Bee foraging algorithm micro-grid optimization scheduling method considering energy storage constraint

Abstract

The invention relates to the technical field of micro-grid scheduling, in particular to a bee foraging algorithm micro-grid optimal scheduling method considering energy storage constraint, which constructs a micro-grid scheduling model with the goal of minimum comprehensive operation cost, and covers distributed power generation cost, energy storage loss, power grid interaction and power shortage penalty. Energy storage charging and discharging power, capacity and SOC dynamic constraints are strictly embedded, an energy storage constraint adaptation factor alpha is introduced, a bee foraging algorithm is improved, and a mapping relation between a honey source and an energy storage state is established; the fitness value is subjected to weighted calculation by the objective function and the constraint satisfaction degree; the neighborhood search step length of employed bees is dynamically adjusted along with alpha, and reconnaissance bees generate a new nectar source based on a constraint boundary; by combining a dynamic crossover mutation operator, local optimum is effectively avoided, an optimal scheduling scheme including photovoltaic, wind power and diesel engine output and an energy storage charging and discharging strategy is finally output, and the method is suitable for intelligent scheduling of a high-proportion renewable energy microgrid.

Description

Technical Field

[0001] This invention relates to the field of microgrid scheduling technology, specifically to a microgrid optimization scheduling method that considers energy storage constraints and employs a bee-foraging algorithm. Background Technology

[0002] As the global energy structure shifts towards cleaner energy sources, the penetration rate of distributed generation (DG) in microgrids, such as solar and wind power, continues to increase. Their intermittent and fluctuating characteristics significantly increase the difficulty of balancing supply and demand in microgrids. Energy storage systems, as core components for mitigating intermittent energy fluctuations and improving power supply reliability, have become standard units in microgrids. However, how to achieve coordinated operation of distributed generation, energy storage, loads, and the main power grid through optimized scheduling has become a key bottleneck for the practical application of microgrid technology.

[0003] The core objective of microgrid optimal scheduling is to minimize overall operating costs while ensuring power supply reliability. Existing technologies are mainly divided into two categories: traditional mathematical methods and intelligent algorithms. Traditional methods, such as linear programming and mixed integer programming, have high computational accuracy, but they are prone to the curse of dimensionality and lack robustness when faced with constraints on energy storage charging and discharging power, dynamic constraints on the state of charge, and nonlinear and nonconvex problems caused by the randomness of wind and solar power output.

[0004] To address these issues, intelligent algorithms such as genetic algorithms and particle swarm optimization have been widely applied, but they still have significant drawbacks: some algorithms have slow convergence speeds and are prone to getting trapped in local optima, making them difficult to adapt to the real-time scheduling requirements of microgrids; more importantly, existing algorithms are not well adapted to energy storage constraints. Most scheduling models only simplify the consideration of the upper and lower limits of charging and discharging power, without combining the dynamic evolution law of SOC, such as the impact of charging and discharging efficiency decay and power fluctuations on SOC, to construct precise constraints. This makes the scheduling scheme prone to overcharging and over-discharging of energy storage, shortening the equipment lifespan.

[0005] Bee-based foraging algorithms have emerged in the scheduling field due to their swarm intelligence optimization characteristics, but existing applications have significant limitations: a mapping relationship between the algorithm and energy storage constraints has not been established, and fixed search step sizes cannot match the dynamic adjustment requirements of energy storage status; furthermore, there is a lack of fitness evaluation mechanisms for energy storage constraints, easily leading to invalid solutions that are cost-optimal but violate constraints. Simultaneously, existing solutions have poor adaptability to both grid-connected and islanded modes, low source-storage coordination efficiency, and a wind-solar integration rate of less than 70%. Summary of the Invention

[0006] Technical problems to be solved To address the shortcomings of existing technologies, this invention provides a microgrid optimization scheduling method that considers energy storage constraints and incorporates a bee foraging algorithm.

[0007] (II) Technical Solution To achieve the above objectives, the present invention provides the following technical solution: A microgrid optimization scheduling method based on a bee foraging algorithm considering energy storage constraints, comprising the following steps: S1: Construct a microgrid optimization scheduling model. The model takes the minimum comprehensive operating cost of the microgrid as the objective function. The constraints include the charging and discharging power constraints, capacity constraints, and SOC constraints of the energy storage system. S2: Improve the bee foraging algorithm by introducing an energy storage constraint adaptation factor and establishing a mapping relationship between nectar source location and energy storage constraint; S3: Based on the improved bee foraging algorithm, initialize the nectar source population, wherein the nectar source corresponds to the microgrid scheduling scheme; S4: Calculate the fitness value of each honey source, which is obtained by weighting the objective function value and the energy storage constraint satisfaction. S5: By employing bees to perform neighborhood searches and adjusting the search step size in conjunction with energy storage constraint adaptation factors, new honey sources are generated. S6: Scout bees eliminate unimproved nectar sources and randomly generate new nectar sources to replace them based on energy storage constraints. S7: Design a dynamic crossover and mutation operator to optimize new honey sources and avoid the algorithm getting trapped in local optima; S8: Determine whether the algorithm meets the convergence condition. If it does, output the scheduling scheme corresponding to the optimal honey source; otherwise, return to step S4. S9: The scheduling scheme includes distributed power output plan, energy storage system charging and discharging time period and power allocation; S10: The optimal scheduling scheme is sent to the microgrid control system to realize the coordinated operation of distributed power sources and energy storage systems.

[0008] Preferably, the objective function in step S1 includes the distributed generation cost, the energy storage system charging and discharging loss cost, the grid interaction cost, and the load shortage penalty cost.

[0009] More preferably, the charging and discharging power constraint of the energy storage system in step S1 satisfies: P charge min≤P charge (t)≤P charge max, P discharge min≤P discharge (t)≤P discharge max, where P charge (t), P discharge (t) represents the energy storage charging and discharging power at time t, P charge min、P charge max represents the upper and lower limits of charging power, P discharge min、P discharge max represents the upper and lower limits of the discharge power.

[0010] Preferably, the SOC constraint of the energy storage system in step S1 satisfies: SOC min ≤SOC t ≤SOC max And SOC t =SOC t−1 - × ×η, where SOC t Let SOC be the stored energy value at time t. min SOC max P represents the upper and lower limits of SOC. charge (t), P discharge (t) represents the charging power and discharging power, respectively; Δt is the time interval in hours; C0 is the rated capacity of the battery in Ah; and η is the charge / discharge efficiency coefficient, such as 0.95.

[0011] Preferably, the energy storage constraint adaptation factor α in step S2 is calculated as follows: α = λ1 × +λ2× +λ3× , where λ1, λ2, and λ3 are weighting coefficients, and λ1+λ2+λ3=1.

[0012] More preferably, the adjustment formula for the neighborhood search step size in step S5 is: Step=Step0×(1+α), where Step0 is the initial step size and α is the energy storage constraint adaptation factor.

[0013] Furthermore, the crossover probability P of the dynamic crossover operator described in step S7 is preferably... c Satisfy: P c =P c0 +(P c1 -P c0 )×(1-iter / iter max ), where P c0 Let P be the initial crossover probability. c1 The maximum crossover probability is given by iter, where iter is the current iteration number. max This represents the maximum number of iterations.

[0014] Preferably, the mutation probability P of the dynamic mutation operator in step S7 m Satisfy: P m =P m0 ×exp(-iter / iter max ), where P m0 This represents the initial mutation probability.

[0015] More preferably, the fitness value calculated in step S4 is: F = ω1 × (1 / (1+J)) + ω2 × β, where J is the objective function value, β is the energy storage constraint satisfaction, ω1 and ω2 are weight coefficients, and ω1 + ω2 = 1. Preferably, the distributed power source includes photovoltaic power, wind power and diesel generator, and the microgrid can switch between grid-connected operation and islanded operation modes.

[0016] (III) Beneficial Effects Compared with existing technologies, this invention provides a microgrid optimization scheduling method based on a bee foraging algorithm that considers energy storage constraints, which has the following beneficial effects: This technical solution constructs a comprehensive cost objective function that includes distributed power generation costs, energy storage losses, grid interaction, and power shortage penalties, and strictly embeds dynamic constraints on the charging and discharging power, capacity, and SOC of the energy storage system to ensure that the dispatching scheme achieves a balance between economic efficiency and physical feasibility.

[0017] The algorithm innovatively introduces an energy storage constraint adaptation factor α, which dynamically couples the honey source search process with the real-time energy storage status. This enables the algorithm to adaptively adjust the step size during neighborhood search, effectively avoiding infeasible solutions caused by violations of SOC or power limits, and significantly improving search efficiency and solution quality.

[0018] By designing a dynamic crossover and mutation operator, the crossover and mutation probabilities are adaptively adjusted with the iteration process, which enhances the global exploration capability in the early stage of the algorithm and strengthens local development in the later stage, significantly alleviating the problem that the traditional bee algorithm is prone to getting trapped in local optima.

[0019] The scout bee mechanism, combined with energy storage constraints, generates new nectar sources, further ensuring population diversity and constraint satisfaction. The final output scheduling scheme accurately covers the output plans of photovoltaic, wind power, and diesel engines, as well as energy storage charging and discharging strategies, supporting grid-connected / islanded mode switching and achieving safe, economical, and reliable operation of the microgrid. The overall method converges quickly and is robust, making it suitable for complex microgrid optimization and scheduling scenarios with a high proportion of renewable energy integration. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the overall method flow framework of the present invention; Figure 2 This is a schematic diagram of the S1 microgrid optimization scheduling model of the present invention; Figure 3 This is a schematic diagram of the core mechanism of the improved bee foraging algorithm S2–S7 of the present invention; Figure 4 This is a schematic diagram illustrating the content and execution of the S9–S10 scheduling scheme of the present invention; Figure 5 This is a schematic diagram illustrating the coupling relationship between the algorithm of this invention and the physical system. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Please see Figure 1-3 The present invention provides a microgrid optimization scheduling method considering energy storage constraints based on a bee foraging algorithm, comprising the following steps: S1: Construct a microgrid optimization scheduling model. The model takes the minimum comprehensive operating cost of the microgrid as the objective function. The constraints include the charging and discharging power constraints, capacity constraints, and SOC constraints of the energy storage system. S2: Improve the bee foraging algorithm by introducing an energy storage constraint adaptation factor and establishing a mapping relationship between nectar source location and energy storage constraint; S3: Based on the improved bee foraging algorithm, initialize the nectar source population, wherein the nectar source corresponds to the microgrid scheduling scheme; S4: Calculate the fitness value of each honey source, which is obtained by weighting the objective function value and the energy storage constraint satisfaction. S5: By employing bees to perform neighborhood searches and adjusting the search step size in conjunction with energy storage constraint adaptation factors, new honey sources are generated. S6: Scout bees eliminate unimproved nectar sources and randomly generate new nectar sources to replace them based on energy storage constraints. S7: Design a dynamic crossover and mutation operator to optimize new honey sources and avoid the algorithm getting trapped in local optima; S8: Determine whether the algorithm meets the convergence condition. If it does, output the scheduling scheme corresponding to the optimal honey source; otherwise, return to step S4. S9: The scheduling scheme includes distributed power output plan, energy storage system charging and discharging time period and power allocation; S10: The optimal scheduling scheme is sent to the microgrid control system to realize the coordinated operation of distributed power sources and energy storage systems.

[0023] This technical solution deeply integrates the multi-dimensional constraints of microgrid energy storage systems with an improved bee-foraging algorithm. Through a closed-loop process of constraint modeling, algorithm adaptation, intelligent optimization, and scheduling execution, it achieves the dual goals of minimizing the overall operating cost of the microgrid and ensuring the safe and stable operation of the energy storage system. Essentially, it leverages the swarm intelligence optimization characteristics of the bee-foraging algorithm, combined with personalized adaptation improvements to energy storage constraints, to address the pain points of traditional scheduling algorithms, such as low optimization accuracy, slow convergence speed, and poor constraint adaptation. Ultimately, it outputs a collaborative scheduling scheme for distributed power sources and energy storage systems, adaptable to both grid-connected and islanded operation modes of the microgrid.

[0024] Microgrid Optimal Scheduling Model (S1) Objective function: Minimize overall operating cost By quantifying the cost composition of all aspects of microgrid operation, a multi-dimensional cost aggregation objective function is constructed to ensure the optimal economic efficiency of the scheduling scheme.

[0025] Distributed power generation cost: covers the operation and maintenance costs of photovoltaic / wind power, but excludes fuel costs, fuel costs and operation and maintenance costs of diesel generators, and is calculated by summing the unit output cost × actual output. Energy storage system charging and discharging loss cost: Based on the charging and discharging efficiency η, the loss in the energy conversion process is quantified. For example, the actual stored energy during charging = input energy × η, and the output energy during discharging = stored energy × η. The cost is calculated by converting the energy loss × unit electricity price. Grid interaction cost: The cost incurred when a microgrid exchanges power with the main grid. For example, the purchase price of electricity is calculated based on the main grid's electricity sales price, and the electricity sales price is calculated based on the grid connection price. This avoids cost waste caused by wind and solar power curtailment or excessive electricity purchase. Load power shortage penalty cost: When the output of distributed power sources and energy storage cannot meet the load demand, the penalty is calculated as power shortage multiplied by the unit penalty coefficient, and the forced algorithm prioritizes the reliability of power supply.

[0026] Constraints of Energy Storage Systems Charge / discharge power constraint formula: P charge min≤P charge (t)≤P charge max, P discharge min≤P discharge (t)≤P discharge max: Based on the physical characteristics of energy storage devices, such as the maximum charge and discharge current of the battery and the power limit of the inverter, the charge and discharge power boundaries at any given time t are defined. This avoids overheating, lifespan degradation, or failure caused by over-power charging and discharging, ensuring the safe operation of the energy storage system. For example, the maximum charging power of a lithium battery is typically 0.5C-1C of its rated capacity; this constraint prevents overcurrent damage to the battery.

[0027] State of charge (SOC) constraint, State of charge constraint Constraint Formula: SOC min ≤SOC t ≤SOC max SOC t =SOC t−1 - × ×η; State of Charge (SOC) is a normalized representation of the remaining electrical energy in an energy storage system. min =20%, SOC max =80%, overcharging is avoided by upper and lower limit constraints. Exceeding the upper limit of SOC will cause battery swelling and over-discharge, while exceeding the lower limit of SOC will cause permanent capacity loss. In the dynamic SOC calculation formula, The term is used to quantify the impact of charging and discharging power fluctuations on power loss. The greater the fluctuation, the greater the loss. Δt is the scheduling time interval, such as 1 hour, to ensure the continuity of the time sequence. C0 is the rated capacity of the battery. Ah is the basic parameter. η is the charging and discharging efficiency coefficient, such as 0.95, to correct for energy conversion loss and ultimately achieve accurate dynamic tracking of SOC.

[0028] Distributed power sources and operating mode constraints Distributed power sources include photovoltaics (output depends on sunlight intensity), wind power (output depends on wind speed), and diesel generators (controllable output). As backup power sources, the algorithm needs to adapt to their output characteristics: wind and solar power are intermittent power sources, while diesel generators are controllable power sources. Microgrids support grid-connected / islanded mode switching. When connected to the grid, they can complement the main grid for peak shaving. When islanded, they need to rely on energy storage and diesel generators to ensure power supply to the load. The algorithm is constrained to adjust the power output ratio in different modes.

[0029] Improved core design of bee foraging algorithm (S2, S5, S7) Energy storage constraint adaptation factor α (S2) Calculation formula: α = λ1 × +λ2× +λ3× , where λ1, λ2, and λ3 are weighting coefficients, and λ1+λ2+λ3=1.

[0030] A mapping bridge is constructed between the energy storage state and the algorithm optimization strategy. The value of α ranges from [0,1], comprehensively reflecting the available power margin, charging power margin, and discharging power utilization of the energy storage system. SOC normalized value: the closer it is to 1, the more power is left and the greater the charging margin. : Charging power margin. The closer it is to 1, the lower the current charging power is to the upper limit, and the charging power can be increased. : Discharge power utilization rate. The closer it is to 1, the closer the current discharge power is to the upper limit. Over-power discharge should be avoided. The weighting coefficients λ1, λ2, and λ3 can be adjusted according to actual needs. For example, if power safety is emphasized, λ1 can be increased, and if charging efficiency is emphasized, λ2 can be increased, so that the algorithm prioritizes searching for high-quality solutions that meet the constraints.

[0031] Dynamic adjustment of neighborhood search step size (S5) Calculation formula: Step = Step0 × (1 + α), where Step0 is the initial step size and α is the energy storage constraint adaptation factor; Traditional bee foraging algorithms use a fixed search step size, which can lead to slow convergence in the early stages due to overly fine searching and excessively coarse searching in the later stages, making them prone to getting trapped in local optima. This scheme dynamically adjusts the step size using α: When α is large, the energy storage state is good, and the constraint satisfaction is high, Step increases, and the algorithm performs a large-scale global search to quickly traverse potential high-quality scheduling schemes. When α is small, the energy storage is close to the constraint boundary, such as when the SOC is close to the lower limit and the discharge power is close to the upper limit, Step decreases, and the algorithm performs a refined local search to avoid the generation of out-of-bounds solutions and improve the optimization accuracy.

[0032] Dynamic crossover and mutation operator (S7) Dynamic crossover probability PC Calculation formula: P c =P c0 +(P c1 -P c0 )×(1-iter / iter max ), where P c0 Let P be the initial crossover probability. c1 The maximum crossover probability is given by iter, where iter is the current iteration number. max This represents the maximum number of iterations. The crossover operator is used to fuse the advantageous genes of different nectar source scheduling schemes, dynamic P C Adaptation algorithm iteration phase: Early stage of iteration (iter / iter) max (small), P c Approaching P c1 This increases the crossover probability, promotes population diversity, and prevents the algorithm from converging to a local optimum too early. Late iteration (iter / iter) max (Large), P c Approaching P c0 This reduces the probability of crossover, stabilizes the genes of high-quality solutions, and improves convergence accuracy.

[0033] Dynamic mutation probability Pm Calculation formula: Pm =P m0 ×exp(-iter / iter max ), where P m0 The initial mutation probability; Mutation operators are used to randomly modify nectar-producing genes, preventing the population from getting trapped in local optima. Dynamic P m Design logic: Initial iteration: P m The algorithm is relatively large, and new solution spaces are explored through high-frequency mutation to prevent premature convergence. Late iteration: P m The algorithm decays exponentially with the number of iterations, reducing the damage of mutations to high-quality solutions and ensuring that the algorithm converges stably to the global optimum.

[0034] Fitness value calculation (S4) F = ω1 × (1 / (1+J)) + ω2 × β, where J is the objective function value, β is the energy storage constraint satisfaction, ω1 and ω2 are weight coefficients, and ω1 + ω2 = 1; Fitness score is a core indicator for evaluating the merits of honey source scheduling schemes. By weighted fusion of economic efficiency and constraint satisfaction, it avoids algorithms violating energy storage constraints in pursuit of minimum cost. 1 / (1+J) transforms the goal of minimizing cost into maximizing fitness. The smaller J is, the larger this value is, which is consistent with the logic of bee foraging algorithms in selecting nectar sources. β: Energy storage constraint satisfaction. If the constraint is fully satisfied, β=1. If any constraint is violated, β decreases linearly to 0 according to the degree of violation, forcing the algorithm to prioritize the solution that satisfies the constraint. The weights ω1 and ω2 can be flexibly adjusted. For example, if the grid side prioritizes economy, ω1 can be increased, and if the energy storage side prioritizes safety, ω2 can be increased to balance the scheduling objectives.

[0035] Detailed workflow steps Constructing a microgrid optimal scheduling model (S1) Define the microgrid composition: including distributed power sources such as photovoltaic / wind power / diesel generators, energy storage systems, loads and grid interaction interfaces, and support grid-connected / islanded mode switching; Define the objective function: quantify the generation cost of distributed power sources, the charging and discharging loss cost of energy storage, the grid interaction cost, and the load shortage penalty cost, and construct an objective function to minimize the overall cost; Set constraints: Input parameters such as the upper and lower limits of the energy storage system's charge and discharge power, the upper and lower limits of its state of charge (SOC), the charge and discharge efficiency η, and the rated capacity C0, and define the constraint boundaries; At the same time, the output range of distributed power sources is limited, such as the maximum output of photovoltaic power being determined by the predicted value of sunlight.

[0036] Improved bee foraging algorithm (S2) Introducing an energy storage constraint adaptation factor α: Based on the energy storage SOC state and charge / discharge power state, combined with weighting coefficients λ1, λ2, and λ3, a real-time calculation logic for α is established; A mapping relationship between nectar sources and scheduling schemes is established. Each honey source corresponds to a set of scheduling schemes, including the output of distributed power sources, the charging and discharging power of energy storage, and the power of grid interaction at each time.

[0037] Initialize the nectar source population (S3) Based on microgrid operation constraints, such as the upper limit of distributed power generation output and the initial SOC value of energy storage (SOC0), N feasible honey source scheduling schemes are randomly generated to ensure that the initial population covers the main region of the solution space. Algorithm parameters are initialized: initial step size Step0 and initial crossover probability P. c0 Maximum crossover probability P c1 Initial mutation probability P m0 Maximum number of iterations (iter) max Population size N, etc. Calculate the nectar source fitness value (S4) For each honey source scheduling scheme, substitute the objective function to calculate J, and calculate the overall operating cost; Check whether the scheme meets the energy storage constraints, charge / discharge power constraints, and SOC constraints, and calculate the constraint satisfaction degree β; The fitness value F of each nectar source is calculated using the formula F=ω1×(1 / (1+J))+ω2×β. After sorting, the top 50% of high-quality nectar sources with the highest fitness are retained.

[0038] Hire bees to search the neighborhood and generate new honey sources (S5) Hired bees perform neighborhood searches for high-quality nectar sources, with the search step size dynamically adjusted according to Step = Step0 × (1 + α), and α calculated in real time; for example, the energy storage charging power P of a certain nectar source at time t. charge (t), according to the step size Step in P charge min,P charge Adjustments are made within the maximum range to generate a new scheduling scheme; ensure that the new honey source meets the energy storage constraints, and automatically correct to the constraint boundary if it exceeds the limit.

[0039] Scout bees eliminate inferior nectar sources and replace them (S6). Scout bees monitor the fitness changes of all nectar sources. If the fitness of a nectar source does not improve after K consecutive iterations (K is a preset threshold), it is judged as a poor quality nectar source. Based on energy storage constraints, such as SOC at SOC min SOC max Within their respective upper and lower limits of charging and discharging power, new nectar sources are randomly generated to replace inferior nectar sources, thus maintaining population diversity.

[0040] Dynamic crossover mutation optimizes new honey sources (S7) Crossover operation: For newly generated honey sources, crossover is performed in pairs according to the dynamic crossover probability Pc, which is adjusted with the number of iterations, to integrate the advantages of both parties, such as the photovoltaic power output scheme of honey source A and the energy storage charging and discharging scheme of honey source B. Mutation operation: For the nectar sources after crossover, random mutation is performed according to the dynamic mutation probability Pm, which decays exponentially with the number of iterations. For example, the output of the diesel generator at a certain moment is slightly adjusted to avoid the population getting trapped in local optima. New nectar sources were selected and optimized, and populations with the top N fitness levels were retained.

[0041] Determine the convergence condition (S8) Convergence condition: If the current iteration number iter = iter max If the change in the optimal fitness value over 10 consecutive iterations is less than a preset threshold, then... If so, the algorithm converges; If converged: output the scheduling scheme corresponding to the optimal honey source; if not converged: iter=iter+1, return to step 4 to recalculate the fitness value, and enter the next iteration.

[0042] Generate a detailed scheduling plan (S9) The optimal scheduling scheme includes: the output plan of photovoltaic / wind power / diesel generators at each time, the charging and discharging time and power allocation of energy storage system, and the interaction power plan between microgrid and large power grid; For example, during periods of ample sunlight, such as 10:00-14:00, photovoltaic power is at full capacity, with some of the excess electricity going to the load, some charging the energy storage, and the remainder being sold to the main grid; during peak load periods at night, such as 19:00-22:00, energy storage discharges and diesel generators supplement the power to meet load demand and avoid excessive electricity purchase costs.

[0043] Scheduling scheme execution and coordinated operation (S10) The optimal scheduling scheme is sent to the microgrid control system, including distributed power controllers, energy storage PCS, and grid interaction switches. The control system monitors the operating status in real time: in grid-connected mode, it coordinates with the main power grid according to the interactive power plan; in islanded mode, it switches to energy storage and diesel generator as backup power supply to ensure continuous power supply to the load. Dynamic adjustment: If the actual wind and solar power output deviates significantly from the predicted value, the algorithm is triggered to quickly recalculate and update the scheduling scheme to ensure operational stability.

[0044] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A microgrid optimal scheduling method considering energy storage constraints and a bee foraging algorithm, characterized in that, Includes the following steps: S1: Construct a microgrid optimization scheduling model. The model takes the minimum comprehensive operating cost of the microgrid as the objective function. The constraints include the charging and discharging power constraints, capacity constraints, and SOC constraints of the energy storage system. S2: Improve the bee foraging algorithm by introducing an energy storage constraint adaptation factor and establishing a mapping relationship between nectar source location and energy storage constraint; S3: Based on the improved bee foraging algorithm, initialize the nectar source population, wherein the nectar source corresponds to the microgrid scheduling scheme; S4: Calculate the fitness value of each honey source, which is obtained by weighting the objective function value and the energy storage constraint satisfaction. S5: By employing bees to conduct neighborhood searches and adjusting the search step size in conjunction with energy storage constraint adaptation factors, new honey sources are generated. S6: Scout bees eliminate unimproved nectar sources and randomly generate new nectar sources to replace them based on energy storage constraints. S7: Design a dynamic crossover and mutation operator to optimize new honey sources and avoid the algorithm getting trapped in local optima; S8: Determine whether the algorithm meets the convergence condition. If it does, output the scheduling scheme corresponding to the optimal honey source; otherwise, return to step S4. S9: The scheduling scheme includes distributed power output plan, energy storage system charging and discharging time period and power allocation; S10: The optimal scheduling scheme is sent to the microgrid control system to realize the coordinated operation of distributed power sources and energy storage systems.

2. The microgrid optimal scheduling method considering energy storage constraints based on the bee foraging algorithm according to claim 1, characterized in that, The objective function mentioned in step S1 includes the generation cost of distributed power sources, the charging and discharging loss cost of energy storage systems, the grid interaction cost, and the load shortage penalty cost.

3. The microgrid optimal scheduling method considering energy storage constraints based on the bee foraging algorithm according to claim 1, characterized in that, The charging and discharging power constraint of the energy storage system in step S1 satisfies: P charge min≤P charge (t)≤P charge max, P discharge min≤P discharge (t)≤P discharge max, where P charge (t), P discharge (t) represents the energy storage charging and discharging power at time t, P charge min、P charge max represents the upper and lower limits of charging power, P discharge min、P discharge max represents the upper and lower limits of the discharge power.

4. The microgrid optimal scheduling method considering energy storage constraints based on the bee foraging algorithm according to claim 1, characterized in that, The SOC constraint of the energy storage system in step S1 satisfies: SOC min ≤SOC t ≤SOC max And SOC t =SOC t−1 - × ×η, where SOC t Let SOC be the stored energy value at time t. min SOC max The upper and lower limits of SOC are given by Pcharge(t) and Pdischarge(t), respectively, where Pcharge(t) and Pdischarge(t) are the charging power and discharging power, respectively. Δt is the time interval in hours, C0 is the rated capacity of the battery in Ah, and η is the charge / discharge efficiency coefficient, such as 0.

95.

5. A microgrid optimal scheduling method considering energy storage constraints based on a bee foraging algorithm according to claim 1, characterized in that, The formula for calculating the energy storage constraint adaptation factor α in step S2 is: α = λ1 × +λ2× +λ3× , where λ1, λ2, and λ3 are weighting coefficients, and λ1+λ2+λ3=1.

6. A microgrid optimal scheduling method considering energy storage constraints based on a bee foraging algorithm according to claim 1, characterized in that, The formula for adjusting the neighborhood search step size in step S5 is: Step = Step0 × (1 + α), where Step0 is the initial step size and α is the energy storage constraint adaptation factor.

7. A microgrid optimal scheduling method considering energy storage constraints based on a bee foraging algorithm according to claim 1, characterized in that, The crossover probability P of the dynamic crossover operator in step S7 c Satisfy: P c =P c0 +(P c1 -P c0 )×(1-iter / iter max ), where P c0 Let P be the initial crossover probability. c1 The maximum crossover probability is given by iter, where iter is the current iteration number. max This represents the maximum number of iterations.

8. A microgrid optimal scheduling method considering energy storage constraints based on a bee foraging algorithm according to claim 1, characterized in that, The mutation probability P of the dynamic mutation operator described in step S7 m Satisfy: P m =P m0 ×exp(-iter / iter max ), where P m0 This represents the initial mutation probability.

9. A microgrid optimal scheduling method considering energy storage constraints based on a bee foraging algorithm according to claim 1, characterized in that, The fitness value calculated in step S4 is: F=ω1×(1 / (1+J))+ω2×β, where J is the objective function value, β is the energy storage constraint satisfaction, ω1 and ω2 are weight coefficients, and ω1+ω2=1.

10. A microgrid optimal scheduling method considering energy storage constraints based on a bee foraging algorithm according to claim 1, characterized in that, The distributed power sources include photovoltaic power sources, wind power sources, and diesel generators. The microgrid can switch between grid-connected operation and islanded operation modes.

Images