A power distribution optimization scheduling method and system suitable for a commercial complex microgrid

Abstract

The application discloses a power distribution optimization scheduling method and system suitable for a commercial complex microgrid. In the modeling, a modular layered structure is introduced, and the whole system is divided into a main power source scheduling layer, a renewable output prediction response layer and an energy storage flexible adjustment layer. In the solving, a collaborative mechanism of mathematical optimization + reinforcement learning is adopted, an MILP model is used for accurate optimization on a deterministic part, and a reinforcement learning algorithm is used for online strategy generation on an uncertain part, so that fast iterative update and feedback control of the scheduling result are realized. Meanwhile, through a boundary coupling coordination algorithm, synchronous update of state variables among the sub-models and global feasibility of the scheduling solution are guaranteed, so that the response speed of the whole scheduling system, the solution quality and the system robustness are improved.

Description

Technical Field

[0001] This invention belongs to the interdisciplinary field of microgrid optimization scheduling and intelligent control, specifically relating to a power distribution optimization scheduling method and system applicable to microgrids in commercial complexes. Background Technology

[0002] With the development of renewable energy technologies and the advancement of urban energy structure transformation, an increasing number of commercial complexes, smart parks, and cultural tourism areas are deploying localized microgrid systems to enhance their energy autonomy and resilience. These urban load centers are typically equipped with photovoltaic power generation, energy storage systems, and emergency backup power supplies to support independent operation in the event of grid peak-valley price differences, power curtailment, maintenance, or faults. Commercial complex microgrids are characterized by high load density, highly regular but volatile electricity demand, and frequent grid-connected / off-grid switching. Furthermore, their complex energy system structure and dynamically changing operating environment place higher demands on the real-time performance, economy, and robustness of the dispatch system.

[0003] Existing microgrid optimization and dispatch methods are significantly inadequate when dealing with the characteristics of urban commercial complex microgrid systems, such as "high load uncertainty, high volatility, and complex multi-energy coordination." They struggle to quickly generate high-quality dispatch schemes while simultaneously achieving supply-demand balance, operating cost control, and efficient utilization of renewable energy. The main technical bottlenecks are as follows:

[0004] First, the optimization models are highly complex and struggle to adapt to the dynamic re-optimization requirements within the rolling horizontal optimization (RHO) domain. Most current scheduling models employ a centralized scheduling architecture, simultaneously considering multiple power sources such as photovoltaics (PV), wind power (WT), diesel generators (DG), and energy storage systems (ESS), as well as various variables including load forecasting, power balance, and state constraints. They also encompass temporal coupling, multi-energy flow coordination, and network constraints. These models are often constructed as mixed-integer linear programming (MILP) or nonlinear programming (NLP) forms. In the context of multiple time periods, multiple devices, and multiple objectives, the number of decision variables and constraints rapidly expands, leading to a significant increase in solution time. This is particularly problematic under the rolling horizontal optimization (RHO) framework, making it difficult to respond in real-time to changing operating conditions and reducing the model's engineering practicality.

[0005] Secondly, the lack of effective hierarchical modeling and coupling / decoupling strategies easily leads to problems such as the "curse of dimensionality" and "computational bottlenecks." In typical independently operating microgrids in commercial complexes, there are complex operational logics and nonlinear constraints among various distributed energy resources (DERs), such as the start-stop control of diesel or natural gas combined heat and power equipment, the state of charge (SOC) limitations of energy storage systems, and the uncontrollability of photovoltaic output. Traditional single-layer centralized modeling methods fail to fully consider the heterogeneity among different energy units in terms of control accuracy, response delay, and scheduling cycle, resulting in excessively high overall system coupling and rapid expansion of the optimization problem dimension, which seriously affects the feasibility and convergence efficiency of the global solution. In addition, current methods generally lack boundary coupling handling mechanisms for key scheduling nodes, making it difficult to effectively coordinate the synchronous updates of state variables between energy storage, power sources, and loads in actual urban operation scenarios, further limiting the model's ability to be implemented and promoted in commercial complex microgrids.

[0006] Furthermore, the algorithm design lacks a coordinated control mechanism for real-time performance and robustness. In systems with significant uncertainties, existing scheduling strategies often rely on historical data for static modeling, making it difficult to effectively address forecast errors in renewable energy output or load fluctuations. On the one hand, while precise algorithms based on MILP or dynamic programming (DP) offer high accuracy, they are slow to respond to changes in system state and cannot output feasible solutions in a timely manner within the scheduling window. On the other hand, while heuristic algorithms based on particle swarm optimization (PSO) and genetic algorithms (GA) can provide fast approximate solutions, the lack of constraint feasibility guarantees leads to poor stability and low operability, failing to meet the safety and reliability requirements of system scheduling. Summary of the Invention

[0007] The purpose of this invention is to propose a power distribution optimization scheduling method and system suitable for microgrids in commercial complexes. This application introduces a modular, layered structure in modeling, dividing the overall system into a main power generation scheduling layer (DieselGenerator and ESS Dispatch Layer), a renewable output prediction and matching layer, and an energy storage flexibility layer. In the solution process, a collaborative mechanism of "mathematical optimization + reinforcement learning (RL)" is adopted. For the deterministic part, a MILP model is used for precise optimization, while for the uncertain part, a reinforcement learning algorithm is used for online generation, achieving rapid iterative updates and feedback control of the scheduling results. Simultaneously, a boundary coupling coordination algorithm ensures the synchronous update of state variables among the sub-models and the global feasibility of the scheduling solution, thereby improving the response speed, solution quality, and system robustness of the entire scheduling system.

[0008] To achieve the objectives of this invention, the technical solution provided by this invention is as follows:

[0009] First aspect

[0010] This invention provides a power distribution optimization scheduling method suitable for microgrids in commercial complexes, comprising the following steps:

[0011] Step S1: Establish a hierarchical decoupled microgrid scheduling model;

[0012] Step S2: Decompose the microgrid dispatch model into three sub-models and solve them in parallel using a hybrid optimization algorithm. Use a boundary coupling coordination algorithm to perform global feasibility verification and synchronization on the solution results to obtain the dispatch plan. The three sub-models include a renewable energy prediction and response layer, a main power dispatch layer, and an energy storage flexible control layer.

[0013] Step S3: Based on the rolling time-domain optimization mechanism, perform closed-loop execution and dynamic feedback updates on the scheduling plan generated in step S2, and output the final control instruction set after coordination and verification.

[0014] Second aspect

[0015] Corresponding to the above method, the present invention also provides a power distribution optimization and scheduling system suitable for microgrids in commercial complexes, comprising the following units: a model establishment unit, a scheduling plan acquisition unit, and a final control instruction set output unit;

[0016] The model building unit is used to build a hierarchical and decoupled microgrid scheduling model;

[0017] The scheduling plan obtaining unit is used to decompose the microgrid scheduling model into three sub-models, and solve them in parallel using a hybrid optimization algorithm. The solution results are then used to perform global feasibility verification and synchronization using a boundary coupling coordination algorithm to obtain the scheduling plan. The three sub-models include a renewable energy prediction and response layer, a main power supply scheduling layer, and an energy storage flexible control layer.

[0018] The final control instruction set output unit is used to perform closed-loop execution and dynamic feedback update of the scheduling plan generated by the scheduling plan obtaining unit based on the rolling time domain optimization mechanism, and output the final control instruction set after coordination and verification.

[0019] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0020] This invention proposes a distribution optimization scheduling solution method suitable for microgrids in commercial complexes. Addressing the high uncertainty and volatility of energy loads in these microgrids, a hierarchical decoupled scheduling model is constructed, and a hybrid solution framework combining reinforcement learning and optimization algorithms is introduced. This method divides the overall problem into three sub-layers: renewable energy prediction and response, main power supply scheduling, and energy storage regulation. These are solved using mixed integer programming, probabilistic modeling, and reinforcement learning, respectively, effectively reducing computational complexity. A global coordination mechanism integrates the results from each sub-module, enabling real-time scheduling updates in the rolling time domain, thus improving the system's economy, robustness, and response speed. This method is applicable to various urban commercial complex microgrid system scenarios. Attached Figure Description

[0021] Figure 1 A schematic diagram of a power distribution optimization scheduling method applicable to microgrids in commercial complexes is provided as an embodiment of the present invention;

[0022] Figure 2 This is a schematic diagram comparing the solution time in a comparative experiment between the method of this invention and the traditional method;

[0023] Figure 3 This is a schematic diagram comparing the supply and demand imbalance rate in a comparative experiment using the method of this invention and the traditional method.

[0024] Figure 4 This is a schematic diagram comparing the fluctuation range of the objective function in a comparative experiment using the method of this invention and a traditional method;

[0025] Figure 5 This is a schematic diagram comparing the stability of energy storage SOC in a comparative experiment using the method of this invention and a traditional method. Detailed Implementation

[0026] 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 of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0027] It should be noted that this application is particularly applicable to the optimization problems of independently operating microgrids in urban commercial complexes in terms of power supply coordination, load forecasting, and energy storage dispatch.

[0028] like Figure 1 As shown, this invention provides a power distribution optimization scheduling method suitable for microgrids in commercial complexes, comprising the following steps:

[0029] Step S1: Establish a hierarchical decoupled microgrid scheduling model;

[0030] Specifically, step S1 includes the following:

[0031] Step S1.1: Define the decision variables for the microgrid dispatch model, including the following:

[0032] Diesel generator at time ; output power; Energy storage systems at all times The charging power; Energy storage systems at all times The discharge power; The state of charge of the energy storage system at time t;

[0033] Step S1.2: Define the parameters for each level of the microgrid dispatch model, including the following:

[0034] Total number of time periods in the scheduling cycle; : The fuel cost function of a diesel generator at time t; : Charging efficiency of energy storage systems; : Discharge efficiency of the energy storage system; The predicted output power of the photovoltaic system at time t; The predicted output power of the wind power system at time t; Load demand at time t; The total network loss of the microgrid system at time t;

[0035] Step S1.3: Establish the objective function of the microgrid scheduling model, including the following:

[0036] ;

[0037] In the formula: Fuel costs of diesel generators; Energy storage lifespan loss cost; The cost of penalties for abandoning wind and solar power; Costs resulting from supply and demand imbalance.

[0038] Step S1.4: Establish the constraints for the microgrid dispatch model, including the following:

[0039] (1.4.1) Power balance constraints:

[0040] ;

[0041] (1.4.2) Diesel engine output limit:

[0042] ;

[0043] In the formula: Lower limit of diesel generator output power; : Maximum output power of diesel generator;

[0044] (1.4.3) Dynamic updates of energy storage SOC:

[0045] ;

[0046] In the formula: The state of charge of the energy storage system at time t+1; It is a quantity that changes over time;

[0047] (1.4.4) Energy storage safety range constraints:

[0048] ;

[0049] In the formula: : Lower limit of SOC for energy storage systems; : Upper limit of SOC of energy storage system;

[0050] (1.4.5) Charge-discharge mutual exclusion constraint:

[0051] .

[0052] Step S2: Decompose the microgrid dispatch model into three sub-models and solve them in parallel using a hybrid optimization algorithm. Use a boundary coupling coordination algorithm to perform global feasibility verification and synchronization on the solution results to obtain the dispatch plan. The three sub-models include a renewable energy prediction and response layer, a main power dispatch layer, and an energy storage flexible control layer.

[0053] The renewable energy forecasting response layer is an ultra-short-term forecasting layer based on LSTM. This layer is a forecasting model that provides high-precision data support for other layers. The model formula for the renewable energy forecasting response layer is as follows:

[0054] ;

[0055] ;

[0056] ;

[0057] ;

[0058] In the formula: : The input vector at time t; The hidden state at time t; The hidden state at time t-1; , , These are the forget gate, input gate, and output gate, respectively. : The state of t cells at any given time; : Cell state at time t-1; Forget gate weight matrix; Forget gate bias vector; : Candidate cell state at time t; W o : Weight matrix of the output layer; b o : The bias vector of the output layer; σ: The sigmoid activation function; Hadamard product.

[0059] The steps for solving the model formula for the renewable energy prediction response layer include the following:

[0060] Input: Real-time collected historical time-series data;

[0061] Calculation: Input historical time series data into the trained LSTM network and perform forward propagation calculation;

[0062] Output (Solution Results): Predicted photovoltaic ultra-short-term point values ​​for the next 15 minutes to 4 hours. And wind power ultra-short-term point forecast value And the confidence interval of the predicted value.

[0063] The main power supply dispatching layer is a long-term economic dispatching model based on MILP. This layer formulates a daily microgrid dispatching plan based on the long-term economic efficiency of the system, according to the output of the predictive response layer. The model formula of the main power supply dispatching layer includes the following:

[0064] (2.1) Model decision variables: Diesel generator at time ; output power; Energy storage systems at all times The charging power; Energy storage systems at all times The discharge power; : The start-stop state of the diesel generator at time t; SOC(t): The state of charge of the energy storage system at time t;

[0065] (2.2) The objective function of the model is as follows:

[0066] ;

[0067] In the formula: T represents the time period; Diesel fuel cost; Diesel generator operation and maintenance cost coefficient;

[0068] Diesel generator start-up costs; The starting flag of the diesel generator at time t; Power deficit penalty cost coefficient; The power deficit at time t;

[0069] (2.3) Model constraints, including the following:

[0070] (2.3.1) Power balance constraint, the formula is as follows:

[0071] ;

[0072] in: This is the load forecast value; For photovoltaic power prediction; This is the output for wind force prediction.

[0073] (2.3.2) Diesel generator constraints

[0074] Output upper and lower limit constraints:

[0075] ;

[0076] In the formula: : The start / stop status of the diesel generator at time t, 0 or 1;

[0077] Slope rate constraint:

[0078] ;

[0079] ;

[0080] In the formula: , These are the upper and lower limits of the gradeability of the diesel generator; Diesel generator during the period -1 output power;

[0081] Minimum start-stop time constraints:

[0082] ;

[0083] ;

[0084] In the formula: , These represent the minimum start and stop times, respectively. At time t-1, the diesel generator has been running continuously for [time period]. The diesel generator has been continuously stopped for the duration of time t-1. The start / stop status at time t-1 is 0 or 1;

[0085] Startup flag constraint:

[0086] ;

[0087] (2.3.3) Constraints of Energy Storage Systems

[0088] Charge and discharge power limits:

[0089] ;

[0090] ;

[0091] In the formula, This represents the maximum charging power. The charging start flag is set at time t; This represents the maximum discharge power. This is the discharge start flag at time t;

[0092] Charge and discharge mutual exclusion constraint:

[0093] ;

[0094] SOC security range constraints:

[0095] ;

[0096] (2.3.4) Definition of power deficit

[0097]

[0098] .

[0099] In addition, the steps for solving the model formulas of the main power scheduling layer are as follows:

[0100] Inputs: 24-hour load forecast curve, wind and solar forecast curve, current initial SOC value of energy storage, equipment parameters and cost coefficient;

[0101] Modeling: Construct the above objective function and constraints into a MILP model;

[0102] Solution: Use commercial solvers such as Gurobi and CPLEX to solve the problem;

[0103] Output (Solution Results): Diesel generator start-stop schedule and output schedule curves for the entire day, energy storage system's planned charge-discharge curves for the entire day, and energy storage system's reference SOC trajectory for the entire day. .

[0104] The energy storage flexible control layer is a real-time optimization control based on DDPG. This layer models the scheduling problem as a Markov decision process (MDP) and learns the optimal strategy to smooth out fluctuations in real time. The model formula of the energy storage flexible control layer includes the following:

[0105] State space ):

[0106] ;

[0107] In the formula, Information for higher-level planning;

[0108] Action space ):

[0109] ;

[0110] In the formula, This refers to the adjustment amount relative to the upper-level plan;

[0111] Reward function ):

[0112] ;

[0113] in: , , , These are the weighting coefficients for each reward item; It is the power value at time t;

[0114] State transition: determined by the dynamic equations of energy storage SOC in the environment:

[0115] ;

[0116] In the formula: This refers to the rated capacity of the energy storage.

[0117] The solution steps for the model formula of the energy storage flexible control layer are as follows:

[0118] Input: The current state obtained from the environment. ;

[0119] Strategy evaluation: [Regarding the state] Input the trained DDPG algorithm Actor network ;

[0120] Action selection: The Actor network outputs the current optimal action. in For noise;

[0121] Execution and Observation: Executing Actions The environment shifts to a new state and receive a reward ;

[0122] Output (solution result): Real-time power control command for energy storage in the next control cycle. .

[0123] It should be noted that the boundary coupling coordination algorithm is used to ensure the consistency of solutions at each layer and global feasibility.

[0124] The coordination algorithm is as follows:

[0125] Coordination Logic:

[0126] Obtain the actual current SOC of energy storage and the planned main power layer. And prediction after the DDPG layer action is executed. ;

[0127] Check for conflict, specifically as follows: Check if... ;in, For safety margin;

[0128] If a conflict occurs, a coordination signal is generated:

[0129] For the MILP layer: Add constraints

[0130]

[0131] in, Given a correction amount;

[0132] This triggers a re-evaluation;

[0133] For the DDPG layer: Modify its reward function to increase the penalty weight for SOC over-limit terms. ;

[0134] Solution steps:

[0135] Input: Collect the inputs, outputs, and actual system measurements for all layers.

[0136] Calculation: Execute the coordination algorithm.

[0137] Output (solution results):

[0138] A globally feasible set of predictive control instructions is issued to physical devices.

[0139] When a conflict arises in the predictive control instruction set, the output is re-optimized with new boundary conditions.

[0140] Step S3: Based on the rolling time-domain optimization mechanism, perform closed-loop execution and dynamic feedback updates on the scheduling plan generated in step S2, and output the final control instruction set after coordination and verification.

[0141] Step S3 specifically includes the following:

[0142] Step S3.1: Start the rolling optimization window at a fixed period, collect the latest actual status data of the microgrid system (load, wind and solar output, SOC), and trigger the renewable energy forecast response layer to update the ultra-short-term forecast;

[0143] Step S3.2: Using the latest actual state data of the microgrid system as input, re-execute the solution process of step S2: For the main power dispatch layer, re-solve the long-term economic dispatch model based on MILP with the current state as the initial value to generate an updated day-ahead plan; for the energy storage flexible control layer, call the DDPG strategy network to generate real-time control commands according to the updated microgrid system state.

[0144] Step S3.3: Call the boundary coupling coordination algorithm to verify the consistency of the output results of each layer after resolving.

[0145] If a conflict is found, immediately backtrack to the coordination process for readjustment until a feasible solution is obtained;

[0146] Step S3.4: Output the final control command set after coordination and verification, including the near-term output of the diesel generator.

[0147] Commands, real-time charging and discharging power commands of the energy storage system, and wind and solar power integration strategies form a closed-loop control system of optimization, execution, feedback, and re-optimization.

[0148] In addition, the present invention also includes the step of verifying the final control instruction set through a simulation platform, as follows:

[0149] The final control instruction set was verified using a simulation platform (Matlab / Simulink) to ensure power balance.

[0150] Equipment safety is ensured. If wind and solar power fluctuations exceed the predicted range, the reinforcement learning module is triggered to regenerate the energy storage strategy. In extreme scenarios (such as typhoon weather), the system switches to robust mode to prioritize power supply to critical loads.

[0151] To fully evaluate the solution speed, optimization effect, and stability in practical operation of the method proposed in this invention, a set of comparative experiments were designed. Traditional mixed-integer linear programming (MILP), the ROME method based on robust optimization, and the method proposed in this invention were compared and analyzed. The experiments focused on the optimization problem size, system robustness under renewable energy fluctuations, objective function stability, and energy storage system operational safety, comprehensively verifying the performance of the proposed method under multiple dimensions.

[0152] Figure 2 The solution time performance of the three methods is shown under different optimization problem sizes. As the problem size gradually increases, the solution time of the MILP and ROME methods increases exponentially, quickly approaching or reaching the set maximum solution time (approximately 10^6 seconds). 4 (seconds). In contrast, the method of this invention has significant advantages for small-scale problems, with solution time remaining at a low level, and it quickly stabilizes even for large-scale problems, indicating that it has significant advantages in computational efficiency and scalability, and has the potential to handle larger-scale problems.

[0153] Figure 3 The comparison of supply and demand imbalance rates (wind and solar fluctuations greater than 30%) shows that, considering the highly volatile nature of renewable energy, the three methods were compared at uncertainty levels of ±0%, ±15%, and ±30%. The results show that the method of this invention has significantly lower supply and demand imbalance rates than MILP and ROME at all fluctuation levels, demonstrating stronger robustness and system coordination capabilities.

[0154] Figure 4 The graph shows a comparison of the fluctuation range of the objective function, reflecting the stability of the objective function obtained by the three methods. MILP and ROME exhibit large fluctuations in their objective functions in different experiments, while the results of this invention show the smallest fluctuation range, indicating that it possesses stronger optimization stability and can maintain high operational consistency when facing disturbances or uncertainties.

[0155] like Figure 5As shown in the comparison of energy storage SOC stability, the intraday variation curves of the state of charge (SOC) of the energy storage system under the three methods are compared. The SOC curve under the control of the present invention fluctuates more smoothly and always remains within the safe operating range, while the SOC under the MILP and ROME schemes has larger fluctuations and even fluctuations outside the critical range. This shows that the method of the present invention is more reliable and has more practical application value in terms of system dynamic control.

[0156] In addition, corresponding to the above method, the present invention also provides a power distribution optimization and scheduling system suitable for microgrids in commercial complexes, including the following units: a model establishment unit, a scheduling plan acquisition unit, and a final control instruction set output unit;

[0157] The model building unit is used to build a hierarchical and decoupled microgrid scheduling model;

[0158] The scheduling plan obtaining unit is used to decompose the microgrid scheduling model into three sub-models, and solve them in parallel using a hybrid optimization algorithm. The solution results are then used to perform global feasibility verification and synchronization using a boundary coupling coordination algorithm to obtain the scheduling plan. The three sub-models include a renewable energy prediction and response layer, a main power supply scheduling layer, and an energy storage flexible control layer.

[0159] The final control instruction set output unit is used to perform closed-loop execution and dynamic feedback update of the scheduling plan generated by the scheduling plan obtaining unit based on the rolling time domain optimization mechanism, and output the final control instruction set after coordination and verification.

[0160] Specifically, the model building unit is used to perform the following:

[0161] Step S1.1: Define the decision variables for the microgrid dispatch model, including the following:

[0162] Diesel generator at time ; output power; Energy storage systems at all times The charging power; Energy storage systems at all times The discharge power; The state of charge of the energy storage system at time t;

[0163] Step S1.2: Define the parameters for each level of the microgrid dispatch model, including the following:

[0164] Total number of time periods in the scheduling cycle; : The fuel cost function of a diesel generator at time t; : Charging efficiency of energy storage systems; : Discharge efficiency of the energy storage system; The predicted output power of the photovoltaic system at time t; The predicted output power of the wind power system at time t; Load demand at time t; The total network loss of the microgrid system at time t;

[0165] Step S1.3: Establish the objective function of the microgrid scheduling model, including the following:

[0166] ;

[0167] In the formula: Fuel costs of diesel generators; Energy storage lifespan loss cost; The cost of penalties for abandoning wind and solar power; Costs resulting from supply and demand imbalance.

[0168] Step S1.4: Establish the constraints for the microgrid dispatch model, including the following:

[0169] (1.4.1) Power balance constraints:

[0170] ;

[0171] (1.4.2) Diesel engine output limit:

[0172] ;

[0173] In the formula: Lower limit of diesel generator output power; : Maximum output power of diesel generator;

[0174] (1.4.3) Dynamic updates of energy storage SOC:

[0175] ;

[0176] In the formula: The state of charge of the energy storage system at time t+1; It is a quantity that changes over time;

[0177] (1.4.4) Energy storage safety range constraints:

[0178] ;

[0179] In the formula: : Lower limit of SOC for energy storage systems; : Upper limit of SOC of energy storage system;

[0180] (1.4.5) Charge-discharge mutual exclusion constraint:

[0181] .

[0182] The renewable energy prediction response layer is an ultra-short-term prediction based on LSTM, and the model formula for the renewable energy prediction response layer is as follows:

[0183] ;

[0184] ;

[0185] ;

[0186] ;

[0187] In the formula: : The input vector at time t; The hidden state at time t; The hidden state at time t-1; , , These are the forget gate, input gate, and output gate, respectively. : The state of t cells at any given time; : Cell state at time t-1; Forget gate weight matrix; Forget gate bias vector; : Candidate cell state at time t; W o : Weight matrix of the output layer; b o : The bias vector of the output layer; σ: The sigmoid activation function; Hadamard product.

[0188] The main power supply scheduling layer is a long-term economic scheduling model based on MILP, and the model formula of the main power supply scheduling layer includes the following:

[0189] (2.1) Model decision variables: Diesel generator at time ; output power; Energy storage systems at all times The charging power; Energy storage systems at all times The discharge power; : The start-stop state of the diesel generator at time t; SOC(t): The state of charge of the energy storage system at time t;

[0190] (2.2) The objective function of the model is as follows:

[0191] ;

[0192] In the formula: T represents the time period; Diesel fuel cost; Diesel generator operation and maintenance cost coefficient;

[0193] Diesel generator start-up costs; The starting flag of the diesel generator at time t; Power deficit penalty cost coefficient; The power deficit at time t;

[0194] (2.3) Model constraints, including the following:

[0195] (2.3.1) Power balance constraint, the formula is as follows:

[0196] ;

[0197] in: This is the load forecast value; For photovoltaic power prediction; This is the output for wind force prediction.

[0198] (2.3.2) Diesel generator constraints

[0199] Output upper and lower limit constraints:

[0200] ;

[0201] In the formula: : The start / stop status of the diesel generator at time t, 0 or 1;

[0202] Slope rate constraint:

[0203] ;

[0204] ;

[0205] In the formula: , These are the upper and lower limits of the gradeability of the diesel generator; Diesel generator during the period -1 output power;

[0206] Minimum start-stop time constraints:

[0207] ;

[0208] ;

[0209] In the formula: , These represent the minimum start and stop times, respectively. At time t-1, the diesel generator has been running continuously for [time period]. The diesel generator has been continuously stopped for the duration of time t-1. The start / stop status at time t-1 is 0 or 1;

[0210] Startup flag constraint:

[0211] ;

[0212] (2.3.3) Constraints of Energy Storage Systems

[0213] Charge and discharge power limits:

[0214] ;

[0215] ;

[0216] In the formula, This represents the maximum charging power. The charging start flag is set at time t; This represents the maximum discharge power. This is the discharge start flag at time t;

[0217] Charge and discharge mutual exclusion constraint:

[0218] ;

[0219] SOC security range constraints:

[0220] ;

[0221] (2.3.4) Definition of power deficit

[0222]

[0223] .

[0224] Finally, it should be noted that the above embodiments are merely illustrative and explanatory of the present invention, and are not intended to limit the present invention to the scope of the described embodiments. Furthermore, those skilled in the art will understand that the present invention is not limited to the above embodiments, and many more variations and modifications can be made based on the teachings of the present invention, all of which fall within the scope of protection claimed by the present invention.

Claims

1. A power distribution optimization scheduling method suitable for microgrids in commercial complexes, characterized in that, Includes the following steps: Step S1: Establish a hierarchical decoupled microgrid scheduling model; Step S2: Decompose the microgrid dispatch model into three sub-models and solve them in parallel using a hybrid optimization algorithm. Use a boundary coupling coordination algorithm to perform global feasibility verification and synchronization on the solution results to obtain the dispatch plan. The three sub-models include a renewable energy prediction and response layer, a main power dispatch layer, and an energy storage flexible control layer. Step S3: Based on the rolling time-domain optimization mechanism, perform closed-loop execution and dynamic feedback update on the scheduling plan generated in step S2, and output the final control instruction set after coordination and verification; In step S2, the renewable energy prediction response layer is an ultra-short-term prediction based on LSTM. The model formula for the renewable energy prediction response layer is as follows: ； ； ； ； In the formula: : The input vector at time t; The hidden state at time t; The hidden state at time t-1; , , These are the forget gate, input gate, and output gate, respectively. : The state of t cells at any given time; : Cell state at time t-1; Forget gate weight matrix; Forget gate bias vector; : Candidate cell state at time t; W o : Weight matrix of the output layer; b o : The bias vector of the output layer; σ: The sigmoid activation function; Hadamard product; In step S2, the main power supply scheduling layer is a long-term economic scheduling model based on MILP, and the model formula of the main power supply scheduling layer includes the following: (2.1) Model decision variables: Diesel generator at time ; output power; Energy storage systems at all times The charging power; Energy storage systems at all times The discharge power; : The start-stop state of the diesel generator at time t; SOC(t): The state of charge of the energy storage system at time t; (2.2) The objective function of the model is as follows: ； In the formula: T represents the time period; Diesel fuel cost; Diesel generator operation and maintenance cost coefficient; Diesel generator start-up costs; The starting state of the diesel generator at time t; Power deficit penalty cost coefficient; Power deficit at time t.

2. The power distribution optimization and scheduling method for microgrids in commercial complexes according to claim 1, characterized in that, Step S1 specifically includes the following: Step S1.1: Define the decision variables for the microgrid dispatch model, including the following: Diesel generator at time ; output power; Energy storage systems at all times The charging power; Energy storage systems at all times The discharge power; The state of charge of the energy storage system at time t; Step S1.2: Define the parameters for each level of the microgrid dispatch model, including the following: Total number of time periods in the scheduling cycle; : The fuel cost function of a diesel generator at time t; : Charging efficiency of energy storage systems; : Discharge efficiency of the energy storage system; The predicted output power of the photovoltaic system at time t; The predicted output power of the wind power system at time t; Load demand at time t; The total network loss of the microgrid system at time t; Step S1.3: Establish the objective function of the microgrid scheduling model, including the following: ； In the formula: Fuel costs of diesel generators; Energy storage lifespan loss cost; The cost of penalties for abandoning wind and solar power; Supply and demand imbalance costs; Step S1.4: Establish the constraints for the microgrid dispatch model, including the following: (1.4.1) Power balance constraints: ； (1.4.2) Diesel engine output limit: ； In the formula: Lower limit of diesel generator output power; : Maximum output power of diesel generator; (1.4.3) Dynamic updates of energy storage SOC: ； In the formula: The state of charge of the energy storage system at time t+1; It is a quantity that changes over time; (1.4.4) Energy storage safety range constraints: ； In the formula: : Lower limit of SOC for energy storage systems; : Upper limit of SOC of energy storage system; (1.4.5) Charge-discharge mutual exclusion constraint: 。 3. The power distribution optimization and scheduling method for microgrids in commercial complexes according to claim 2, characterized in that, In step S2, the main power supply scheduling layer is a long-term economic scheduling model based on MILP, and the model formula of the main power supply scheduling layer also includes the following: (2.3) Model constraints, including the following: (2.3.1) Power balance constraint, the formula is as follows: ； in: This is the load forecast value; For photovoltaic power prediction; This is the output for wind force prediction. (2.3.2) Diesel generator constraints Output upper and lower limit constraints: ； In the formula: : The start / stop status of the diesel generator at time t, 0 or 1; Slope rate constraint: ； ； In the formula: , These are the upper and lower limits of the gradeability of the diesel generator; Diesel generator during the period -1 output power; Minimum start-stop time constraints: ； ； In the formula: , These represent the minimum start and stop times, respectively. At time t-1, the diesel generator has been running continuously for [time period]. The diesel generator has been continuously stopped for the duration of time t-1. The start / stop status at time t-1 is 0 or 1; Startup flag constraint: ； (2.3.3) Constraints of Energy Storage Systems Charge and discharge power limits: ； ； In the formula, This represents the maximum charging power. The charging start flag is set at time t; This represents the maximum discharge power. This is the discharge start flag at time t; Charge and discharge mutual exclusion constraint: ； SOC security range constraints: ； (2.3.4) Definition of power deficit ； 。 4. The power distribution optimization and scheduling method for microgrids in commercial complexes according to claim 3, characterized in that, In step S2, the energy storage flexible control layer is a real-time optimized control based on DDPG, and the model formula of the energy storage flexible control layer includes the following: State space ): ； In the formula, Information for higher-level planning; Action space ): ； In the formula, This refers to the adjustment amount relative to the upper-level plan; Reward function ): ； in: , , , These are the weighting coefficients for each reward item; It is the power value at time t; State transition: determined by the dynamic equations of energy storage SOC in the environment: ； In the formula: This refers to the rated capacity of the energy storage.

5. The power distribution optimization and scheduling method for microgrids in commercial complexes according to claim 4, characterized in that, Step S3 specifically includes the following: Step S3.1: Start the rolling optimization window at a fixed period, collect the latest actual state data of the microgrid system, and trigger the renewable energy prediction response layer to update the ultra-short-term forecast; Step S3.2: Using the latest actual state data of the microgrid system as input, re-execute the solution process of step S2: For the main power dispatch layer, re-solve the long-term economic dispatch model based on MILP with the current state as the initial value to generate an updated day-ahead plan; for the energy storage flexible control layer, call the DDPG strategy network to generate real-time control commands according to the updated microgrid system state. Step S3.3: Call the boundary coupling coordination algorithm to verify the consistency of the output results of each layer after resolving. sex; If a conflict is found, the process is immediately reverted back to the coordination process for readjustment until a feasible solution is found. Step S3.4: Output the final control command set after coordination and verification, including the near-term output of the diesel generator. Commands, real-time charging and discharging power commands of the energy storage system, and wind and solar power integration strategies form a closed-loop control system of optimization, execution, feedback, and re-optimization.

6. A power distribution optimization dispatching system suitable for microgrids in commercial complexes, wherein the system is applied in the power distribution optimization dispatching method for microgrids in commercial complexes as described in any one of claims 1-5, characterized in that, It includes the following units: model building unit, scheduling plan acquisition unit, and final control instruction set output unit; The model building unit is used to build a hierarchical and decoupled microgrid scheduling model; The scheduling plan obtaining unit is used to decompose the microgrid scheduling model into three sub-models, and solve them in parallel using a hybrid optimization algorithm. The solution results are then used to perform global feasibility verification and synchronization using a boundary coupling coordination algorithm to obtain the scheduling plan. The three sub-models include a renewable energy prediction and response layer, a main power supply scheduling layer, and an energy storage flexible control layer. The final control instruction set output unit is used to perform closed-loop execution and dynamic feedback update of the scheduling plan generated by the scheduling plan obtaining unit based on the rolling time domain optimization mechanism, and output the final control instruction set after coordination and verification.

7. A power distribution optimization and dispatching system suitable for microgrids in commercial complexes according to claim 6, characterized in that, The model building unit is specifically used to perform the following: Step S1.1: Define the decision variables for the microgrid dispatch model, including the following: Diesel generator at time ; output power; Energy storage systems at all times The charging power; Energy storage systems at all times The discharge power; The state of charge of the energy storage system at time t; Step S1.2: Define the parameters for each level of the microgrid dispatch model, including the following: Total number of time periods in the scheduling cycle; : The fuel cost function of a diesel generator at time t; : Charging efficiency of energy storage systems; : Discharge efficiency of the energy storage system; The predicted output power of the photovoltaic system at time t; The predicted output power of the wind power system at time t; Load demand at time t; The total network loss of the microgrid system at time t; Step S1.3: Establish the objective function of the microgrid scheduling model, including the following: ； In the formula: Fuel costs of diesel generators; Energy storage lifespan loss cost; The cost of penalties for abandoning wind and solar power; Supply and demand imbalance costs; Step S1.4: Establish the constraints for the microgrid dispatch model, including the following: (1.4.1) Power balance constraints: ； (1.4.2) Diesel engine output limit: ； In the formula: Lower limit of diesel generator output power; : Maximum output power of diesel generator; (1.4.3) Dynamic updates of energy storage SOC: ； In the formula: The state of charge of the energy storage system at time t+1; It is a quantity that changes over time; (1.4.4) Energy storage safety range constraints: ； In the formula: : Lower limit of SOC for energy storage systems; : Upper limit of SOC of energy storage system; (1.4.5) Charge-discharge mutual exclusion constraint: 。 8. A power distribution optimization and dispatching system suitable for microgrids in commercial complexes according to claim 7, characterized in that, The renewable energy forecasting response layer is an ultra-short-term forecast based on LSTM. The model formula for the renewable energy forecasting response layer is as follows: ； ； ； ； In the formula: : The input vector at time t; The hidden state at time t; The hidden state at time t-1; , , These are the forget gate, input gate, and output gate, respectively. : The state of t cells at any given time; : Cell state at time t-1; Forget gate weight matrix; Forget gate bias vector; : Candidate cell state at time t; W o : Weight matrix of the output layer; b o : The bias vector of the output layer; σ: Sigmoid activation function; Hadamard product.

9. A power distribution optimization and dispatching system suitable for microgrids in commercial complexes according to claim 8, characterized in that, The main power supply scheduling layer is based on the MILP long-term economic scheduling model, and the model formula of the main power supply scheduling layer includes the following: (2.1) Model decision variables: Diesel generator at time ; output power; Energy storage systems at all times The charging power; Energy storage systems at all times The discharge power; : The start-stop state of the diesel generator at time t; SOC(t): The state of charge of the energy storage system at time t; (2.2) The objective function of the model is as follows: ； In the formula: T represents the time period; Diesel fuel cost; Diesel generator operation and maintenance cost coefficient; Diesel generator start-up costs; The starting state of the diesel generator at time t; Power deficit penalty cost coefficient; The power deficit at time t; (2.3) Model constraints, including the following: (2.3.1) Power balance constraint, the formula is as follows: ； in: This is the load forecast value; For photovoltaic power prediction; This is the output for wind force prediction. (2.3.2) Diesel generator constraints Output upper and lower limit constraints: ； In the formula: : The start / stop status of the diesel generator at time t, 0 or 1; Slope rate constraint: ； ； In the formula: , These are the upper and lower limits of the gradeability of the diesel generator; Diesel generator during the period -1 output power; Minimum start-stop time constraints: ； ； In the formula: , These represent the minimum start and stop times, respectively. At time t-1, the diesel generator has been running continuously for [time period]. The diesel generator has been continuously stopped for the duration of time t-1. The start / stop status at time t-1 is 0 or 1; Startup flag constraint: ； (2.3.3) Constraints of Energy Storage Systems Charge and discharge power limits: ； ； In the formula, This represents the maximum charging power. The charging start flag is set at time t; This represents the maximum discharge power. This is the discharge start flag at time t; Charge and discharge mutual exclusion constraint: ； SOC security range constraints: ； (2.3.4) Definition of power deficit ； 。

Images