Distribution and micro-grid cooperative scheduling method considering honeycomb micro-grid group electric energy sharing

Abstract

The invention discloses a power distribution and micro-grid cooperative scheduling method considering honeycomb micro-grid group electric energy sharing, and the method comprises the steps: building a micro-grid group honeycomb model based on a non-rectangular coordinate system under the background that renewable energy participates in the daily operation of a power distribution network through a micro-grid, and positioning each unit in a system; secondly, considering the geographic position and the source load characteristic of each MG in the cluster, and dividing the whole micro-grid cluster into several micro-grid groups by controlling the starting or stopping of an intelligent base station in the system and the upper limit of the power transmission capacity of the intelligent base station; thirdly, realizing electric energy sharing in each micro-grid group by using a multi-dimensional improved alternating direction multiplier method; finally, through cooperative operation of the micro-grid group and the power distribution network, it is proved that the method can fully absorb renewable energy sources, meanwhile, the carbon emission of the power distribution network is reduced, and the environmental benefits of the power distribution network are improved. Therefore, support is provided for novel power system dispatching under the background that a large number of renewable energy microgrid clusters are accessed to the power distribution network, and the method has certain engineering use value.

Description

TECHNICAL FIELD

[0001] The application relates to a micro-grid collaborative scheduling method considering electric energy sharing of a honeycomb micro-grid group, and belongs to the technical field of power system scheduling. BACKGROUND

[0002] A renewable energy source (RES) and a traditional generator set are complementarily operated in an environment-friendly architecture of a power system. This process faces multiple challenges: ensuring stable and reliable power supply, fully absorbing renewable energy, and achieving long-term sustainable development. This makes the modern power grid adjust its scheduling system and improve energy utilization to meet the demand of high proportion of renewable energy connected to the grid.

[0003] Existing research focuses more on the scenario of multiple microgrids connected to the grid and participating in distribution network scheduling, but the behavior of microgrid autonomous clustering is also noteworthy. Microgrids in cities and surrounding areas have the basis of electric energy sharing due to similar energy structures and adjacent geographical distribution. The mechanism of microgrid clustering and sharing before being connected to the grid can improve resource utilization and operating domain margin, and can also realize complementary cooperation with the distribution network and fully release the mutual potential of the two. At this time, the source and load characteristics of different microgrids, geographical conditions, and the effects of these factors on electric energy sharing are worth studying. In addition, the organic combination of microgrid clusters and distribution networks under high RES penetration needs to be considered, Figure 1 a micro-grid system framework. A quantitative model is established to explore the environmental and economic benefits brought by micro-grid collaborative operation, and to analyze its promoting effect on renewable energy development, especially the advantages of micro-grid group wind-solar-storage collaborative optimization and energy interconnection. SUMMARY

[0004] The technical problem to be solved by the application is to provide a micro-grid collaborative scheduling method considering electric energy sharing of a honeycomb micro-grid group, to realize collaborative scheduling of multiple microgrids and distribution networks considering renewable energy consumption, micro-grid resource sharing, and system economy and environmental protection.

[0005] The application adopts the following technical solutions to solve the above technical problems:

[0006] A micro-grid collaborative scheduling method considering electric energy sharing of a honeycomb micro-grid group, comprising the following steps:

[0007] Step 1, combining microgrids with renewable energy sources and closed power supply areas with intelligent base stations having power transmission and information exchange functions to form a honeycomb structure of a power system topology, and constructing a micro-grid cluster honeycomb model of the power system topology based on a non-orthogonal coordinate system;

[0008] Step 2, considering the geographical position and source load characteristics of each micro-grid in the system, the entire micro-grid cluster is divided into several micro-grid groups by controlling whether the intelligent base station is enabled and the upper limit of its power transmission capacity;

[0009] Step 3, on the basis of the group division result in step 2, considering the local consumption of renewable energy, a micro-grid group electric energy sharing model is constructed, and a multi-dimensional improved alternating direction multiplier method is used for solving;

[0010] Step 4, on the basis of the electric energy sharing result of each micro-grid group in step 3, a micro-grid collaborative operation model is constructed and solved to minimize the power distribution network loss, and the collaborative operation of the power distribution network and the micro-grid cluster is realized.

[0011] Compared with the prior art, the above technical scheme has the following technical effects:

[0012] 1, the source load scale characteristics and geographical position distribution of the renewable energy micro-grid are considered, and the micro-grid cluster is modeled by using a non-right-angle coordinate system. The micro-grid is divided into groups by pinching optimization of the transmission capacity of the intelligent base station. The electric energy sharing of the micro-grid group is realized by using a multi-dimensional improved ADMM. On this basis, the micro-grid collaborative operation is realized, the micro-grid electric energy supply is guaranteed, the renewable energy is fully utilized, and the environmental benefit of the power distribution network is improved.

[0013] 2, the present application can provide a large number of micro-grid cluster scene plane topological model, a relatively reasonable micro-grid grouping method. The multi-dimensional improved ADMM used has obvious advantages in convergence performance and solving efficiency when facing a series of distributed problems of electric energy sharing, and the micro-grid collaborative scheduling is beneficial to the development of renewable energy, thereby providing support for the economic and green development of the new type of power system, and has certain engineering use value. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 is a schematic diagram of the micro-grid system framework of the present application;

[0015] Figure 2 is a schematic diagram of the micro-grid cluster honeycomb modeling;

[0016] Figure 3 is a flow chart for determining the micro-grid grouping by iterative optimization of the transmission capacity of the intelligent base station;

[0017] Figure 4 is a flow chart for solving the problem by using a multi-dimensional improved ADMM;

[0018] Figure 5 is the micro-grid and power distribution network load data in the embodiment, wherein (a) is the micro-grid load, and (b) is the power distribution network load;

[0019] Figure 6 is the renewable energy output data of each microgrid in the embodiment;

[0020] Figure 7 is the smart base station transmission capacity optimization process and microgrid clustering result in the embodiment;

[0021] Figure 8 is the microgrid system topology graph in the embodiment;

[0022] Figures 9(a)-9(e) is the scheduling result of microgrid 1-microgrid 5 in the embodiment;

[0023] Figure 10 is the part of the power of the power grid node in the embodiment;

[0024] Figure 11 is the power grid node voltage comparison in the embodiment, (a) is the microgrid cooperative operation voltage, and (b) is the power grid alone operation voltage;

[0025] Figure 12 is the multi-dimensional improved ADMM residual convergence in the embodiment. DETAILED DESCRIPTION

[0026] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings. The embodiments described below by referring to the accompanying drawings are exemplary and are only used to explain the present application, and cannot be explained as a limitation of the present application.

[0027] The present application proposes a microgrid cooperative scheduling method considering the electric energy sharing of the honeycomb microgrid cluster, and the specific steps are as follows:

[0028] Step 1, for the microgrid mainly based on renewable energy, considering the fact that there is a power supply boundary and geographical location distribution in reality, the microgrid power supply area is regarded as a closed hexagon, and the cluster formed by multiple microgrids is regarded as the covered area after the hexagonal close packing. At the junction of each microgrid, i.e. on the vertex of the hexagon, the smart base station with power transmission and information exchange capability is pre-installed, which has power transmission and information exchange capability. The smart base station and the microgrid form a structure similar to a honeycomb.

[0029] The above structure similar to a honeycomb is further standardized on the plane topology, and a non-right-angle coordinate system is constructed, so that the position of any microgrid or smart base station in the system can be represented by a specific vector or coordinate, i.e. (x MG ,y MG ) and (x IBS ,y IBS). Where subscript MG represents microgrid, and subscript IBS represents intelligent base station. For a microgrid with coordinate (m, n), the coordinates of the surrounding intelligent base stations form a set D = [(m-1, n), (m+1, n), (m-1, n+1), (m, n+1), (m, n-1), (m+1, n-1)], as shown in Figure 2 .

[0030] Step 2, considering the source-load characteristics of each microgrid in the cluster, the entire cluster is divided into several microgrid groups by controlling whether the intelligent base station is enabled and the upper limit of its power transmission capacity.

[0031] The daily net output peak value of microgrid i is introduced to describe the source-load characteristics of the microgrid:

[0032]

[0033] In the formula, is the predicted value of the RES output inside microgrid i at t, is the load inside microgrid i at t, T is the scheduling period, T = 24;

[0034] The objective function is set to minimize the upper limit of the safety power of the microgrid in the intelligent base station Based on the requirements of engineering practicability and safety margin of power systems, the objective function is integer processed:

[0035]

[0036] In the formula, k is an integer, and η is the unit safety power standard value, that is, the minimum step of the objective function in optimization.

[0037] By controlling the enabled intelligent base station, the members contained in each microgrid group after division are determined, and each microgrid is required to be connected to only one intelligent base station, and there is an intelligent base station around each microgrid whose transmission power capacity exceeds the net output of itself. Taking the microgrid with coordinate (m, n) as an example, the following constraints are obtained:

[0038]

[0039] u IBS (m-1, n) + u IBS (m+1, n) + u IBS (m-1, n+1) + u IBS (m, n+1) + u IBS (m, n-1) + u IBS (m+1, n-1) = 1

[0040] In the formula, u IBSThis is the smart base station activation status bit; a value of 1 indicates that the smart base station is in use, and a value of 0 indicates that the smart base station is temporarily disabled. μ is the safety margin or backup coefficient to cope with prediction errors.

[0041] According to the microgrid coordinates (x MG ,y MG ) and smart base station coordinates (x IBS ,y IBS To obtain the connection relationship between the microgrid and the smart base station l:

[0042]

[0043] In the formula, This is the connection status bit. A value of 1 indicates that the microgrid is connected to the smart base station l, and a value of 0 indicates that it is not connected.

[0044] When grouping microgrids based on this connectivity, efforts should be made to ensure that the net output or net load of different groups are similar in size to avoid areas with heavy loads. Coupling the geographical location of the microgrid with source-load characteristics yields the tie-line safety power constraints for smart base station l:

[0045]

[0046] In the formula, Ω MG This is a collection of microgrids. This formula ensures that the sum of the net output of all microgrids connected to the smart base station l is within the tie-line safety constraints. Furthermore, by combining this with the unit safe power η, the safety constraints are standardized, making the sizes of different groups as similar as possible.

[0047] Because the objective function for transmission capacity is a minimization-maximization problem, and the model contains u IBS Given multiple binary variables, and the overall structure being non-convex, we introduce a maximal positive number M, and set variable lower and upper bounds ξ and ν for the feasible region. This linearizes the model while iteratively solving the objective function. satisfy:

[0048]

[0049] By continuously modifying ξ and ν, the objective function is solved in a squeeze-out manner. The values ​​have been integerized; by restricting ξ and ν to integers, this discrete iterative process can stably find the optimal value. Finding the optimal value... The optimal value u will be calculated simultaneously. IBS and By grouping all microgrids connected to the same smart base station into a single group, we can obtain the microgrid cluster partitioning result. For example... Figure 3 As shown.

[0050] Step 3: According to the micro-grid cluster division result obtained in step 2, the multi-dimensional improved ADMM is used to realize the power sharing of micro-grids in the cluster, and the renewable energy consumption rate is ensured.

[0051] The day-ahead operation cost is constructed for each micro-grid The day-ahead operation model of the objective function is as follows:

[0052]

[0053] In the formula, π RES and π aba respectively represent the RES operation and maintenance cost and the wind and light abandonment penalty cost, is the actual RES output value of the micro-grid i actually put into use; π cut and π tr respectively represent the unit price of compensation for user disconnection and transferred load, and respectively represent the disconnected and transferred load; π ES is the unit life loss cost of the energy storage, and respectively represent the charging and discharging power; π EX is the unit price of the power over network fee paid to the base station operator when the micro-grid shares power, ij,t is the power exchanged between the micro-grid i and the micro-grid j; and π sell respectively represent the micro-grid purchase price and the micro-grid power selling price to the distribution network, and respectively represent the micro-grid power purchase quantity and the micro-grid power selling quantity to the distribution network, π cr is the carbon emission unit price calculated according to the micro-grid power purchase quantity.

[0054] During the daily operation of the micro-grid, the actual renewable energy output put into use should meet the following constraints:

[0055]

[0056] The user load of the micro-grid includes flexible load that can participate in demand response, which is divided into price-type demand response and incentive-type demand response based on the response mechanism. The former promotes the user to shift the electricity demand within the dispatching period according to the change of the electricity price, while the latter advocates direct reduction of electricity consumption to optimize the load curve and reduce the operation cost. The two behaviors need to meet the following constraints:

[0057]

[0058]

[0059] In the formula, δ cut∈ [-1, 0] and δ tr ∈ [0, 1] are the proportionality coefficients of the curtailable and shiftable electrical loads, respectively, and Δt is the time interval. cut is the set of time periods in which demand response is allowed by curtailment of loads. Both demand response resources have upper and lower limits, and the total amount of electrical load shifted in a dispatch period is 0.

[0060] On this basis, the internal load of the microgrid after demand response can be expressed as:

[0061]

[0062] To implement the operation strategy of preferential consumption of renewable energy, the model adds a renewable energy surplus flag at time t

[0063]

[0064] Correspondingly, the operation constraints of the microgrid energy storage system need to be adjusted:

[0065]

[0066] where η loss is the self-loss coefficient of the energy storage system; η ch and η dic are the charging and discharging efficiencies, respectively; and are the upper and lower limits of the state of charge of the energy storage system, respectively; and are the upper limits of the charging and discharging powers, respectively; and are binary variables representing the charging and discharging states, respectively, and represent charging, and represent discharging. The above equations specify that the state of charge of the energy storage system is conserved at the beginning and end of the stage and that simultaneous charging and discharging is not allowed, and that the microgrid preferentially uses local RES resources.

[0067] Limited by the safety constraints of the smart base station tie line, the purchase and sale of electricity between the microgrid and the distribution network must satisfy the following constraints:

[0068]

[0069] where C is the upper limit of the transmission capacity of the smart base station to which the microgrid cluster G belongs, Ω G is the set of microgrids under cluster G, and binary variables and ​The state bit of buying and selling electricity of the micro-grid, the above formula limits the micro-grid from buying and selling electricity at the same time.

[0070] The power of the power transmission between the micro-grids in any period should also be within the corresponding safety constraint :

[0071]

[0072] In addition, the micro-grid needs to meet its own power balance when running:

[0073]

[0074] An auxiliary variable P ji,t The decision variable P ij,t The problem is decomposed at each micro-grid, in addition, the Lagrange multiplier λ ij,t , the penalty factor ρ controls the iteration step size, and the augmented Lagrangian function is constructed for each micro-grid:

[0075]

[0076] In the formula, I is a micro-grid set, which contains n micro-grids.

[0077] Assuming that the decision variables of other micro-grids are fixed in turn in the k+1 iteration, the decision variable of a single micro-grid is solved according to the augmented Lagrangian function, and the Lagrange multiplier is updated after all variables are iterated. According to this, the following form is formed:

[0078]

[0079] The multi-dimensional improved ADMM used in the application is improved in four dimensions of the decision variable, the Lagrange multiplier, the penalty factor and the ADMM iteration mechanism in the Lagrangian function on the basis of the traditional ADMM. In order to avoid the problem of too many iteration times caused by slow updating of the Lagrange multiplier, the solving process of the multi-micro-grid power sharing model is improved based on the PR splitting method, and the quadratic updating of the Lagrange multiplier is realized. The following form is formed:

[0080]

[0081] In the formula, α ∈ (0, 1) is a relaxation factor to ensure convergence. In the formula, the P ij,t After solving, the corresponding λ is updated by half; when the auxiliary variable P ji,t of the micro-grid j to the micro-grid i is solved, the update of λ in this iteration is completed.

[0082] After each iteration is completed, the original residual ro and the dual residual r d , and the iteration is determined to be converged.

[0083]

[0084] where ε o and ε d are small positive numbers that control the upper bounds of the primal and dual residuals, respectively.

[0085] The convergence of ADMM is affected by the penalty factor, and if it is not properly selected, it will cause slow speed or even not converge. Since λ is constantly reselected in the iteration process, ρ also needs to be updated with the iteration. The present application adjusts ρ in this way:

[0086]

[0087] where δ>1 is used to determine the proportional relationship between the primal and dual residuals. The scaling factor ω>1 increases the step size when the primal residual is too large, limits the gap between the coupled variables P ij,t and P ji,t , thereby speeding up the convergence of the primal residual; and when the dual residual is too large, the step size is reduced to speed up the convergence of the dual residual and reduce the oscillation of the Lagrangian function.

[0088] The microgrid energy sharing model constructed by the present application needs to perform distributed optimization on 2 or 3 microgrid subjects, and accordingly generates 2 or 3 pairs of separable operators. When 3 microgrids are encountered for energy sharing, the directly extended ADMM cannot guarantee the convergence of this type of problem. Therefore, a Gaussian back substitution process is added after the iteration is completed to correct it:

[0089]

[0090] Taking 3 microgrids as an example:

[0091]

[0092] where E is an identity matrix, and a suitable small amount τ∈[0.9, 0.95] is a correction coefficient for the Gaussian back substitution. In this way, the convergence of the algorithm can be guaranteed.

[0093] Setting a dynamic convergence precision divides the overall iteration process into segments. The initial convergence precision is determined according to the initial iteration residual, and the convergence precision is reduced by equal ratio and the iteration is restarted every time the convergence criterion is met until the final convergence requirement is met. For example, the first iteration residual is 2.0424×10 4 , the initial convergence precision is 1×10 4 . When the iteration residual is less than the current convergence precision, the iteration is stopped, and the current convergence precision is reduced to 1×10​2 When the iteration residual 1 x 10 2 the current convergence precision is reduced to 1 x 10 0 The process is repeated until the convergence precision is reduced to the target precision and the iteration residual is less than the convergence precision. Each iteration inherits λ and ρ, and does not need to start from a fixed point each time. Thus, the multi-segment warm start of the overall algorithm is achieved, effectively avoiding the problem of too long single-iteration time in the later stage of solving. The solving process is shown in Figure 4 .

[0094] Step 4, on the basis of realizing power sharing of each micro-grid group, realize the collaborative operation of all micro-grids and distribution networks. Including the following steps:

[0095] Step 41, construct a micro-grid and distribution network collaborative operation target function with the minimum distribution network loss as the target, and use Cplex solver to solve:

[0096]

[0097] In the formula, I de,t is the current of branch de at time t, r de is the resistance of branch de, Ω L is the branch set.

[0098] The node power balance constraint generated after the micro-grid group is connected is as follows:

[0099]

[0100] In the formula, and are the active and reactive power flowing into node e from the line at time t, and are the active and reactive power demand of node e at time t, and are the active and reactive power flowing out of node e at time t, x de is the reactance of branch de, Ω cl is the micro-grid group access node set.

[0101] For other nodes and branches, after the second-order cone relaxation, the power flow constraint of the distribution network can be expressed as:

[0102]

[0103] In the formula, Ω de is the set of nodes at the head of the branch with e as the end node, and are the active and reactive power flowing through branch de, Ω efThen it is the set of end nodes of the branch with e as the first end node.

[0104] The Ohm's law constraint that the distribution network must satisfy can be expressed as:

[0105]

[0106] In the formula, It is the square of the voltage magnitude at node d.

[0107] To ensure power quality, the system must still meet node voltage constraints and branch current constraints after the microgrid cluster is connected:

[0108]

[0109] In the formula, U max,i and U min,i These are the upper and lower limits of the voltage amplitude, respectively; I max,de This is the upper limit of the current amplitude of the branch circuit.

[0110] The invention will be illustrated by an example below:

[0111] First, five renewable energy microgrids and an IEEE 33-node distribution network were selected as case studies. Figure 5 (a) and (b) in the middle and Figure 6 Based on the renewable energy and load forecast data shown, a distribution-microgrid collaborative scheduling model under the power sharing of a honeycomb microgrid group is constructed, and the corresponding problems are solved using the transmission capacity squeeze optimization and multidimensional improved ADMM of this invention.

[0112] The iterative process for solving the cluster partitioning results is as follows: Figure 7 As shown, a total of 12 iterations were performed. The initial upper limit ν was set relatively large, and in the first three iterations, the lower limit ξ was mainly increased in steps of 15. The clustering strategy at this stage was to group MG1, MG2, and MG4, which had relatively high net power output, into one group, while MG3 and MG5 were automatically grouped together. Starting from the fourth iteration, the upper limit was reduced, and the clustering strategy was to divide the entire microgrid cluster into one group of MG1 and MG2, and another group of MG3, MG4, and MG5, making the net power output levels of the two microgrid groups comparable. Obviously, this is a more reasonable clustering method. Finally, both the upper and lower limits were determined to be 17, and the smart base station transmission capacity was 850kW.

[0113] During the whole iteration process, the upper and lower limits are constantly close to each other, and finally reach an agreement. In the whole algorithm, the upper limit v determines the direction and amplitude of the lower limit ξ update, and is constantly reduced with the lower limit update; the lower limit ξ is used to generate the actual solution and participate in the judgment; and the constraint condition helps the objective function to find the minimum transmission capacity that can meet the power transmission demand of all connected microgrids, and also takes into account the uniform distribution of microgrid size. The three work together to form a set of grouping methods suitable for the renewable energy microgrid cluster accessing the distribution network under the architecture of the honeycomb microgrid group.

[0114] After the microgrid is divided into two groups A and B, it is accessed to the distribution network through the intelligent base station at node 8 and node 29 respectively, and the formed micro-system is as shown in Figure 8 The day-ahead scheduling is carried out for cluster A and cluster B respectively, so as to promote the power sharing between microgrids. The operation results of each microgrid subsystem in cluster A and cluster B are as shown in Figures 9(a)-9(e)

[0115] The scenario of each microgrid running independently and interacting with the distribution network is compared with the day-ahead operation cost of the microgrid cluster and micro-system cooperative mode of the present application. The operation cost of cluster A is reduced by 18.93%, the cost of cluster B is reduced by 7.40%, and the total operation cost of all microgrids is reduced by 12.496%.

[0116] Table 1 shows the cost composition of each microgrid, where the positive value is the cost of consumption, and the negative value is the income of transaction. Thanks to the micro-system cooperation, the renewable energy consumption rate of each microgrid is close to 100%, and there is no penalty cost for abandoning wind and light. Renewable energy is fully utilized. From the perspective of the distribution network, microgrid power sharing actually reduces the power demand of the distribution network, reduces the operation pressure of the distribution network, and also equivalent to reducing the carbon emissions of the whole micro-system, and effectively improves the environmental benefits.

[0117] Table 1

[0118]

[0119] The cooperative operation of microgrids and distribution networks enables them to respond to each other and make scheduling according to their own power and load conditions, showing a mutually beneficial effect. Figure 10 The active load of each node of the distribution network except the balance node 1 in several time periods is given, where the positive value represents the demand and the negative value represents the supply. Figure 10 ​The power injected by the generator into the distribution network is also given. The distribution network plays a role of supporting the micro-grid on one hand, and makes full use of the clean energy that the micro-grid itself cannot accommodate on the other hand. In the process of micro-grid cooperation, the source and load resources have higher inclusiveness and carrying capacity. In addition, the carbon emission is reduced from 21347.43 kg to 19821.68 kg. The access of the micro-grid cluster indeed shares part of the user load of the distribution network as a whole, and the environmental benefits of the distribution network are more significant in the context of encouraging the development of clean energy. In addition, the network loss of the distribution network is reduced by 4.67%.

[0120] The voltage distribution of the distribution network before and after the micro-grid cooperation is compared, as shown in (a) and (b) of FIG. 6. Figure 11 The micro-grid cooperation also has a boosting effect on the system voltage, and makes the voltage span of the system relatively concentrated, and the voltage offset degree is lower, which reduces the risk of voltage out-of-limit and avoids the possibility of voltage deficiency.

[0121] In the example adopted in the application, MG1 and MG5 can completely independently run, but at this time, the renewable energy consumption rate of the two micro-grids is 85.12% and 81.15% respectively, and a high penalty cost of abandoned wind and light needs to be paid. The remaining MG needs to be supported by the distribution network. After grouping, cluster A can independently run, but the renewable energy consumption rate is 94.72%, which is still wasted. Cluster B needs to be supported by the micro-grid in part of the time period. Therefore, the role of the distribution network in supporting the micro-grid and assisting in consumption cannot be ignored.

[0122] Figure 12 The convergence process of the dual residual is given, and the logarithmic coordinate axis is attached to the figure to more clearly show the change of the residual. The multi-dimensional improved ADMM is iterated for 4 rounds, and the decision variable is recalculated at the beginning of each iteration, so the dual residual rapidly decreases after a sharp increase, which is the symbol of hot start. In each iteration, the dual residual basically presents a monotonous decreasing trend by virtue of the quadratic update of the Lagrange multiplier and the adaptive scaling of the penalty factor, which shows the advantages of the improved algorithm in convergence performance and solving efficiency.

[0123] The statistical solving time is shown in Table 2, and it can be found that the average single iteration time will be prolonged with the increase of the iteration number, so that the iteration segmentation can effectively reduce the solving time.

[0124] Table 2

[0125]

[0126] The application improves the ADMM from four dimensions, uses the Gauss back substitution to ensure that P ij,tThe convergence of the algorithm is guaranteed, the update efficiency of and is ensured by using half-step update and adaptive scaling, and the overall time consumption of the algorithm is reduced by using multi-section warm start. The multi-dimensional improved ADMM of the application has shorter solving time, less iteration times and higher overall calculation efficiency in the solution of the multi-microgrid.

[0127] Based on the same inventive concept, the embodiment of the application provides a computer device, which comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the microgrid collaborative scheduling method considering the electric energy sharing of the honeycomb-shaped microgrid group when executing the computer program.

[0128] Based on the same inventive concept, the embodiment of the application provides a computer readable storage medium, which stores a computer program, and the computer program implements the steps of the microgrid collaborative scheduling method considering the electric energy sharing of the honeycomb-shaped microgrid group when executed by a processor.

[0129] Those skilled in the art will understand that the embodiments of the application can be provided as a method, a system, or a computer program product. Therefore, the application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0130] The application is described with reference to flowcharts and / or block diagrams of the method, device (system), and computer program product according to the embodiments of the application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device that implements the flowcharts and / or block diagrams. Figure 1 The function specified in one flow or multiple flows and / or blocks. Figure 1 The device that implements the function specified in one block or multiple blocks.

[0131] These computer program instructions can also be stored in a computer-readable memory that can guide the computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer-readable memory produce a product comprising instruction devices that implement the flowcharts and / or block diagrams. Figure 1 The function specified in one flow or multiple flows and / or blocks. Figure 1 The device that implements the function specified in one block or multiple blocks.

[0132] These computer program instructions can also be loaded into a computer or other programmable data processing devices, so that a series of operational steps are performed on the computer or other programmable data processing devices to generate a computer implemented process, so that the instructions executed on the computer or other programmable data processing devices provide a process for implementing the flowchart Figure 1 one flow or a plurality of flows and / or the functions specified in the block Figure 1 one block or a plurality of blocks.

[0133] The above embodiments are only to illustrate the technical idea of the present application, and cannot limit the protection scope of the present application. Any modification made according to the technical idea of the present application on the basis of the technical scheme falls within the protection scope of the present application.

Claims

1. A microgrid cooperative scheduling method considering power sharing of a honeycomb microgrid cluster, characterized in that, The method comprises the following steps: Step 1, combining microgrids with renewable energy and smart base stations with power transmission and information exchange functions in a closed power supply area to form a honeycomb structure of power system topology, and constructing a microgrid cluster honeycomb model of the power system topology based on a non-right-angle coordinate system; Step 2, considering the geographical location and source load characteristics of each microgrid in the system, dividing the entire microgrid cluster into several microgrid groups by controlling whether the smart base station is enabled and the upper limit of the power transmission capacity; Step 3, based on the group division result of step 2, considering the local consumption of renewable energy, constructing a microgrid group electric energy sharing model, and solving it using a multi-dimensional improved alternating direction multiplier method; Step 4, based on the electric energy sharing result of each microgrid group in step 3, constructing a distribution and microgrid collaborative operation model to minimize the loss of the distribution network and solving it to realize the collaborative operation of the distribution network and the microgrid cluster.

2. The microgrid collaborative scheduling method considering the electric energy sharing of the honeycomb microgrid group according to claim 1, characterized in that, The specific process of step 1 is as follows: The microgrid power supply area with renewable energy is regarded as a closed hexagon, and smart base stations with power transmission and information exchange functions are preinstalled at the junctions of each microgrid, i.e. the vertices of the hexagon, so that the microgrid and the smart base station form a honeycomb structure of power system topology; Based on the non-right-angle coordinate system in the Cartesian coordinate system, the microgrid group honeycomb model of the power system topology is constructed, the position of any microgrid or smart base station in the power system is converted into a coordinate representation, and for a microgrid with coordinates (m, n), the set of coordinates of the smart base stations around it is D = [(m-1, n), (m+1, n), (m-1, n+1), (m, n+1), (m, n-1), (m+1, n-1)].

3. The microgrid collaborative scheduling method considering the electric energy sharing of the honeycomb microgrid group according to claim 2, characterized in that, The specific process of step 2 is as follows: Step 21, introduce the daily net output peak to describe the source load characteristics of the microgrid: wherein, is the daily net output peak value of the micro-grid i, is the predicted value of the renewable energy output inside the micro-grid i at t, is the load inside the micro-grid i at t, and T is the dispatching period, T = 24; Step 22, set the objective function of the micro-grid cluster division as minimizing the upper limit of the safe power of the micro-connection line in the lth intelligent base station P l f integer processing is performed on the objective function: Where k is an integer, and η is the unit safety power standard value; By controlling the enabled smart base station, determine the members of each microgrid group after division, require each microgrid to be connected to only one smart base station, and require each microgrid to have a smart base station around it with transmission power capacity exceeding its net output; the corresponding constraints are as follows: Wherein, u IBS is the enable state bit of the intelligent base station, u IBS = 1 indicates that the intelligent base station is put into use, u IBS = 0 indicates that the intelligent base station is temporarily disabled; μ is a safety margin or backup coefficient for coping with prediction errors; represents the daily net output peak of the micro-grid with coordinates (m, n). Step 23, according to the coordinates of the microgrid and the smart base station, determine the connection relationship between the microgrid and the smart base station l: wherein, is a connection relationship status bit, represents that the micro-grid is connected with the smart base station l, represents not connected;(x MG ,y MG ) are coordinates of the micro-grid, and (x IBS ,y IBS ) are coordinates of the smart base station l; To ensure that the sum of the net outputs of all microgrids connected to the smart base station l is within the tie line safety constraint, and the upper limit of the safety power of all smart base stations is uniform, the tie line safety power constraint of the smart base station is obtained as: wherein Ω IBS is a set of intelligent base stations; Step 24, introduce a very large positive number M, set the variable feasible region lower limit ξ and upper limit v, both of which are integers, and use the way of pinching to iteratively solve the objective function set in step 22, that is, the objective function of step 22 is satisfies: By constantly correcting ξ and ν, the optimal value is obtained synchronously The optimal value corresponds to u IBS and All microgrids connected to the same intelligent base station are divided into a group to obtain the microgrid cluster division result.

4. The microgrid collaborative scheduling method considering the electric energy sharing of the honeycomb microgrid group according to claim 3, characterized in that, In step 24, the specific process of iterative solving of the objective function is as follows: Step 241, initialize ξ and v, ξ is 0, set step size as is a positive integer; Step 242, solve the power transmission capacity of each intelligent base station according to the constraint condition of step 22, and judge whether the power transmission capacity of each intelligent base station meets the interconnection line safety power constraint of step 23, if yes, enter step 249, otherwise, let Enter step 243; Step 243, solve the power transmission capacity of each smart base station according to the constraint condition of step 22, and judge whether the power transmission capacity of each smart base station meets the tie line safety power constraint of step 23; Step 244, if yes, go to step 245, otherwise add a step to the current feasible region lower limit ξ and go back to step 243; Step 245, take the first ξ1 that meets the tie line safety power constraint of step 23 as ν, calculate ν / 2 and take the upper integer as the new lower limit of the feasible region ξ. Step 246, solve the power transmission capacity of each smart base station according to the constraint condition of step 22, and judge whether the power transmission capacity of each smart base station meets the interconnection line safety power constraint of step 23; Step 247, if yes, go to step 248, otherwise let ξ = ξ + (v-ξ) / 2 and return to step 246; Step 248, judge whether the value of v-ξ is less than or equal to 1, if yes, go to step 249, otherwise take the current ξ as the new v, take the first ξ1 meeting the interconnection line safety power constraint of step 23 as the new lower limit of the feasible region ξ, and return to step 246; Step 249, output the power transmission capacity of each smart base station, determine the connection relationship between the smart base station and the micro-grid, and obtain the micro-grid cluster division result.

5. The microgrid collaborative scheduling method considering the electric energy sharing of the honeycomb microgrid group according to claim 3, characterized in that, The specific process of step 3 is as follows: Step 31, construct for each microgrid a day-ahead operation cost Day-ahead operation model for the objective function: wherein, π RES and π aba are the renewable energy operation cost and the wind and light curtailment penalty cost, respectively, is the actual renewable energy output value of the microgrid i actually put into use; π cut and π tr are the unit prices for compensating the user cut-off and the transferred load, and are the cut-off and the transferred load of the user; π ES is the unit life loss cost of the energy storage, and are the charging and discharging power; π EX is the unit price of the electricity over-network fee paid to the base station operator when the microgrid shares electricity, ij,t is the electricity exchanged between the microgrid i and the microgrid j; and π sell are the microgrid electricity purchase price and the microgrid electricity sale price to the distribution network, respectively, and are the microgrid electricity purchase quantity and the microgrid electricity sale quantity to the distribution network, respectively, π cr is the carbon emission unit price calculated according to the microgrid electricity purchase quantity; The actual renewable energy output in use of the micro-grid in daily operation meets the following constraints: The load cut off and transferred to the user meets the following constraints: wherein δ cut and δ tr are the proportional coefficients of the curable and shiftable electrical load, respectively, δ cut ∈ [-1, 0], δ tr ∈ [0, 1], Δt is the time interval, T cut is the set of time periods in which demand response is allowed by shedding load. Demand response post-microgrid internal load is represented as: To give priority to the consumption of renewable energy, a renewable energy surplus flag at time t is introduced in the model Then the operation constraints of the micro-grid energy storage system need to meet: wherein, So and So-1 are the state of charge stored at time t and t-1, respectively, η loss η is the self-loss coefficient of the energy storage system; η ch and η dic are the charging and discharging efficiencies, respectively; and So max and So min are the upper and lower limits of the state of charge of the energy storage system, respectively; and P max and P min are the upper limits of the charging and discharging power, respectively; and So and So-1 are the state of charge stored at time t and t-1, respectively, η and denotes charging, and denotes discharging; The power purchase and sale between the micro-grid and the distribution network meets the following constraints: wherein, is the upper limit of transmission capacity of the intelligent base station to which the microgrid cluster G belongs, Ω G is the set of microgrids under the cluster G, binary variable and are the purchase and sale of electricity state bits of the microgrid, respectively; The power for the electric energy transmission between microgrids should be in the corresponding safety constraints In: The micro-grid meets its own power balance in operation as follows: Step 32, introduce auxiliary variable P ji,t and Lagrange multiplier λ ij,t , set the penalty factor p to control the iteration step length, and construct the augmented Lagrange function for each microgrid: Wherein, n is the number of micro-grids in each micro-grid group, and I represents the set of micro-grids in each micro-grid group; The augmented Lagrange function constructed is solved by using a multi-dimensional improved alternating direction multiplier method, and at the k+1th iteration, the Lagrange multiplier is updated based on the PR splitting method: Wherein, α is a relaxation factor, α∈(0,1); At the same time, the penalty factor is updated: wherein, ρ k+1 , ρ k are the penalty factors of the k+1, k iteration, respectively, ε o and ε d are positive numbers to control the upper limit of the primal and dual residuals, respectively; δ > 1 is used to determine the proportional relationship of the primal and dual residuals; ω is a scaling factor, ω > 1; After each iteration is completed, the original residual r o and the dual residual r d When r o and r d satisfy the following constraint condition, the iteration converges: After the iteration is completed, the Gauss back substitution process is added to correct the power transmission between the micro-grids: Wherein, E is a unit matrix, τ is a correction coefficient of the Gauss back substitution, τ∈[0.9,0.95]; The overall iteration process is segmented by setting a dynamic convergence precision, the initial convergence precision is determined according to the initial iteration residual, the convergence precision is reduced by equal ratio and the iteration is restarted after each iteration meets the iteration convergence condition, and the final convergence requirement is met.

6. The microgrid collaborative scheduling method considering the electric energy sharing of the honeycomb microgrid group according to claim 1, characterized in that, The specific process of step 4 is as follows: Step 41, a micro-grid and distribution collaborative operation objective function is constructed with the minimum distribution network loss as the target: wherein I de,t is the current of branch de at time t, A de is the resistance of branch de, Ω L is the set of branches of the distribution network; The node power balance constraint generated after the micro-grid group is accessed is as follows: wherein, and are the active and reactive power flowing from the line into the distribution network node e at time t, and are the active and reactive power demand at node e at time t, and are the active and reactive power flowing from node e at time t, x de is the reactance of branch de, Ω cl is the set of nodes where the microgrid clusters are connected to the distribution network; For other nodes and branches, after the second-order cone relaxation, the distribution network power flow constraint is represented as: wherein Ω de is the set of head nodes of branches with e as a tail node, and are the active and reactive power flowing over branch de, respectively, Ω ef is the set of tail nodes of branches with e as a head node, and are the active and reactive power flowing over branch ef, respectively. The Ohm's law constraint that the distribution network needs to meet is represented as: where U e,t is the voltage amplitude at node e, U d,t is the voltage amplitude at node d; After the micro-grid group is accessed, the system still needs to meet the node voltage constraint and the branch current constraint: where U max,d and U min,d are the upper and lower voltage limits, respectively; I max,de is the upper branch de current limit; The Cplex solver is used to solve the micro-grid and distribution collaborative operation objective function, and the collaborative operation of the distribution network and the micro-grid cluster is realized.

7. A computer device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that, The processor executes the computer program to realize the steps of the micro-grid and distribution collaborative scheduling method considering the electric energy sharing of the honeycomb micro-grid group.

8. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 7. The computer program is executed by the processor to realize the steps of the micro-grid and distribution collaborative scheduling method considering the electric energy sharing of the honeycomb micro-grid group.

Images