Micro-grid multi-source collaborative optimization method based on mixed game and regret value constraint

Abstract

The invention belongs to the technical field of micro-grid energy management, and particularly relates to a micro-grid multi-source collaborative optimization method based on a mixed game and regret value constraint. According to the method, a double-layer collaborative framework of'upper-layer master-slave game-lower-layer cooperative game 'is constructed, a power distribution network operator is used as an upper-layer leader to formulate a dynamic electricity price strategy to realize global optimization, and an energy storage system is used as a local leader at the lower layer to coordinate distributed energy such as wind energy and photovoltaic energy, so that a multi-agent collaborative optimization model is formed. In addition, according to the method, the robustness level of the system is dynamically adjusted by introducing the regret value constraint, the balance between economy and reliability is achieved, and the energy storage operation efficiency is improved in combination with a dynamic SOC balance mechanism and a life loss model. According to the method, iterative optimization is carried out through strategies of adjusting an electricity price strategy, optimizing energy storage coordination, enhancing regret value constraint, improving an SOC balance mechanism and the like, and finally an optimal solution meeting efficient, economical and reliable operation requirements of the micro-grid is output.

Description

Technical Field

[0001] The present invention belongs to the technical field of microgrid energy management, and in particular relates to a microgrid multi-source collaborative optimization method based on hybrid game and regret value constraints. Background Art

[0002] As the penetration of renewable energy in microgrids continues to increase, traditional centralized optimization methods are no longer able to adapt to the decentralized decision-making characteristics of distributed energy resources. Single game models, however, cannot address the dual needs of global optimization and local coordination. Existing technologies suffer from three key issues: First, while the traditional master-slave game framework can guide electricity prices, it ignores the synergistic potential among distributed energy resources. Second, conventional optimization models lack consideration of the behavioral characteristics of decision-makers, particularly the potential for strategic reversals when actual returns fall short of expectations. Finally, energy storage system operations often focus solely on short-term economic viability, ignoring the impact of long-term lifespan losses on system reliability. Summary of the Invention

[0003] The present invention aims to solve one of the problems existing in the background technology.

[0004] To this end, the present invention provides a microgrid multi-source collaborative optimization method based on hybrid game and regret value constraint.

[0005] The technical solution adopted by the present invention to solve its technical problem is:

[0006] A microgrid multi-source collaborative optimization method based on hybrid game and regret value constraint, including:

[0007] Step 1: Construct a two-layer collaborative framework. A two-layer collaborative optimization framework consisting of an upper-layer master-slave game and a lower-layer cooperative game is established to perform hierarchical coordinated control of the microgrid multi-source system.

[0008] Step 2: Introduce a regret value constraint mechanism. Introduce regret value constraints into the game model, combine the dynamic SOC equilibrium mechanism and the life loss model to ensure the rationality and stability of decision-making;

[0009] Step 3: Develop a dynamic electricity pricing strategy. The distribution network operator will develop a dynamic electricity pricing strategy based on the game model to guide the optimal scheduling of distributed energy resources.

[0010] Step 4: Coordinate the operation of distributed energy resources, using the energy storage system as a local leader to coordinate the output distribution of distributed energy resources such as photovoltaic and wind power;

[0011] Step 5: Dynamically adjust the system robustness by adjusting the robustness control parameters in real time according to the system operating status to improve the energy storage system's operating efficiency and system stability;

[0012] Step 6: Build a multi-agent collaborative optimization model and solve the optimal balance between economy and reliability through hybrid game theory;

[0013] Step 7: Simulation Verification and Strategy Adjustment: Conduct simulation experiments to verify the results. If the results do not meet expectations, adjust the electricity price strategy, optimize the energy storage coordination plan, strengthen the regret value constraint, or improve the SOC balancing mechanism.

[0014] Step 8: Output the optimal solution. When the simulation verification meets expectations, the system outputs the global optimal solution after coordinated optimization of the hybrid game and regret value constraints.

[0015] Furthermore, in step 2, the regret value constraint condition is: Among them, P ESS (t) represents the actual charging and discharging power of the energy storage system at time t, represents the optimal charging and discharging power of the energy storage system at time t, It represents the maximum allowed regret value, which is used to limit the deviation between the actual decision and the optimal decision.

[0016] Furthermore, in step 2, the dynamic SOC balancing mechanism is: SOC ESS (t+1)=SOC ESS (t)+η·(P ESS (t)-P LOAD (t)), where SOC ESS (t) represents the state of charge of the energy storage system at time t, η represents the SOC change rate, which is related to the charge and discharge efficiency of the energy storage system; P LOAD (t) represents the load demand of the microgrid at time t.

[0017] Furthermore, in the step 2, the life loss model is: L ESS (t) = L ESS (t-1)+ΔL ESS (t), where L ESS (t) represents the cumulative life loss of the energy storage system at time t, ΔL ESS (t) represents the incremental life loss of the energy storage system within time t, which is usually related to the charge and discharge depth and the number of cycles.

[0018] Furthermore, in step 4, the objective function for optimizing the output distribution of distributed energy can be expressed as: Among them C ESS (t) represents the operating cost of the energy storage system at time t, C PV (t) represents the operating cost of the photovoltaic power generation system at time t, C WT (t) represents the operating cost of the wind power generation system at time t, CMT (t) represents the operating cost of the micro gas turbine at time t, C DE (t) represents the operating cost of the diesel generator at time t; the output of the energy storage system needs to meet: P ESS,min ≤P ESS (t)≤P ESS,max , where P ESS,min and P ESS,max are the minimum and maximum charge and discharge power of the energy storage system respectively; for the output distribution of distributed energy such as photovoltaic, wind power, micro gas turbine, diesel generator, etc., it is necessary to meet the power balance constraint: P NET (t) = P LOAD (t)+P ESS (t)-P PV (t)-P WT (t)-P MT (t)-P DE (t), where P NET (t) represents the net load of the microgrid at time t, P LOAD (t) represents the total power load in the microgrid, P ESS (t) represents the power of the energy storage system, P PV (t), P MT (t) represents the power generated by photovoltaic and wind power, P MT (t) represents the power generation of the micro gas turbine, P DE (t) represents the power generated by the diesel generator.

[0019] Furthermore, in step five, the robustness control parameter is adjusted in real time according to the system operating status, θ(t+1)=θ(t)+α·Δθ(t), where θ(t) represents the robustness control parameter at the current time t, α represents the adjustment step size, which is used to control the speed of parameter adjustment, and Δθ(t) represents the adjustment amount based on the system operating status.

[0020] Furthermore, the adjustment amount Δθ(t) is defined based on the charge and discharge power deviation of the energy storage system: Among them, P ESS (t) represents the actual charging and discharging power of the energy storage system at time t, It represents the optimal charging and discharging power of the energy storage system at time t, where T is the optimized time range.

[0021] Furthermore, in step six, the objective function of the multi-agent collaborative optimization model is:

[0022] Furthermore, the reliability cost R(t) is defined based on the deviation of the system net load: R(t) = |P NET (t)-P LOAD(t)|.

[0023] Furthermore, the optimal balance between economy and reliability can be solved through hybrid game theory using the following formula:

[0024] The beneficial effect of this invention lies in its ability to achieve collaborative optimization of microgrid multi-source systems across multiple dimensions, including economic benefits, operational reliability, and equipment lifespan, by constructing a two-layer collaborative framework: an upper-layer master-slave game and a lower-layer cooperative game. This innovatively introduces a regret value constraint mechanism, combined with dynamic SOC balancing and a lifespan loss model. This approach effectively overcomes the limited adaptability of traditional methods in complex scenarios, providing a superior solution for microgrids with a high proportion of renewable energy access. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The present invention will be further described below with reference to the accompanying drawings and examples.

[0026] Figure 1 It is a method flow chart of the microgrid multi-source collaborative optimization method based on hybrid game and regret value constraint in the present invention.

[0027] Figure 2 This is an energy utilization distribution diagram after the microgrid is optimized and managed using the microgrid multi-source collaborative optimization method in Example 1 of the present invention.

[0028] Figure 3 This is a diagram of energy storage system state changes after the microgrid is optimized and managed using the microgrid multi-source collaborative optimization method in Example 1 of the present invention. DETAILED DESCRIPTION

[0029] The present invention will now be described in further detail with reference to the accompanying drawings, which are simplified schematic diagrams that illustrate the basic structure of the present invention in a schematic manner.

[0030] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, features defined as "first" or "second" may explicitly or implicitly include one or more of such features. In the description of the present invention, unless otherwise specified, "multiple" means two or more.

[0031] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed, detachable, or integral connections; mechanical or electrical connections; direct or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.

[0032] A microgrid multi-source collaborative optimization method based on hybrid game and regret value constraint includes the following steps:

[0033] Step 1: Construct a two-layer collaborative framework. Establish a two-layer collaborative optimization framework including the upper-layer master-slave game and the lower-layer cooperative game to realize the hierarchical coordinated control of the microgrid multi-source system.

[0034] A two-layer collaborative optimization framework is established. The upper layer uses a master-slave game model to define the game relationship between distribution network operators (leaders) and distributed energy clusters (followers); the lower layer uses a cooperative game model to describe the collaborative relationship between energy storage systems and distributed energy resources such as wind power and photovoltaics.

[0035] ① Upper-level master-slave game model

[0036] The objective function of the leader (distribution network operator):

[0037]

[0038] Among them, p L represents the decision variables of the distribution network operator, X L (p L ) represents the cost function of the distribution network operator, which usually includes the purchase cost, transmission cost, etc. L (pL ,p F ) represents the reliability cost, and the decision p of the distributed energy cluster F Related, λ represents the weight coefficient of reliability cost.

[0039] Follower (distributed energy cluster) objective function:

[0040]

[0041] Among them, p F represents the decision variables of the distributed energy cluster, C F (p F represents the cost function of the distributed energy cluster, which usually includes operation cost and maintenance cost, R F (p L ,p F ) represents the reliability cost associated with the distributed energy cluster, and the decision-making process of the distribution network operator L is related, μ represents the weight coefficient of reliability cost.

[0042] ②Lower-level cooperative game model

[0043] The objective function of the cooperative game is:

[0044]

[0045] Among them, p S represents the decision variables of the energy storage system, p D represents the decision variables of distributed energy, C ESS (t), C PV (t), C WT (t), C MT (t), C DE (t) represents the operating cost of the energy storage system, photovoltaic, wind power, micro gas turbine, and diesel generator at time t, respectively, and R(t) represents the reliability cost of the system at time t.

[0046] Step 2: Introduce the regret value constraint mechanism. Introduce the regret value constraint condition into the game model, combine the dynamic SOC equilibrium mechanism and life loss model to ensure the rationality and stability of decision-making.

[0047] A regret value constraint mechanism is introduced, and regret value constraints are introduced into the game model. Combined with the dynamic SOC equilibrium mechanism and life loss model, the rationality and stability of decision-making are ensured.

[0048] ① Regret value constraints

[0049]

[0050] Among them, PESS (t) represents the actual charging and discharging power of the energy storage system at time t, represents the optimal charging and discharging power of the energy storage system at time t, It represents the maximum allowed regret value, which is used to limit the deviation between the actual decision and the optimal decision.

[0051] ② Dynamic SOC balancing mechanism

[0052] SOC ESS (t+1)=SOC ESS (t)+η·(P ESS (t)-P LOAD (t))

[0053] Among them, SOC ESS (t) represents the state of charge of the energy storage system at time t, η represents the SOC change rate, which is related to the charge and discharge efficiency of the energy storage system; P LOAD (t) represents the load demand of the microgrid at time t.

[0054] ③Life loss model

[0055] L ESS (t) = L ESS (t-1)+ΔL ESS (t)

[0056] Among them, L ESS (t) represents the cumulative life loss of the energy storage system at time t, ΔL ESS (t) represents the incremental life loss of the energy storage system within time t, which is usually related to the charge and discharge depth and the number of cycles.

[0057] Thus, the objective function of the leader (distribution network operator) in the upper master-slave game model is:

[0058]

[0059] Follower (distributed energy cluster) objective function:

[0060]

[0061] The objective function of the cooperative game in the lower-level cooperative game model is:

[0062]

[0063] Among them, γ L , γ F is the life loss penalty term in the corresponding objective function, γ represents the weight coefficient of life loss, L ESS (t) represents the cumulative life loss of the energy storage system at time t.

[0064] Step 3: Develop a dynamic electricity pricing strategy. The distribution network operator will develop a dynamic electricity pricing strategy based on the game model to guide the optimal scheduling of distributed energy.

[0065] The distribution network operator formulates a dynamic electricity price strategy based on the game model to guide the optimal scheduling of distributed energy.

[0066] ①The goal of dynamic electricity pricing strategy

[0067] Distribution network operators formulate dynamic electricity pricing strategies based on game models, aiming to guide the optimal scheduling of distributed energy while balancing the grid operation costs and user electricity costs.

[0068] ②Formula for dynamic electricity pricing strategy

[0069] Dynamic electricity price strategies can be adjusted according to grid load conditions and the output of distributed energy resources.

[0070] The specific formula is as follows:

[0071]

[0072] Among them, p 电价 (t) represents the dynamic electricity price at time t; β(t) represents the grid load rate at time t, which is expressed as Among them, P lOAD (t) represents the load demand at time t, P MAX It represents the maximum capacity of the power grid; a1, a2, a3, and a4 represent the electricity prices during off-peak, normal, peak, and on-peak periods, respectively; l1, l2, l3, and l4 represent the critical values of different load rate intervals, which are used to divide electricity price periods.

[0073] ③ Dynamic electricity pricing strategy

[0074] Through dynamic electricity pricing strategies, distribution network operators can adjust electricity prices in real time based on grid load conditions, guiding distributed energy resources to increase output during off-peak hours and reduce output during peak hours, thereby optimizing grid operations. This strategy not only helps reduce grid operating costs but also improves the utilization efficiency of distributed energy resources.

[0075] Step 4: Coordinate the operation of distributed energy, using the energy storage system as a local leader to coordinate the output distribution of distributed energy such as photovoltaic and wind power.

[0076] Coordinate the operation of distributed energy, use the energy storage system as a local leader, and coordinate the output distribution of distributed energy such as photovoltaics, wind power, micro gas turbines, and diesel generators.

[0077] ①Objective function

[0078] Taking the energy storage system as the local leader, the objective function for optimizing the output distribution of distributed energy can be expressed as:

[0079] Among them C ESS (t) represents the operating cost of the energy storage system at time t, C PV (t) represents the operating cost of the photovoltaic power generation system at time t, C WT (t) represents the operating cost of the wind power generation system at time t, C MT (t) represents the operating cost of the micro gas turbine at time t, C DE (t) represents the operating cost of the diesel generator at time t.

[0080] ② Output coordination formula of energy storage system

[0081] As a local leader, the output of the energy storage system must meet the following constraints:

[0082] P EsS,min ≤P ESS (t)≤P ESS,max

[0083] Among them, P eSS,min and P ESS,max are the minimum and maximum charge and discharge power of the energy storage system, respectively.

[0084] ③Distributed energy output allocation formula

[0085] For the output distribution of distributed energy sources such as photovoltaics, wind power, micro gas turbines, and diesel generators, the power balance constraint must be met: P NET (t) = P LOAD (t)+P ESS (t)-P PV (t)-P WT (t)-P MT (t)-P DE (t), where P NET (t) represents the net load of the microgrid at time t, P LOAD (t) represents the total power load in the microgrid, P ESS (t) represents the power of the energy storage system, P PV (t), P WT (t) represents the power generated by photovoltaic and wind power, P MT (t) represents the power generation of the micro gas turbine, P DE (t) represents the power generated by the diesel generator.

[0086] Step five: Dynamically adjust the system robustness and adjust the robustness control parameters in real time according to the system operating status to improve the operating efficiency and stability of the energy storage system.

[0087] Dynamically adjust system robustness and adjust robustness control parameters in real time according to system operating status to improve energy storage system operating efficiency and system stability.

[0088] ① Robust control parameter adjustment formula

[0089] Assuming that the robustness control parameter is θ(t), the formula for dynamically adjusting the robustness control parameter according to the system operating status can be expressed as:

[0090] θ(t+1)=θ(t)+α·Δθ(t)

[0091] Where θ(t) represents the robust control parameter at the current time t, α represents the adjustment step size, which is used to control the speed of parameter adjustment, and Δθ(t) represents the adjustment amount based on the system operating status.

[0092] ②Calculation formula for adjustment amount Δθ(t)

[0093] Defined based on the charge and discharge power deviation of the energy storage system:

[0094]

[0095] Among them, P ESS (t) represents the actual charging and discharging power of the energy storage system at time t, It represents the optimal charging and discharging power of the energy storage system at time t, where T is the optimized time range.

[0096] ③System stability improvement formula

[0097] To improve system stability, a stability index S(t) can be introduced as part of the optimization objective. The stability index S(t) is introduced into both the upper-level master-slave game model and the lower-level cooperative game model. In the upper-level model, it is added to the objective function of the leader or follower to reflect global stability responsibility. In the lower-level model, it is used as a cooperative objective to coordinate local behaviors. The stability index can be defined as:

[0098] ④ Regret value constraint

[0099] In order to ensure the robustness of the optimization results, the regret value constraint also needs to be met:

[0100]

[0101] in, represents the optimal charging and discharging power of the energy storage system at time t, Indicates the maximum allowed regret value.

[0102] Step 6: Establish a collaborative optimization model, construct a multi-agent collaborative optimization model, and solve the optimal balance between economy and reliability through hybrid game theory.

[0103] Establish a collaborative optimization model, construct a multi-agent collaborative optimization model, and solve the optimal balance between economy and reliability through hybrid game theory.

[0104] ① Collaborative optimization objective function

[0105] When building a multi-agent collaborative optimization model, it is necessary to consider both economy and reliability. The objective function can be expressed as:

[0106]

[0107] ②Reliability cost formula

[0108] The reliability cost R(t) can be defined based on the system operating status and reliability requirements. It can be defined based on the deviation of the system net load:

[0109] R(t)=|P NET (t)-P LOAD (t)|

[0110] ③Mixed game solution formula

[0111] The following formula can be used to solve the optimal balance between economy and reliability through mixed game theory:

[0112]

[0113] Step 7: Simulation verification and strategy adjustment: Conduct simulation experiments to verify the results. If the results do not meet expectations, adjust the electricity price strategy, optimize the energy storage coordination plan, strengthen the regret value constraint, or improve the SOC balancing mechanism.

[0114] Step 8: Output the optimal solution. When the simulation verification meets expectations, the system outputs the global optimal solution after the coordinated optimization of the hybrid game and the regret value constraint. The output results include the dynamic electricity price strategy formulated by the distribution network operator, the optimized output plan of distributed energy such as wind power and photovoltaics, the energy storage system control parameters based on the dynamic SOC balance and life loss model, and the system operation threshold configuration after robustness adjustment. This optimal solution achieves the comprehensive optimization of system operation reliability and energy storage efficiency while ensuring economy through the two-layer coordination mechanism of the upper master-slave game and the lower cooperative game, providing a complete solution for the multi-source coordinated operation of microgrids.

[0115] Example 1

[0116] This application studies a microgrid system that covers multiple energy generation methods and energy storage devices. Its data sources are extensive and diverse. Taking a microgrid that includes photovoltaic power generation, wind power generation, gas turbine power generation, diesel power generation, energy storage discharge, and energy storage charging systems as an example, traditional scheduling methods generally adopt rule-based scheduling and static optimization, which may lead to insufficient renewable energy absorption, high rates of curtailment of solar and wind power, rapid degradation of energy storage life, poor economy, and insufficient stability.

[0117] After the energy scheduling is carried out by the microgrid multi-source collaborative optimization method based on hybrid game and regret value constraint in this application, the results are as follows: Figure 2 、 3 As shown:

[0118] Figure 2 The microgrid's energy balance exhibits significant fluctuations over a 24-hour period. During the morning peak period (7:00-9:00), total load rises significantly. During this period, photovoltaic power generation has not yet reached its peak, while wind power contribution remains relatively stable. Gas turbines and diesel engines, as the primary power sources, carry the majority of the load, while the energy storage system remains in a discharging state to fill the power gap. During the daytime (8:00-16:00), photovoltaic power generation increases significantly, becoming the primary power source alongside wind power, and the energy storage system shifts to a charging state to store excess energy. During the evening peak period (19:00-22:00), total load rises again. At this time, photovoltaic power generation drops to zero, while wind power and gas turbines become the primary power sources, and the energy storage system resumes discharging to relieve grid pressure. During the nighttime (22:00-6:00), load is lower, primarily supplied by gas turbines, wind power, and a small number of diesel engines, while the energy storage system remains in a charging or idle state. Overall, the microgrid effectively balances supply and demand at different times through the coordinated scheduling of multiple energy sources and the flexible charging and discharging of energy storage.

[0119] Figure 3The graph shows the state of charge (SOC) of the microgrid energy storage system over a 24-hour period, with the state of charge (SOC) on the vertical axis and time on the horizontal axis. This graph illustrates the system's scheduling strategy: from midnight to 1 a.m., the SOC rapidly charges from an initial 70% to over 80%, remaining stable from 1 a.m. to 6 a.m., actively absorbing energy during the nighttime off-peak period. From 6 a.m. to 7 a.m., the SOC decreases slightly, then rapidly decreases from over 80% to around 70% from 7 a.m. to 9 a.m., corresponding to continuous discharge during the morning peak. From 9 a.m. to 10 a.m., the SOC increases slightly, remaining between 70% and 75% from 10 p.m. to 1 p.m., corresponding to moderate charging during the photovoltaic peak period. From 1 p.m. to 3 p.m., the SOC rapidly decreases to around 50%, remaining stable from 15 p.m. to 17 p.m., then slightly decreases from 17 p.m. to 18 p.m., prioritizing discharge during the evening peak period, keeping the SOC relatively low from 18 p.m. to 22 p.m.. From 22 p.m. to 10 p.m., the SOC rapidly recovers to over 80%, with active charging again during the nighttime off-peak period. The curve fluctuates by approximately 40% (from 45% to 85%), indicating that energy storage is deeply involved in system regulation. The SOC remains within a safe range (20% to 90%), preventing overcharge and overdischarge, ensuring battery life and regulatory capacity.

[0120] With the above-described preferred embodiments of the present invention as a guide, and with reference to the above description, relevant personnel are fully capable of making various changes and modifications without departing from the technical spirit of this invention. The technical scope of this invention is not limited to the contents of the specification, but must be determined according to the scope of the claims.

Claims

1. A microgrid multi-source collaborative optimization method based on hybrid game and regret value constraint, characterized in that: include, Step 1: Construct a two-layer collaborative framework. A two-layer collaborative optimization framework consisting of an upper-layer master-slave game and a lower-layer cooperative game is established to perform hierarchical coordinated control of the microgrid multi-source system. Step 2: Introduce a regret value constraint mechanism. Introduce regret value constraints into the game model, combine the dynamic SOC equilibrium mechanism and the life loss model to ensure the rationality and stability of decision-making; Step 3: Develop a dynamic electricity pricing strategy. The distribution network operator will develop a dynamic electricity pricing strategy based on the game model to guide the optimal scheduling of distributed energy resources. Step 4: Coordinate the operation of distributed energy resources, using the energy storage system as a local leader to coordinate the output distribution of distributed energy resources such as photovoltaic and wind power; Step 5: Dynamically adjust the system robustness by adjusting the robustness control parameters in real time according to the system operating status to improve the energy storage system's operating efficiency and system stability; Step 6: Build a multi-agent collaborative optimization model and solve the optimal balance between economy and reliability through hybrid game theory; Step 7: Simulation Verification and Strategy Adjustment: Conduct simulation experiments to verify the results. If the results do not meet expectations, adjust the electricity price strategy, optimize the energy storage coordination plan, strengthen the regret value constraint, or improve the SOC balancing mechanism. Step 8: Output the optimal solution. When the simulation verification meets expectations, the system outputs the global optimal solution after coordinated optimization of the hybrid game and regret value constraints.

2. The microgrid multi-source collaborative optimization method based on hybrid game and regret value constraint according to claim 1 is characterized in that: In step 2, the regret value constraint condition is: Among them, P ESS (t) represents the actual charging and discharging power of the energy storage system at time t, represents the optimal charging and discharging power of the energy storage system at time t, It represents the maximum allowed regret value, which is used to limit the deviation between the actual decision and the optimal decision.

3. The microgrid multi-source collaborative optimization method based on hybrid game and regret value constraint according to claim 2 is characterized in that: In step 2, the dynamic SOC balancing mechanism is: SOC ESS (t+1)=SOC ESS (t)+η·(P ESS (t)-P LOAD (t)), where SOC ESS (t) represents the state of charge of the energy storage system at time t, η represents the SOC change rate, which is related to the charge and discharge efficiency of the energy storage system; P LOAD (t) represents the load demand of the microgrid at time t.

4. The microgrid multi-source collaborative optimization method based on hybrid game and regret value constraint according to claim 3 is characterized in that: In the step 2, the life loss model is: L ESS (t) = L ESS (t-1)+ΔL ESS (t), where L ESS (t) represents the cumulative life loss of the energy storage system at time t, ΔL ESS (t) represents the incremental life loss of the energy storage system within time t, which is usually related to the charge and discharge depth and the number of cycles.

5. The microgrid multi-source collaborative optimization method based on hybrid game and regret value constraint according to claim 1 is characterized in that: In step 4, the objective function for optimizing the output distribution of distributed energy can be expressed as: Among them C ESS (t) represents the operating cost of the energy storage system at time t, C PV (t) represents the operating cost of the photovoltaic power generation system at time t, C WT (t) represents the operating cost of the wind power generation system at time t, C MT (t) represents the operating cost of the micro gas turbine at time t, C DE (t) represents the operating cost of the diesel generator at time t; the output of the energy storage system needs to meet: P ESS,min ≤P ESS (t)≤P ESS,max , where P ESS,min and P ESS,max are the minimum and maximum charge and discharge power of the energy storage system respectively; for the output distribution of distributed energy sources such as photovoltaic, wind power, micro gas turbines, and diesel generators, the power balance constraint must be met: P NET (t) = P LOAD (t)+P ESS (t)-P PV (t)-P WT (t)-P MT (t)-P DE (t), where P NET (t) represents the net load of the microgrid at time t, P LOAD (t) represents the total power load in the microgrid, P ESS (t) represents the power of the energy storage system, P PV (t), P WT (t) represents the power generated by photovoltaic and wind power, P MT (t) represents the power generation of the micro gas turbine, P DE (t) represents the power generated by the diesel generator.

6. The microgrid multi-source collaborative optimization method based on hybrid game and regret value constraint according to claim 1 is characterized in that: In step five, the robustness control parameter is adjusted in real time according to the system operating status, θ(t+1)=θ(t)+α·Δθ(t), where θ(t) represents the robustness control parameter at the current time t, α represents the adjustment step size, which is used to control the speed of parameter adjustment, and Δθ(t) represents the adjustment amount based on the system operating status.

7. The microgrid multi-source collaborative optimization method based on hybrid game and regret value constraint according to claim 6 is characterized in that: The adjustment value Δθ(t) is defined based on the charge and discharge power deviation of the energy storage system: Among them, P ESS (t) represents the actual charging and discharging power of the energy storage system at time t, It represents the optimal charging and discharging power of the energy storage system at time t, where T is the optimized time range.

8. The microgrid multi-source collaborative optimization method based on hybrid game and regret value constraint according to claim 1 is characterized in that: In step 6, the objective function of the multi-agent collaborative optimization model is:

9. The microgrid multi-source collaborative optimization method based on hybrid game and regret value constraint according to claim 8, characterized in that: The reliability cost R(t) is defined based on the deviation of the system net load: R(t) = |P NET (t)-P LOAD (t)|.

10. The microgrid multi-source collaborative optimization method based on hybrid game and regret value constraint according to claim 9, characterized in that: The following formula can be used to solve the optimal balance between economy and reliability through mixed game theory:

Images