CN117791610A - Multi-microgrid energy scheduling optimization method based on ADMM algorithm - Google Patents

Abstract

The invention discloses an ADMM algorithm-based multi-micro-grid energy scheduling optimization method. The method aims at optimizing the multi-micro-grid energy management system so as to reduce the running cost of each micro-grid. The method comprises the following implementation steps: (1) Collecting wind-solar power generation data and required load data of a micro-grid, and preprocessing the data; (2) constructing a multi-micro-grid energy scheduling framework; (3) establishing a mathematical model of the micro-grid; (4) establishing an energy scheduling optimization strategy of the micro-grid; (5) establishing an energy scheduling optimization strategy of the multi-micro-grid; (6) Adopting an ADMM algorithm to optimize the energy scheduling framework of the multi-micro power grid; (7) verifying the validity of the proposed algorithm. According to the invention, the energy sharing among the micro-grids is realized by carrying out mathematical modeling on the multiple micro-grids and constructing an energy scheduling optimization strategy, so that the economic cost is saved, the utilization rate of renewable energy power generation is improved, and finally the energy scheduling optimization of the multiple micro-grids is realized.

Description

Multi-microgrid energy scheduling optimization method based on ADMM algorithm

Technical Field

The invention belongs to the technical field of energy scheduling of micro-grids, in particular to the field of energy scheduling optimization of interconnected multi-micro-grids, and particularly relates to a multi-micro-grid energy scheduling optimization method based on an ADMM algorithm.

Background

The rapid development of renewable energy power generation technology brings great benefits to economic development, ecological environment and the like of various countries; the micro-grid is used as a small power generation and distribution system integrating distributed power sources, energy storage systems, energy conversion devices, loads, communication devices and other devices, and generally utilizes renewable energy sources such as solar energy, wind energy and the like to generate electric power. The aim is to provide reliable, efficient and flexible electric power service, and reduce the dependence on a central power grid; the development and extension of the micro-grid can fully promote the large-scale access of the distributed power supply and the renewable energy sources, realize the high-reliability supply of various energy forms of loads, and be an effective way for realizing an active power distribution network, so that the traditional power grid is transited to the intelligent power grid;

as the number of micro-grids is rapidly increased, the internal load demand of the micro-grids is expanded; to increase communication and interaction with nearby micro-grids, each micro-grid in the nearby area forms an integrated, flexible network comprising a plurality of independent micro-grids; these micro-grids are geographically close and connected to a distribution bus. The development of multiple micro-grids can reduce the operation cost of the micro-grids, so that a large amount of local renewable energy sources can be timely consumed, and resource waste is avoided; moreover, real-time energy sharing can be realized between micro-grids, the burden of a main grid is reduced, the congestion of a distribution line is relieved, and the reliability of the micro-grids and the main grid is improved;

the micro-grid energy management system forms a control center to carry out global control on each device of the micro-grid through information flow transmission in the line. The operation purpose is mainly to acquire and analyze energy input information, optimize according to a preset operation target to obtain a scheduling plan, and continuously control the energy flow in the system; each independent micro-grid is equivalent to an independent intelligent body, and can perform energy interaction with a main power grid and also perform energy sharing with nearby micro-grids; therefore, the energy management of the micro-grids is more focused on realizing the optimal energy management in a faster and more efficient manner among the micro-grids, so that the economic benefit is improved and the renewable energy utilization rate is improved.

Disclosure of Invention

The invention aims to provide an ADMM (adaptive modulation) algorithm-based multi-micro-grid energy scheduling optimization method, which comprises the steps of constructing a mathematical model of each device in a micro-grid, establishing an optimal scheduling objective function of the micro-grid, constructing a distributed multi-micro-grid structure, and performing energy scheduling optimization on the multi-micro-grid by adopting the ADMM algorithm; the method aims at reducing the economic cost of multi-microgrid operation, effectively improving the energy sharing efficiency among the interconnected multi-microgrids and improving the utilization rate of renewable energy.

In order to achieve the above purpose, the method for optimizing multi-micro grid energy scheduling based on ADMM algorithm comprises the following implementation steps:

1) Collecting wind-solar power generation data and load data of a micro-grid, and preprocessing; preprocessing acquired sample data;

2) Constructing an energy scheduling frame of a plurality of micro-grids, wherein the energy scheduling frame of the plurality of micro-grids is a distributed type interconnected multi-micro-grid system;

the little electric wire netting of many comprises a plurality of little electric wire netting, carries out information interaction through communication equipment between the little electric wire netting, little electric wire netting mainly includes: controllable power generation equipment, renewable energy power generation equipment, a battery energy storage system and a load;

3) Establishing a mathematical model of each micro-grid, performing mathematical modeling on each structure inside each micro-grid, and establishing constraint conditions of each device;

4) An energy scheduling optimization strategy of the micro-grid is established, and the purposes of reducing the utilization rate of non-renewable energy sources, prolonging the service life of energy storage equipment and improving the satisfaction of users are achieved by optimizing each part in the micro-grid;

5) Establishing an energy scheduling optimization strategy of the multiple micro-grids, so that adjacent micro-grids in an energy scheduling frame of the multiple micro-grids perform information interaction, and the micro-grids with surplus electricity can perform energy interaction with the micro-grids with insufficient surrounding electricity generation capacity to support load requirements in the electricity consumption peak period, thereby finally achieving the aims of improving the utilization rate of new energy and saving economic cost;

6) Optimizing an energy scheduling framework of the multi-micro grid by adopting an ADMM algorithm, and optimizing a distributed mode by using an alternate direction multiplier method (Alternating Direction Method of Multipliers, ADMM); the method is applied to the energy scheduling process of the multi-micro power grid, so that the energy scheduling process is more efficient, and a better energy sharing effect is realized.

Collecting wind-light power generation data and load data of a micro-grid, wherein the collecting of wind-light power generation data and photovoltaic power generation data in 24h of the micro-grid and consumed electricity quantity, namely load data, in 24h of users in the area; and maximum storage capacity of energy storage equipment in the micro-grid and maximum power generation data of each power generation equipment;

the preprocessing of the acquired sample data comprises the following steps:

1) Processing the abnormal data by adopting a statistical discrimination method of a 3 sigma criterion;

the statistical discrimination method of the 3 sigma criterion comprises the following steps:

the acquired sample data is y= { Y ₁ ,y ₂ ,…,y _n Mean value of } isDeviation of->Standard deviation is obtained according to the Bessel formula:

if a certain sample data y _i Deviation |e of (2) _i 3 sigma, then consider this data unreasonable, which should be culled from the sample data; the comparison is cycled until all sample data is detected.

The controllable power generation equipment is traditional equipment for generating power by using non-renewable energy sources of diesel and natural gas, and comprises a diesel generator set and a micro gas turbine;

the renewable energy power generation equipment comprises wind power generation equipment and photovoltaic power generation equipment;

the battery energy storage system is energy storage equipment;

the load comprises an uncontrollable load and a controllable load, wherein the uncontrollable load is a type of load necessary for production and life; the controllable compliance can be adjusted in real time according to the user's power demand, and user satisfaction must be considered when adjusting with load shedding behavior.

The step of establishing a mathematical model of the micro-grid is as follows:

1) Power generation model of micro gas turbine:

when the renewable energy generating capacity in the micro-grid is insufficient, the micro-gas turbine is used as an auxiliary call resource to generate electricity, and the relation between the output power and the cost can be expressed as follows:

C _DG (P _DG )＝α _DG (P _DG ) ² +β _DG P _DG +γ _DG ；

wherein: p (P) _DG Alpha is the output power of the micro gas turbine _DG 、β _DG 、γ _DG Are all the cost coefficients of power generation;

2) Power generation model of wind power generation equipment:

the relationship between the output power of the wind generator (WT) and the wind speed V can be expressed as:

wherein: p (P) ^wind And P _WN The actual output power and rated power of the wind driven generator are respectively; v _R ，v _o And v _J The rated wind speed, cut-out wind speed and cut-in wind speed of the wind driven generator are respectively;

3) Generating model of photovoltaic power generation equipment:

the output power of a photovoltaic generator set (PV) is expressed as:

wherein: p (P) _PVN At a standard temperature T _std (25 ℃) and solar irradiance I _std (1000W/m ² ) Lower power; i _act Is the actual solar irradiance; alpha is a coefficient; t (T) _PVC Is the temperature of the PV cell;

4) Energy storage device model:

in the micro-grid energy storage system, the battery can age due to cyclic charge and discharge, so the life loss of the battery is expressed as:

wherein ρ is the calculated percentage of battery life loss; e (E) _o Effectively accumulating throughput for a certain period of time; e (E) _o,all Effectively accumulating throughput under the whole life cycle; calculation E _o The method comprises the following steps:

τ _SOC ＝α _ESS ·SOC+β _ESS ；

wherein: τ _SOC Alpha is the effective weight _ESS And beta _ESS Is an experience coefficient; SOC is the state of charge of the battery, constrained by:

SOC _min ≤SOC≤SOC _max ；

wherein: SOC (State of Charge) _min And SOC (System on chip) _max A safe upper and lower boundary for battery energy storage; further, the power output limit of the battery energy storage system is:

wherein:and->Respectively expressed as upper and lower limits of maximum allowable charge and discharge power of the battery, when P _ESS Positive, indicating that the battery is discharging; when negative, the battery is charged;

W _k +Z _k ≤1，W _k ,Z _k e {0,1} is a binary integer variable, which represents a charge-discharge state, so that the battery is prevented from being charged and discharged simultaneously;

5) Load model:

in the micro-grid scheduling process, the user required load model can be expressed as:

P _L ＝P _LU +P _LC ；

wherein P is _LU ，P _LC Representing the resectable load and the residual load, respectively.

The energy scheduling optimization strategy of the micro-grid is established, and the method comprises the following steps:

since wind power generation and photovoltaic power generation are clean energy sources, in order to minimize the power generation cost, renewable energy sources are fully scheduled, and the optimal scheduling objective function of the ith micro grid can be expressed as:

minC _i (P _i )＝min(h _i (P _i )+ξ(P _i ^exc )；

h _i (P _i )＝C _i,DG (P _i ^DG )+C _i,ESS (P _i,ESS )+C _i,G (P _i,G )；

combining the two formulas to obtain:

minC _i (P _i )＝min[C _i,DG (P _i ^DG )+C _i,ESS (P _i,ESS )+C _i,G (P _i,G )+ξ(P _i ^exc )]；

the establishment constraint conditions are as follows:

wherein, xi (P) _i ^exc ) Network transmission cost generated for the i-th micro-grid and other micro-grids interactive power; c (C) _i (P _i ) The total cost comprises the power generation cost of the controllable power generation equipment, the battery equipment cost of the energy storage system and the cost of energy scheduling between the ith micro power grid and the main power grid; p (P) _i ^exc In the positive, the micro-grid purchases electric energy from the power grid; otherwise, the electric energy is sold to the power grid; wherein the constraints include a power balance constraint, a maximum output power that limits the micro gas turbine, a maximum charge of the battery in the energy storage system, and a battery state of charge constraint.

The energy scheduling optimization strategy for the multi-micro-grid is established, and the method comprises the following steps:

in the multi-microgrid interconnected with the microgrids, the microgrids are independent autonomous microgrids, a plurality of microgrids form the interconnected multi-microgrid together, and an energy management system of the microgrid can perform information interaction with a main power grid and other microgrids; the multi-micro-grid optimization strategy is mainly to formulate an electric energy dispatching optimization strategy based on the topological structure of the multi-micro-grid system, so that the minimum total running cost of the multi-micro-grid is realized, and the utilization rate of renewable energy sources is improved; the optimal scheduling objective function for a multi-microgrid consisting of N microgrids can be expressed as:

its constraints can be expressed as:

wherein: v _i ＝[P _iG ,P _ij ]Exchanging power vectors for the ith and the main grids, the ith and the jth micro grids, P _i ^exc Is a vector of the power exchanged between the individual micro-grids.

The method adopts an ADMM algorithm to optimize the energy scheduling framework of the multi-micro-grid, and comprises the following steps:

the standard form for distributed optimization using the alternate direction multiplier method is as follows:

the standard form of ADMM algorithm is:

min[m(P)+w(v)]；

s.t. QP+Yv＝c；

P ^exc -Yv＝0；

wherein: p epsilon R ⁱ ；v∈R ^j ；Q∈R ^p×i ；Y∈R ^p×j ；c∈R ^p The method comprises the steps of carrying out a first treatment on the surface of the Y represents a connection relation matrix of the micro-grid, c is a constant matrix, and represents a difference value between the load of the micro-grid and the power generation amount of renewable energy sources; decomposing a multi-microgrid energy scheduling optimization problem objective function into two parts m (P), w (v), wherein the two parts are convex functions related to P and v; when m approaches R ⁱ →R∪{+∞}W approximates R ^j R { + in the case of infinity }, the ADMM algorithm converges and, obtaining an optimal solution for the optimization problem;

the iteration process of the ADMM algorithm is that the sub-problems are alternately performed according to the sequence, the solving result of the former sub-problem is substituted into the latter sub-problem to carry out optimization solving, and after one iteration of all the sub-problems is finished, the Lagrange multiplier is updated; the distributed optimization iterative process can be expressed as:

wherein k is the iteration number; psi is penalty parameter; lambda (lambda) ^k Is the lagrangian multiplier at the kth iteration.

Based on the energy scheduling problem of the multi-micro power grid is optimized by adopting an ADMM algorithm, and the algorithm has higher convergence speed and can find a solution with higher quality; the algorithm can ensure convergence to the global optimal solution, and has good stability and robustness; in addition, the algorithm can be effectively applied to a plurality of micro-grid distributed systems, allows a plurality of computing units to respectively process respective data, and strengthens cooperation of the various parts through information interaction.

The invention discloses an energy scheduling optimization method for a multi-micro-grid based on an ADMM algorithm, which has the following intentional effects: the method constructs the multi-micro power grid into a distributed structure to carry out communication interaction; the computational complexity during energy scheduling is reduced, and the reliability and the robustness of the micro-grid are improved; in addition, the structure of each part of the micro-grid is considered in the energy scheduling optimization process, so that the running cost of equipment is reduced, the utilization rate of renewable energy sources is improved, and the cycle life of batteries in an energy storage system is prolonged; the optimization of each part in the micro-grid dispatching process is fully considered, and the solution is carried out by simplifying an objective function and alternating iteration based on an ADMM algorithm. Compared with the prior art, the method has better convergence, and is a simple and efficient algorithm.

Drawings

FIG. 1 is a schematic flow chart;

FIG. 2 is a schematic diagram of a multi-microgrid structure;

FIG. 3 is a graph of wind and solar power generation data within three micro-grids 24 h;

FIG. 4 is a graph of the number of iterations using the ADMM algorithm;

FIG. 5 is a graph comparing load data before and after optimization of three microgrid algorithms;

fig. 6 is an energy scheduling result diagram of the micro grid 1;

fig. 7 is an energy scheduling result diagram of the micro grid 2;

fig. 8 is an energy scheduling result diagram of the micro grid 3;

fig. 9 is a graph of three microgrid energy interaction results.

Detailed Description

Example 1

The invention discloses a multi-micro-grid energy scheduling optimization method based on an ADMM algorithm, which is shown in figure 1 and comprises the following steps:

1. collecting wind-solar power generation data and load data of a micro-grid, and preprocessing; preprocessing acquired sample data;

collecting the power generation data of the micro-grids comprises collecting wind power generation data and photovoltaic power generation data in 24h of each independent micro-grid and the consumed electricity quantity, namely load data, in 24h of users in the region; maximum storage capacity of energy storage equipment in the micro-grid, maximum power generation of each power generation equipment and other data;

preprocessing the obtained sample data, and processing the abnormal data by adopting a statistical discrimination method of a 3 sigma criterion;

the acquired sample data is y= { Y ₁ ,y ₂ ,…,y _n Mean value of } isDeviation of->Standard deviation is obtained according to the Bessel formula:

if a certain sample data y _i Deviation |e of (2) _i 3 sigma, then consider this data unreasonable, which should be culled from the sample data; the comparison is circularly carried out until all sample data are detected;

2. constructing an energy scheduling frame of a plurality of micro-grids, wherein the energy scheduling frame of the plurality of micro-grids is a distributed type interconnected multi-micro-grid system;

the little electric wire netting of many comprises a plurality of little electric wire netting, carries out information interaction through communication equipment between the little electric wire netting, little electric wire netting mainly includes: controllable power generation equipment, renewable energy power generation equipment, a battery energy storage system and a load;

the controllable power generation device: in the micro-grid, the controllable power generation equipment refers to traditional equipment for generating power by using non-renewable energy sources such as diesel, natural gas and the like, such as a diesel generator set, a micro gas turbine and the like; the purpose of using the equipment in the micro-grid is to ensure the safe and stable operation of various primary power supply systems (such as medical equipment power supply, emergency communication systems and the like); in addition, the controllable power generation equipment can also cope with instability and the like caused by the fluctuation of the renewable energy power generation equipment;

renewable energy power generation equipment: one big characteristic of the micro-grid is that the flexible and efficient application of the distributed power supply is realized, and the large-scale access of renewable energy sources is promoted; therefore, the renewable energy power generation equipment is connected with the micro-grid to supply power, so that not only can clean and environment-friendly energy supply be realized, but also the economic cost can be reduced, and the development of the intelligent grid is promoted;

the renewable energy power generation equipment comprises wind power generation equipment and photovoltaic power generation equipment;

the battery energy storage system: because of the fluctuation and uncertainty of renewable energy power generation, redundant renewable energy is stored to timely supplement energy demand gaps, and the supply and demand balance in the running process of the micro-grid is ensured;

the load demand generally shows a relatively stable change trend in the historical data, so that the load data in a specific area can be predicted; the load types comprise uncontrollable loads and controllable loads, wherein the uncontrollable loads refer to loads which are necessary for production and life, and the loads meet the requirements; the controllable coincidence can be adjusted in real time according to the electricity demand of the user, and the user satisfaction degree must be considered when the load cutting behavior is used for adjustment;

3. establishing a mathematical model of each micro-grid, performing mathematical modeling on each structure inside each micro-grid, and establishing constraint conditions of each device; the step of establishing a mathematical model of the micro-grid is as follows:

1) Power generation model of micro gas turbine:

when the renewable energy generating capacity in the micro-grid is insufficient, the micro-gas turbine is used as an auxiliary call resource to generate electricity, and the relation between the output power and the cost can be expressed as follows:

C _DG (P _DG )＝α _DG (P _DG ) ² +β _DG P _DG +γ _DG ；

wherein: p (P) _DG Alpha is the output power of the micro gas turbine _DG 、β _DG 、γ _DG Are all the cost coefficients of power generation;

2) Power generation model of wind power generation equipment:

the relationship between the output power of the wind generator (WT) and the wind speed V can be expressed as:

wherein: p (P) ^wind And P _WN The actual output power and rated power of the wind driven generator are respectively; v _R ，v _o And v _J Rated for wind power generatorsWind speed, cut-out wind speed and cut-in wind speed;

3) Generating model of photovoltaic power generation equipment:

the output power of a photovoltaic generator set (PV) is expressed as:

wherein: p (P) _PVN At a standard temperature T _std (25 ℃) and solar irradiance I _std (1000W/m ² ) Lower power; i _act Is the actual solar irradiance; alpha is a coefficient; t (T) _PVC Is the temperature of the PV cell;

4) Energy storage device model:

in the micro-grid energy storage system, the battery can age due to cyclic charge and discharge, so the life loss of the battery is expressed as:

wherein ρ is the calculated percentage of battery life loss; e (E) _o Effectively accumulating throughput for a certain period of time; e (E) _o,all Effectively accumulating throughput under the whole life cycle; calculation E _o The method comprises the following steps:

τ _SOC ＝α _ESS ·SOC+β _ESS ；

wherein: τ _SOC Alpha is the effective weight _ESS And beta _ESS Is an experience coefficient; SOC is the state of charge of the battery, constrained by:

SOC _min ≤SOC≤SOC _max ；

wherein: SOC (State of Charge) _min And SOC (System on chip) _max A safe upper and lower boundary for battery energy storage; further, the power output limit of the battery energy storage system is:

wherein:and->Respectively expressed as upper and lower limits of maximum allowable charge and discharge power of the battery, when P _ESS Positive, indicating that the battery is discharging; when negative, the battery is charged;

W _k +Z _k ≤1，W _k ,Z _k e {0,1} is a binary integer variable, which represents a charge-discharge state, so that the battery is prevented from being charged and discharged simultaneously;

5) Load model:

in the micro-grid scheduling process, the user required load model can be expressed as:

P _L ＝P _LU +P _LC ；

wherein P is _LU ，P _LC Represent the resectable load and the residual load, respectively;

4. an energy scheduling optimization strategy of the micro-grid is established, and the purposes of reducing the utilization rate of non-renewable energy sources, prolonging the service life of energy storage equipment and improving the satisfaction of users are achieved by optimizing each part in the micro-grid;

since wind power generation and photovoltaic power generation are clean energy sources, in order to minimize the power generation cost, renewable energy sources are fully scheduled, and the optimal scheduling objective function of the ith micro grid is expressed as:

minC _i (P _i )＝min(h _i (P _i )+ξ(P _i ^exc )；

h _i (P _i )＝C _i,DG (P _i ^DG )+C _i,ESS (P _i,ESS )+C _i,G (P _i,G )；

combining the two formulas to obtain:

min C _i (P _i )＝min[C _i,DG (P _i ^DG )+C _i,ESS (P _i,ESS )+C _i,G (P _i,G )+ξ(P _i ^exc )]；

the establishment constraint conditions are as follows:

wherein, xi (P) _i ^exc ) Network transmission cost generated for the i-th micro-grid and other micro-grids interactive power; c (C) _i (P _i ) The total cost comprises the electricity generation cost of controllable electricity generation equipment, the cost of battery equipment of an energy storage system and the like, and the cost of energy scheduling between an ith micro-grid and a main grid; p (P) _i ^exc In the positive, the micro-grid purchases electric energy from the power grid; otherwise, the electric energy is sold to the power grid; wherein the constraints include a power balance constraint, a maximum output power limiting the micro gas turbine, a maximum charge of a battery in the energy storage system, and a battery state of charge constraint;

5. establishing an energy scheduling optimization strategy of the multiple micro-grids, so that adjacent micro-grids in an energy scheduling frame of the multiple micro-grids perform information interaction, and the micro-grids with surplus electricity can perform energy interaction with the micro-grids with insufficient surrounding electricity generation capacity to support load requirements in the electricity consumption peak period, thereby finally achieving the aims of improving the utilization rate of new energy and saving economic cost;

in the micro-grid interconnection multi-micro-grid structure, each micro-grid is an independent self-known micro-grid, a plurality of micro-grids jointly form an interconnection multi-micro-grid system, and an energy management system of the micro-grid performs information interaction with a main grid and other micro-grids. The multi-micro-grid optimization strategy is mainly to formulate an electric energy dispatching optimization strategy based on the topological structure of the multi-micro-grid system, so that the minimum total running cost of the multi-micro-grid is realized, and the utilization rate of renewable energy sources is improved; the optimal scheduling objective function for the multiple micro-grids consisting of N micro-grids is expressed as:

its constraints can be expressed as:

wherein: v _i ＝[P _iG ,P _ij ]Exchanging power vectors for the ith and the main grids, the ith and the jth micro grids, P _i ^exc Is a vector of the power exchanged between the individual micro-grids;

6. optimizing an energy scheduling framework of the multi-micro grid by adopting an ADMM algorithm, and optimizing a distributed mode by using an alternate direction multiplier method (Alternating Direction Method of Multipliers, ADMM); the method is applied to the energy scheduling process of the multi-micro power grid, so that the energy scheduling process is more efficient, and a better energy sharing effect is realized;

for the distributed optimization problem, the method is a simple, efficient and robust optimization algorithm, has good convergence, and does not require that the objective function of the optimization problem is a strict convex function; ADMM integrates the decomposability of a dual rising method and the better convergence of a multiplier method, and the core idea is that a big problem is decomposed into a plurality of small problems, and the small problems are alternately iterated to solve, so that the original target and the dual variable are converged together, and the standard form is as follows:

the standard form of ADMM algorithm is:

min[m(P)+w(v)]；

s.t. QP+Yv＝c；

P ^exc -Yv＝0；

wherein: p epsilon R ⁱ ；v∈R ^j ；Q∈R ^p×i ；Y∈R ^p×j ；c∈R ^p The method comprises the steps of carrying out a first treatment on the surface of the Y represents a connection relation matrix of the micro-grid, and c represents a constant matrix, which represents the difference value between the load of the micro-grid and the power generation amount of renewable energy sources. The multi-microgrid energy scheduling optimization problem objective function is decomposed into two parts m (P), w (v), where the two parts are convex functions for P and v. When m approaches R ⁱ R and U { ++ infinity is }, w approximates R ^j R { + in the case of infinity }, ADMM algorithm receiptConverging and optimizing the problem to obtain an optimal solution;

the iteration process of the ADMM algorithm is that the sub-problems are alternately performed according to the sequence, the solving result of the former sub-problem is substituted into the latter sub-problem to carry out optimization solving, and after one iteration of all the sub-problems is finished, the Lagrange multiplier is updated; the distributed optimization iterative process can be expressed as:

wherein k is the iteration number; psi is penalty parameter; lambda (lambda) ^k Is the lagrangian multiplier at the kth iteration.

Example 2

The invention discloses a multi-micro-grid energy scheduling optimization method based on an ADMM algorithm, which comprises the following implementation steps:

constructing a multi-micro-grid structure comprising three micro-grids, wherein the first micro-grid adopts a fan to generate electricity, the second micro-grid and the third micro-grid adopt a photovoltaic panel to generate electricity, the generating capacity data of the multi-micro-grid structure are shown in figure 3, and the figure shows the specific generating capacity of renewable energy sources within the three micro-grids 24 h; the configuration parameters are shown in table 1; the energy storage device parameter configurations of the three micro-grids are shown in table 2;

table 1 multiple microgrid system configuration parameters

Table 2 microgrid energy storage system parameters

Table 3 micro grid time-of-use electricity purchase price

Type(s)	Peak time electricity price	Level of charge at ordinary times	Electricity price at valley time
				Electricity purchasing (Yuan/KW)	1.241	0.779	0.488
Electricity vending (Yuan/KW)	0.620	0.387	0.244

In the embodiment, matlab2023a is adopted for experimental simulation, mathematical modeling is carried out after a distributed structure of three micro-grids is constructed, then an ADMM algorithm is adopted for optimizing the three micro-grids, and FIG. 4 shows the convergence condition of the algorithm, which shows the effectiveness of the method; FIG. 5 shows a comparison of three microgrid load optimizations before and after, respectively; it can be seen that the load requirements of the three micro-grids are relatively reduced after optimization. On the premise of ensuring the satisfaction degree of users, the load demands of the users are timely adjusted, and the purposes of saving energy and cost are achieved; FIGS. 6-8 are each optimized energy scheduling for three micro-grids; in the micro grid 1, it can be seen that in the peak electricity consumption period 9:00-17: in 00, the power generation amount of renewable energy sources in the period of the micro power grid 1 is far smaller than the load demand, and after optimization, the micro power grid in the period does not need to purchase power to the power grid any more, the user demand can be met by interacting with other two micro power grids, and the purposes of saving cost and sharing energy sources are achieved; the micro-grid 2 and the micro-grid 3 share surplus electric quantity to the micro-grid 1 or sell the surplus electric quantity to the power grid in the period, so that the full utilization of renewable energy sources is realized; fig. 9 shows the case of three microgrids energy interactions after optimization by the ADMM algorithm.

Claims (7)

1. An ADMM algorithm-based multi-microgrid energy scheduling optimization method is characterized by comprising the following steps of: the implementation steps are as follows:

1) Collecting wind-solar power generation data and load data of a micro-grid, and preprocessing; preprocessing acquired sample data;

2) Constructing an energy scheduling frame of a plurality of micro-grids, wherein the energy scheduling frame of the plurality of micro-grids is a distributed type interconnected multi-micro-grid system;

the little electric wire netting of many comprises a plurality of little electric wire netting, carries out information interaction through communication equipment between the little electric wire netting, little electric wire netting mainly includes: controllable power generation equipment, renewable energy power generation equipment, a battery energy storage system and a load;

3) Establishing a mathematical model of each micro-grid, performing mathematical modeling on each structure inside each micro-grid, and establishing constraint conditions of each device;

4) An energy scheduling optimization strategy of the micro-grid is established, and the purposes of reducing the utilization rate of non-renewable energy sources, prolonging the service life of energy storage equipment and improving the satisfaction of users are achieved by optimizing each part in the micro-grid;

5) Establishing an energy scheduling optimization strategy of the multiple micro-grids, so that adjacent micro-grids in an energy scheduling frame of the multiple micro-grids perform information interaction, and the micro-grids with surplus electricity can perform energy interaction with the micro-grids with insufficient surrounding electricity generation capacity to support load requirements in the electricity consumption peak period, thereby finally achieving the aims of improving the utilization rate of new energy and saving economic cost;

6) And optimizing an energy scheduling framework of the multi-micro power grid by adopting an ADMM algorithm, and optimizing a distributed type by using an alternate direction multiplier method.

2. The multi-microgrid energy scheduling optimization method based on the ADMM algorithm as claimed in claim 1, wherein the method comprises the following steps: collecting wind-light power generation data and load data of a micro-grid, wherein the collecting of wind-light power generation data and photovoltaic power generation data in 24h of the micro-grid and consumed electricity quantity, namely load data, in 24h of users in the area; and maximum storage capacity of energy storage equipment in the micro-grid and maximum power generation data of each power generation equipment;

the preprocessing of the acquired sample data comprises the following steps:

1) Processing the abnormal data by adopting a statistical discrimination method of a 3 sigma criterion;

the statistical discrimination method of the 3 sigma criterion comprises the following steps:

the acquired sample data is y= { Y ₁ ,y ₂ ,…,y _n Mean value of } isDeviation of->Standard deviation is obtained according to the Bessel formula:

if a certain sample data y _i Deviation |e of (2) _i 3 sigma, then consider this data unreasonable, which should be culled from the sample data; the comparison is cycled until all sample data is detected.

3. The multi-microgrid energy scheduling optimization method based on the ADMM algorithm as claimed in claim 2, wherein the method comprises the following steps: the controllable power generation equipment is traditional equipment for generating power by using non-renewable energy sources of diesel and natural gas, and comprises a diesel generator set and a micro gas turbine;

the renewable energy power generation equipment comprises wind power generation equipment and photovoltaic power generation equipment;

the battery energy storage system is energy storage equipment;

the load comprises an uncontrollable load and a controllable load, wherein the uncontrollable load is a type of load necessary for production and life; the controllable compliance can be adjusted in real time according to the user's power demand, and user satisfaction must be considered when adjusting with load shedding behavior.

4. The multi-microgrid energy scheduling optimization method based on the ADMM algorithm as claimed in claim 3, wherein the method comprises the following steps: the step of establishing a mathematical model of the micro-grid is as follows:

1) Power generation model of micro gas turbine:

when the renewable energy generating capacity in the micro-grid is insufficient, the micro-gas turbine is used as an auxiliary call resource to generate electricity, and the relation between the output power and the cost can be expressed as follows:

C _DG (P _DG )＝α _DG (P _DG ) ² +β _DG P _DG +γ _DG ；

wherein: p (P) _DG Alpha is the output power of the micro gas turbine _DG 、β _DG 、γ _DG Are all the cost coefficients of power generation;

2) Power generation model of wind power generation equipment:

the relationship between the output power of the wind generator (WT) and the wind speed V can be expressed as:

wherein: p (P) ^wind And P _WN The actual output power and rated power of the wind driven generator are respectively; v _R ，v _o And v _J The rated wind speed, cut-out wind speed and cut-in wind speed of the wind driven generator are respectively;

3) Generating model of photovoltaic power generation equipment:

the output power of a photovoltaic generator set (PV) is expressed as:

wherein: p (P) _PVN At a standard temperature T _std (25 ℃) and solar irradiance I _std (1000W/m ² ) Lower power; i _act Is the actual solar irradiance; alpha is a coefficient; t (T) _PVC Is the temperature of the PV cell;

4) Energy storage device model:

in the micro-grid energy storage system, the battery can age due to cyclic charge and discharge, so the life loss of the battery is expressed as:

wherein ρ is the calculated percentage of battery life loss; e (E) _o Effectively accumulating throughput for a certain period of time; e (E) _o,all Effectively accumulating throughput under the whole life cycle; calculation E _o The method comprises the following steps:

τ _SOC ＝α _ESS ·SOC+β _ESS ；

wherein: τ _SOC Alpha is the effective weight _ESS And beta _ESS Is an experience coefficient; SOC is the state of charge of the battery, constrained by:

SOC _min ≤SOC≤SOC _max ；

wherein: SOC (State of Charge) _min And SOC (System on chip) _max A safe upper and lower boundary for battery energy storage; further, the power output limit of the battery energy storage system is:

wherein:and->Respectively expressed as upper and lower limits of maximum allowable charge and discharge power of the battery, when P _ESS Positive, indicating that the battery is discharging; when negative, the battery is charged;

W _k +Z _k ≤1，W _k ,Z _k e {0,1} is a binary integer variable, which represents a charge-discharge state, so that the battery is prevented from being charged and discharged simultaneously;

5) Load model:

in the micro-grid scheduling process, the user required load model can be expressed as:

P _L ＝P _LU +P _LC ；

wherein P is _LU ，P _LC Representing the resectable load and the residual load, respectively.

5. The multi-microgrid energy scheduling optimization method based on the ADMM algorithm as claimed in claim 4, wherein the method comprises the following steps: the energy scheduling optimization strategy of the micro-grid is established, and the method comprises the following steps:

since wind power generation and photovoltaic power generation are clean energy sources, in order to minimize the power generation cost, renewable energy sources are fully scheduled, and the optimal scheduling objective function of the ith micro grid can be expressed as:

minC _i (P _i )＝min(h _i (P _i )+ξ(P _i ^exc )；

h _i (P _i )＝C _i,DG (P _i ^DG )+C _i,ESS (P _i,ESS )+C _i,G (P _i,G )；

combining the two formulas to obtain:

minC _i (P _i )＝min[C _i,DG (P _i ^DG )+C _i,ESS (P _i,ESS )+C _i,G (P _i,G )+ξ(P _i ^exc )]；

the establishment constraint conditions are as follows:

wherein, xi (P) _i ^exc ) Network transmission cost generated for the i-th micro-grid and other micro-grids interactive power; c (C) _i (P _i ) The total cost comprises the power generation cost of the controllable power generation equipment, the battery equipment cost of the energy storage system and the cost of energy scheduling between the ith micro power grid and the main power grid; p (P) _i ^exc In the positive, the micro-grid purchases electric energy from the power grid; otherwise, the electric energy is sold to the power grid; wherein the constraints include a power balance constraint, a maximum output power that limits the micro gas turbine, a maximum charge of the battery in the energy storage system, and a battery state of charge constraint.

6. The multi-microgrid energy scheduling optimization method based on the ADMM algorithm as claimed in claim 5, wherein the method comprises the following steps: the energy scheduling optimization strategy for the multi-micro-grid is established, and the method comprises the following steps:

in the multi-microgrid interconnected with the microgrids, the microgrids are independent autonomous microgrids, a plurality of microgrids form the interconnected multi-microgrid together, and an energy management system of the microgrid can perform information interaction with a main power grid and other microgrids; the multi-micro-grid optimization strategy is mainly to formulate an electric energy dispatching optimization strategy based on the topological structure of the multi-micro-grid system, so that the minimum total running cost of the multi-micro-grid is realized, and the utilization rate of renewable energy sources is improved; the optimal scheduling objective function for a multi-microgrid consisting of N microgrids can be expressed as:

its constraints can be expressed as:

wherein: v _i ＝[P _iG ,P _ij ]Exchanging power vectors for the ith and the main grids, the ith and the jth micro grids, P _i ^exc Is a vector of the power exchanged between the individual micro-grids.

7. The multi-microgrid energy scheduling optimization method based on the ADMM algorithm as claimed in claim 6, wherein the method comprises the following steps: the method adopts an ADMM algorithm to optimize the energy scheduling framework of the multi-micro-grid, and comprises the following steps:

the standard form for distributed optimization using the alternate direction multiplier method is as follows:

the standard form of ADMM algorithm is:

min[m(P)+w(v)]；

s.t.QP+Yv＝c；

P ^exc -Yv＝0；

wherein: p epsilon R ⁱ ；v∈R ^j ；Q∈R ^p×i ；Y∈R ^p×j ；c∈R ^p The method comprises the steps of carrying out a first treatment on the surface of the Y represents a connection relation matrix of the micro-grid, c is a constant matrix, and represents a difference value between the load of the micro-grid and the power generation amount of renewable energy sources; decomposing a multi-microgrid energy scheduling optimization problem objective function into two parts m (P), w (v), wherein the two parts are convex functions related to P and v; when m approaches R ⁱ R and U { ++ infinity is }, w approximates R ^j R { + in the case of infinity }, the ADMM algorithm converges and, obtaining an optimal solution for the optimization problem;

the iteration process of the ADMM algorithm is that the sub-problems are alternately performed according to the sequence, the solving result of the former sub-problem is substituted into the latter sub-problem to carry out optimization solving, and after one iteration of all the sub-problems is finished, the Lagrange multiplier is updated; the distributed optimization iterative process can be expressed as:

wherein k is the iteration number; psi is penalty parameter; lambda (lambda) ^k Is the lagrangian multiplier at the kth iteration.