model is obtained; c is reactive power compensation equipment set, Q c,t No reactive compensation equipment c for period tA power; p (P) d,t And Q d,t Active power demand and reactive power demand for the t-period load d.
4. The optimization scheduling method for the power distribution network-cloud energy storage system according to claim 1, wherein the cloud energy storage user constraint and the power distribution network tide constraint in the constraint of the lower model in the step 2 are specifically;
1) Cloud energy storage user constraints
The cloud energy storage user constraint comprises an electric automobile constraint, a demand response load constraint and an uninterruptible power supply constraint;
a) Constraint of electric automobile
The electric vehicle constraint comprises an electric vehicle charge-discharge capacity constraint, an electric vehicle charge-discharge power constraint, an electric vehicle minimum charge-discharge time constraint and an electric vehicle charging station charge-discharge total power limit constraint;
wherein: v is an electric automobile set;and->The charging power and the discharging power of the electric automobile v in the t period are respectively; o (O) v,t Is the capacity eta of the electric automobile v in the period t v Charging and discharging for electric automobileElectrical efficiency; />And->The upper limit and the lower limit of the capacity of the electric automobile v are respectively; />The method is characterized in that the method is the initial electric quantity of the electric automobile v when the electric automobile v is connected to the network, < >>The expected electric quantity when the electric automobile v leaves the charging station; />Net charge for electric vehicle v; />And->Respectively representing the upper limit and the lower limit of v discharge power of the electric automobile, < >>The discharge state of the electric automobile v in the t period is 1, and the discharge state is 0, and the discharge state is not; />And->Respectively representing the upper limit and the lower limit of v charging power of the electric automobile, < >>The charging state of the electric automobile v in the t period is 1, and the charging state is not charged when the charging state is 0; />Off-grid time counter for t-period electric automobile after charging >Minimum off-grid time length after the electric automobile is charged; />Off-grid time counter for t-period electric automobile after being powered off>The minimum off-grid time is the minimum off-grid time after the electric automobile is discharged; />On-line charging time counter for t-period electric automobile>The method comprises the steps of (1) enabling an electric automobile to be in a network for a minimum charging time; />For t period electric automobile online discharge duration counter, < >>The on-grid minimum discharge time length of the electric automobile is set; a (Y) is a set of devices in the electric vehicle charging station Y, Y is a set of electric vehicle charging stations,/->And->Maximum charging power and discharging power of the electric vehicle charging station y respectively;
b) Demand response load constraints
The demand response load comprises an interruptible load, and the constraint condition is an interruptible load limit constraint;
wherein:an interrupt load amount of the interruptible load d for the t period; />The participation state of the interruptible load d in the t period is 1, and the interrupt state is represented when the participation state is 0, and the interrupt state is not represented when the participation state is 0; alpha d The ratio of the interruptible load capacity to the total load;
c) Uninterruptible power supply restraint
The uninterrupted power supply constraint comprises uninterrupted power supply capacity constraint and uninterrupted power supply charge-discharge power constraint;
O u,T ＝O u,0
wherein: u is an uninterruptible power supply set;and->The charging power and the latter discharging power of the uninterruptible power supply u in the t period are respectively; o (O) u,t For the capacity of the uninterrupted power supply u in the period of t, eta u The charging and discharging efficiency of the uninterrupted power supply is improved; />And->The upper limit and the lower limit of the u capacity of the uninterrupted power supply are set; o (O) u,T And O u,0 The capacities of the uninterruptible power supply u at the end time and the initial time in the scheduling period are respectively set; />And->Respectively represent the upper limit and the lower limit of the discharge power of the uninterruptible power supply u, < >>The discharge state of the uninterruptible power supply u in the t period is 1, and the discharge state is 0, so that the discharge state is not discharged; />And->Respectively representing the upper limit and the lower limit of the charging power of the uninterruptible power supply u, < >>The charging state of the uninterruptible power supply u in the t period is 1, and the charging state is 0, namely the charging state is not charged;
the sum of the charge and discharge power of each cloud energy storage user in the lower model is equal to the charge and discharge power of each cloud energy storage system obtained by solving the upper model;
wherein: a (e) is a device set of the cloud energy storage system e;
2) Power distribution network tide constraint
Linearizing power flow constraint of a power distribution network by adopting a second order cone relaxation method, wherein the operation constraint of the power distribution network comprises node power balance constraint of the power distribution network, line voltage drop limit constraint of the power distribution network, second order cone constraint and line operation constraint; wherein, the network loss difference value of the upper layer model and the lower layer modelEqual to the estimated net loss value P of the upper model t loss And the actual net loss value of the lower model +.>Is a difference in (2);
wherein: a (i) is a collection of distribution network devices directly connected to node i; l and N bus Respectively isA collection of power transmission lines and nodes of the power distribution network; s (l) and r (l) are a transmitting end node and a receiving end node of the power transmission line l; p (P) l,t And Q l,t Active power flow and reactive power flow of the power transmission line l in the t period are respectively;and->The active network loss and the reactive network loss of the power transmission line l in the t period are calculated; />The network loss difference value between the upper layer model and the lower layer model; omega l,t Is the current square value of the power transmission line l in the t period, and ψ j,t Sum phi i,t The voltage square values of the node j and the node i in the t period are respectively; r is (r) l And x l The resistance value and the reactance value of the power transmission line l are respectively; />And->Square values, ψ, representing upper and lower limits, respectively, of current permitted to flow through transmission line l i max Sum phi i min The square values of the upper and lower voltage limits that node i can withstand are shown, respectively.
5. The optimal scheduling method for the power distribution network-cloud energy storage system according to claim 1, wherein in the step 3, the wind-light output uncertainty model is considered as follows:
wherein: superscript sce indicates the value of the variable under scene sce, and γ represents the set of typical wind and light output scenes; The emergency climbing rate of the gas unit; />Is an adjustable range of substation transmission power in an uncertainty scene.
6. The optimal scheduling method for the power distribution network-cloud energy storage system according to claim 1, wherein in the step 4, in a double-layer optimal scheduling model of the power distribution network-cloud energy storage system
(1) In the upper model, the running cost of the power distribution network in the objective function comprises the use cost of the cloud energy storage system, the electricity purchasing cost of the upper power grid, the electricity generating cost of the gas turbine set and the electricity discarding penalty of the renewable energy, and the expression of the objective function is as follows:
minF 1 ＝F M +F E +F G +F Voll
wherein: f (F) 1 F for the running cost of the distribution network M F, purchasing electricity cost for upper-level power grid of power distribution network E F for the use cost of the cloud energy storage system G F is the power generation cost of the gas unit Voll The cost is punished for renewable energy source waste electricity;
wherein:
wherein:the electricity purchase price corresponding to the t period;
when the power distribution network needs to charge/discharge the cloud energy storage system, the cloud energy storage operator pays corresponding service cost, and the cloud energy storage operator only needs to pay net charge cost to the power distribution network for the net charge part of the cloud energy storage system;
wherein:service unit price for charging/discharging of cloud energy storage system e, < >>A net charging unit price for the cloud energy storage system e;
Wherein: m is M g (. Cndot.) is a heat rate curve,fuel unit price for period t;
wherein:and->The unit price of electricity discarding punishment of the photovoltaic generator set and the wind turbine generator set is respectively;
(2) In the lower model, the main body is a cloud energy storage system, and the profit of a cloud energy storage operator in the objective function comprises the following parts:
max(F 2 -F P )＝F E -(F U +F V +F I )-F P
wherein: f (F) 2 The total profit obtained for the cloud energy storage operator is equal to the cloud energy storage system use cost paid by the power distribution network to the cloud energy storage operator minus the cloud energy storage user scheduling cost paid by the cloud energy storage operator to the cloud energy storage user, F U Scheduling costs for uninterruptible Power supply users, F V Scheduling cost for electric automobile user, F I Scheduling costs for interruptible load users, F P Punishment cost is conducted on the network loss difference value between the upper layer model and the lower layer model;
the cloud energy storage operator pays the discharging/charging service charge to the electric automobile user, and the electric automobile user only needs to pay the net charging charge for the net charging part of the electric automobile;
wherein:the unit price of charge/discharge service of the electric automobile; />The net charging unit price of the electric automobile;
the demand response load only considers the interruptible load of a user, and the cloud energy storage system induces the user to cut down electricity consumption in a compensation mode to be used as a part of the discharge power of the cloud energy storage system;
Wherein:giving a compensation unit price capable of interrupting load to a cloud energy storage system operator;
the cloud energy storage operator pays the discharge/charge service charge to the uninterruptible power supply user;
wherein:the unit price is used for charging/discharging of the uninterrupted power supply;
when solving the output plans of all units in the power distribution network and the charge and discharge plans of the cloud energy storage system, the upper model does not calculate the power flow of the power distribution network, so that the network loss in the active power balance constraint is a predicted value; when a lower model makes a charge and discharge plan of each cloud energy storage user, carrying out power flow calculation of the power distribution network to obtain an actual network loss value; the network loss difference value is the difference value between the actual network loss and the estimated network loss;
wherein: c p Penalty unit price for the net loss difference.
7. The optimal scheduling method for the power distribution network-cloud energy storage system according to claim 1, wherein the process of step 5 is specifically as follows:
based on basic prediction scene of wind power and photovoltaic power generation, wind power output prediction error is assumedAnd photovoltaic output prediction errorMeets the standard normal distribution, i.e.)>Generating an uncertainty output scene by a Monte Carlo simulation method, wherein in the scene sce, the output of wind power and photovoltaic power are respectively as follows:
Wherein:and->The predicted force values of the wind driven generator w and the photovoltaic generator s in the t period under the scene sce are respectively; />And->Respectively predicting force values of the wind driven generator w and the photovoltaic generator s in a t period under a basic prediction scene;and->The output prediction error values of the wind driven generator w and the photovoltaic generator s in the t period under the scene sce are respectively;
the scene is reduced by a synchronous generation method, a plurality of typical scenes are screened out for model solving, so that the aim of improving the solving efficiency is fulfilled, and the scene reduction steps are as follows:
step a: setting SS as an initial scene set; DS is a scene set to be deleted, and the initial state is an empty set; the Euclidean distance between each scene in the initial scene set SS is calculated:
wherein: DT (DT) r,k For the (r) th scene omega r And the kth scene Ω k Euclidean distance between, P t r And P t k The method comprises the steps that a force value is predicted for a wind-solar unit in a t period under a scene r and a scene k respectively; omega shape u Is a set of all scenes;
step b: calculating scene Ω r And all of themProbability distance PD of his scene k (r) the smaller the value, the higher the similarity of the scene with other scenes;
wherein: ρ r Is scene omega r Is a probability of occurrence of (1); r=1, 2,..n, N is the total number of scenes;
Step c: calculating the probability distance of each scene, if one scene omega exists b Probability distance PD from all other scenes k (b)＝minPD k (r); then select scene Ω b As a scene to be deleted;
step d: calculating each scene and scene omega in the initial scene set SS b If DT is the Euclidean distance of b,d ＝minDT b,k K noteqb, b = 1,2, N; then delete scene Ω b Scene omega b The probability of occurrence is added to scene Ω d Applying; updating scene sets and occurrence probabilities of various scenes:
SS＝SS-{Ω b }；DS＝DS+{Ω b }
ρ d ＝ρ d +ρ b ；ρ b ＝0
step e: repeating the steps b-d until the number of scenes meets the target set value.
8. The optimal scheduling method for the power distribution network-cloud energy storage system according to claim 1, wherein the operation parameters of the power distribution network-cloud energy storage system comprise the output cost, climbing capacity, minimum start-stop time, electric load size, power limit transmitted by a power transmission line, current limit of the power transmission line, voltage limit of a node, charge-discharge service cost of the cloud energy storage system, capacity limit of the cloud energy storage system, charge-discharge power limit of the cloud energy storage system, charge-discharge service cost of an electric vehicle user, capacity limit of the electric vehicle user, charge-discharge power limit of the electric vehicle user, charge-discharge service cost of an uninterruptible power supply user, capacity limit of the uninterruptible power supply user, charge-discharge power limit of the uninterruptible power supply user, charge-discharge compensation cost of a demand response load and capacity limit of a demand response load.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311724876.6A CN117713240A (en) | 2023-12-14 | 2023-12-14 | Optimal scheduling method for power distribution network-cloud energy storage system considering wind-solar uncertainty |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311724876.6A CN117713240A (en) | 2023-12-14 | 2023-12-14 | Optimal scheduling method for power distribution network-cloud energy storage system considering wind-solar uncertainty |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117713240A true CN117713240A (en) | 2024-03-15 |
Family
ID=90154817
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311724876.6A Pending CN117713240A (en) | 2023-12-14 | 2023-12-14 | Optimal scheduling method for power distribution network-cloud energy storage system considering wind-solar uncertainty |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117713240A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117937550A (en) * | 2024-03-25 | 2024-04-26 | 宁波市电力设计院有限公司 | Optimized scheduling method and system for comprehensive energy storage system |
-
2023
- 2023-12-14 CN CN202311724876.6A patent/CN117713240A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117937550A (en) * | 2024-03-25 | 2024-04-26 | 宁波市电力设计院有限公司 | Optimized scheduling method and system for comprehensive energy storage system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zheng et al. | Online distributed MPC-based optimal scheduling for EV charging stations in distribution systems | |
| CN110188950B (en) | Multi-agent technology-based optimal scheduling modeling method for power supply side and demand side of virtual power plant | |
| CN109559035B (en) | Urban distribution network double-layer planning method considering flexibility | |
| Kumar et al. | V2G capacity estimation using dynamic EV scheduling | |
| Ghofrani et al. | A framework for optimal placement of energy storage units within a power system with high wind penetration | |
| CN108470239A (en) | The active distribution network multi objective layered programming method of meter and demand side management and energy storage | |
| CN111404206B (en) | Wind-solar energy storage power generation system capacity double-layer planning method considering investment return constraint | |
| Alabi et al. | Improved hybrid inexact optimal scheduling of virtual powerplant (VPP) for zero-carbon multi-energy system (ZCMES) incorporating Electric Vehicle (EV) multi-flexible approach | |
| CN105139147A (en) | Economic scheduling method for micro-grid system | |
| CN117713240A (en) | Optimal scheduling method for power distribution network-cloud energy storage system considering wind-solar uncertainty | |
| CN102593855B (en) | Method for stabilizing fluctuation of output power of renewable energy power supply in power system | |
| Brenna et al. | From virtual power plant (VPP) to sustainable energy microsystem (SEM): An opportunity for buildings energy management | |
| CN115117940A (en) | Wind power, photovoltaic and load uncertainty considered wind-light-water-fire storage system low-carbon scheduling model modeling method | |
| CN111639866A (en) | Energy storage capacity configuration method of energy storage charging station based on genetic algorithm | |
| CN110826210B (en) | Multi-region building virtual power plant modeling and optimization coordination method based on power interconnection | |
| CN112182915A (en) | Optimized scheduling method and system for cooperatively promoting wind power consumption | |
| CN116402307A (en) | Power grid planning capacity analysis method considering operation characteristics of schedulable flexible resources | |
| CN114862163B (en) | Optimized scheduling method of comprehensive energy system | |
| Morrissey et al. | Optimal energy storage schedules for load leveling and ramp rate control in distribution systems | |
| CN114491997B (en) | Virtual power plant operation optimization method and system considering demand response and electric automobile | |
| CN116131318A (en) | Two-stage robust optimization control method and device for toughness-oriented lifting active power distribution network | |
| CN106253356B (en) | Alternating current-direct current mixing microgrid Real-time Economic Dispatch method based on energy storage electricity value assessment | |
| Yao et al. | Determination of a dispatch strategy to maximize income for a wind turbine-BESS power station | |
| Xu et al. | Day-ahead battery scheduling in microgrid considering wind power uncertainty using ordinal optimization | |
| Erdoğan et al. | Coordinated electric vehicle charging strategy in microgrids containing PV system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |