CN105811397A - Multi-energy complementation microgrid scheduling method based on multi-time scales - Google Patents

Abstract

The invention relates to a multi-energy complementation microgrid scheduling method based on multi-time scales. According to the technical points, the method comprises the following steps of: 1, setting a microgrid system scene, and respectively modeling a combined cooling heating and power device, an ice-storage air conditioner and a storage battery device in the microgrid system scene; 2, based on photovoltaic and wind power output scene generation and elimination technologies, analyzing the uncertainties of photovoltaic output and wind power output, and eliminating the fluctuation influences of the output of photovoltaic and wind power renewable energy sources; 3, establishing day-ahead combined optimization scheduling model; 4, establishing a microgrid real-time optimization scheduling model; 5, based on an improved particle swarm optimization algorithm, solving the microgrid day-ahead combined optimization scheduling model and the microgrid real-time optimization scheduling model, and obtaining a multi-energy complementation microgrid operation strategy under multi-time scales. According to the invention, the operation cost of the microgrid is lowered, the utilization efficiency of energy is fully improved, and the method has an important effect on the aspect of peak shaving and load shifting of a power grid.

Description

A kind of microgrid dispatching method of providing multiple forms of energy to complement each other based on Multiple Time Scales

Technical field

The present invention relates to microgrid dispatching technique field, particularly a kind of microgrid dispatching method of providing multiple forms of energy to complement each other based on Multiple Time Scales.

Background technology

At present, microgrid is as the important component part of intelligent grid, in reducing energy consumption, improving Power System Reliability and motility etc., there is great potential, and the microgrid management and running strategy study of providing multiple forms of energy to complement each other under Multiple Time Scales is one of the primary study direction in intelligent grid field.Existing microgrid scheduling strategy does not take into full account the undulatory property of the regenerative resource such as wind-powered electricity generation, photovoltaic, or only considers the impact of regenerative resource undulatory property in scheduling a few days ago, and to its undulatory property research deficiency in Real-Time Scheduling；It addition, ice-storage air-conditioning operational mode is complicated, almost without the micro-capacitance sensor model comprising ice-storage air-conditioning in current research；Situation numerous for equipment in micro-grid system, that constraint is complicated, finds the microgrid Optimization scheduling algorithm being suitable under multiple target Multiple Time Scales, is a pendulum outstanding problem in face of microgrid builder.

Summary of the invention

It is an object of the invention to overcome the deficiencies in the prior art, a kind of microgrid dispatching method of providing multiple forms of energy to complement each other based on Multiple Time Scales is provided, set up the microgrid model of providing multiple forms of energy to complement each other comprising ice-storage air-conditioning, solve the regenerative resource such as wind-powered electricity generation, photovoltaic undulatory property problem under scheduling a few days ago and Real-Time Scheduling.

This invention address that it technical problem is that takes techniques below scheme to realize:

A kind of microgrid dispatching method of providing multiple forms of energy to complement each other based on Multiple Time Scales, comprises the following steps:

Step 1, micro-grid system scene is set and respectively cold, heat and electricity triple supply equipment, ice-storage air-conditioning and accumulator equipment in micro-grid system scene is modeled；

Step 2, generate based on photovoltaic and wind power output scene and elimination technique is analyzed photovoltaic and exerted oneself and the uncertainty of wind power output, eliminate photovoltaic and undulatory property impact that wind-powered electricity generation regenerative resource is exerted oneself；

Step 3, set up with the minimum joint optimal operation model a few days ago for target of microgrid operating cost；

Step 4, set up with real-time exchange power and the scheduled net interchange difference real-time joint optimal operation model of minimum microgrid under the Multiple Time Scales of target；

Step 5, based on improved Particle Swarm Optimization, described microgrid joint optimal operation model a few days ago and the real-time joint optimal operation model of microgrid are solved, obtain the microgrid operation reserve of providing multiple forms of energy to complement each other under Multiple Time Scales.

And, concretely comprising the following steps of described step 1: respectively set up such as drag peace treaty bundle conditional equation: the electric power efficiency model of cold, heat and electricity triple supply equipment, electricity exert oneself and cold exert oneself between relational model and constraining equation；Ice-storage air-conditioning is at air conditioning mode, ice-make mode, ice-melt mode and air-conditioning and ice-melt model and the constraining equation of exerting oneself in composite mode；Accumulator is dispatched and the operation constraining equation under Real-Time Scheduling a few days ago；

(1) the electric power efficiency model of described cold, heat and electricity triple supply equipment, electricity exert oneself and cold exert oneself between relational model and constraining equation be respectively as follows:

1. the electric power efficiency model of cold, heat and power triple supply system:

$E_{CCHP}\left(T\right)=f\left(P_{CCHP}\right(T\left)\right)=f_{ISO}\left(\frac{P_{CCHP}\left(T\right)}{P_{ISO-\max }}\right)\×\frac{E_{max}}{f_{ISO}\left(\frac{P_{max}}{P_{ISO-max}}\right)}$

In formula, P_CCHP(T) represent that the electricity of T moment CCHP is exerted oneself, P_maxWith E_maxRepresenting that under current operating conditions, the maximum electricity of CCHP is exerted oneself and maximum electrical efficiency respectively, ISO represents standard condition, and f is the nonlinear function that corresponding unit is given；E_CCHP(T) it is the electrical efficiency of gas turbine.

2. the electricity of cold, heat and power triple supply system exert oneself and cold exert oneself between relational model:

A. microgrid with electricity refrigeration mode under electricity exert oneself and cold exert oneself between relational model:

Q_CCHP(T)=g (P_CCHP(T))

In formula, Q_CCHP(T) cold the exerting oneself of T moment CCHP, P are represented_CCHP(T) represent that the electricity of T moment CCHP is exerted oneself；Functional relationship between g function representation T moment gas turbine cold is exerted oneself and electricity is exerted oneself.

B. microgrid with under cold power mode processed electricity exert oneself and cold exert oneself between relational model:

P_CCHP(T)=g^-1(P_CCHP(T))

In formula, P_CCHP(T) represent that the electricity of T moment CCHP is exerted oneself；g^-1Functional relationship between function representation T moment gas turbine cold is exerted oneself and electricity is exerted oneself.

3. the constraining equation that micro-grid system meets is:

$\left\{\begin{array}{c}{I}_{CCHP}\left(T\right)=0,T\∈T_{valley}\\ I_{CCHP}\left(T\right)P_{min}\≤P_{CCHP}\left(T\right)\≤I_{CCHP}\left(T\right)P_{max}\\ I_{CCHP}\left(T\right)Q_{\min }\≤Q_{CCHP}\left(T\right)\≤Q_{CCHP}\left(T\right)Q_{max}\end{array}\right.$

Wherein, I_CCHP(T)∈(0,1)

In formula, I_CCHP(T) on off state of T moment CCHP equipment is represented；P_CCHP(T) represent that the electricity of T moment CCHP is exerted oneself；Q_CCHP(T) cold the exerting oneself of T moment CCHP is represented；P_maxWith P_minRepresent under current operating conditions that the maximum electricity of CCHP is exerted oneself and minimum electricity is exerted oneself respectively；Q_maxWith Q_minRepresent under current operating conditions that the maximum cold of CCHP is exerted oneself and minimum cold exerted oneself respectively；T_valleyRepresent electric load paddy period.

(2) described ice-storage air-conditioning exerts oneself model and constraining equation is respectively as follows: air conditioning mode, ice-make mode, ice-melt mode and air-conditioning and ice-melt are in composite mode

1. ice-storage air-conditioning model of exerting oneself under air conditioning mode is:

$P_a\left(T\right)=\frac{Q_a\left(T\right)}{a_1{Q}_a\left(T\right)+a_2}$

In formula, P_aAnd Q (T)_a(T) T moment ice-storage air-conditioning power consumption under air conditioning mode and refrigerating capacity are represented respectively；a₁With a₂For constant；

Its constraints is:

$\left\{\begin{array}{c}{I}_a\left(T\right)=0,T\∈T_{valley}\\ I_a\left(T\right)Q_{a-\min }\≤Q_a\left(T\right)\≤I_a\left(T\right)Q_{a-max}\end{array}\right.$

In formula, I_a(T) on off state of T moment air conditioning mode is represented；Q_a(T) T moment ice-storage air-conditioning refrigerating capacity under air conditioning mode is represented；T_valleyRepresent electric load paddy period；Q_a-maxWith Q_a-minRepresent the minimum and maximum refrigerating capacity under air conditioning mode respectively；

2. ice-storage air-conditioning under ice-make mode model of exerting oneself be:

$P_c\left(T\right)=\frac{Q_c\left(T\right)}{a_1{Q}_c\left(T\right)+a_2}$

In formula, P_cAnd Q (T)_c(T) T moment ice-storage air-conditioning power consumption under air conditioning mode and refrigerating capacity are represented respectively；a₁With a₂For constant；

Its constraints is:

$\left\{\begin{array}{c}{I}_c\left(T\right)=0,T\∉T_{valley}\\ \underset{T=T_0}{\overset{T_0+23}{\Σ}}|I_c\left(T\right)-I_c(T-1)|=2\\ Q_c\left(T\right)=I_c\left(T\right)Q_{a-max}\end{array}\right.$

In formula, I_cAnd I (T)_c(T-1) on off state of T moment and T-1 moment ice-make mode is represented respectively；Q_c(T) T moment ice-storage air-conditioning refrigerating capacity under air conditioning mode is represented；T_valleyRepresent electric load paddy period；T₀Represent the start time of paddy period；Q_a-maxRepresent the maximum cooling capacity under air conditioning mode.

3. ice-storage air-conditioning constraints under ice-melt mode is:

$\left\{\begin{array}{c}{I}_d\left(T\right)=0,T\∈T_{valley}\\ 0\≤Q_d\left(T\right)=I_d\left(T\right)Q_{d-max}\end{array}\right.$

In formula, I_d(T) on off state of T moment ice-melt mode is represented；Q_dAnd Q (T)_d-maxRepresent T moment ice-storage air-conditioning refrigerating capacity under ice-melt mode and maximum cooling capacity respectively；T_valleyRepresent electric load paddy period；

4. ice-storage air-conditioning at the model of exerting oneself in composite mode of air conditioning mode and ice-melt mode is:

IS (T)=(1-η₁)IS(T-1)+η₂Q_c(T)-Q_d(T)

In formula, IS (T) represents the cold energy of storage in T moment Ice Storage Tank；η₁With η₂Represent the refrigerating efficiency under cold energy storage loss factor and air conditioning mode respectively；Q_c(T) T moment ice-storage air-conditioning refrigerating capacity under air conditioning mode is represented；Q_d(T) T moment ice-storage air-conditioning refrigerating capacity under ice-melt mode is represented；

Its constraints is:

${\mathrm{IS}}_{min}\left(T\right)\≤IS\left(T\right)\≤(1-{\mathrm{\η}}_1)IS(T-1),T\∉T_{valley}$

In formula, IS_min(T) the minimum cold energy needing storage in Ice Storage Tank is represented；IS (T) represents the cold energy of storage in T moment Ice Storage Tank；η₁Represent cold energy storage loss factor；T_valleyRepresent electric load paddy period；

(3) described accumulator is dispatched and the operation constraining equation under Real-Time Scheduling a few days ago；

1. accumulator dispatch a few days ago under operation constraining equation:

$\left\{\begin{array}{c}SOC(T+1)=SOC\left(T\right)-\left({\mathrm{\η}}_{b\_c}{I}_{b\_c}\right(T)+I_{b\_d}(T)/{\mathrm{\η}}_{b\_d})P_b\left(T\right)\mathrm{\Δ}t/c_b\\ {\mathrm{\η}}_{b\_c}\left(T\right)P_{c-\min }+I_{b\_d}\left(T\right)P_{d-\min }\≤P_b\left(T\right)\≤I_{b\_c}\left(T\right)P_{c-\max }+I_{b\_d}\left(T\right)P_{d-\max }\\ I_{b\_c}\left(T\right)+I_{b\_d}\left(T\right)\∈(0,1)\\ {\mathrm{SOC}}_{\min }\≤SOC(T+1)\≤{\mathrm{SOC}}_{\max }\end{array}\right.$

In formula, SOC represents state-of-charge；SOC_maxWith SOC_minRepresent the bound of state-of-charge respectively；η_{b_c}With η_{b_d}Represent discharge and recharge coefficient respectively；P_c-maxWith P_c-minRepresent maximum charge power and minimum charge power respectively；P_d-maxWith P_d-minRepresent maximum discharge power and minimum discharge power respectively；I_{b_c}(T) on off state of T moment accumulator charging, I are represented_{b_c}(T)∈(0,1)；I_{b_d}(T) on off state of T moment battery discharging, I are represented_{b_d}(T)∈(0,1)；P_b(T) charge-discharge electric power of T moment accumulator is represented；Δ t represents interval of time；c_bRepresent the capacity of accumulator.

2. the operation constraining equation under accumulator Real-Time Scheduling:

$\left\{\begin{array}{c}SOC(t+i+1|t)=SOC(t+i|t)-\left({\mathrm{\η}}_{b\_c}{I}_{b\_c}\right(t+i|t)+I_{b\_d}(t+i|t)/{\mathrm{\η}}_{b\_d})P_{b\_r}(t+i|t)\mathrm{\Δ}t/c_b\\ I_{b\_c}(t+i|t)P_{c-\min }+I_{b\_d}(t+i|t)P_{d-\min }\≤P_{b\_r}(t+i|t)\≤I_{b\_c}(t+i|t)P_{c-\max }+I_{b\_d}(t+i|t)P_{d-\max }\\ I_{b\_c}(t+i|t)+I_{b\_d}(t+i|t)\∈(0,1)\\ {\mathrm{SOC}}_{\min }\≤SOC(t+i+1|t)\≤{\mathrm{SOC}}_{\max }\end{array}\right.$

In formula, t+i | t represents the dispatch value of i step forward；SOC (t+i+1 | t) and SOC (t+i | t) represent the state-of-charge that accumulator i+1 forward walks and i walks the moment forward respectively；η_{b_c}With η_{b_d}Represent discharge and recharge coefficient respectively；P_c-maxWith P_c-minRepresent maximum charge power and minimum charge power respectively；P_d-maxWith P_d-maxRepresent maximum discharge power and minimum discharge power respectively；I_{b_c}(t+i | t) represent that i walks the on off state of moment accumulator charging forward；I_{b_d}(t+i | t) represent that i walks the on off state of moment battery discharging forward；P_{b_r}(t+i | t) represent that under Real-Time Scheduling, i walks the charge-discharge electric power of moment accumulator forward；Δ t represents interval of time；c_bRepresent the capacity of accumulator.

And, the concrete steps of described step 2 include:

(1) based on wind power output and photovoltaic power generation output forecasting, set up following wind-powered electricity generation and photovoltaic exerted oneself normal distribution probability model:

$P_{pv}\left(T\right)~N({\mathrm{\μ}}_{pv},{\mathrm{\σ}}_{pv}^2)$

$P_{wind}\left(T\right)~N({\mathrm{\μ}}_{wind},{\mathrm{\σ}}_{wind}^2)$

Wherein, σ_pv=0.1 μ_pv；σ_wind=0.1 μ_wind

In formula, μ_pvWith μ_windIt is that photovoltaic is exerted oneself and the predictive value of wind power output respectively；σ_pvWith σ_windIt it is corresponding variance；P_pvAnd P (T)_wind(T) it is T moment photovoltaic respectively and the actual of wind-powered electricity generation is exerted oneself；N represents normal distribution；

(2) use LHS method to generate the photovoltaic and wind power output scene obeying above-mentioned probabilistic model, use scene elimination technique eliminate low probability scene therein and scene strong for dependency merged.

And, described step 2 (2nd) step generates comprising the concrete steps that of photovoltaic and wind power output scene:

1. variable x will often be tieed up after determining sample size HⁱDefinition territory intervalIt is divided into H equal minizone so thatAn original super cube is divided into HⁿIndividual small cubes；

2. generate a H × n matrix A, then every string of matrix A be all ordered series of numbers 1,2, a random fully intermeshing of H}；The corresponding selected little hypercube of each row of A, if randomly generating a sample in each little hypercube, then selecting H sample, obeying wind-powered electricity generation described in step 2 (1st) step and photovoltaic is exerted oneself the photovoltaic of normal distribution probability model and wind power output scene thus generating.

And, the concrete steps of described step 3 include:

(1) with the minimum object function setting up joint optimal operation model a few days ago for target of microgrid day total operating cost:

${\mathrm{minC}}_{\exp }=\underset{s}{\Σ}{p}_s(\underset{T=T_0}{\overset{T_0+23}{\Σ}}{c}_{Grid}\left(T\right)P_{Grid}^s\left(T\right)+\underset{T=T_0}{\overset{T_0+23}{\Σ}}{c}_{Gas}F\left(T\right))$

In formula, p_sIt it is the scene s probability occurred；It is the exchange power of microgrid and major network under scene s；F (T) is gas consumption；c_GridAnd c (T)_GasIt is electricity price and gas price respectively；T₀Represent the start time of paddy period；

(2) constraints that the electric load of microgrid, refrigeration duty joint optimal operation a few days ago runs is set up

1. the constraints that the electric load of microgrid joint optimal operation a few days ago runs

$P_{wind}^s\left(T\right)+P_{pv}^s\left(T\right)+P_{CCHP}\left(T\right)+P_b\left(T\right)+P_{Grid}^s\left(T\right)=P_{load}\left(T\right)+P_a\left(T\right)+P_c\left(T\right)+P_d\left(T\right)$

In formula, P_load(T) it is the electric load in microgrid T moment；P_CCHP(T) represent that the electricity of T moment CCHP is exerted oneself；P_a(T) T moment ice-storage air-conditioning power consumption under air conditioning mode is represented；P_b(T) charge-discharge electric power of T moment accumulator is represented；P_c(T) T moment ice-storage air-conditioning power consumption under air conditioning mode is represented；P_d(T) power consumption in T moment under ice-melt mode is represented；Represent the wind power output in T moment under the s scene；Represent that under the s scene, the photovoltaic in T moment is exerted oneself；Represent the exchange power between T moment microgrid and major network under the s scene.

2. the constraints that the refrigeration duty of microgrid joint optimal operation a few days ago runs

Q_CCHP(T)+Q_a(T)+Q_d(T)=Q_load(T)

In formula, Q_CCHP(T) cold the exerting oneself of T moment CCHP is represented；Q_d(T) T moment ice-storage air-conditioning refrigerating capacity under ice-melt mode is represented；Q_a(T) T moment ice-storage air-conditioning refrigerating capacity under air conditioning mode is represented respectively；Q_load(T) the refrigeration duty demand a few days ago predicting T of lower moment is represented；

And, the concrete steps of described step 4 include:

(1) object function of the real-time joint optimal operation model of microgrid under setting up Multiple Time Scales with scheduled net interchange difference is minimum for target with real-time exchange power:

$Obj=\underset{P_{b\_r}(t+i|t)}{min}\underset{i=0}{\overset{11}{\Σ}}|P_{Grid}\left(T\right)-P_{Grid\_r}(t+i|t)|\×p,t+i\∈T$

In formula, P_{Grid_r}(t+i | t) is the real-time exchange power of microgrid and electrical network, and p is coefficient；P_{b_r}(t+i | t) represent that under Real-Time Scheduling, i walks the charge-discharge electric power of moment accumulator forward；P_Grid(T) the exchange power between T moment microgrid and major network is represented；Obj represents object function.

(2) constraints that the real-time joint optimal operation of electric load of microgrid runs is set up

P_{wind_r}(t+i|t)+P_{pv_r}(t+i|t)+P_{CCHP_r}(T)+P_{b_r}(t+i|t)+P_{Grid_r}(t+i|t)

=P_{load_r}(t+i|t)+P_{a_r}(T)+P_c(T)+P_d(T),t+i∈T

In formula, r represents Real-Time Scheduling；P_{wind_r}(t+i | t) represent that under Real-Time Scheduling, i walks the wind power output in moment forward；P_{pv_r}(t+i | t) represent that the photovoltaic in i step moment forward under Real-Time Scheduling is exerted oneself；P_{CCHP_r}(T) represent that under Real-Time Scheduling, the electricity of T moment gas turbine is exerted oneself；P_{b_r}(t+i | t) represent that under Real-Time Scheduling, i walks the charge-discharge electric power of moment accumulator forward；P_{Grid_r}(t+i | t) represents the real-time exchange power of microgrid and electrical network；P_{load_r}(t+i | t) represent under Real-Time Scheduling forward i walk the electric load predictive value in moment；P_{a_r}(T) power consumption in T moment under Real-Time Scheduling and air conditioning mode is represented；P_c(T) T moment ice-storage air-conditioning power consumption under air conditioning mode is represented；P_d(T) power consumption in T moment under ice-melt mode is represented.

And, the concrete steps of described step 5 include:

(1) choose the variable of applicable improved Particle Swarm Optimization as search particle, set up and be updated to the improved Particle Swarm Optimization model of feature with particle position renewal and region of search and initialize population；

(2) use this improved Particle Swarm Optimization model solution to optimize and Real time optimal dispatch model a few days ago, obtain the microgrid operation reserve of providing multiple forms of energy to complement each other under Multiple Time Scales.

And, described step 5 (2nd) step method particularly includes:

Evaluate the fitness Fit that in each population, particle is individual, if Fit is < pbest_iThen replace pbest with Fit_i；Otherwise, then continue to judge pbest_iWith the relation of gbest, if pbest_i< gbest, then use pbest_iReplace gbest；Otherwise, then particle position and region of search position are updated；Then judging whether calculated all particles and whether met end condition, if meeting, terminating to calculate；

Wherein, pbesti is the optimal solution individuality extreme value that particle itself finds；Gbest is the optimal solution global extremum that whole population is found at present.

Advantages of the present invention and having the benefit effect that

1, the actual features that the present invention runs in conjunction with microgrid, exemplary apparatus in micro-grid system is modeled, especially the impact of ice-storage air-conditioning is considered, use scene generation and elimination technique to solve the undulatory property problem of the regenerative resource such as wind-powered electricity generation, photovoltaic, use the microgrid Optimal Scheduling updating under the Modified particle swarm optimization Algorithm for Solving multiple target multiple constraint Multiple Time Scales being updated to feature with region of search with particle position.The present invention can reduce the operating cost of microgrid, fully improves the service efficiency of the energy, and plays a significant role in the peak load shifting of electrical network.

2, first the equipment such as the cold, heat and electricity triple supply equipment in microgrid, ice-storage air-conditioning, accumulator are modeled by the present invention；The uncertainty with elimination technique research photovoltaic and wind power output is generated secondly based on scene；Then set up with the real-time joint optimal operation model of microgrid under the minimum model of joint optimal operation a few days ago for target of microgrid operating cost and Multiple Time Scales；It is finally based on modified particle swarm optiziation and solves the model of scheduling a few days ago and Real-Time Scheduling.The present invention establishes the microgrid model of providing multiple forms of energy to complement each other comprising ice-storage air-conditioning, take into full account that wind-powered electricity generation, photovoltaic in scheduling a few days ago and the undulatory property problem under Real-Time Scheduling and give full play to Modified particle swarm optimization algorithm advantage in solving the nonlinear optimal problem comprising stochastic variable, solved tradition microgrid dispatching method problem of the model solution under unsatisfactory and multiple target multiple constraint Multiple Time Scales in the undulatory property problem solving regenerative resource.

3, the equipment such as the cold, heat and electricity triple supply equipment in microgrid, ice-storage air-conditioning, accumulator have been carried out detailed modeling by the present invention, take into full account that photovoltaic and wind power output are in scheduling a few days ago and the uncertainty under Real-Time Scheduling, LHS technique study scene is used to generate and elimination technique, eliminate low probability scene and scene strong for dependency is merged, thus eliminating the undulatory property impact of the Renewable resource such as photovoltaic and wind-powered electricity generation.

4, the present invention establishes with microgrid operating cost minimum for target with real-time exchange power and the minimum model of joint optimal operation a few days ago for target of scheduled net interchange difference and real-time joint optimal operation model respectively；On the basis of conventional particle group's algorithm, establishing the Modified particle swarm optimization algorithm model being updated to feature with particle position renewal and region of search, the microgrid scheduling aspect of providing multiple forms of energy to complement each other under Multiple Time Scales achieves good effect.

Accompanying drawing explanation

Fig. 1 is the dispatching method overview flow chart of the present invention；

Fig. 2 is the real-time joint optimal operation flow chart of microgrid refrigeration duty of the present invention；

Fig. 3 is based on improvement particle cluster algorithm and carries out microgrid joint optimal operation a few days ago and the real-time joint optimal operation flow chart of microgrid.

Detailed description of the invention

Below in conjunction with accompanying drawing, the embodiment of the present invention is described in further detail:

A kind of microgrid dispatching method of providing multiple forms of energy to complement each other based on Multiple Time Scales, as it is shown in figure 1, comprise the following steps:

Step 1, micro-grid system scene is set and respectively cold, heat and electricity triple supply equipment, ice-storage air-conditioning and accumulator equipment in micro-grid system scene is modeled.

Concretely comprising the following steps of described step 1: respectively set up such as drag peace treaty bundle conditional equation: the electric power efficiency model of cold, heat and electricity triple supply equipment, electricity exert oneself and cold exert oneself between relational model and constraining equation；Ice-storage air-conditioning is at air conditioning mode, ice-make mode, ice-melt mode and air-conditioning and ice-melt model and the constraining equation of exerting oneself in composite mode；Accumulator is dispatched and the operation constraining equation under Real-Time Scheduling a few days ago；

(1) the electric power efficiency model of described cold, heat and electricity triple supply equipment, electricity exert oneself and cold exert oneself between relational model and constraining equation be respectively as follows:

1. the electric power efficiency model of cold, heat and power triple supply system:

$E_{CCHP}\left(T\right)=f\left(P_{CCHP}\right(T\left)\right)=f_{ISO}\left(\frac{P_{CCHP}}{P_{ISO-\max }}\right)\&times;\frac{E_{\max }}{f_{ISO}\left(\frac{P_{\max }}{P_{ISO-\max }}\right)}$

In formula, P_CCHP(T) represent that the electricity of T moment CCHP is exerted oneself, P_maxWith E_maxRepresenting that under current operating conditions, the maximum electricity of CCHP is exerted oneself and maximum electrical efficiency respectively, ISO represents standard condition, and f is the nonlinear function that corresponding unit is given；E_CCHP(T) it is the electrical efficiency of gas turbine.

2. the electricity of cold, heat and power triple supply system exert oneself and cold exert oneself between relational model:

A. microgrid with electricity refrigeration mode under electricity exert oneself and cold exert oneself between relational model:

Q_CCHP(T)=g (P_CCHP(T))

In formula, Q_CCHP(T) cold the exerting oneself of T moment CCHP, P are represented_CCHP(T) represent that the electricity of T moment CCHP is exerted oneself；Functional relationship between g function representation T moment gas turbine cold is exerted oneself and electricity is exerted oneself.

B. microgrid with under cold power mode processed electricity exert oneself and cold exert oneself between relational model:

P_CCHP(T)=g^-1(P_CCHP(T))

In formula, P_CCHP(T) represent that the electricity of T moment CCHP is exerted oneself；g^-1Functional relationship between function representation T moment gas turbine cold is exerted oneself and electricity is exerted oneself.

3. the constraining equation that micro-grid system meets is:

$\left\{\begin{array}{c}{I}_{CCHP}\left(T\right)=0,T\&Element;T_{valley}\\ I_{CCHP}\left(T\right)P_{min}\&le;P_{CCHP}\left(T\right)\&le;I_{CCHP}\left(T\right)P_{max}\\ I_{CCHP}\left(T\right)Q_{\min }\&le;Q_{CCHP}\left(T\right)\&le;Q_{CCHP}\left(T\right)Q_{max}\end{array}\right.$

Wherein, I_CCHP(T)∈(0,1)

In formula, I_CCHP(T) on off state of T moment CCHP equipment is represented；P_CCHP(T) represent that the electricity of T moment CCHP is exerted oneself；Q_CCHP(T) cold the exerting oneself of T moment CCHP is represented；P_maxWith P_minRepresent under current operating conditions that the maximum electricity of CCHP is exerted oneself and minimum electricity is exerted oneself respectively；Q_maxWith Q_minRepresent under current operating conditions that the maximum cold of CCHP is exerted oneself and minimum cold exerted oneself respectively；T_valleyRepresent electric load paddy period.

(2) described ice-storage air-conditioning exerts oneself model and constraining equation is respectively as follows: air conditioning mode, ice-make mode, ice-melt mode and air-conditioning and ice-melt are in composite mode

1. ice-storage air-conditioning model of exerting oneself under air conditioning mode is:

$P_a\left(T\right)=\frac{Q_a\left(T\right)}{a_1{Q}_a\left(T\right)+a_2}$

In formula, P_aAnd Q (T)_a(T) T moment ice-storage air-conditioning power consumption under air conditioning mode and refrigerating capacity are represented respectively；a₁With a₂For constant；

Its constraints is:

$\left\{\begin{array}{c}{I}_a\left(T\right)=0,T\&Element;T_{valley}\\ I_a\left(T\right)Q_{a-min}\&le;Q_a\left(T\right)\&le;I_a\left(T\right)Q_{a-max}\end{array}\right.$

In formula, I_a(T) on off state of T moment air conditioning mode is represented；Q_a(T) T moment ice-storage air-conditioning refrigerating capacity under air conditioning mode is represented；T_valleyRepresent electric load paddy period；Q_a-maxWith Q_a-minRepresent the minimum and maximum refrigerating capacity under air conditioning mode respectively；

2. ice-storage air-conditioning under ice-make mode model of exerting oneself be:

$P_c\left(T\right)=\frac{Q_c\left(T\right)}{a_1{Q}_c\left(T\right)+a_2}$

In formula, P_cAnd Q (T)_c(T) T moment ice-storage air-conditioning power consumption under air conditioning mode and refrigerating capacity are represented respectively；a₁With a₂For constant；

Its constraints is:

$\left\{\begin{array}{c}{I}_c\left(T\right)=0,T\&NotElement;T_{valley}\\ \underset{T=T_0}{\overset{T_0+23}{\&Sigma;}}|I_c\left(T\right)-I_c(T-1)|=2\\ Q_c\left(T\right)=I_c\left(T\right)Q_{a-max}\end{array}\right.$

In formula, I_cAnd I (T)_c(T-1) on off state of T moment and T-1 moment ice-make mode is represented respectively；Q_c(T) T moment ice-storage air-conditioning refrigerating capacity under air conditioning mode is represented；T_valleyRepresent electric load paddy period；T₀Represent the start time of paddy period；Q_a-maxRepresent the maximum cooling capacity under air conditioning mode.

3. ice-storage air-conditioning constraints under ice-melt mode is:

$\left\{\begin{array}{c}{I}_d\left(T\right)=0,T\&Element;T_{valley}\\ 0\&le;Q_d\left(T\right)=I_d\left(T\right)Q_{d-max}\end{array}\right.$

In formula, I_d(T) on off state of T moment ice-melt mode is represented；Q_dAnd Q (T)_d-maxRepresent T moment ice-storage air-conditioning refrigerating capacity under ice-melt mode and maximum cooling capacity respectively；T_valleyRepresent electric load paddy period；

4. ice-storage air-conditioning at the model of exerting oneself in composite mode of air conditioning mode and ice-melt mode is:

IS (T)=(1-η₁)IS(T-1)+η₂Q_c(T)-Q_d(T)

In formula, IS (T) represents the cold energy of storage in T moment Ice Storage Tank；η₁With η₂Represent the refrigerating efficiency under cold energy storage loss factor and air conditioning mode respectively；Q_c(T) T moment ice-storage air-conditioning refrigerating capacity under air conditioning mode is represented；Q_d(T) T moment ice-storage air-conditioning refrigerating capacity under ice-melt mode is represented；

Its constraints is:

${\mathrm{IS}}_{min}\left(T\right)\&le;IS\left(T\right)\&le;(1-{\mathrm{\&eta;}}_1)IS(T-1),T\&NotElement;T_{valley}$

In formula, IS_min(T) the minimum cold energy needing storage in Ice Storage Tank is represented；IS (T) represents the cold energy of storage in T moment Ice Storage Tank；η₁Represent cold energy storage loss factor；T_valleyRepresent electric load paddy period；

(3) described accumulator is dispatched and the operation constraining equation under Real-Time Scheduling a few days ago；

1. accumulator dispatch a few days ago under operation constraining equation:

$\left\{\begin{array}{c}SOC(T+1)=SOC\left(T\right)-({\mathrm{\&eta;}}_{b\_c}{I}_{b\_c}\left(T\right)+I_{b\_d}\left(T\right)/{\mathrm{\&eta;}}_{b\_d})P_b\left(T\right)\mathrm{\&Delta;}t/c_b\\ I_{b\_c}\left(T\right)P_{c-min}+I_{b\_d}\left(T\right)P_{d-min}\&le;P_b\left(T\right)\&le;I_{b\_c}\left(T\right)P_{c-max}+I_{b\_d}\left(T\right)P_{d-max}\\ I_{b\_c}\left(T\right)+I_{b\_d}\left(T\right)\&Element;(0,1)\\ SO{C}_{\min }\&le;SOC(T+1)\&le;SO{C}_{\max }\end{array}\right.$

In formula, SOC represents state-of-charge；SOC_maxWith SOC_minRepresent the bound of state-of-charge respectively；η_{b_c}With η_{b_d}Represent discharge and recharge coefficient respectively；P_c-maxWith P_c-minRepresent maximum charge power and minimum charge power respectively；P_d-maxWith P_d-minRepresent maximum discharge power and minimum discharge power respectively；I_{b_c}(T) on off state of T moment accumulator charging, I are represented_{b_c}(T)∈(0,1)；I_{b_d}(T) on off state of T moment battery discharging, I are represented_{b_d}(T)∈(0,1)；P_b(T) charge-discharge electric power of T moment accumulator is represented；Δ t represents interval of time；c_bRepresent the capacity of accumulator.

2. the operation constraining equation under accumulator Real-Time Scheduling:

$\left\{\begin{array}{c}SOC(t+i+1|t)=SOC(t+i|t)-\left({\mathrm{\&eta;}}_{b\_c}{I}_{b\_c}\right(t+i|t)+I_{b\_d}(t+i|t)/{\mathrm{\&eta;}}_{b\_d})P_{b\_r}(t+i|t)\mathrm{\&Delta;}t/c_b\\ I_{b\_c}(t+i|t)P_{c-\min }+I_{b\_d}(t+i|t)P_{d-\min }\&le;P_{b\_r}(t+i|t)\&le;I_{b\_c}(t+i|t)P_{c-\max }+I_{b\_d}(t+i|t)P_{d-\max }\\ I_{b\_c}(t+i|t)+I_{b\_d}(t+i|t)\&Element;(0,1)\\ {\mathrm{SOC}}_{\min }\&le;SOC(t+i+1|t)\&le;{\mathrm{SOC}}_{\max }\end{array}\right.$

In formula, t+i | t represents the dispatch value of i step forward；SOC (t+i+1 | t) and SOC (t+i | t) represent the state-of-charge that accumulator i+1 forward walks and i walks the moment forward respectively；η_{b_c}With η_{b_d}Represent discharge and recharge coefficient respectively；P_c-maxWith P_c-minRepresent maximum charge power and minimum charge power respectively；P_d-maxWith P_d-maxRepresent maximum discharge power and minimum discharge power respectively；I_{b_c}(t+i | t) represent that i walks the on off state of moment accumulator charging forward；I_{b_d}(t+i | t) represent that i walks the on off state of moment battery discharging forward；P_{b_r}(t+i | t) represent that under Real-Time Scheduling, i walks the charge-discharge electric power of moment accumulator forward；Δ t represents interval of time；c_bRepresent the capacity of accumulator.

Step 2, generate based on photovoltaic and wind power output scene and elimination technique is analyzed photovoltaic and exerted oneself and the uncertainty of wind power output, eliminate photovoltaic and undulatory property impact that wind-powered electricity generation regenerative resource is exerted oneself.

The concrete steps of described step 2 include:

(1) based on wind power output and photovoltaic power generation output forecasting, set up following wind-powered electricity generation and photovoltaic exerted oneself normal distribution probability model:

$P_{pv}\left(T\right)~N({\mathrm{\&mu;}}_{pv},{\mathrm{\&sigma;}}_{pv}^2)$

$P_{wind}\left(T\right)~N({\mathrm{\&mu;}}_{wind},{\mathrm{\&sigma;}}_{wind}^2)$

Wherein, σ_pv=0.1 μ_pv；σ_wind=0.1 μ_wind

In formula, μ_pvWith μ_windIt is that photovoltaic is exerted oneself and the predictive value of wind power output respectively；σ_pvWith σ_windIt it is corresponding variance；P_pvAnd P (T)_wind(T) it is T moment photovoltaic respectively and the actual of wind-powered electricity generation is exerted oneself；N represents normal distribution；

(2) use LHS method to generate the photovoltaic and wind power output scene obeying above-mentioned probabilistic model, use scene elimination technique eliminate low probability scene therein and scene strong for dependency merged.

Described step 2 (2nd) step is used LHS method to generate and is obeyed wind-powered electricity generation described in step 2 (1st) step and photovoltaic is exerted oneself the photovoltaic of normal distribution probability model and comprising the concrete steps that of wind power output scene:

1. variable x will often be tieed up after determining sample size HⁱDefinition territory intervalIt is divided into H equal minizone so thatAn original super cube is divided into HⁿIndividual small cubes；

2. generate a H × n matrix A, then every string of matrix A be all ordered series of numbers 1,2, a random fully intermeshing of H}；The corresponding selected little hypercube of each row of A, if randomly generating a sample in each little hypercube, then selecting H sample, obeying wind-powered electricity generation described in step 2 (1st) step and photovoltaic is exerted oneself the photovoltaic of normal distribution probability model and wind power output scene thus generating.

Step 3, set up with the minimum joint optimal operation model a few days ago for target of microgrid operating cost.

The concrete steps of described step 3 include:

(1) with the minimum object function setting up joint optimal operation model a few days ago for target of microgrid day total operating cost:

${\mathrm{minC}}_{\exp }=\underset{s}{\&Sigma;}{p}_s\left(\underset{T=T_0}{\overset{T_0+23}{\&Sigma;}}{c}_{Grid}\right(T\left)P_{Grid}^s\right(T)+\underset{T=T_0}{\overset{T_0+23}{\&Sigma;}}{c}_{Gas}F(T\left)\right)$

In formula, p_sIt it is the scene s probability occurred；It is the exchange power of microgrid and major network under scene s；F (T) is gas consumption；c_GridAnd c (T)_GasIt is electricity price and gas price respectively；T₀Represent the start time of paddy period；

(2) constraints that the electric load of microgrid, refrigeration duty joint optimal operation a few days ago runs is set up

1. the constraints that the electric load of microgrid joint optimal operation a few days ago runs

$P_{wind}^s\left(T\right)+P_{pv}^s\left(T\right)+P_{CCHP}\left(T\right)+P_b\left(T\right)+P_{Grid}^s\left(T\right)=P_{load}\left(T\right)+P_a\left(T\right)+P_c\left(T\right)+P_d\left(T\right)$

In formula, P_load(T) it is the electric load in microgrid T moment；P_CCHP(T) represent that the electricity of T moment CCHP is exerted oneself；P_a(T) T moment ice-storage air-conditioning power consumption under air conditioning mode is represented；P_b(T) charge-discharge electric power of T moment accumulator is represented；P_c(T) T moment ice-storage air-conditioning power consumption under air conditioning mode is represented；P_d(T) power consumption in T moment under ice-melt mode is represented；Represent the wind power output in T moment under the s scene；Represent that under the s scene, the photovoltaic in T moment is exerted oneself；Represent the exchange power between T moment microgrid and major network under the s scene.

2. the constraints that the refrigeration duty of microgrid joint optimal operation a few days ago runs

Q_CCHP(T)+Q_a(T)+Q_d(T)=Q_load(T)

In formula, Q_CCHP(T) cold the exerting oneself of T moment CCHP is represented；Q_d(T) T moment ice-storage air-conditioning refrigerating capacity under ice-melt mode is represented；Q_a(T) T moment ice-storage air-conditioning refrigerating capacity under air conditioning mode is represented respectively；Q_load(T) the refrigeration duty demand a few days ago predicting T of lower moment is represented；

Step 4, set up with real-time exchange power and the scheduled net interchange difference real-time joint optimal operation model of minimum microgrid under the Multiple Time Scales of target.

The concrete steps of described step 4 include:

(1) object function of the real-time joint optimal operation model of microgrid under setting up Multiple Time Scales with scheduled net interchange difference is minimum for target with real-time exchange power:

$Obj=\underset{P_{b\_r}(t+i|t)}{min}\underset{i=0}{\overset{11}{\&Sigma;}}|P_{Grid}\left(T\right)-P_{Grid\_r}(t+i|t)|\&times;p,t+i\&Element;T$

In formula, P_{Grid_r}(t+i | t) is the real-time exchange power of microgrid and electrical network, and p is coefficient；P_{b_r}(t+i | t) represent that under Real-Time Scheduling, i walks the charge-discharge electric power of moment accumulator forward；P_Grid(T) the exchange power between T moment microgrid and major network is represented；Obj represents object function.

(2) constraints that the real-time joint optimal operation of electric load of microgrid runs is set up

P_{wind_r}(t+i|t)+P_{pv_r}(t+i|t)+P_{CCHP_r}(T)+P_{b_r}(t+i|t)+P_{Grid_r}(t+i|t)

=P_{load_r}(t+i|t)+P_{a_r}(T)+P_c(T)+P_d(T),t+i∈T

In formula, r represents Real-Time Scheduling；P_{wind_r}(t+i | t) represent that under Real-Time Scheduling, i walks the wind power output in moment forward；P_{pv_r}(t+i | t) represent that the photovoltaic in i step moment forward under Real-Time Scheduling is exerted oneself；P_{CCHP_r}(T) represent that under Real-Time Scheduling, the electricity of T moment gas turbine is exerted oneself；P_{b_r}(t+i | t) represent that under Real-Time Scheduling, i walks the charge-discharge electric power of moment accumulator forward；P_{Grid_r}(t+i | t) represents the real-time exchange power of microgrid and electrical network；P_{load_r}(t+i | t) represent under Real-Time Scheduling forward i walk the electric load predictive value in moment；P_{a_r}(T) power consumption in T moment under Real-Time Scheduling and air conditioning mode is represented；P_c(T) T moment ice-storage air-conditioning power consumption under air conditioning mode is represented；P_d(T) power consumption in T moment under ice-melt mode is represented.

Wherein, the flow process of the real-time joint optimal operation of microgrid refrigeration duty is as shown in Figure 2, it concretely comprises the following steps: first calculates the variable quantity of microgrid real-time cooling load and judges that whether this variable quantity is less than 0, if this variable quantity is less than 0, then continue to determine whether that electric energy flows to major network from CCHP, if there being electric energy to flow to major network from CCHP, then preferentially reduce Q_CCHP(T), if flowing to major network without electric energy from CCHP, then Q is preferentially reduced_a(T)；If the variable quantity of microgrid real-time cooling load is more than zero, then continuing to determine whether that electric energy flows to major network from CCHP, if there being electric energy to flow to major network from CCHP, then preferentially having reduced Q_a(T), if flowing to major network without electric energy from CCHP, then Q is preferentially reduced_CCHP(T), try to achieve real-time cooling load finally according to refrigeration duty balance to exert oneself Q_{a_r}And Q (T)_CCHP-r(T)。

As shown in Figure 2, the operation principle of the real-time joint optimal operation of microgrid refrigeration duty is: the change that the fluctuation of refrigeration duty is exerted oneself by ice-storage air-conditioning and gas turbine electricity balances, and refrigeration duty scheduling and new forms of energy go out fluctuation and finally all realized by the balance of real-time electric load.When real-time cooling is load unbalanced, the situation about lacking with gas turbine electric load that is full of according to refrigeration duty, determine this moment gas turbine and the cold adjustment order exerted oneself of ice-storage air-conditioning.Such as, if real-time cooling is underload and gas turbine has electric power more than needed to flow to major network, then preferentially adjust that gas turbine is cold exerts oneself, otherwise preferentially adjust ice-storage air-conditioning refrigerating capacity under air conditioning mode.

Step 5, based on improved Particle Swarm Optimization, described microgrid joint optimal operation model a few days ago and the real-time joint optimal operation model of microgrid are solved, obtain the microgrid operation reserve of providing multiple forms of energy to complement each other under Multiple Time Scales.

The concrete steps of described step 5, as shown in Figure 3:

(1) choose the variable of applicable improved Particle Swarm Optimization as search particle, set up and be updated to the improved Particle Swarm Optimization model of feature with particle position renewal and region of search and initialize population；

(2) use this improved Particle Swarm Optimization model solution to optimize and Real time optimal dispatch model a few days ago, obtain the microgrid operation reserve of providing multiple forms of energy to complement each other under Multiple Time Scales.

Particle position correction has been solution coupling multiple constraint problem, and in an iterative process, the forcibly position of more new particle, to guarantee that particle is always in feasible zone；The renewal of region of search is namely in the starting stage of search, each particle scans for the bigger region of search, quickly to determine the approximate location of target, carrying out along with iteration, constantly reduce the space of search, quickly to realize being accurately positioned of target, the computing formula that region of search radius changes with iteration is as follows:

D=d₀/(1+exp((i-0.7i_max)/5))

In formula, d₀For constant, generally take d₀=50；I is current iteration number of times；i_maxIt is maximum iteration time, generally takes i_max=200；Exp represents exponential function.

Its method particularly includes: first evaluate the fitness Fit that in each population, particle is individual, if Fit is < pbest_iThen replace pbest with Fit_i；Otherwise, then continue to judge pbest_iWith the relation of gbest, if pbest_i< gbest, then use pbest_iReplace gbest；Otherwise, then particle position and region of search position are updated；Then judging whether calculated all particles and whether met end condition, if meeting, terminating to calculate.

Wherein, pbesti is the optimal solution individuality extreme value that particle itself finds；Gbest is the optimal solution global extremum that whole population is found at present.

It is emphasized that; embodiment of the present invention is illustrative; rather than it is determinate; therefore the present invention includes the embodiment that is not limited to described in detailed description of the invention; every other embodiments drawn according to technical scheme by those skilled in the art, also belong to the scope of protection of the invention.

Claims (8)

1. the microgrid dispatching method of providing multiple forms of energy to complement each other based on Multiple Time Scales, it is characterised in that comprise the following steps:

Step 1, micro-grid system scene is set and respectively cold, heat and electricity triple supply equipment, ice-storage air-conditioning and accumulator equipment in micro-grid system scene is modeled；

Step 2, generate based on photovoltaic and wind power output scene and elimination technique is analyzed photovoltaic and exerted oneself and the uncertainty of wind power output, eliminate photovoltaic and undulatory property impact that wind-powered electricity generation regenerative resource is exerted oneself；

Step 3, set up with the minimum joint optimal operation model a few days ago for target of microgrid operating cost；

Step 4, set up with real-time exchange power and the scheduled net interchange difference real-time joint optimal operation model of minimum microgrid under the Multiple Time Scales of target；

Step 5, based on improved Particle Swarm Optimization, described microgrid joint optimal operation model a few days ago and the real-time joint optimal operation model of microgrid are solved, obtain the microgrid operation reserve of providing multiple forms of energy to complement each other under Multiple Time Scales.

2. a kind of microgrid dispatching method of providing multiple forms of energy to complement each other based on Multiple Time Scales according to claim 1, it is characterised in that: concretely comprising the following steps of described step 1: respectively set up such as drag peace treaty bundle conditional equation: the electric power efficiency model of cold, heat and electricity triple supply equipment, electricity exert oneself and cold exert oneself between relational model and constraining equation；Ice-storage air-conditioning is at air conditioning mode, ice-make mode, ice-melt mode and air-conditioning and ice-melt model and the constraining equation of exerting oneself in composite mode；Accumulator is dispatched and the operation constraining equation under Real-Time Scheduling a few days ago；

(1) the electric power efficiency model of described cold, heat and electricity triple supply equipment, electricity exert oneself and cold exert oneself between relational model and constraining equation be respectively as follows:

1. the electric power efficiency model of cold, heat and power triple supply system:

$E_{CCHP}\left(T\right)=f\left(P_{CCHP}\right(T\left)\right)=f_{ISO}\left(\frac{P_{CCHP}\left(T\right)}{P_{ISO-\max }}\right)\&times;\frac{E_{\max }}{f_{ISO}\left(\frac{P_{\max }}{P_{ISO-\max }}\right)}$

In formula, P_CCHP(T) represent that the electricity of T moment CCHP is exerted oneself, P_maxWith E_maxRepresenting that under current operating conditions, the maximum electricity of CCHP is exerted oneself and maximum electrical efficiency respectively, ISO represents standard condition, and f is the nonlinear function that corresponding unit is given；E_CCHP(T) it is the electrical efficiency of gas turbine；

2. the electricity of cold, heat and power triple supply system exert oneself and cold exert oneself between relational model:

A. microgrid with electricity refrigeration mode under electricity exert oneself and cold exert oneself between relational model:

Q_CCHP(T)=g (P_CCHP(T))

In formula, Q_CCHP(T) cold the exerting oneself of T moment CCHP, P are represented_CCHP(T) represent that the electricity of T moment CCHP is exerted oneself；Functional relationship between g function representation T moment gas turbine cold is exerted oneself and electricity is exerted oneself；

B. microgrid with under cold power mode processed electricity exert oneself and cold exert oneself between relational model:

P_CCHP(T)=g^-1(P_CCHP(T))

In formula, P_CCHP(T) represent that the electricity of T moment CCHP is exerted oneself；g^-1Functional relationship between function representation T moment gas turbine cold is exerted oneself and electricity is exerted oneself；

3. the constraining equation that micro-grid system meets is:

$\left\{\begin{array}{c}{I}_{CCHP}\left(T\right)=0,T\&Element;T_{valley}\\ I_{CCHP}\left(T\right)P_{min}\&le;P_{CCHP}\left(T\right)\&le;I_{CCHP}\left(T\right)P_{max}\\ I_{CCHP}\left(T\right)Q_{min}\&le;Q_{CCHP}\left(T\right)\&le;Q_{CCHP}\left(T\right)Q_{max}\end{array}\right.$

Wherein, I_CCHP(T)∈(0,1)

In formula, I_CCHP(T) on off state of T moment CCHP equipment is represented；P_CCHP(T) represent that the electricity of T moment CCHP is exerted oneself；Q_CCHP(T) cold the exerting oneself of T moment CCHP is represented；P_maxWith P_minRepresent under current operating conditions that the maximum electricity of CCHP is exerted oneself and minimum electricity is exerted oneself respectively；Q_maxWith Q_minRepresent under current operating conditions that the maximum cold of CCHP is exerted oneself and minimum cold exerted oneself respectively；T_valleyElectric load paddy period；

(2) described ice-storage air-conditioning exerts oneself model and constraining equation is respectively as follows: air conditioning mode, ice-make mode, ice-melt mode and air-conditioning and ice-melt are in composite mode

1. ice-storage air-conditioning model of exerting oneself under air conditioning mode is:

$P_a\left(T\right)=\frac{Q_a\left(T\right)}{a_1{Q}_a\left(T\right)+a_2}$

In formula, P_aAnd Q (T)_a(T) T moment ice-storage air-conditioning power consumption under air conditioning mode and refrigerating capacity are represented respectively；a₁With a₂For constant；

Its constraints is:

$\left\{\begin{array}{c}{I}_a\left(T\right)=0,T\&Element;T_{valley}\\ I_a\left(T\right)Q_{a-min}\&le;Q_a\left(T\right)\&le;I_a\left(T\right)Q_{a-max}\end{array}\right.$

In formula, I_a(T) on off state of T moment air conditioning mode is represented；Q_a(T) T moment ice-storage air-conditioning refrigerating capacity under air conditioning mode is represented；T_valleyRepresent electric load paddy period；Q_a-maxWith Q_a-minRepresent the minimum and maximum refrigerating capacity under air conditioning mode respectively；

2. ice-storage air-conditioning under ice-make mode model of exerting oneself be:

$P_c\left(T\right)=\frac{Q_c\left(T\right)}{a_1{Q}_c\left(T\right)+a_2}$

In formula, P_cAnd Q (T)_c(T) T moment ice-storage air-conditioning power consumption under air conditioning mode and refrigerating capacity are represented respectively；a₁With a₂For constant；

Its constraints is:

$\left\{\begin{array}{c}{I}_c\left(T\right)=0,T\&NotElement;T_{valley}\\ \underset{T=T_0}{\overset{T_0+23}{\&Sigma;}}|I_c\left(T\right)-I_c(T-1)|=2\\ Q_c\left(T\right)=I_c\left(T\right)Q_{a-max}\end{array}\right.$

In formula, I_cAnd I (T)_c(T-1) on off state of T moment and T-1 moment ice-make mode is represented respectively；Q_c(T) T moment ice-storage air-conditioning refrigerating capacity under air conditioning mode is represented；T_valleyRepresent electric load paddy period；T₀Represent the start time of paddy period；Q_a-maxRepresent the maximum cooling capacity under air conditioning mode；

3. ice-storage air-conditioning constraints under ice-melt mode is:

$\left\{\begin{array}{c}{I}_d\left(T\right)=0,T\&Element;T_{valley}\\ 0\&le;Q_d\left(T\right)=I_d\left(T\right)Q_{d-max}\end{array}\right.$

In formula, I_d(T) on off state of T moment ice-melt mode is represented；Q_dAnd Q (T)_d-maxRepresent T moment ice-storage air-conditioning refrigerating capacity under ice-melt mode and maximum cooling capacity respectively；T_valleyRepresent electric load paddy period；

4. ice-storage air-conditioning at the model of exerting oneself in composite mode of air conditioning mode and ice-melt mode is:

IS (T)=(1-η₁)IS(T-1)+η₂Q_c(T)-Q_d(T)

In formula, IS (T) represents the cold energy of storage in T moment Ice Storage Tank；η₁With η₂Represent the refrigerating efficiency under cold energy storage loss factor and air conditioning mode respectively；Q_c(T) T moment ice-storage air-conditioning refrigerating capacity under air conditioning mode is represented；Q_d(T) T moment ice-storage air-conditioning refrigerating capacity under ice-melt mode is represented；

Its constraints is:

${\mathrm{IS}}_{\min }\left(T\right)\&le;\mathrm{IS}\left(T\right)\&le;(1-{\mathrm{\&eta;}}_1)\mathrm{IS}(T-1),T\&NotElement;T_{\mathrm{valley}}$

In formula, IS_min(T) the minimum cold energy needing storage in Ice Storage Tank is represented；IS (T) represents the cold energy of storage in T moment Ice Storage Tank；η₁Represent cold energy storage loss factor；T_valleyRepresent electric load paddy period；

(3) described accumulator is dispatched and the operation constraining equation under Real-Time Scheduling a few days ago；

1. accumulator dispatch a few days ago under operation constraining equation:

$\left\{\begin{array}{c}SOC(T+1)=SOC\left(T\right)-\left({\mathrm{\&eta;}}_{b\_c}{I}_{b\_c}\right(T)+I_{b\_d}\left(T\right)/{\mathrm{\&eta;}}_{b\_d})P_b\left(T\right)\mathrm{\&Delta;}t/c_b\\ I_{b\_c}\left(T\right)P_{c-\min }+I_{b\_d}\left(T\right)P_{d-\min }\&le;P_b\left(T\right)\&le;I_{b\_c}\left(T\right)P_{c-\max }+I_{b\_d}\left(T\right)P_{d-\max }\\ I_{b\_c}\left(T\right)+I_{b\_d}\left(T\right)\&Element;(0,1)\\ {\mathrm{SOC}}_{\min }\&le;SOC(T+1)\&le;{\mathrm{SOC}}_{\max }\end{array}\right.$

In formula, SOC represents state-of-charge；SOC_maxWith SOC_minRepresent the bound of state-of-charge respectively；η_{b_c}With η_{b_d}Represent discharge and recharge coefficient respectively；P_c-maxWith P_c-minRepresent maximum charge power and minimum charge power respectively；P_d-maxWith P_d-minRepresent maximum discharge power and minimum discharge power respectively；I_{b_c}(T) on off state of T moment accumulator charging, I are represented_{b_c}(T)∈(0,1)；I_{b_d}(T) on off state of T moment battery discharging, I are represented_{b_d}(T)∈(0,1)；P_b(T) charge-discharge electric power of T moment accumulator is represented；Δ t represents interval of time；c_bRepresent the capacity of accumulator；

2. the operation constraining equation under accumulator Real-Time Scheduling:

$\left\{\begin{array}{c}SOC(t+i+1|t)=SOC(t+i|t)-\left({\mathrm{\&eta;}}_{b\_c}{I}_{b\_c}\right(t+i\left|t\right)+I_{b\_d}(t+i|t)/{\mathrm{\&eta;}}_{b\_d})P_{b\_r}(t+i|t)\mathrm{\&Delta;}t/c_b\\ I_{b\_c}(t+i|t)P_{c-\min }+I_{b\_d}(t+i|t)P_{d-\min }\&le;P_{b\_r}(t+i|t)\&le;I_{b\_c}(t+i|t)P_{c-\max }+I_{b\_d}(t+i|t)P_{d-\max }\\ I_{b\_c}(t+i|t)+I_{b\_d}(t+i|t)\&Element;(0,1)\\ {\mathrm{SOC}}_{\min }\&le;SOC(t+i+1|t)\&le;{\mathrm{SOC}}_{\max }\end{array}\right.$

In formula, t+i | t represents the dispatch value of i step forward；SOC (t+i+1 | t) and SOC (t+i | t) represent the state-of-charge that accumulator i+1 forward walks and i walks the moment forward respectively；η_{b_c}With η_{b_d}Represent discharge and recharge coefficient respectively；P_c-maxWith P_c-minRepresent maximum charge power and minimum charge power respectively；P_d-maxWith P_d-maxRepresent maximum discharge power and minimum discharge power respectively；I_{b_c}(t+i | t) represent that i walks the on off state of moment accumulator charging forward；I_{b_d}(t+i | t) represent that i walks the on off state of moment battery discharging forward；P_{b_r}(t+i | t) represent that under Real-Time Scheduling, i walks the charge-discharge electric power of moment accumulator forward；Δ t represents interval of time；c_bRepresent the capacity of accumulator.

3. a kind of microgrid dispatching method of providing multiple forms of energy to complement each other based on Multiple Time Scales according to claim 1 and 2, it is characterised in that: the concrete steps of described step 2 include:

(1) based on wind power output and photovoltaic power generation output forecasting, set up following wind-powered electricity generation and photovoltaic exerted oneself normal distribution probability model:

$P_{pv}\left(T\right)~N({\mathrm{\&mu;}}_{pv},{\mathrm{\&sigma;}}_{pv}^2)$

$P_{wind}\left(T\right)~N({\mathrm{\&mu;}}_{wind},{\mathrm{\&sigma;}}_{wind}^2)$

Wherein, σ_pv=0.1 μ_pv；σ_wind=0.1 μ_wind

In formula, μ_pvWith μ_windIt is that photovoltaic is exerted oneself and the predictive value of wind power output respectively；σ_pvWith σ_windIt it is corresponding variance；P_pvAnd P (T)_wind(T) it is T moment photovoltaic respectively and the actual of wind-powered electricity generation is exerted oneself；N represents normal distribution；

(2) use LHS method to generate the photovoltaic and wind power output scene obeying above-mentioned probabilistic model, use scene elimination technique eliminate low probability scene therein and scene strong for dependency merged.

4. a kind of microgrid dispatching method of providing multiple forms of energy to complement each other based on Multiple Time Scales according to claim 3, it is characterised in that: described step (2) generates comprising the concrete steps that of photovoltaic and wind power output scene:

1. variable x will often be tieed up after determining sample size HⁱDefinition territory intervalIt is divided into H equal minizone so thatAn original super cube is divided into HⁿIndividual small cubes；

2. generate a H × n matrix A, then every string of matrix A be all ordered series of numbers 1,2, a random fully intermeshing of H}；The corresponding selected little hypercube of each row of A, if randomly generating a sample in each little hypercube, then selecting H sample, obeying wind-powered electricity generation described in step 2 (1st) step and photovoltaic is exerted oneself the photovoltaic of normal distribution probability model and wind power output scene thus generating.

5. a kind of microgrid dispatching method of providing multiple forms of energy to complement each other based on Multiple Time Scales according to claim 1 and 2, it is characterised in that: the concrete steps of described step 3 include:

(1) with the minimum object function setting up joint optimal operation model a few days ago for target of microgrid day total operating cost:

$\min C_{\exp }=\underset{s}{\mathrm{\&Sigma;}}{p}_s\left(\underset{T=T_0}{\overset{T_0+23}{\mathrm{\&Sigma;}}}{c}_{Grid}\right(T\left)P_{Grid}^s\right(T)+\underset{T=T_0}{\overset{T_0+23}{\mathrm{\&Sigma;}}}{c}_{Gas}F(T\left)\right)$

In formula, p_sIt it is the scene s probability occurred；It is the exchange power of microgrid and major network under scene s；F (T) is gas consumption；c_GridAnd c (T)_GasIt is electricity price and gas price respectively；T₀Represent the start time of paddy period；

(2) constraints that the electric load of microgrid, refrigeration duty joint optimal operation a few days ago runs is set up

1. the constraints that the electric load of microgrid joint optimal operation a few days ago runs

$P_{wind}^s\left(T\right)+P_{pv}^s\left(T\right)+P_{CCHP}\left(T\right)+P_b\left(T\right)+P_{Grid}^s\left(T\right)=P_{load}\left(T\right)+P_a\left(T\right)+P_c\left(T\right)+P_d\left(T\right)$

In formula, P_load(T) it is the electric load in microgrid T moment；P_CCHP(T) represent that the electricity of T moment CCHP is exerted oneself；P_a(T) T moment ice-storage air-conditioning power consumption under air conditioning mode is represented；P_b(T) charge-discharge electric power of T moment accumulator is represented；P_c(T) T moment ice-storage air-conditioning power consumption under air conditioning mode is represented；P_d(T) power consumption in T moment under ice-melt mode is represented；Represent the wind power output in T moment under the s scene；Represent that under the s scene, the photovoltaic in T moment is exerted oneself；Represent the exchange power between T moment microgrid and major network under the s scene；

2. the constraints that the refrigeration duty of microgrid joint optimal operation a few days ago runs

Q_CCHP(T)+Q_a(T)+Q_d(T)=Q_load(T)

In formula, Q_CCHP(T) cold the exerting oneself of T moment CCHP is represented；Q_d(T) T moment ice-storage air-conditioning refrigerating capacity under ice-melt mode is represented；Q_a(T) T moment ice-storage air-conditioning refrigerating capacity under air conditioning mode is represented respectively；Q_load(T) the refrigeration duty demand a few days ago predicting T of lower moment is represented.

6. a kind of microgrid dispatching method of providing multiple forms of energy to complement each other based on Multiple Time Scales according to claim 1 and 2, it is characterised in that: the concrete steps of described step 4 include:

(1) object function of the real-time joint optimal operation model of microgrid under setting up Multiple Time Scales with scheduled net interchange difference is minimum for target with real-time exchange power:

$Obj=\underset{P_{b\_r}(t+i|t)}{min}\underset{i=0}{\overset{11}{\&Sigma;}}|P_{Grid}\left(T\right)-P_{Grid\_r}(t+i|t)|\&times;p,t+i\&Element;T$

In formula, P_{Grid_r}(t+i | t) is the real-time exchange power of microgrid and electrical network, and p is coefficient；P_{b_r}(t+i | t) represent that under Real-Time Scheduling, i walks the charge-discharge electric power of moment accumulator forward；P_Grid(T) the exchange power between T moment microgrid and major network is represented；Obj represents object function；

(2) constraints that the real-time joint optimal operation of electric load of microgrid runs is set up

P_{wind_r}(t+i|t)+P_{pv_r}(t+i|t)+P_{CCHP_r}(T)+P_{b_r}(t+i|t)+P_{Grid_r}(t+i|t)

=P_{load_r}(t+i|t)+P_{a_r}(T)+P_c(T)+P_d(T),t+i∈T

In formula, r represents Real-Time Scheduling；P_{wind_r}(t+i | t) represent that under Real-Time Scheduling, i walks the wind power output in moment forward；P_{pv_r}(t+i | t) represent that the photovoltaic in i step moment forward under Real-Time Scheduling is exerted oneself；P_{CCHP_r}(T) represent that under Real-Time Scheduling, the electricity of T moment gas turbine is exerted oneself；P_{b_r}(t+i | t) represent that under Real-Time Scheduling, i walks the charge-discharge electric power of moment accumulator forward；P_{Grid_r}(t+i | t) represents the real-time exchange power of microgrid and electrical network；P_{load_r}(t+i | t) represent under Real-Time Scheduling forward i walk the electric load predictive value in moment；P_{a_r}(T) power consumption in T moment under Real-Time Scheduling and air conditioning mode is represented；P_c(T) T moment ice-storage air-conditioning power consumption under air conditioning mode is represented；P_d(T) power consumption in T moment under ice-melt mode is represented.

7. a kind of microgrid dispatching method of providing multiple forms of energy to complement each other based on Multiple Time Scales according to claim 5, it is characterised in that: the concrete steps of described step 5 include:

(1) choose the variable of applicable improved Particle Swarm Optimization as search particle, set up and be updated to the improved Particle Swarm Optimization model of feature with particle position renewal and region of search and initialize population；

(2) use this improved Particle Swarm Optimization model solution to optimize and Real time optimal dispatch model a few days ago, obtain the microgrid operation reserve of providing multiple forms of energy to complement each other under Multiple Time Scales.

8. a kind of microgrid dispatching method of providing multiple forms of energy to complement each other based on Multiple Time Scales according to claim 6, it is characterised in that: described step 5 (2nd) step method particularly includes:

Evaluate the fitness Fit that in each population, particle is individual, if Fit is < pbest_iThen replace pbest with Fit_i；Otherwise, then continue to judge pbest_iWith the relation of gbest, if pbest_i< gbest, then use pbest_iReplace gbest；Otherwise, then particle position and region of search position are updated；Then judging whether calculated all particles and whether met end condition, if meeting, terminating to calculate；

Wherein, pbesti is the optimal solution individuality extreme value that particle itself finds；Gbest is the optimal solution global extremum that whole population is found at present.