Microgrid dynamic energy dispatching method, microgrid as storage medium

Abstract

The present invention discloses a method for dynamic energy dispatching of microgrids, microgrids and storage devices, including the following steps: S1, defining decision variables, decision variable types and setting constraints, establishing an objective function, and quantitatively simulating the optimal dispatching of microgrids with multiple constraints; S2, importing prediction variables, and using the characteristics of the source, network, load and storage of the next period of time obtained by the time series prediction algorithm based on the long short-term memory network of energy data; S3, running the precise algorithm of mixed integer programming / discrete optimization to obtain the feasible domain of all solutions that meet the constraints. The present invention can realize the flexible allocation of energy according to the dynamic changes of real-time power consumption, power generation and cost factors, and greatly reduce the unnecessary loss of energy; and by considering the equipment constraints in a refined manner, the service life of the equipment is greatly extended, and the frequency and cost of operation and maintenance are reduced; finally, relying on the optimization algorithm, the system operation efficiency and economic benefits are given equal weight, and the microgrid is promoted to operate stably, greenly and efficiently in all directions.

Description

Technical Field

[0001] The present invention relates to the technical field of smart microgrids, and in particular to a microgrid dynamic energy scheduling method, a microgrid and a storage medium. Background Art

[0002] Microgrid is a small power generation and distribution system composed of distributed power generation devices (such as distributed photovoltaic and wind power), power loads (such as production power and charging pile power), power grid power supply, distributed energy storage devices (hereinafter referred to as "source, grid, load and storage"), energy conversion devices, and monitoring and protection devices. The system has the ability to reduce electricity costs and improve economic benefits, and is an autonomous regional power system.

[0003] Energy costs for Chinese companies account for about 15% of total production costs. Some companies have reduced electricity costs by installing distributed photovoltaic power generation systems and matching them with energy storage equipment. They have the conditions for operating regional microgrids and have generated a demand for intelligent optimization of microgrid operations. However, the vast majority of companies have not yet achieved intelligent management of sources, networks, loads, and storage, resulting in energy waste and low efficiency, and there is still huge room for optimization of electricity costs. The market urgently needs a microgrid management system based on digitalization and intelligence to automatically and dynamically coordinate sources, networks, loads, and storage to achieve the goal of minimizing corporate electricity costs.

[0004] CN117237031A discloses a method and device for intelligently guiding regional energy consumption based on heuristic multi-objective optimization, which includes: constructing a multi-objective and multi-energy collaborative control model for microgrids, deploying energy consumption and energy storage sensing modules on each type of microgrid for each user, collecting real-time data on the amount of energy generated and stored, using a fusion cooperative co-evolution method, a multi-population dynamic reorganization method, and an asynchronous parallel optimization method to solve the multi-objective and multi-energy collaborative control model for microgrids, and outputting the optimal price allocation that meets the current constraints and objectives. The invention supports the power grid to conduct price guidance and intelligent microgrid control to avoid the problem of the model falling into local optimality too early, and supports the power grid to conduct price guidance and intelligent microgrid control. However, this method is guided by price and ignores the complexity of the actual power consumption environment, and fails to perform dynamic regulation according to changes in the actual power consumption state. There is still a lot of room for improvement in the control optimization of intelligent microgrids. Summary of the invention

[0005] In order to solve the above problems of the prior art solutions, the present invention proposes a microgrid dynamic energy scheduling method, a microgrid and a storage medium to solve the above technical problems.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a microgrid dynamic energy scheduling method, characterized in that it includes the following steps:

[0007] S1. Define decision variables, decision variable types and set constraints, establish objective functions, and quantitatively simulate the optimal scheduling of microgrids with multiple constraints;

[0008] S2. Import the prediction variables and use the time series prediction algorithm based on the long short-term memory network of energy data to obtain the characteristics of the source, network, load and storage in the next period, including but not limited to photovoltaic power generation, load power consumption, real-time electricity price, and corresponding parameters of each device, such as the rated capacity of energy storage;

[0009] S3. Run the exact algorithm of mixed integer programming / discrete optimization to obtain the feasible domain consisting of all solutions that satisfy the constraints. The solution that makes the objective function reach the optimal value in the feasible domain is the decision variable for model optimization.

[0010] Furthermore, the decision variables are the coordination strategies of each device at each moment, specifically unknown variables that need to be solved, including but not limited to the charge and discharge power of energy storage;

[0011] The objective function is the optimization goal under different energy management strategies, including but not limited to minimizing electricity cost, smoothing fluctuations, and orderly charging;

[0012] The constraints include the load demand limit that the output of the equipment at each moment should meet the power load demand, and the equipment's own limitations, including but not limited to the constraints of energy storage to prevent overcharging and over-discharging, and climbing constraints.

[0013] Furthermore, the microgrid optimization scheduling is to dynamically optimize the comprehensive energy consumption model in real time, and continuously adjust the coordinated strategy of photovoltaic power generation, energy storage, and power load according to the dynamic changes of power load data, photovoltaic output, energy storage charging and discharging costs, and local electricity price policies.

[0014] Furthermore, the time series prediction algorithm based on the energy data long short-term memory network is trained with historical energy data to improve the accuracy of predicting the source, network, load and storage characteristics for the next time period.

[0015] Furthermore, the precise algorithm of the mixed integer programming / discrete optimization adopts an iterative calculation method to gradually screen out solutions that meet the constraints and make the objective function optimal.

[0016] Furthermore, the defined decision variables include, but are not limited to, the output power of one or more distributed energy resources in each scheduling period, such as solar energy, wind energy, energy storage equipment, and diesel generators. For continuously adjustable energy equipment, its output power can be a continuous variable; for some equipment with discrete operating states (such as the on / off state of a diesel generator), it is a discrete variable. For example, Psolar(t) is set to the output power of the solar panel in period t (continuous variable), and Udiesel(t) is the on / off state of the diesel generator in period t (Udiesel(t) is a discrete variable, Udiesel(t) = 0 means off, and Udiesel(t) = 1 means on).

[0017] Furthermore, when determining the objective function, the goal is to minimize the operating cost or maximize the energy utilization efficiency, where the operating cost objective function includes but is not limited to one or more of the fuel cost (such as the fuel consumption cost of the diesel generator) and the loss cost of the energy storage device; assuming that the fuel cost of the diesel generator is linearly related to the output power, the unit fuel cost is C fuel , then the fuel cost of the diesel generator in time period t is C fuel ×P diesel (t)×u diesel (t), the total operating cost objective function can be expressed as the sum of the costs of various energy equipment in all time periods, that is, where is the total number of time slots in the scheduling period.

[0018] Furthermore, the constraints include but are not limited to power balance constraints, equipment capacity constraints and energy storage equipment constraints; wherein,

[0019] Power balance constraint: In each scheduling period, the total power generation of the microgrid (including various distributed energy sources) must be equal to the total load demand plus the transmission loss. The function is as follows: P solar (t)+P wind (t)+P diesel (t)×u diesel (t)+P storage (t) = P load (t)+P loss (t), where P solar (t) is the power generated by the solar panel, P wind (t) is the wind power generation power, P diesel (t)×u diesel (t) is the power generated by the diesel generator, P storage (t) is the output power of the energy storage device, P load (t) is the load demand power, P loss (t) is the transmission loss power;

[0020] Equipment capacity constraints: The output power of each energy device must be within its rated capacity; for example, for solar panels, in is the rated output power of the solar panel.

[0021] Energy storage device constraints: These include the charging and discharging power limits and energy capacity limits of energy storage devices. When discharging And the energy state E of the energy storage device storage (t) Need to be satisfied in and are the minimum and maximum energy capacities of the energy storage device, respectively.

[0022] Furthermore, it also includes an iterative calculation process, wherein the iterative calculation process is:

[0023] (1) Initialization:

[0024] Set the initial solution. For discrete variables (such as the on / off state of the equipment), the initial value can be set randomly. For continuous variables, the value can be set within its feasible interval. For example, the initial value is (the diesel generator is initially turned off), which is half of the predicted solar power in the current period, etc.

[0025] Calculate the initial objective function value, substitute the initial solution into the objective function and constraints, and check whether the constraints are met. If not, reinitialize;

[0026] (2) Iteration steps:

[0027] Select variables to adjust: In each iteration, one or more decision variables are selected for adjustment. Certain strategies can be adopted, such as according to the degree of influence of the variables on the objective function (determined by sensitivity analysis) or random selection. For example, if it is found that the on / off state of the diesel generator has a greater impact on the cost, the adjustment is selected in a certain iteration.

[0028] Adjust variable values ​​and check constraints: For continuous variables, adjust them in a certain step size within their feasible interval; for discrete variables, change their discrete values ​​(such as from 0 to 1). Then check whether the new solution meets all constraints. If not, you need to go back to the previous step or make appropriate corrections; for example, after adjusting the output power of the diesel generator, check whether the power balance constraint and equipment capacity constraint are met;

[0029] Calculate the new objective function value: Substitute the new solution that meets the constraints into the objective function and calculate the objective function value; compare the new objective function value with the objective function value of the previous iteration. If the new value is better (such as lower operating cost), accept the new solution as the current optimal solution; otherwise, abandon the new solution and continue to the next iteration;

[0030] (3) Termination conditions:

[0031] Set an upper limit for the number of iterations, and stop the iteration when this limit is reached; for example, set a maximum of 100 iterations;

[0032] Or when the objective function value changes very little (less than a certain set threshold) in multiple consecutive iterations, it is considered that it has converged to the optimal solution and the iteration is stopped. For example, when the objective function value changes less than 0.01% in 5 consecutive iterations, the iteration is terminated.

[0033] Furthermore, the branch and bound method is adopted, and the steps are as follows:

[0034] (1) Branching step: For discrete variables, such as the on / off state of a diesel generator, when it is found that there may be multiple discrete values ​​at a certain node (iteration stage), the problem is branched into multiple sub-problems; for example, in a certain iteration, the on / off state of the diesel generator is uncertain whether it is 0 or 1, so two branches are created, one forcing u diesel (t) = 0, another mandatory u diesel (t) = 1;

[0035] (2) Bounding step: Calculate an upper or lower bound of the objective function value for each branch (subproblem). You can quickly calculate an estimated bound by relaxing certain constraints (such as treating discrete variables as continuous variables). For example, when calculating the lower bound of a branch, assume that all devices operate in the most economical way (ignoring the limitations of discrete variables) to obtain a lower bound estimate of the cost. Then compare the bounds of each branch and prioritize branches that are likely to obtain the optimal solution. If the bound of a branch is worse than the currently known optimal solution, you can abandon this branch, thereby reducing the search space.

[0036] (3) Branch and bound in the iterative process: In the iterative process, branch and bound operations are continuously performed to gradually narrow the search range until the optimal solution is found or it is determined that there is no better solution. For example, as the iteration proceeds, the boundaries of each branch are continuously updated, and the branch to be explored is selected according to the boundary conditions, and finally a solution that satisfies the constraints and makes the objective function optimal is obtained.

[0037] The present invention also provides a microgrid, which is dispatched and controlled according to any of the microgrid dynamic energy dispatching methods described above.

[0038] The present invention also provides a computer-readable storage medium containing a computer program, characterized in that when the computer program is executed by one or more processors, the microgrid dynamic energy scheduling method described in any one of the above items is implemented.

[0039] Compared with traditional energy scheduling methods, the dynamic energy scheduling algorithm of the present invention has obvious advantages. It can flexibly allocate energy according to the dynamic changes of real-time electricity consumption, power generation and cost factors, and significantly reduce unnecessary energy loss. It also greatly extends the service life of equipment and reduces the frequency and cost of operation and maintenance by carefully considering equipment constraints. Finally, relying on the optimization algorithm, it gives equal importance to system operation efficiency and economic benefits, and promotes the stable, green and efficient operation of microgrids in all aspects. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 This is a schematic diagram of a microgrid according to the present invention;

[0041] Figure 2 Schematic diagram of the decision results of the energy scheduling algorithm. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] like Figure 1 and Figure 2 As shown, a microgrid dynamic energy scheduling method is characterized by comprising the following steps:

[0044] S1. Define decision variables, decision variable types and set constraints, establish objective functions, and quantitatively simulate the optimal scheduling of microgrids with multiple constraints;

[0045] S2. Import the prediction variables and use the time series prediction algorithm based on the long short-term memory network of energy data to obtain the characteristics of the source, network, load and storage in the next period, including but not limited to photovoltaic power generation, load power consumption, real-time electricity price, and corresponding parameters of each device, such as the rated capacity of energy storage;

[0046] S3. Run the exact algorithm of mixed integer programming / discrete optimization to obtain the feasible domain consisting of all solutions that satisfy the constraints. The solution that makes the objective function reach the optimal value in the feasible domain is the decision variable for model optimization.

[0047] The decision variables are the coordination strategies of each device at each moment, specifically unknown variables that need to be solved, including but not limited to the charge and discharge power of energy storage;

[0048] The objective function is the optimization goal under different energy management strategies, including but not limited to minimizing electricity cost, smoothing fluctuations, and orderly charging;

[0049] The constraints include the load demand limit that the output of the equipment at each moment should meet the power load demand, and the equipment's own limitations, including but not limited to the constraints of energy storage to prevent overcharging and over-discharging, and climbing constraints.

[0050] The microgrid optimization scheduling is to dynamically optimize the comprehensive energy consumption model in real time, and continuously adjust the coordinated strategy of photovoltaic power generation, energy storage, and power load according to the dynamic changes of power load data, photovoltaic output, energy storage charging and discharging costs, and local electricity price policies.

[0051] The time series prediction algorithm based on the energy data long short-term memory network is trained with historical energy data to improve the accuracy of predicting the source, network, load and storage characteristics in the next time period.

[0052] The precise algorithm of the mixed integer programming / discrete optimization adopts an iterative calculation method to gradually screen out solutions that meet the constraints and make the objective function optimal.

[0053] The defined decision variables include but are not limited to the output power of one or more distributed energy resources in each scheduling period, such as solar energy, wind energy, energy storage equipment, and diesel generators. For continuously adjustable energy equipment, its output power can be a continuous variable; for some equipment with discrete operating states (such as the on / off state of a diesel generator), it is a discrete variable. For example, Psolar(t) is set to the output power of the solar panel in period t (continuous variable), and Udiesel(t) is the on / off state of the diesel generator in period t (Udiesel(t) is a discrete variable, Udiesel(t) = 0 means off, and Udiesel(t) = 1 means on).

[0054] When determining the objective function, the goal is to minimize the operating cost or maximize the energy utilization efficiency, where the operating cost objective function includes but is not limited to one or more of the fuel cost (such as the fuel consumption cost of the diesel generator) and the loss cost of the energy storage device; assuming that the fuel cost of the diesel generator is linearly related to the output power, the unit fuel cost is C fuel , then the fuel cost of the diesel generator in time period t is C fuel ×P diesel (t)×u diesel (t), the total operating cost objective function can be expressed as the sum of the costs of various energy equipment in all time periods, that is, where is the total number of time slots in the scheduling period.

[0055] The constraints include but are not limited to power balance constraints, equipment capacity constraints and energy storage equipment constraints; wherein,

[0056] Power balance constraint: In each scheduling period, the total power generation of the microgrid (including various distributed energy sources) must be equal to the total load demand plus the transmission loss. The function is as follows: P solar (t)+P wind (t)+P diesel (t)×u diesel (t)+P storage (t) = P load (t)+P loss (t), where P solar (t) is the power generated by the solar panel, P wind (t) is the wind power generation power, P diesel (t)×u diesel (t) is the power generated by the diesel generator, P storage (t) is the output power of the energy storage device, P load (t) is the load demand power, P loss (t) is the transmission loss power;

[0057] Equipment capacity constraints: The output power of each energy device must be within its rated capacity; for example, for solar panels, in is the rated output power of the solar panel.

[0058] Energy storage device constraints: These include the charging and discharging power limits and energy capacity limits of energy storage devices. When discharging And the energy state E of the energy storage device storage (t) Need to be satisfied in and are the minimum and maximum energy capacities of the energy storage device, respectively.

[0059] The present invention also includes an iterative calculation process, which is:

[0060] (1) Initialization:

[0061] Set the initial solution. For discrete variables (such as the on / off state of the equipment), the initial value can be set randomly. For continuous variables, the value can be set within its feasible interval. For example, the initial value is (the diesel generator is initially turned off), which is half of the predicted solar power in the current period, etc.

[0062] Calculate the initial objective function value, substitute the initial solution into the objective function and constraints, and check whether the constraints are met. If not, reinitialize;

[0063] (2) Iteration steps:

[0064] Select variables to adjust: In each iteration, one or more decision variables are selected for adjustment. Certain strategies can be adopted, such as according to the degree of influence of the variables on the objective function (determined by sensitivity analysis) or random selection. For example, if it is found that the on / off state of the diesel generator has a greater impact on the cost, the adjustment is selected in a certain iteration.

[0065] Adjust variable values ​​and check constraints: For continuous variables, adjust them in a certain step size within their feasible interval; for discrete variables, change their discrete values ​​(such as from 0 to 1). Then check whether the new solution meets all constraints. If not, you need to go back to the previous step or make appropriate corrections; for example, after adjusting the output power of the diesel generator, check whether the power balance constraint and equipment capacity constraint are met;

[0066] Calculate the new objective function value: Substitute the new solution that satisfies the constraints into the objective function and calculate the objective function value; compare the new objective function value with the objective function value of the previous iteration. If the new value is better (such as lower operating cost), accept the new solution as the current optimal solution; otherwise, abandon the new solution and continue to the next iteration;

[0067] (3) Termination conditions:

[0068] Set an upper limit for the number of iterations, and stop the iteration when this limit is reached; for example, set a maximum of 100 iterations;

[0069] Or when the objective function value changes very little (less than a certain set threshold) in multiple consecutive iterations, it is considered that it has converged to the optimal solution and the iteration is stopped. For example, when the objective function value changes less than 0.01% in 5 consecutive iterations, the iteration is terminated.

[0070] In practical applications, the branch and bound method can also be used, and its steps are:

[0071] (1) Branching step: For discrete variables, such as the on / off state of a diesel generator, when it is found that there may be multiple discrete values ​​at a certain node (iteration stage), the problem is branched into multiple sub-problems; for example, in a certain iteration, the on / off state of the diesel generator is uncertain whether it is 0 or 1, so two branches are created, one forcing u diesel (t) = 0, another mandatory u diesel (t) = 1;

[0072] (2) Bounding step: Calculate an upper or lower bound of the objective function value for each branch (subproblem). You can quickly calculate an estimated bound by relaxing certain constraints (such as treating discrete variables as continuous variables). For example, when calculating the lower bound of a branch, assume that all devices operate in the most economical way (ignoring the limitations of discrete variables) to obtain a lower bound estimate of the cost. Then compare the bounds of each branch and prioritize branches that are likely to obtain the optimal solution. If the bound of a branch is worse than the currently known optimal solution, you can abandon this branch, thereby reducing the search space.

[0073] (3) Branch and bound in the iterative process: In the iterative process, branch and bound operations are continuously performed to gradually narrow the search range until the optimal solution is found or it is determined that there is no better solution. For example, as the iteration proceeds, the boundaries of each branch are continuously updated, and the branch to be explored is selected according to the boundary conditions, and finally a solution that satisfies the constraints and makes the objective function optimal is obtained.

[0074] The present invention also provides a microgrid, which is dispatched and controlled according to any of the microgrid dynamic energy dispatching methods described above.

[0075] The present invention also provides a computer-readable storage medium containing a computer program, characterized in that when the computer program is executed by one or more processors, the microgrid dynamic energy scheduling method described in any one of the above items is implemented.

[0076] Compared with traditional energy scheduling methods, the dynamic energy scheduling algorithm of the present invention has obvious advantages. It can realize the flexible allocation of energy according to the dynamic changes of real-time power consumption, power generation and cost factors, and greatly reduce the unnecessary energy loss; and by considering the equipment constraints in a refined manner, it can greatly extend the service life of the equipment and reduce the frequency and cost of operation and maintenance; finally, relying on the optimization algorithm, the system operation efficiency and economic benefits are given equal weight, and the microgrid is promoted to operate steadily, greenly and efficiently in an all-round way. Short-term and ultra-short-term power forecasts are carried out for power sources and loads; the energy scheduling algorithm can solve the high-frequency, nonlinear, and multi-constrained high-dimensional optimization problems faced by microgrids; through edge computing technology, strategy optimization and coordination can be carried out quickly. Through actual case calculations, the system can increase the revenue by more than 30% based on the original source and storage, greatly reduce energy costs, and maximize the economic benefits of users.

[0077] The present invention supports a variety of custom control strategies to help users flexibly adjust according to their own needs. For example, the peak shaving and valley filling strategy can reduce the electricity bill expenditure during peak periods by adjusting the electricity consumption balance demand during peak and off-peak periods; demand control avoids exceeding the upper limit of the grid load by reasonably scheduling the load. The system also supports functions such as dynamic capacity expansion, backup power supply, fluctuation smoothing, orderly charging and reverse power protection to ensure the safety and stability of system operation under complex grid conditions. It provides users with more flexible and safe electricity options.

[0078] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for dynamic energy dispatching of a microgrid, characterized in that: The following steps are involved: S1. Define decision variables, decision variable types and set constraints, establish objective functions, and quantitatively simulate the optimal scheduling of microgrids with multiple constraints; S2. Import the prediction variables and use the time series prediction algorithm based on the long short-term memory network of energy data to obtain the characteristics of the source, network, load and storage in the next period, including but not limited to photovoltaic power generation, load power consumption, real-time electricity price, and corresponding parameters of each device; S3. Run the exact algorithm of mixed integer programming / discrete optimization to obtain the feasible domain of all solutions that satisfy the constraints.

2. The microgrid dynamic energy dispatching method according to claim 1 is characterized in that: The decision variables are the coordination strategies of each device at each moment, specifically unknown variables that need to be solved, including but not limited to the charge and discharge power of energy storage; The objective function is the optimization goal under different energy management strategies, including but not limited to minimizing electricity cost, smoothing fluctuations, and orderly charging; The constraints include the load demand limit that the output of the equipment at each moment should meet the power load demand, and the equipment's own limitations, including but not limited to the constraints of energy storage to prevent overcharging and over-discharging, and climbing constraints.

3. The microgrid dynamic energy dispatching method according to claim 2 is characterized in that: The microgrid optimization scheduling is to dynamically optimize the comprehensive energy consumption model in real time, and continuously adjust the coordinated strategy of photovoltaic power generation, energy storage, and power load according to the dynamic changes of power load data, photovoltaic output, energy storage charging and discharging costs, and local electricity price policies.

4. The microgrid dynamic energy dispatching method according to claim 3 is characterized in that: The time series prediction algorithm based on the energy data long short-term memory network is trained with historical energy data to improve the accuracy of predicting the source, network, load and storage characteristics for the next time period.

5. The microgrid dynamic energy dispatching method according to claim 4 is characterized in that: The precise algorithm of the mixed integer programming / discrete optimization adopts an iterative calculation method to gradually screen out solutions that meet the constraints and make the objective function optimal.

6. The microgrid dynamic energy dispatching method according to claim 5 is characterized in that: The defined decision variables include but are not limited to the output power of one or more distributed energy resources such as solar energy, wind energy, energy storage equipment, and diesel generators in each scheduling period. For continuously adjustable energy equipment, its output power can be a continuous variable; for some equipment with discrete operating states, it is a discrete variable.

7. The microgrid dynamic energy dispatching method according to claim 5 is characterized in that: When determining the objective function, the goal is to minimize the operating cost or maximize the energy utilization efficiency, where the operating cost objective function includes but is not limited to one or more of the fuel cost and the loss cost of the energy storage device; assuming that the fuel cost of the diesel generator is linearly related to the output power, the unit fuel cost is C fuel , then the fuel cost of the diesel generator in time period t is C fuel ×P diesel (t))×u diesel (t), the total operating cost objective function can be expressed as the sum of the costs of various energy equipment in all time periods, that is, where is the total number of time slots in the scheduling period.

8. The microgrid dynamic energy dispatching method according to claim 7 is characterized in that: The constraints include but are not limited to power balance constraints, equipment capacity constraints and energy storage equipment constraints; wherein, Power balance constraint: In each scheduling period, the total power generation of the microgrid must be equal to the total load demand plus the transmission loss. The function is as follows: P solar (t)+P wind (t)+P diesel (t)×u diesel (t)+P storage (t) = P load (t)+P loss (t), where P solar (t) is the power generated by the solar panel, P wind (t) is the wind power generation power, P diesel (t)×u diesel (t) is the power generated by the diesel generator, P storage (t) is the output power of the energy storage device, P load (t) is the load demand power, P loss (t) is the transmission loss power; Equipment capacity constraints: The output power of each energy device must be within its rated capacity; for example, for solar panels, in is the rated output power of the solar panel; Energy storage device constraints: including charging and discharging power limits and energy capacity limits of energy storage devices; including, when charging When discharging And the energy state E of the energy storage device storage (t) Need to be satisfied in and are the minimum and maximum energy capacities of the energy storage device, respectively.

9. The microgrid dynamic energy dispatching method according to any one of claims 1 to 8, characterized in that: It also includes an iterative calculation process, which is: (1) Initialization: Set the initial solution. For discrete variables, the initial value can be set randomly. For continuous variables, the initial value can be set within its feasible interval. Calculate the initial objective function value, substitute the initial solution into the objective function and constraints, and check whether the constraints are met. If not, reinitialize; (2) Iteration steps: Select variables for adjustment: In each iteration, one or more decision variables are selected for adjustment. Certain strategies can be adopted, such as selecting variables according to their influence on the objective function or randomly. For example, if it is found that the on / off state of the diesel generator has a greater impact on the cost, the variable can be adjusted in a certain iteration. Adjust variable values ​​and check constraints: For continuous variables, adjust them in a certain step size within their feasible interval; for discrete variables, change their discrete values; then check whether the new solution satisfies all constraints; if not, go back to the previous step or make appropriate corrections; Calculate the new objective function value: Substitute the new solution that satisfies the constraints into the objective function and calculate the objective function value; compare the new objective function value with the objective function value of the previous iteration. If the new value is better, accept the new solution as the current optimal solution; otherwise, abandon the new solution and continue to the next iteration; (3) Termination conditions: Set an upper limit for the number of iterations, and stop the iteration when this limit is reached; for example, set a maximum of 100 iterations; Or when the objective function value changes less than the set value in multiple consecutive iterations, it is considered to have converged to the optimal solution and the iteration is stopped.

10. The microgrid dynamic energy dispatching method according to any one of claims 1 to 8, characterized in that: The branch and bound method is used, and the steps are: (1) Branching step: For discrete variables, such as the on / off state of a diesel generator, when it is found that there may be multiple discrete values ​​at a certain node, the problem is branched into multiple sub-problems; (2) Bounding step: Calculate an upper or lower bound on the objective function value for each branch; an estimated bound can be quickly calculated by relaxing certain constraints; (3) Branch and bound in the iterative process: In the iterative process, branch and bound operations are continuously performed to gradually narrow the search range until the optimal solution is found or it is determined that there is no better solution.

11. A microgrid, characterized in that: The dispatching control is performed according to the microgrid dynamic energy dispatching method described in any one of claims 1-9.

12. A computer-readable storage medium containing a computer program, characterized in that: When the computer program is executed by one or more processors, the microgrid dynamic energy scheduling method described in any one of claims 1 to 9 is implemented.

Images