WO2019128012A1

WO2019128012A1 - Robust optimal coordinated dispatching method for alternating-current and direct-current hybrid micro-grid

Info

Publication number: WO2019128012A1
Application number: PCT/CN2018/084942
Authority: WO
Inventors: 顾伟; 邱海峰; 周苏洋; 吴志
Original assignee: 东南大学
Priority date: 2017-12-28
Filing date: 2018-04-27
Publication date: 2019-07-04
Also published as: CN108108846A; CN108108846B

Abstract

Disclosed is a robust optimal coordinated dispatching method for an alternating-current and direct-current hybrid micro-grid. The method comprises the following steps: step 10) acquiring source load prediction data and creating an uncertainty set; step 20) acquiring an objective function of an optimal dispatching model; step 30) creating constraint conditions of the optimal dispatching model; and step 40) solving a robust optimal dispatching problem: using a column constraint generation algorithm to solve a robust optimization problem to obtain a robust coordinated operation manner of an alternating-current and direct-current hybrid micro-grid. The method takes the uncertainty of multiple source loads in an alternating-current and direct-current hybrid micro-grid into consideration, can realize the robost optimal dispatching of the alternating-current and direct-current hybrid micro-grid and provides guidance and assistance to work out an operation manner of the alternating-current and direct-current hybrid micro-grid.

Description

Robust optimization coordinated scheduling method for AC/DC hybrid microgrid

Technical field

The invention belongs to the technical field of microgrid optimization scheduling and energy management, and in particular relates to a robust optimization coordinated scheduling method for AC/DC hybrid microgrid.

Background technique

Due to the depletion of fossil energy and its high pollution to the ecological environment, natural resources such as wind energy and solar energy have received extensive attention for their clean and renewable characteristics. At present, more and more renewable energy power generation has been connected. Grid. As a small autonomous system that collects distributed power generation, energy storage and load, microgrid has become an effective technology and an important way to make rational use of renewable energy in the field of power systems. In order to ensure the reliable operation of the microgrid, energy management of the microgrid is required. According to the predictive data and system information of renewable energy such as scenery, and considering the long-term operational efficiency of the microgrid, an operational scheduling plan is formulated for it.

Renewable energy is random and intermittent due to the influence of natural conditions, and the load volatility is strong, which leads to more uncertainties in the microgrid, which brings great challenges to the optimal scheduling of the microgrid. AC/DC hybrid microgrid as a new type of microgrid structure, which connects AC bus and DC bus through bidirectional converter to realize district and power supply of AC and DC, but the current uncertainty in the optimal scheduling of AC/DC hybrid microgrid The stochastic optimization method is adopted, and the robust optimization scheduling is less, and the operating cost, operation characteristics and related constraints of each unit in the AC/DC hybrid microgrid have not been fully considered.

Summary of the invention

Technical Problem: The technical problem to be solved by the present invention is to provide a robust optimization coordinated scheduling method for AC/DC hybrid microgrid, which can realize AC/DC hybrid micro-in consideration of source-load uncertainty in AC/DC hybrid microgrid. The robust optimization scheduling of the network provides guidance and assistance for the development of the AC/DC hybrid microgrid.

Technical Solution: The method for robust optimization coordinated scheduling of an AC/DC hybrid microgrid according to the present invention comprises the following steps:

Step 10) acquiring source load prediction data and constructing an uncertainty set;

Step 20) acquiring an operating cost coefficient of each device in the system, and constructing an objective function of a robust optimal scheduling model in the form of min-max-min;

Step 30) Obtain an operation limit of each device in the system, and establish a constraint condition of a robust optimization scheduling model in the form of min-max-min;

Step 40) Solving the robust optimization scheduling problem, that is, using the column constraint generation algorithm to solve the robust optimization problem, and obtaining the robust coordinated operation mode of the AC-DC hybrid microgrid.

Further, in the method of the present invention, in step 10), the source load prediction data is a predicted nominal value, an upper and lower deviation value, and a time period budget parameter of the renewable energy output and load in the AC/DC hybrid microgrid, and the source load prediction is performed. The data is substituted into the following equations (1) to (4), that is, the uncertainty set is constructed:

In the formula, for the wind turbine output uncertainty set W, w _t ,

They are the actual value of the maximum output power of the fan during the t period, the predicted nominal value, the predicted upper deviation value and the predicted lower deviation value; Π _w is the time period budget parameter of the fan output uncertainty;

with

The parameters are introduced for the upper deviation of the fan output uncertainty and the lower deviation is introduced; N _t is the total time period of a scheduling period; P, L _dc and L _ac are the uncertainty set of the photovoltaic output and the uncertainty of the DC load, respectively. Uncertainty set of set and AC load; p _t ,

They are the actual value of the maximum output power of PV in t period, the predicted nominal value, the predicted upper deviation value and the predicted lower deviation value; Π _p is the time period budget parameter of the photovoltaic output uncertainty;

with

Introducing parameters for the upper deviation of the uncertainty of photovoltaic output, and introducing the parameters for the lower deviation; l _dc,t ,

They are the actual value of the maximum power of the DC load in the t period, the predicted nominal value, the predicted upper deviation value and the predicted lower deviation value; Π _{l, dc} is the time period budget parameter of the DC load uncertainty;

And κ _-dc,t are the upper deviation introduction parameters and the lower deviation introduction parameters of DC load uncertainty; l _ac,t ,

l _-ac,t are the actual value of the maximum power of the AC load during the t period, the predicted nominal value, the predicted upper deviation value and the predicted lower deviation value; Π _{l, ac} is the time period budget parameter of the AC load uncertainty;

And κ _-ac,t introduce parameters for the upper deviation of the AC load uncertainty and the lower deviation.

Further, in the method of the present invention, in step 20), the operating cost coefficient of each device in the system includes all operating cost coefficients related to diesel generators, energy storage, bidirectional converters, fans, photovoltaics, and AC and DC loads, Substituting the running cost coefficient into the following formula (5), the objective function of the robust optimal scheduling model in the form of min-max-min is established:

The relevant parameters in equation (5) are calculated according to the following formula:

In the formula,

with

The starting cost and shutdown cost of the diesel generator are respectively;

The fuel cost for diesel generators;

with

The operation and maintenance costs of diesel generators, energy storage, bidirectional converters, fans and photovoltaics;

Cost for energy storage;

Punish the cost of power cuts,

with

They are the starting and shutting down cost coefficients of the diesel generators respectively; I _{DE, t} is the starting sign of the diesel generator in the t period, 1 means that the diesel generator is started in the t period, 0 means the diesel generator is not started in the t period ;M _DE,t is the shutdown flag of the diesel generator in the t period, 1 means that the diesel generator is shut down during the t period, 0 means that the diesel generator is not shut down during the t period; U _DE,t represents the diesel fuel in the t period The running state of the generator, when the value is 1, indicates that the diesel generator is in the on state during the t period, and the value of 0 indicates that the diesel generator is in the stop state during the t period; U _{acdcBC, t} is the t period bidirectional converter To the commutation operation status flag, 1 indicates that there is a forward commutation in the t period, 0 indicates that there is no forward commutation in the t period, U _{dcacBC, t} is the t- _zone bidirectional converter negative commutation operation status flag, 1 It means that there is a negative commutation in the t period, and 0 means that there is no negative commutation in the t period;

The fuel cost coefficient of the diesel generator; a _DE and b _DE are the fuel consumption characteristic cost coefficients of the diesel generator; P _DE,t is the operating power of the diesel generator during the t period;

Is the rated power of the diesel generator; Δt is the time interval of two periods;

with

The operating and maintenance cost factors for diesel generators, energy storage, bidirectional converters, fans and photovoltaics;

Cost coefficient for energy storage loss;

The cost coefficient for the power cut-off penalty; P _chES,t and P _disES,t are the charging power and the discharging power of the energy storage in the t period respectively; P _acdcBC,t is the bidirectional converter from the alternating current bus to the DC bus in the t period Forward commutating power; P _{dcacBC, t} is the negative commutating power of the bidirectional converter from the DC bus to the AC bus during the t period; P _{WT, t} and P _{PV, t} are the power generation of the fan and photovoltaic respectively during the t period Power; P _{cut, acL, t} and P _{cut, dcL, t} respectively represent the load power of the AC zone cut off during the t period and the load power of the DC zone cut off; P _{tran, acL, t} and P _{tran, dcL, t} are t The time period communication can schedule the load operation power and the DC schedulable load operation power.

Further, in the method of the present invention, in step 30), the operating limit of each device in the system is all operating limits related to diesel generators, energy storage, bidirectional converters, fans, photovoltaics, and AC and DC loads, Substituting the running limit into the following equations (10) to (22), the constraints of the robust optimal scheduling model in the form of min-max-min are established:

0 ≤ P _{WT, t} ≤ w _t _{, 0} ≤ P _{PV, t} ≤ p _t (40)

I _DE,t +M _DE,t ≤1,I _DE,t -M _DE,t =U _DE,t -U _DE,t-1 (43)

Equation (10) is the power generation constraint of the fan and photovoltaic; the equations (11)-(13) are the minimum continuous startup time, minimum continuous shutdown time and maximum continuous startup time constraint of the diesel generator.

with

They are the minimum continuous start-up period limit of the diesel generator, the minimum continuous shutdown period limit and the maximum continuous start-up period limit; the formula (14) is the upper and lower limits of the diesel generator operating power and the climbing speed constraint.

with

For the upper and lower limits of the operating power of the diesel generator in the on state,

with

It is the rate limit for the downhill and uphill slopes of the diesel generator in the unit time period; the formula (15) is the maximum charge and discharge power constraint of the energy storage.

with

The maximum charge and discharge power limit for energy storage; Equation (16) is the energy storage state constraint, and S ^min and S ^max are the lower and upper limits of the energy storage allowable state, S(t) and S(t-1) is the state of charge of energy storage in the period t and t-1, η _C and η _D are the charging and discharging efficiency limits of energy storage, and S(0) is the initial state of charge of energy storage. , S(N _t ) is the state of charge of the stored energy at the end of the scheduling period; equations (17)-(18) are the commutation power and power fluctuation constraints of the bidirectional converter.

with

Indicates the operating power limits for forward commutation and negative commutation,

with

The lower limit value and the upper limit value of the power fluctuation of the bidirectional converter in the adjacent time period; the equations (19)-(20) are the operating power of the AC and DC cut-off load and the schedulable load, and the AC-DC schedulable load for each time period. Power constraints, P _{cut, ac, maxL, t} and P _{cut, dc, maxL, t} are the maximum _removable load power limits for AC and DC during t periods, P _{tran, ac, maxL, t} and P _{tran, dc, maxL,t} is the maximum operating power limit for AC and DC schedulable loads in t-period, [t ^ac,1 ,t ^ac,end ] is the operating time interval limit for AC schedulable loads, [t ^dc,1 ,t ^{dc , end} ] is the operating time interval limit of the DC schedulable load,

with

Is the planned power consumption limit for AC and DC schedulable loads; Equations (21)-(22) are power balance constraints for DC and AC zones,

with

The forward and negative commutation efficiency limits for the bidirectional converter.

Further, in the method of the present invention, the specific content of step 40) includes:

Step 401): Write the robust optimal scheduling model in the form of min-max-min represented by equations (1)-(22) into the following matrix representation:

s.t.A·x≤b, B·x=e,x∈{0,1} (54)

D·y≤f, E·y=g, (55)

F·y≤h-G·x, (56)

J·y≤w, K·y≤p, (57)

M·y=l _dc , N·y=l _ac (58)

Where x is the 0-1 state variable of the first phase of equation (5), y is the second phase power variable, and w, p, l _dc , l _ac are the set of the second phase uncertainty set variables, c d is a constant matrix in the objective function; Equation (24) represents a constraint condition only related to x, A, b, B, and e are constant matrices in the constraint; Equation (25) represents only y related Constraints, D, f, E, g are the constant matrices in the constraint; Equation (26) represents the constraints associated with x and y, F, h, G are the constant matrices in the constraint; Represents constraints associated with w, p, and y, where J, w, K, and p are constant matrices in the constraint; Eq. (28) represents constraints associated with l _dc , l _{ac ,} and y , M, N Both are constant matrices in this constraint.

Step 402): Based on the model of the matrix representation in step 401), the sub-problem of the robust optimal scheduling model in the form of min-max-min using the column constraint generation algorithm is as follows:

Wherein α, β, χ, γ, ψ, μ _dc and μ _ac are the dual variables of y in the formulae (25)-(28).

Step 403): Based on the model of the matrix representation in step 401) and the sub-problem of step 402), the main problem of using the column constraint generation algorithm to form the robust optimal scheduling model in the form of min-max-min is as follows:

Where l is the total number of iterations, k is the current number of iterations, and the x optimized by the main problem is sub-problem as a known variable, w _k , p _k , l _dc,k , l _ac,k are after the kth iteration The optimization result of w, p, l _dc and l _ac in the sub-problem, y _k is the optimization result of y in the sub-problem after the k-th iteration; η is the optimization variable related to the objective function value of the sub-problem.

Step 404): Using the integer optimization modeling toolbox YALMIP to call the solver SCIP iteratively solve the sub-problem of step 402) and the main problem of step 403) to obtain a robust coordinated operation mode of the AC-DC hybrid microgrid.

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

The existing AC/DC hybrid microgrid optimization scheduling research is optimized for the scenarios of renewable energy output and load power determination. However, the actual renewable energy output and load power prediction often have deviations, so the possible scenarios are not certain. The deterministic optimization scheduling scheme cannot satisfy multiple source-load uncertain scenarios, and cannot be effectively applied in actual engineering. The invention proposes a two-stage robust optimization scheduling model for the source-load uncertainty in the AC-DC hybrid microgrid. The first stage optimizes the start and stop of the diesel generator and the operating state of the bidirectional converter for all uncertain scenarios. The second stage optimizes the operating power of each equipment unit, and the column constraint generation algorithm can quickly solve the worst problem. Minimum operating cost and piconet operation plan under the scenario. The robust optimal scheduling model uses the interval uncertainty set to describe all the uncertainties in the microgrid. The obtained operational plan under the worst scenarios can meet the operational constraints in all other scenarios, ensuring the piconet in any scenario. The safe and economic operation of the system effectively solves the uncertainty problem in the optimal scheduling of AC/DC hybrid microgrid.

DRAWINGS

Figure 1 is a flow chart of an embodiment of the present invention;

2 is a topological structural diagram of an AC-DC hybrid microgrid according to an embodiment of the present invention.

Detailed ways

The technical solutions of the embodiments of the present invention are further described below with reference to the accompanying drawings.

As shown in FIG. 1, the topology of the AC/DC hybrid microgrid is shown in FIG. 2 as an embodiment of the method of the present invention. The method includes the following steps:

Preferably, the step 10) specifically includes: obtaining a predicted nominal value, an upper and lower deviation value, and a time period budget parameter of the renewable energy output and load in the AC/DC hybrid microgrid, and substituting the source load prediction data into the following formula: (1) In equation (4), the construct is obtained with an uncertainty set:

In the formula, for the wind turbine output uncertainty set W, w _t ,

with

Introducing parameters for the upper deviation of the DC load uncertainty and introducing the parameters for the lower deviation; l _ac,t ,

They are the actual value of the maximum power of the AC load during the t period, the predicted nominal value, the predicted upper deviation value and the predicted lower deviation value; Π _{l, ac} is the time period budget parameter of the AC load uncertainty;

Preferably, the step 20) specifically includes: acquiring operating cost coefficients of each device in the system, including all operating costs related to diesel generators, energy storage, bidirectional converters, fans, photovoltaics, and AC and DC loads. The coefficient, and the running cost coefficient is substituted into the following formula (5), that is, the objective function of the robust optimal scheduling model in the form of min-max-min is established:

In the formula,

with

The starting cost and shutdown cost of the diesel generator are respectively;

The fuel cost for diesel generators;

with

Cost for energy storage;

Punish the cost of power cuts,

with

Cost coefficient for energy storage loss;

Preferably, the step 30) specifically includes: obtaining operating limits of each device in the system, including all operating limits related to diesel generators, energy storage, bidirectional converters, fans, photovoltaics, and AC and DC loads. Value, the operational limit is substituted into the following equations (10) to (22), that is, the constraints of the robust optimal scheduling model in the form of min-max-min are established:

0 ≤ P _{WT, t} ≤ w _t _{, 0} ≤ P _{PV, t} ≤ p _t (65)

I _DE,t +M _DE,t ≤1,I _DE,t -M _DE,t =U _DE,t -U _DE,t-1 (68)

with

As a preferred solution, the step 40) specifically includes:

s.t.A·x≤b, B·x=e,x∈{0,1} (79)

D·y≤f, E·y=g, (80)

F·y≤h-G·x, (81)

J·y≤w, K·y≤p, (82)

M·y=l _dc , N·y=l _ac (83)

Where x is the 0-1 state variable of the first stage of equation (5), y is the second stage power variable, w, p, l _dc , l _ac is the set of the second stage uncertainty set variable, c d is a constant matrix in the objective function; Equation (24) represents a constraint condition only related to x, A, b, B, and e are constant matrices in the constraint; Equation (25) represents only y related Constraints, D, f, E, g are the constant matrices in the constraint; Equation (26) represents the constraints associated with x and y, F, h, G are the constant matrices in the constraint; Represents constraints associated with w, p, and y, where J, w, K, and p are constant matrices in the constraint; Eq. (28) represents constraints associated with l _dc , l _{ac ,} and y , M, N Both are constant matrices in this constraint.

Step 404): Using the integer optimization modeling toolbox YALMIP to call the solver SCIP iteratively solve the sub-problem of step 402) and the main problem of step 403) to obtain a robust coordinated operation mode of the AC-DC hybrid microgrid. YALMIP is an integer optimization modeling toolbox in the company's mathematical software MATLAB.

The method of the embodiment of the invention proposes a two-stage robust optimization scheduling model and a solution method for the AC-DC hybrid micro-network, considering the source-load uncertainty, and the first stage determines the start-stop and the bidirectional converter of the diesel generator. The operating state, the second phase optimizes the operating power of each equipment unit, and uses the column constraint generation algorithm to quickly solve the minimum operating cost and micro-grid operation plan in the worst case scenario.

The basic principles, main features and advantages of the present invention have been shown and described above. It should be understood by those skilled in the art that the present invention is not limited by the foregoing embodiments, and the description of the present invention and the description of the present invention are only intended to further illustrate the principles of the present invention without departing from the spirit and scope of the invention. There are various changes and modifications of the invention which fall within the scope of the invention as claimed. The scope of the invention is defined by the claims and their equivalents.

Claims

A method for robust optimization coordinated scheduling of AC/DC hybrid microgrid, characterized in that the method comprises the following steps:

Step 10) acquiring source load prediction data and constructing an uncertainty set;

Step 20) acquiring an operating cost coefficient of each device in the system, and constructing an objective function of a robust optimal scheduling model in the form of min-max-min;

Step 30) Obtain an operation limit of each device in the system, and establish a constraint condition of a robust optimization scheduling model in the form of min-max-min;

Step 40) Solving the robust optimization scheduling problem, that is, using the column constraint generation algorithm to solve the robust optimization problem, and obtaining the robust coordinated operation mode of the AC-DC hybrid microgrid.
The method for robust optimization coordinated scheduling of an AC/DC hybrid microgrid according to claim 1, wherein in the step 10), the source-charge prediction data is a predictive indicator of the output and load of the renewable energy in the AC-DC hybrid microgrid. The weighing value, the upper and lower deviation value and the time period budget parameter are substituted into the following formulas (1) to (4), that is, the uncertainty set is constructed:

In the formula, for the wind turbine output uncertainty set W, w t ,
They are the actual value of the maximum output power of the fan during the t period, the predicted nominal value, the predicted upper deviation value and the predicted lower deviation value; Π w is the time period budget parameter of the fan output uncertainty;
with
The parameters are introduced for the upper deviation of the fan output uncertainty and the lower deviation is introduced; N t is the total time period of a scheduling period; P, L dc and L ac are the uncertainty set of the photovoltaic output and the uncertainty of the DC load, respectively. Uncertainty set of set and AC load; p t ,
They are the actual value of the maximum output power of PV in t period, the predicted nominal value, the predicted upper deviation value and the predicted lower deviation value; Π p is the time period budget parameter of the photovoltaic output uncertainty;
with
Introducing parameters for the upper deviation of the uncertainty of photovoltaic output, and introducing the parameters for the lower deviation; l dc,t ,
l -dc,t are the actual value of the maximum power of the DC load in the t period, the predicted nominal value, the predicted upper deviation value and the predicted lower deviation value; Π l, dc is the time period budget parameter of the DC load uncertainty;
And κ -dc,t are the upper deviation introduction parameters and the lower deviation introduction parameters of DC load uncertainty; l ac,t ,
l -ac,t are the actual value of the maximum power of the AC load during the t period, the predicted nominal value, the predicted upper deviation value and the predicted lower deviation value; Π l, ac is the time period budget parameter of the AC load uncertainty;
And κ -ac,t introduce parameters for the upper deviation of the AC load uncertainty and the lower deviation.
The method for robust optimization coordinated scheduling of an AC/DC hybrid microgrid according to claim 2, wherein in the step 20), the operating cost coefficient of each device in the system includes a diesel generator, an energy storage, and a two-way commutation. All operating cost factors related to the load, fan, photovoltaic and AC and DC loads, the operating cost coefficient is substituted into the following formula (5), that is, the objective function of the robust optimal scheduling model in the form of min-max-min is established:

The relevant parameters in equation (5) are calculated according to the following formula:

In the formula,
with
The starting cost and shutdown cost of the diesel generator are respectively;
The fuel cost for diesel generators;
with
The operation and maintenance costs of diesel generators, energy storage, bidirectional converters, fans and photovoltaics;
Cost for energy storage;
Punish the cost of power cuts,
with
They are the starting and shutting down cost coefficients of the diesel generators respectively; I DE, t is the starting sign of the diesel generator in the t period, 1 means that the diesel generator is started in the t period, 0 means the diesel generator is not started in the t period ;M DE,t is the shutdown flag of the diesel generator in the t period, 1 means that the diesel generator is shut down during the t period, 0 means that the diesel generator is not shut down during the t period; U DE,t represents the diesel fuel in the t period The running state of the generator, when the value is 1, indicates that the diesel generator is in the on state during the t period, and the value of 0 indicates that the diesel generator is in the stop state during the t period; U acdcBC, t is the t period bidirectional converter To the commutation operation status flag, 1 indicates that there is a forward commutation in the t period, 0 indicates that there is no forward commutation in the t period, U dcacBC, t is the t- zone bidirectional converter negative commutation operation status flag, 1 It means that there is a negative commutation in the t period, and 0 means that there is no negative commutation in the t period;
The fuel cost coefficient of the diesel generator; a DE and b DE are the fuel consumption characteristic cost coefficients of the diesel generator; P DE,t is the operating power of the diesel generator during the t period;
Is the rated power of the diesel generator; Δt is the time interval of two periods;
with
The operating and maintenance cost factors for diesel generators, energy storage, bidirectional converters, fans and photovoltaics;
Cost coefficient for energy storage loss;
The cost coefficient for the power cut-off penalty; P chES,t and P disES,t are the charging power and the discharging power of the energy storage in the t period respectively; P acdcBC,t is the bidirectional converter from the alternating current bus to the DC bus in the t period Forward commutating power; P dcacBC, t is the negative commutating power of the bidirectional converter from the DC bus to the AC bus during the t period; P WT, t and P PV, t are the power generation of the fan and photovoltaic respectively during the t period Power; P cut, acL, t and P cut, dcL, t respectively represent the load power of the AC zone cut off during the t period and the load power of the DC zone cut off; P tran, acL, t and P tran, dcL, t are t The time period communication can schedule the load operation power and the DC schedulable load operation power.
The method for robust optimization coordinated scheduling of an AC/DC hybrid microgrid according to claim 3, wherein in the step 30), the operating limit of each device in the system is a diesel generator, an energy storage, and a two-way commutation. All operating limits related to the load, fan, photovoltaic and AC and DC loads, and the operating limits are substituted into the following equations (10) to (22), that is, the constraints of the robust optimal scheduling model in the form of min-max-min are established. condition:

0 ≤ P WT, t ≤ w t , 0 ≤ P PV, t ≤ p t (10)

I DE,t +M DE,t ≤1,I DE,t -M DE,t =U DE,t -U DE,t-1 (13)

Equation (10) is the power generation constraint of the fan and photovoltaic; the equations (11)-(13) are the minimum continuous startup time, minimum continuous shutdown time and maximum continuous startup time constraint of the diesel generator.
with
They are the minimum continuous start-up period limit of the diesel generator, the minimum continuous shutdown period limit and the maximum continuous start-up period limit; the formula (14) is the upper and lower limits of the diesel generator operating power and the climbing speed constraint.
with
For the upper and lower limits of the operating power of the diesel generator in the on state,
with
It is the rate limit for the downhill and uphill slopes of the diesel generator in the unit time period; the formula (15) is the maximum charge and discharge power constraint of the energy storage.
with
The maximum charge and discharge power limit for energy storage; Equation (16) is the energy storage state constraint, and S min and S max are the lower and upper limits of the energy storage allowable state, S(t) and S(t-1) is the state of charge of energy storage in the period t and t-1, η C and η D are the charging and discharging efficiency limits of energy storage, and S(0) is the initial state of charge of energy storage. , S(N t ) is the state of charge of the stored energy at the end of the scheduling period; equations (17)-(18) are the commutation power and power fluctuation constraints of the bidirectional converter.
with
Indicates the operating power limits for forward commutation and negative commutation,
with
The lower limit value and the upper limit value of the power fluctuation of the bidirectional converter in the adjacent time period; the equations (19)-(20) are the operating power of the AC and DC cut-off load and the schedulable load, and the AC-DC schedulable load for each time period. Power constraints, P cut, ac, maxL, t and P cut, dc, maxL, t are the maximum removable load power limits for AC and DC during t periods, P tran, ac, maxL, t and P tran, dc, maxL,t is the maximum operating power limit for AC and DC schedulable loads in t-period, [t ac,1 ,t ac,end ] is the operating time interval limit for AC schedulable loads, [t dc,1 ,t dc , end ] is the operating time interval limit of the DC schedulable load,
with
Is the planned power consumption limit for AC and DC schedulable loads; Equations (21)-(22) are power balance constraints for DC and AC zones,
with
The forward and negative commutation efficiency limits for the bidirectional converter.
The method of claim 4, wherein the specific content of the step 40) comprises:

Step 401): Write the robust optimal scheduling model in the form of min-max-min represented by equations (1)-(22) into the following matrix representation:

s.t.A·x≤b, B·x=e,x∈{0,1} (24)

D·y≤f, E·y=g, (25)

F·y≤h-G·x, (26)

J·y≤w, K·y≤p, (27)

M·y=l dc , N·y=l ac (28)

Where x is the 0-1 state variable of the first phase of equation (5), y is the second phase power variable, and w, p, l dc , l ac are the set of the second phase uncertainty set variables, c d is a constant matrix in the objective function; Equation (24) represents a constraint condition only related to x, A, b, B, and e are constant matrices in the constraint; Equation (25) represents only y related Constraints, D, f, E, g are the constant matrices in the constraint; Equation (26) represents the constraints associated with x and y, F, h, G are the constant matrices in the constraint; Represents constraints associated with w, p, and y, where J, w, K, and p are constant matrices in the constraint; Eq. (28) represents constraints associated with l dc , l ac , and y , M, N Both are constant matrices in this constraint.

Step 402): Based on the model of the matrix representation in step 401), the sub-problem of the robust optimal scheduling model in the form of min-max-min using the column constraint generation algorithm is as follows:

Wherein α, β, χ, γ, ψ, μ dc and μ ac are the dual variables of y in the formulae (25)-(28).

Step 403): Based on the model of the matrix representation in step 401) and the sub-problem of step 402), the main problem of using the column constraint generation algorithm to form the robust optimal scheduling model in the form of min-max-min is as follows:

Where l is the total number of iterations, k is the current number of iterations, and the x optimized by the main problem is sub-problem as a known variable, w k , p k , l dc,k , l ac,k are after the kth iteration The optimization result of w, p, l dc and l ac in the sub-problem, y k is the optimization result of y in the sub-problem after the k-th iteration; η is the optimization variable related to the objective function value of the sub-problem.

Step 404): Using the integer optimization modeling toolbox YALMIP to call the solver SCIP iteratively solve the sub-problem of step 402) and the main problem of step 403) to obtain a robust coordinated operation mode of the AC-DC hybrid microgrid.