Finite-time economic dispatching control method for wind-solar-diesel storage load micro-grid

Abstract

The invention provides an economic dispatching control algorithm for a wind-solar-diesel storage load micro-grid in finite time. A micro-grid economic dispatching model considering the uncertainty of wind power generation and photovoltaic power generation, diesel power generation and energy storage battery factors is established. And meanwhile, the pollution of non-renewable energy power generation to the environment is considered, so that the constructed target function is minimum. An objective function problem is converted into a dual problem, a distributed algorithm based on the consistency principle is introduced, the optimal power of a generator is searched in finite time, and the power generation efficiency of the micro-grid is improved.

Description

Technical Field

[0001] The present invention relates to the field of power system planning, and in particular to a microgrid economic dispatch control algorithm based on finite time consistency. Background Art

[0002] Due to the presence of new energy equipment such as wind turbines in microgrids, their output is easily affected by unstable weather factors. In addition, the scheduling complexity brought by batteries in microgrids and the nonlinear factors of diesel generators have brought great challenges to the economic scheduling of microgrids. In response to these problems, this paper establishes an expected value model for microgrid economic scheduling that takes into account the uncertainty of wind power generation and factors such as batteries and diesel generators. On this basis, a piecewise linearization method is proposed to solve the global optimal solution for economic scheduling.

[0003] In order to build microgrids, make them operate smoothly, and improve the economy and reliability of microgrids, many planning methods have been proposed, such as [LI Z, CHENG Z, SI J, et al., 2022. Distributed Event-Triggered Hierarchical Control to Improve Economic Operation of Hybrid AC / DC Microgrids[J / OL].IEEE Transactions on Power Systems, 37(5):3653-3668.DOI: 10.1109 / TPWRS.2021.3133487 .] A new distributed event-triggered hierarchical control strategy is proposed to improve the economic operation of AC / DC hybrid motors. For the primary control, the distributed local control of AC generators, DC generators and interconnected converters is realized by using the droop control method. For the secondary control, a distributed economic dispatch algorithm, a distributed average bus voltage discovery algorithm and a distributed proportional power sharing algorithm are first proposed; based on the developed algorithms, the control objectives of voltage, frequency recovery and economic operation of the AC / DC hybrid variable frequency speed regulation system are achieved.

[0004] The above literature provides valuable research ideas for the economic dispatch of microgrids, but there is little research on microgrids that include wind, solar, diesel and load storage. Wind and solar energy have strong randomness and volatility, which are difficult to accurately predict and model. In addition, the uncertainty of wind and solar energy will also affect the stability and reliability of the power system, increasing the difficulty of economic dispatch. At the same time, most of the existing economic dispatch models and algorithms are based on deterministic assumptions, and the uncertainty of wind and solar energy is insufficiently considered. Compared with the prior art, the present invention has the following beneficial effects:

[0005] The present invention takes into account the uncertainty of wind and solar power generation due to factors such as weather, first forecasts demand, calculates the power to be allocated, and then allocates power through a finite time consistency economic dispatch algorithm. In addition, environmental pollution is taken into account, and the pollution emission cost of generator power generation is calculated, which has practical significance for energy conservation and emission reduction;

[0006] The expression of the stabilization time is determined by the control parameters in the control algorithm and the second smallest eigenvalue of the Laplace matrix. Therefore, the upper limit of the stabilization time can be set artificially, and the convergence time is independent of the initial state of the generator. This will provide more flexibility for the system. Summary of the invention

[0007] The present invention aims to solve the problems in the prior art and proposes an economic dispatch control algorithm for wind-solar-diesel-load-storage microgrid based on finite-time consistency theory to solve the economic dispatch problem of distributed microgrids. The microgrid model is constructed using three types of non-renewable energy, renewable energy and energy storage batteries. At the same time, considering the cost of pollutant emissions, the optimal solution of different generators is found within 24 hours. Finally, the convergence of the distributed algorithm is analyzed. The experimental results show that the optimal solution can well meet the supply and demand balance equality constraints and capacity inequality constraints in each time slot.

[0008] The technical solution used in the present invention is an economic dispatch control algorithm for a wind-solar-diesel-load-storage microgrid based on finite-time consistency theory. The specific steps are as follows:

[0009] Step 1: Construct a mathematical model of economic dispatch in the microgrid, which includes the wind and solar output in the wind-solar-diesel-storage-load microgrid, the total power generation cost of all generators, the supply and demand balance between the power generation end and the power consumption end, and the power generation power limit of the generator; according to the Lagrange multiplier method, the consistency variable λ and the optimal power generation P are obtained. i ;

[0010] Step 2: Design an economic dispatch algorithm based on finite-time consistency to update the incremental cost λ of each generator i ;

[0011] Step 3: Generators should calculate their optimal incremental costs And the optimal power generation corresponding to its own parameters calculation

[0012] Step 4: Determine whether the optimal output power of each generator is within the power limit range. If so, the optimal power is obtained and economic dispatch is completed; BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a schematic diagram of the process of the present invention; DETAILED DESCRIPTION

[0014] First, as shown in the figure, the communication topology of each generator in the smart grid can be represented by an undirected graph To indicate that and Represent the node set, edge set and adjacency matrix of an undirected graph respectively. ij =(i,j) means i and j are adjacent nodes and they can exchange information with each other. If (i,j)∈ε, then aij=1, otherwise aij=0.

[0015] According to step one, a mathematical model of economic dispatch is established in the smart grid, wherein the model includes the total power generation cost of all generators in the smart grid, the supply and demand balance between the power generation end and the power consumption end, and the power generation power constraint of the generator; and the consistency variable λ and the optimal power generation are obtained according to the Lagrange multiplier method.

[0016] The various parts of a microgrid can be divided into three categories: power generation equipment, energy storage systems, and load components. Photovoltaic power generation systems are affected by factors such as temperature and light intensity; wind power generation systems are affected by factors such as wind speed and weather, and there is uncertainty in output. With the continuous increase in distributed power sources, the penetration rate of distributed power sources in microgrids is also increasing. Therefore, in order to ensure the universality of the research, modeling of different power sources is indispensable.

[0017] 1. Generators and energy storage equipment

[0018] 1.1 Diesel generator

[0019] The cost function of a diesel generator can be expressed as the following quadratic polynomial:

[0020]

[0021] Where a DG ,b DG ,c DG is the cost coefficient of diesel generator set, P DG (t) represents the output power of the diesel generator at time t.

[0022] 1.2 Battery

[0023] The storage system is represented by a battery. It has the ability to charge and discharge and shares incremental cost information with other generators. When the power generated by conventional energy and renewable energy is insufficient or excessive, the battery pack can achieve stable power supply and demand through charging and discharging. The cost function of the energy storage battery pack model can be expressed as:

[0024]

[0025] Where a ESS ,bESS ,c ESS is the battery pack cost coefficient, P ESS (t) represents the output power of the battery pack at time t.

[0026] 1.3 Wind power generation

[0027] Wind power generation is a clean energy. After the wind passes through the wind turbine, part of the kinetic energy is converted into mechanical energy and then into electrical energy. The wind turbine output power P WT (t) wind speed v(t) has a certain relationship, the specific function relationship is as follows:

[0028]

[0029] where v in Indicates the minimum wind speed for power generation to start. out P is the upper limit of wind speed for safe power generation. N is the rated output power, v N is the corresponding wind speed, and v(t) is the wind speed at time t.

[0030] The cost function of a wind turbine is as follows:

[0031] C WT (t) = B WT P WT (t)

[0032] Among them C WT (t) is the cost of wind power generation, which is proportional to the power generated by the wind turbine. WT is the wind power generation cost coefficient.

[0033] 1.4 Photovoltaic power generation

[0034] Photovoltaic power generation is a clean energy source. After sunlight passes through the photovoltaic panel, part of the light energy is converted into electrical energy. The output power of the photovoltaic panel is approximately proportional to the light intensity, and is also affected by factors such as temperature. The specific functional relationship is as follows:

[0035] P pv =3.24(M pv (1-0.0041(T t -8))*I s

[0036] Where M PV ,T t ,I s They are solar panel capacity, temperature and solar radiation intensity.

[0037] The cost function of a photovoltaic cell is as follows:

[0038] C PV(t) = B PV P PV (t)

[0039] Where C PV (t) is the cost of photovoltaic power generation, B PV is the photovoltaic power generation cost coefficient.

[0040] 2. Environmental cost function

[0041] In the process of traditional energy power generation, raw materials such as coal are often consumed, and some gases that are harmful to the atmosphere and human health are inevitably produced. In order to measure and reduce the impact of pollutants produced by traditional energy, environmental emissions are considered here, and the environmental cost function is as follows:

[0042]

[0043] where α g,i , β g,i and γ g,i It is the pollution emission coefficient when traditional generators generate electricity.

[0044] 3. Problem Constraints

[0045] 3.1 Supply and demand constraints

[0046] Considering that in actual scenarios, the sum of the power generated by each generator should not be less than the total load demand of the entire microgrid. Therefore, the supply and demand constraints can be expressed as:

[0047]

[0048] t∈{1,2,...,T}

[0049] Where T is 24, which means that the supply and demand constraints can be met at every moment, DG is the conventional generator, ESS is the energy storage, RE is the renewable energy (considered as a renewable generator), N is the total number of conventional generators, renewable energy or energy storage, and P d (t) is the total load demand of the microgrid at time t.

[0050] 3.2 Capacity Constraints

[0051] To ensure that each generator can operate within the normal range, the capacity constraint needs to be met, which can be expressed as:

[0052]

[0053] in and Represent the lower and upper limits of the power generation capacity of the i-th generator, respectively. Indicates the lower and upper limits of battery capacity. The lower and upper limits for renewable energy generation.

[0054] 3.3 Battery SOC Constraint

[0055] SOC min ≤SOC(t)≤SOC max

[0056] In the formula, SOC(t) represents the state-of-charge (SOC) of the battery at that moment. min and SOC max They are the upper and lower limits of the state of charge. min The value range is 0.1 to 0.2, SOC max The value range is 0.8 to 0.9. In this article, it is set to 0.2 and is set to 0.9. The state of charge value of the battery changes evenly at all times. At this time, the state of charge can be expressed as:

[0057]

[0058] Where P Bat (t) is the output power of the battery during the change time. Positive values ​​represent discharge, negative values ​​represent charge, E B is the battery capacity. The frequency of battery state changes will affect its service life. The number of times a battery is charged in a cycle cannot exceed N1 times, and the number of times it is discharged cannot exceed N2 times. In this study, both N1 and N2 are set to 12.

[0059] The individual cost of each power generation contributes to the total cost of the microgrid. In the economic dispatch problem, the goal is to minimize the total operating cost of the microgrid and manage the controllable loads to ensure power balance. The goal of the EDP of the present invention is to determine the power generation of each generator and battery to minimize the total cost of power generation while satisfying the total demand constraint. Unlike single-objective optimization problems, operating costs and emissions need to be considered simultaneously, and we handle these two conflicting functions through weight factors. Then, the EDP can be expressed as:

[0060] minδ1C cost +δ2E emission

[0061] Where δ1 and δ2 are the weights of the corresponding functions, satisfying δ1+δ2=1, so the merged objective function is:

[0062]

[0063]

[0064] Where P i (t) = [PDG (1),...,P DG (N),P ESS (1)], represents the power output of dispatchable generators and batteries at time t. Based on the convexity of each local cost function and the total demand constraint, we can derive a dual problem with respect to the total demand constraint from the above problem.

[0065] The dual optimal set corresponds to the optimal set of the economic dispatch problem. Therefore, we can design a method that can solve the corresponding dual problem and obtain the optimal value of the economic dispatch problem.

[0066] Transformed into a Lagrangian problem

[0067] We can define the problem as the Lagrangian function at each moment:

[0068]

[0069] Where P is defined as P i (t) = [P DG (1),...,P DG (N),P ESS (1)] represents the power output of the dispatchable generator and battery at time t. And it satisfies the constraint P i min ≤P i (t)≤P i max , d represents the remaining power demand after removing renewable energy generation, that is, P d (t) is the total power demand at the user side.

[0070] Derivative of the Lagrangian function yields:

[0071]

[0072] where a i ,b i is the cost coefficient, α i ,β i is the pollution coefficient, so the optimal incremental cost is

[0073] λ=δ1(2a i P i +b i )+δ2(2α i P i +β i )

[0074] Therefore, the relationship between the optimal power generation of each generator and the optimal incremental cost is:

[0075]

[0076] According to step 2, a wind-solar-diesel-load-storage microgrid economic dispatch control algorithm based on finite-time consistency is designed to update the optimal incremental cost of each generator:

[0077]

[0078] In the formula, λ i is the optimal incremental cost of the ith generator, a ij is the value of the adjacency matrix, sgn represents the sign function, α and η are gain coefficients, is the gradient, expressed as:

[0079]

[0080] P min and P max The upper and lower limits of the generator power.

[0081] The method to determine the optimal dispatch of each agent to minimize the operating cost of the microgrid is the application of consensus theory. In consensus optimization, each generator shares the incremental cost information with its neighbors in a distributed manner. This incremental cost information is used to update the incremental cost of the generator itself and adjust the power associated with each updated incremental cost. This process is repeated until a consensus is reached so that all generators have the same incremental cost.

[0082] Using the above finite-time consistent wind-solar-diesel-load-storage microgrid economic dispatch control algorithm, we prove that the derivative of the Lyapunov function V(t) satisfy It can be obtained that the incremental cost reaches the optimal value in a limited time.

[0083] The Lyapunov function V(t) is selected as follows:

[0084]

[0085] available

[0086]

[0087] There are We can get

[0088]

[0089] consider We can scale it down to

[0090]

[0091] in D(t) is the incidence matrix.

[0092]

[0093] Also because

[0094]

[0095] You can get:

[0096]

[0097] The above method can achieve economic dispatch in a limited time, where the convergence time satisfies:

[0098]

[0099] After time Ts, λ i Converge to the optimal incremental cost The optimal power generation can be calculated:

[0100]

[0101] And determine whether the generated power is within the generator power limit:

[0102]

[0103] At the next moment, jump to step 2, calculate the optimal incremental cost for the needs of different user sides, and then complete steps 3 and 4.

Claims

1. A finite-time wind-solar-diesel-load-storage microgrid economic dispatch control method, which establishes a microgrid economic dispatch model that takes into account the uncertainty of wind power generation and photovoltaic power generation, diesel power generation and energy storage battery factors, and is characterized by: The following steps are involved: Step 1: Construct a mathematical model of economic dispatch in microgrids, establish the objective function of economic dispatch problem, and obtain the consistency variable λ and the optimal power generation power P according to the Lagrange multiplier method. i The mathematical relationship between Step 2: Design an economic dispatch algorithm based on finite-time consistency to update the incremental cost λ of each generator i , and converges to the optimal incremental cost after a finite time Step 3: Each generator is calculated based on the optimal incremental cost And the optimal power generation corresponding to its own parameters calculation Step 4: Determine whether the optimal output power of each generator is within the power limit range. If so, the optimal power is obtained and economic dispatch is completed; In step 1, a mathematical model of economic dispatch in a microgrid is constructed, which is characterized by including wind power generation, photovoltaic power generation, diesel power generation, and battery models, and obtaining corresponding cost functions and environmental cost models; The objective function of the economic dispatch problem is established in step 1, wherein the objective function of the economic dispatch problem can be expressed as: minδ1C cost +δ2E emission SOC min ≤SOC(t)≤SOC max Where δ1 and δ2 are cost coefficients and δ1+δ2=1, C cost With E emission is the cost and environmental cost of each power generation equipment, DG is conventional generator, ESS is energy storage, RE is renewable energy (considered as a renewable generator), N is the total number of conventional generators, renewable energy or energy storage, P d (t) is the electricity demand on the user side at time t; and are the upper and lower limits of the capacity of conventional generator i, and represents the upper and lower limits of the capacity of energy storage device i, and are the upper and lower limits of renewable energy i capacity; Step 1: According to the Lagrange multiplier method, the consistency variable λ and the optimal power generation power P are obtained. i The mathematical relationship is characterized by the consistency variable λ and the optimal power generation P i The mathematical relationship is: where a i ,b i is the cost coefficient, α i ,β i is the pollution coefficient; Step 2 is to design an economic dispatch algorithm based on finite time consistency to update the incremental cost of each generator, wherein the incremental cost update algorithm of each generator is designed as follows: In the formula, λ i is the optimal incremental cost of the ith generator, a ij is the value of the adjacency matrix, sgn represents the sign function, α and η are gain coefficients, is the gradient; the variable λ of each generator i and P i are the dual state variables and primal state variables of the economic dispatch problem, λ i The consistency allows the generator to exchange information only with adjacent generators, and the Lyapunov function V(t) is selected as: Final simplification The convergence time is: After a finite time λ i Converge to optimal incremental cost In step 3, each generator is calculated based on the optimal incremental cost. And the optimal power generation corresponding to its own parameters calculation It is characterized in that It is calculated by the following formula: In the formula, is the optimal incremental cost, a i and b i represents the cost coefficient of generator i, α i and β i is the environmental pollution emission coefficient of generator i; Step 4 determines whether the optimal output power of each generator is within the power limit range. If so, the optimal power is obtained and economic dispatch is completed. The characteristic is that the optimal power obtained is: At the next moment, jump to step 2, calculate the optimal incremental cost for the needs of different user sides, and then complete steps 3 and 4.

Images