Multi-target dispatching method and system for alternating-current and direct-current hybrid micro-grid based on digital twinning

Abstract

The invention provides an AC / DC hybrid microgrid multi-target scheduling method and system based on digital twinning, and relates to the technical field of microgrid scheduling, and the method comprises the following steps: building an AC / DC hybrid microgrid digital twinning model, and dynamically integrating the related data of a power grid; establishing a comprehensive total cost model of the small AC / DC hybrid micro-grid system; according to the method, three power generation modes of photovoltaic power generation, photo-thermal power generation and wind power generation and an energy storage device are combined, and an equipment dynamic efficiency model is established based on the influence of factors such as a local load and an external large power grid environment on the efficiency of AC / DC hybrid micro-grid equipment; various twin data such as a local load and an external large power grid environment are obtained through a digital twin model, a multi-target particle swarm algorithm based on an improved archive maintenance strategy and a fuzzy decision method is proposed with economic benefits and the health degree of the energy storage device as targets, the output condition of the energy storage device is dispatched and optimized in real time, and the output efficiency of the energy storage device is improved. And the multi-energy complementation and conversion capability of the system is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of microgrid scheduling, and particularly to a multi-objective scheduling method and system for AC-DC hybrid microgrids based on digital twins. Background Art

[0002] With the rapid development of renewable energy, AC-DC hybrid microgrids have become an important research direction in the energy system due to their flexibility and high efficiency. However, the following problems exist in the existing technologies:

[0003] Simplified equipment modeling: Traditional AC-DC hybrid microgrid scheduling models usually ignore the real-time impact of equipment dynamic performance and environmental factors (such as wind speed and irradiance), resulting in insufficient scheduling accuracy;

[0004] Lagged information interaction: Existing methods rely on offline data and cannot respond in real time to local meteorological conditions, load fluctuations, and grid electricity price changes, affecting the timeliness of scheduling strategies;

[0005] Insufficient multi-objective optimization: Existing algorithms mostly focus on a single economic objective and ignore the health of energy storage devices (such as the impact of charge-discharge cycles on life), resulting in an increase in the long-term operating cost of the system;

[0006] Limitations in algorithm performance: Traditional multi-objective optimization algorithms (such as NSGA-II and MOPSO) are prone to falling into local optima and cannot effectively balance the convergence and distribution of solutions;

[0007] Therefore, the present invention proposes a multi-objective scheduling method and system for AC-DC hybrid microgrids based on digital twins to solve the problems existing in the existing technologies. Summary of the Invention

[0008] In view of the above problems, the present invention proposes a multi-objective scheduling method and system for AC-DC hybrid microgrids based on digital twins. The method and system perform real-time scheduling optimization on the output of energy storage devices, enhancing the economy and environmental performance of the system and improving the multi-energy complementarity and conversion ability of the system.

[0009] To achieve the objectives of the present invention, the present invention is realized through the following technical solutions: A multi-objective scheduling method for AC-DC hybrid microgrids based on digital twins, comprising the following steps:

[0010] S1: Establish a digital twin model of the AC-DC hybrid microgrid and dynamically integrate grid-related data;

[0011] S2: Establish a comprehensive total cost model for the small AC-DC hybrid microgrid system, taking into account the impact of the charge-discharge state of charge (SOC) cycle interval on life, and using an exponential function to quantify the health;

[0012] S3: Establish the constraint conditions for the AC-DC hybrid microgrid system;

[0013] S4: Based on the established multi-objective optimal scheduling mathematical model of the system, obtain the basic parameters of the system;

[0014] S5: Construct an improved multi-objective particle swarm optimization algorithm, and introduce a density estimation index based on displacement and a fuzzy decision-making method;

[0015] S6: Combine time-of-use electricity price and renewable energy output to dynamically adjust the energy storage charge and discharge strategy;

[0016] S7: Output the optimal scheduling plan and synchronously update the parameters of the digital twin model.

[0017] The further improvement lies in that in the S1, the AC-DC hybrid microgrid digital twin model includes a wind power generation model, a photovoltaic power generation model, a solar thermal power generation model, and an energy storage device model; dynamically integrate grid-related data, including wind speed, irradiance, load demand, and time-of-use electricity price data.

[0018] The further improvement lies in that the wind power generation model is:

[0019]

[0020] where v t is the instantaneous wind speed, v in , v r and v out are the cut-in wind speed, rated wind speed, and cut-out wind speed respectively, and P r is the rated power of the fan;

[0021] The photovoltaic power generation model is:

[0022]

[0023] where P PV is the output power of the photovoltaic panel, G t is the solar irradiance, G STC is the irradiance under standard test conditions, is the temperature coefficient -0.0047, T STC is the surface temperature of the photovoltaic panel under standard test conditions of 25 °C, and P STC is the output power of the photovoltaic panel under standard measurement.

[0024] The further improvement lies in that the solar thermal power generation model is:

[0025] Collector part:

[0026]

[0027] where is the collector power of solar thermal power generation, is the heat collection efficiency of the solar thermal power station, is the mirror field area, and GI is the solar irradiance;

[0028] Power generation part:

[0029]

[0030] Among them, is the output power of solar thermal power generation, is the heat collection power of solar thermal power generation, is the power generation efficiency of solar thermal power generation;

[0031] The model of the energy storage device includes the models of the battery and the heat storage device as:

[0032]

[0033] Among them, SOC (t + 1) and SOC (t) are the state of charge of the battery at times t + 1 and t, and are respectively the charge and discharge powers of the battery, and are respectively the charge and discharge efficiencies of the battery;

[0034]

[0035] Among them, S HS (t + 1) and S HS (t) are the capacities of the heat storage device at times t + 1 and t, and are respectively the heat charging and discharging powers of the heat storage device, and are respectively the heat charging and discharging efficiencies of the heat storage device.

[0036] The further improvement lies in: in the S2, the comprehensive total cost model of the small AC-DC hybrid microgrid system includes the power generation cost, the energy storage operation and maintenance cost, and the grid interaction cost, and at the same time takes into account the influence of the SOC cycle interval on the life based on the state of charge, and uses an exponential function to quantify the health:

[0037]

[0038] Among them, —— Operating cost; Z —— Number of sub - grids; N —— Total number of time periods; s1 —— Operating cost per unit time of a wind power generation unit, yuan / h; s2 —— Operating cost per unit time of a photovoltaic power generation unit, yuan / h; s3 —— Operating cost per unit time of a solar thermal power generation unit, yuan / h; s4 —— Operating cost per unit time of an energy storage device, yuan / h; δ —— Unit time interval, h; c(t) —— Purchase and sale electricity price of the large power grid at time t, yuan / kWh; P load (t) —— Grid power consumption load at time t, kW; f2 —— Change range of the charge state of the energy storage device, %; —— Charge state of the energy storage device at time t; —— Average charge state of the energy storage device, %;

[0039] The further improvement lies in that: in the above S3, the constraint conditions specifically include:

[0040] The operation of the AC - DC hybrid micro - grid system needs to meet the power constraints within the AC - DC hybrid micro - grid:

[0041]

[0042] Among them, —— Power of the i - th sub - grid at time t, kW; —— Photovoltaic power of the i - th sub - grid at time t, kW; —— Wind power generation power of the i - th sub - grid at time t, kW —— Solar thermal power generation power of the i - th sub - grid at time t, kW; —— Power of the energy storage device in the i - th sub - grid at time t, kW; —— Load power of the i - th sub - grid at time t, kW;

[0043] The operation of the AC - DC hybrid micro - grid system needs to meet the power constraints between the AC - DC hybrid micro - grids:

[0044]

[0045] The operation of the AC - DC hybrid micro - grid system needs to meet the charge state constraints of the energy storage device and the charge - discharge power constraints of the energy storage device:

[0046]

[0047] In the formula: —— Charge state of the i - th sub - grid at the initial moment, %; —— Charge state of the i - th sub - grid at the termination moment, %; —— Minimum charge amount of the energy storage device, %; —— Maximum charge amount of the energy storage device, %;

[0048] The operation of the AC-DC hybrid microgrid system needs to satisfy the tie-line power constraint:

[0049]

[0050] In the formula: —— Lower limit value, kW; —— Upper limit value, kW; —— The power of sub-network i at time t, kW;

[0051] The operation of the AC-DC hybrid microgrid system needs to satisfy the external power purchase and sale constraint:

[0052]

[0053] In the formula: —— Upper limit of external power sale, kW; —— Upper limit of external power purchase, kW; —— External power purchase quantity at time t, kW.

[0054] A further improvement lies in that: in S4, based on the system model, objective function and constraint conditions proposed in S1-S3, a mathematical model for multi-objective optimal scheduling of the system is established, initial microgrid load data is obtained, annual meteorological historical data and time-of-use electricity price data are obtained, and basic parameters such as equipment operation efficiency, rated power, and start-stop time are obtained.

[0055] A further improvement lies in that: in S5, an improved multi-objective particle swarm optimization algorithm is designed and the archive maintenance strategy is improved; the displacement-based density estimation ISDE index is introduced to screen non-dominated solutions with both convergence and distribution; a fuzzy decision-making method is introduced to quantify the satisfaction degree of each objective through membership functions and select the optimal scheduling scheme.

[0056] A further improvement lies in that: in S6, the energy storage charge and discharge strategy is as follows:

[0057] When the electricity price is in the peak period, priority is given to selling electricity to earn profits. When the power generation of the AC-DC hybrid microgrid is higher than the local load, the energy storage device discharges as much as possible within the allowable range; when the power generation of the AC-DC hybrid microgrid is lower than the local load, the energy storage device discharges to meet the electricity demand. When the energy storage device cannot meet the demand, power is considered to be purchased from the grid;

[0058] When the electricity price is in the valley period, the power purchase strategy is given priority. Within the range of the charge state of the energy storage device, power is preferentially purchased from the large power grid. At this time, the purchased power simultaneously meets the charging demand of the energy storage device and the local load gap. When the charge state of the energy storage device reaches the upper limit, the energy storage ends. At this time, the purchased power only meets the local load gap. When the electricity price is in the flat stage, the priority is to maintain the charge state of the energy storage device. When the power generation of the AC-DC hybrid microgrid is higher than the local load, the energy storage device is preferentially charged. When the local load exceeds the generating capacity of the AC-DC hybrid microgrid's generating units, the energy storage device fills the gap;

[0059] The multi-objective scheduling system for the AC-DC hybrid microgrid based on digital twin includes a power system and a digital twin platform. The power system is used to provide power. The power system consists of three energy utilization forms: photovoltaic power generation, solar thermal power generation, and wind power generation, as well as energy storage devices;

[0060] The digital twin platform is used for data collection, providing an interaction interface, and building a digital twin model. Based on the multi-objective particle swarm optimization algorithm with an improved archive maintenance strategy and a fuzzy decision-making method, the output situation of the power system is optimized in real-time scheduling.

[0061] The beneficial effects of the present invention are as follows:

[0062] 1. The present invention combines three power generation forms: photovoltaic power generation, solar thermal power generation, and wind power generation, as well as energy storage devices. Based on the impacts of factors such as local load and external large power grid environment on the equipment efficiency of the AC-DC hybrid microgrid, an equipment dynamic efficiency model is established. Through the digital twin model, various twin data such as local load and external large power grid environment are obtained. With economic benefits and the health of the energy storage device as the goals, a multi-objective particle swarm optimization algorithm based on an improved archive maintenance strategy and a fuzzy decision-making method is proposed to optimize the output situation of the energy storage device in real-time scheduling, enhancing the economy and environmental performance of the system and improving the multi-energy complementarity and conversion ability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 It is the structural block diagram of the power system of the present invention;

[0064] Figure 2 It is the structure diagram of the digital twin platform of the present invention;

[0065] Figure 3 It is the schematic diagram of the energy storage charge and discharge strategy of the present invention;

[0066] Figure 4 It is the basic algorithm framework diagram of the MOPSO-IAFD of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0067] To deepen the understanding of the present invention, the following will further elaborate on the present invention in combination with embodiments. These embodiments are only used to explain the present invention and do not constitute a limitation on the protection scope of the present invention.

[0068] Embodiment 1

[0069] According to Figure 1 、 2 、3, 4, the present embodiment proposes a multi-objective scheduling method for an AC-DC hybrid microgrid based on digital twin, including the following steps:

[0070] Build equipment models in the AC-DC hybrid microgrid system, including a wind power generation model, a photovoltaic power generation model, a solar thermal power generation model, and a energy storage device model, specifically including:

[0071] (1) Wind power generation model:

[0072]

[0073] Among them, v t is the instantaneous wind speed, v in , v r and v out are the cut-in wind speed, rated wind speed, and cut-out wind speed respectively, and P r is the rated power of the fan.

[0074] (2) Photovoltaic power generation model:

[0075]

[0076] Among them, P PV is the output power of the photovoltaic panel, G t is the solar irradiance, G STC is the irradiance under standard test conditions, is the temperature coefficient -0.0047, T STC is the surface temperature of the photovoltaic panel under standard test conditions, 25 °C, and P STC is the output power of the photovoltaic panel under standard test.

[0077] (3) The solar thermal power generation model is:

[0078] Collector part:

[0079]

[0080] Among them, is the collector power of solar thermal power generation, is the collector efficiency of the solar thermal power plant, is the mirror field area, and GI is the solar irradiance.

[0081] Power generation part:

[0082]

[0083] Among them, is the output power of solar thermal power generation, is the heat collection power of solar thermal power generation, is the power generation efficiency of solar thermal power generation.

[0084] (4) Energy storage device, including a storage battery and a heat storage device, and the model is:

[0085]

[0086] Among them, SOC (t + 1) and SOC (t) are the state of charge of the storage battery at times t + 1 and t, and are the charge and discharge powers of the storage battery respectively, and are the charge and discharge efficiencies of the storage battery respectively.

[0087]

[0088] Among them, S HS (t + 1) and S HS (t) are the capacities of the heat storage device at times t + 1 and t, and are the heat charge and discharge powers of the heat storage device respectively, and are the heat charge and discharge efficiencies of the heat storage device respectively.

[0089] Establish a comprehensive total cost model for a small AC-DC hybrid microgrid system that includes power generation cost, energy storage operation and maintenance cost, and grid interaction cost. At the same time, considering the impact of the charge state (SOC) cycle interval on the lifespan, an exponential function is used to quantify the health degree.

[0090]

[0091] Among them, —— Operating cost, ; Z—— Number of subnets; N—— Total number of time periods; s1—— Operating cost per unit time of the wind turbine generator, yuan / h; s2—— Operating cost per unit time of the photovoltaic generator set, yuan / h; s3—— Operating cost per unit time of the solar thermal generator set, yuan / h; s4—— Operating cost per unit time of the energy storage device, yuan / h; δ—— Unit time interval, h; c(t)—— Purchase and sale electricity price of the large power grid at time t, yuan / kWh; P load (t)—— Grid power consumption load at time t, kW; f2—— Change range of the charge state of the energy storage device, %. —— State of charge of the energy storage device at time t; —— Average state of charge of the energy storage device, %.

[0092] Establish the constraints of the AC-DC hybrid microgrid system, specifically including:

[0093] (1) The operation of the AC-DC hybrid microgrid system needs to satisfy the power constraints within the AC-DC hybrid microgrid:

[0094]

[0095] Among them: —— Power of the i-th subnet at time t, kW; —— Photovoltaic power of the i-th subnet at time t, kW; —— Wind power generation power of the i-th subnet at time t, kW —— Solar thermal power generation power of the i-th subnet at time t, kW; —— Power of the energy storage device in the i-th subnet at time t, kW; —— Load power of the i-th subnet at time t, kW.

[0096] (2) The operation of the AC-DC hybrid microgrid system needs to satisfy the power constraints between the AC-DC hybrid microgrids:

[0097]

[0098] (3) The operation of the AC-DC hybrid microgrid system needs to satisfy the state-of-charge constraint of the energy storage device and the charge-discharge power constraint of the energy storage device:

[0099]

[0100] In the formula: —— State of charge of the i-th subnet at the initial moment, %; —— State of charge of the i-th subnet at the termination moment, %; —— Minimum charge amount of the energy storage device, %; —— Maximum charge amount of the energy storage device, %.

[0101] (4) The operation of the AC-DC hybrid microgrid system needs to satisfy the tie-line power constraint:

[0102]

[0103] In the formula: —— Lower limit value, kW; —— Upper limit value, kW; —— Power of subnet i at time t, kW.

[0104] (5) The operation of the AC / DC hybrid microgrid system needs to meet the external power purchase and sale constraints:

[0105]

[0106] Where: —— The upper limit of external power sales, kW; —— The upper limit of external power purchases, kW; —— The external power purchase volume at time t, kW.

[0107] Based on the proposed system model, objective function, and constraint conditions, establish a mathematical model for the multi-objective optimal scheduling of the system; obtain the initial load data of the microgrid; obtain the annual meteorological historical data and time-of-use electricity price data; obtain basic parameters such as equipment operation efficiency, rated power, start-stop time, etc.

[0108] Design an improved multi-objective particle swarm optimization algorithm (MOPSO-IAFD): Improve the archive maintenance strategy: Introduce the displacement-based density estimation (ISDE) index to screen non-dominated solutions with both convergence and distribution; Fuzzy decision-making method: Quantify the satisfaction of each objective through membership functions and select the optimal scheduling plan.

[0109] Optimize the energy storage charge and discharge strategy in real time, combine the time-of-use electricity price and renewable energy output, and dynamically adjust the power purchase and sale plan and energy storage power;

[0110] Specifically, when the electricity price is at the peak period, give priority to selling electricity to earn profits. If the power generation of the AC / DC hybrid microgrid is higher than the local load, the energy storage device will discharge as much as possible within the allowable range; if the power generation of the AC / DC hybrid microgrid is lower than the local load, the energy storage device will discharge to meet the electricity demand. When the energy storage device cannot meet the demand, power needs to be purchased from the grid.

[0111] When the electricity price is at the valley period, the power purchase strategy should be given priority. Within the state of charge range of the energy storage device, purchase electricity from the large grid as much as possible. At this time, the power purchase power simultaneously meets the charging demand of the energy storage device and the local load gap. When the state of charge of the energy storage device reaches the upper limit, the energy storage ends. At this time, the power purchase power only meets the local load gap. When the electricity price is at the flat stage, give priority to maintaining the state of charge of the energy storage device. When the power generation of the AC / DC hybrid microgrid is higher than the local load, give priority to charging the energy storage device. If the local load exceeds the generating capacity of the AC / DC hybrid microgrid's generating units, the energy storage device will fill the gap.

[0112] Output the optimal scheduling plan and synchronously update the parameters of the digital twin model.

[0113] The scheduling of AC-DC hybrid microgrids requires determining the charging and discharging powers of energy storage devices in i subgrids over 24 time periods simultaneously, and i×24 decision variables need to be determined. Therefore, a matrix encoding method is adopted, that is:

[0114]

[0115] In the formula: —— The power of the energy storage device in the i-th subgrid at the first time period, kW.

[0116] In the multi-objective particle swarm optimization algorithm, a significant feature is the use of archives to retain elite solutions. The results obtained by such algorithms are often a set containing multiple solutions rather than a single solution. Therefore, an efficient archive management strategy is required to store high-quality optimal solution sets. However, the current archive management method based on the Pareto dominance principle only focuses on whether a solution is not dominated by other solutions. It should be noted that the Pareto dominance relationship only belongs to the category of qualitative evaluation. Whether a solution is in a non-dominated state completely depends on whether its objective values are superior to those of other solutions in the archive, that is, it performs better in at least one objective. For non-dominated solutions, they are regarded as being at the same level in terms of convergence. Nevertheless, the convergence efficiency of particles is also deeply affected by the superiority of their objective values relative to other particles. Consider a scenario in a two-dimensional objective space. Suppose there are 3 non-dominating particles: X(1.0, 0.5), Y(1.2, 0.4), and Z(0.9, 0.6). In this context, particle Z performs significantly better than X and Y in the first objective, thus showing more prominent convergence characteristics. However, according to the Pareto dominance theory, X, Y, and Z are regarded as having the same convergence efficiency, which ignores the actual performance differences among them and fails to fully reflect their true advantages and disadvantages. In view of the above challenges, an archive maintenance strategy based on quantitative evaluation of solution quality is conceived. The first step is to introduce a quantitative evaluation method that can accurately judge based on the superiority of the objective values of solutions compared to other solutions. Subsequently, a step-by-step reduction strategy is designed, the core of which is to gradually remove relatively weak solutions from the candidate solution set until the preset solution set capacity limit is reached. When the archive exceeds the limit, the solutions with the smallest performance evaluation indicators are removed from the archive in turn.

[0117] After obtaining a series of candidate scheduling schemes, this study adopts a strategy based on fuzzy comprehensive evaluation to select the optimal scheduling scheme. This strategy is different from traditional decision-making methods and does not require prior determination of the clear weight distribution among various objectives. First, define the membership function of the i-th objective of the k-th scheduling scheme. The higher the membership function value of the scheduling scheme on this objective, the higher the satisfaction of this objective. Then, for the k-th candidate scheduling scheme, calculate the normalized membership function, and the scheduling scheme with the largest value is the optimal scheduling scheme.

[0118] Example 2

[0119] According to Figure 1 、 2 、Figures 3 and 4, this embodiment proposes a multi-objective scheduling system for AC-DC hybrid microgrids based on digital twins. In the AC-DC hybrid microgrid, each subnet consists of a wind power generation device, a photovoltaic power generation device, an energy storage device, a local load, and an interface converter that connects the bus and each power source. The common DC bus uses an interconnection converter to connect each subnet. Through a static switch, the connection between the common DC bus and the large power grid is realized to achieve power interaction.

[0120] The power system includes a wind turbine generator for converting wind energy into electrical energy; a photovoltaic unit and a solar thermal unit for converting light energy into electrical energy, and an external power grid; an energy storage system, including a storage battery, etc.

[0121] As Figure 2 shown, through the digital twin model, real-time monitoring, data analysis, and optimal scheduling of the physical system can be realized, providing strong support for the operation and management of the microgrid and other physical systems. The data of the real model is transmitted to the digital twin platform, and the digital twin platform will establish a digital model. The digital model performs real-time calculation and analysis, and the obtained results are fed back to the digital twin platform. The conceptual model of the AC-DC hybrid microgrid based on digital twins consists of a physical model and a digital model. The physical model (including various devices and sensors in the AC-DC hybrid microgrid) transmits real-time environment, load data, and device information to the digital model. Real-time optimization of the microgrid is realized in the digital model.

[0122] Example 3

[0123] According to Figure 1 、 2 、Figures 3 and 4, in this embodiment, the improved charge and discharge strategy is as Figure 3 shown. This strategy dynamically adjusts the energy storage operation mode based on the time-of-use electricity price mechanism, specifically as follows:

[0124] (1) Peak period scheduling:

[0125] When the electricity price is in the peak range, give priority to selling electricity to the power grid to obtain benefits;

[0126] If the system power generation exceeds the local load demand, the energy storage device discharges maximally within the power limit;

[0127] If the power generation is insufficient, the energy storage discharges to make up the load gap;

[0128] When the energy storage capacity is exhausted and still cannot meet the demand, start purchasing electricity from the power grid.

[0129] (2) Valley period scheduling:

[0130] During low electricity price periods, give priority to purchasing electricity from the grid to charge the energy storage, while covering the load gap.

[0131] When the state of charge (SOC) of the energy storage reaches the upper limit, stop charging, and the purchased electricity is only used for load power supply.

[0132] (3) Normal period scheduling:

[0133] Aiming to maintain the stability of the energy storage SOC:

[0134] If the system's generated electricity is surplus, give priority to charging the energy storage;

[0135] If the load exceeds the generation capacity, the energy storage discharges to fill the power difference;

[0136] By dynamically balancing the charging and discharging behaviors, reduce the deep cycling of the energy storage and extend the equipment life.

[0137] Technical effects:

[0138] This strategy realizes the collaborative optimization of maximizing economic benefits and the health of the energy storage through real-time matching of electricity price signals with generation-load, while improving the renewable energy consumption rate.

[0139] Embodiment 4

[0140] According to Figure 1 , 2 , 3, and 4, in this embodiment, for the AC-DC hybrid microgrid multi-objective optimization model proposed for optimal solution, a multi-objective particle swarm optimization algorithm based on an improved archive maintenance strategy and fuzzy decision method (multi-objective particle swarm optimization based on improved archive and fuzzy decision, MOPSO-IAFD) is proposed. First, an improved archive maintenance strategy based on the indicator based on shift for density estimation (ISDE) index is proposed, which not only maintains solutions with good convergence, but also keeps the solutions in the archive with good distribution, thus achieving a complete record of the entire Pareto dominance front; on the other hand, a best scheduling scheme selection strategy based on fuzzy decision is introduced to effectively solve the defect that traditional multi-objective particle swarm algorithms can only solve one non-dominated solution set. The basic algorithm framework of MOPSO-IAFD is as Figure 4 shown.

[0141] The scheduling of the AC / DC hybrid microgrid needs to simultaneously determine the charging and discharging powers of the energy storage devices in i sub-grids during 24 time periods, and i×24 decision variables need to be determined. Therefore, a matrix encoding method is adopted, that is:

[0142]

[0143] In the formula: —— The power of the energy storage device in the i-th sub-grid at the first time period, kW.

[0144] In the multi-objective particle swarm optimization algorithm, a significant feature is the use of archives to retain elite solutions. The results obtained by such algorithms are often a set containing multiple solutions rather than a single solution. Therefore, an efficient archive management strategy is needed to store high-quality optimal solution sets. However, the current archive management method based on the Pareto dominance principle only focuses on whether a solution is not dominated by other solutions. It should be noted that the Pareto dominance relationship only belongs to the category of qualitative evaluation. Whether a solution is in a non-dominated state entirely depends on whether its objective values are superior to those of other solutions in the archive, that is, it performs better in at least one objective. For non-dominated solutions, they are regarded as being at the same level in terms of convergence. Nevertheless, the convergence efficiency of particles is also deeply affected by the superiority of their objective values relative to other particles. Consider a scenario in a two-dimensional objective space. Suppose there are 3 non-dominant particles: X(1.0, 0.5), Y(1.2, 0.4), and Z(0.9, 0.6). In this context, particle Z performs significantly better than X and Y in the first objective, thus showing more prominent convergence characteristics. However, according to the Pareto dominance theory, X, Y, and Z are regarded as having the same convergence efficiency, which ignores the actual performance differences among them and fails to fully reflect their true advantages and disadvantages. In view of the above challenges, an archive maintenance strategy based on quantitative evaluation of solution quality is conceived. The first step is to introduce a quantitative evaluation method that can accurately judge based on the superiority of the objective values of the solutions compared to other solutions. Subsequently, a step-by-step reduction strategy is designed, the core of which is to gradually remove relatively weak solutions from the candidate solution set until the preset solution set capacity limit is reached. When the archive exceeds the limit, the solutions with the smallest performance evaluation indicators are removed from the archive in turn.

[0145] After obtaining a series of candidate scheduling schemes, this study adopts a strategy based on fuzzy comprehensive evaluation to select the optimal scheduling scheme. This strategy is different from traditional decision-making methods and does not require prior determination of the clear weight distribution among various objectives. First, define the membership function of the i-th objective of the k-th scheduling scheme. The higher the membership function value of the scheduling scheme on this objective, the higher the satisfaction of this objective. Then, for the k-th candidate scheduling scheme, calculate the normalized membership function, and the scheduling scheme with the largest value is the optimal scheduling scheme.

[0146] Example 5

[0147] According to Figure 1 、 2 As shown in Figures 3 and 4, in this embodiment, the physical model of the AC-DC hybrid microgrid with multiple subnets to be simulated is built on the Cloudpss platform, which includes 2 AC subnets and 3 DC subnets. The parameters of each generator set and energy storage device within the subnets are kept consistent. In this experiment, the time-of-use electricity price mechanism is adopted, and the electricity purchase and sale prices for each period are shown in Table 1. The digital twin technology is used to predict the output of renewable energy generator sets. First, the predicted values of the wind speed, irradiance, and battery surface temperature at each moment of the day are obtained through the physical model built on the Cloudpss platform; then, the power of the wind power, photovoltaic, and solar thermal generator sets is calculated according to the mathematical model of distributed power sources; finally, the daily prediction curve of the generator sets is obtained. Regarding MOPSO-IAFD: the inertia weight ω is set to 3.2, the learning factor c1 is set to 1.8, while the learning factor c2 is 1.6, the population size is 100, and it is executed five times repeatedly. The electricity purchase and sale conditions within each subnet and the state of charge of the storage battery can be obtained.

[0148] Table 1 Time-of-Use Electricity Price Table

[0149]

[0150] The present invention combines three power generation forms of photovoltaic power generation, solar thermal power generation, and wind power generation, as well as energy storage devices. Based on the impacts of factors such as local load and external large power grid environment on the equipment efficiency of the AC-DC hybrid microgrid, an equipment dynamic efficiency model is established. Through the digital twin model, various twin data such as local load and external large power grid environment are obtained. With economic benefits and the health of the energy storage device as the goals, a multi-objective particle swarm optimization algorithm based on an improved archive maintenance strategy and fuzzy decision-making method is proposed to perform real-time scheduling optimization on the output of the energy storage device, enhancing the economy and environmental performance of the system and improving the multi-energy complementarity and conversion ability of the system. Moreover, the present invention comprehensively considers factors such as local meteorological conditions, local load, and external large power grid environment, establishes a digital model of the AC-DC hybrid microgrid, then proposes a multi-objective particle swarm optimization algorithm MOPSO-IAFD based on an improved archive maintenance strategy and fuzzy decision-making method, takes economic benefits and the health of the energy storage device as the optimization goals, performs real-time scheduling optimization on the output of the energy storage device, and finally, uses the Cloudpss platform to generate a large amount of data to construct a digital twin body.

[0151] The basic principles, main features and advantages of the present invention have been shown and described above. Those skilled in the art should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification only illustrates the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A multi-objective scheduling method for AC / DC hybrid microgrids based on digital twins, characterized in that, It includes the following steps: S1: Establish a digital twin model of the AC-DC hybrid microgrid and dynamically integrate relevant grid data; S2: Establish a comprehensive total cost model of the small AC-DC hybrid microgrid system, taking into account the impact of the charge state SOC cycle interval on the lifespan, and using an exponential function to quantify the health status; S3: Establish the constraint conditions of the AC-DC hybrid microgrid system; S4: Based on the above-established mathematical model of multi-objective optimal scheduling of the system, obtain the basic parameters of the system; S5: Construct an improved multi-objective particle swarm optimization algorithm, and introduce a displacement-based density estimation index and a fuzzy decision-making method; S6: Combine time-of-use electricity prices and renewable energy output, and dynamically adjust the energy storage charge and discharge strategy; S7: Output the optimal scheduling plan and synchronously update the parameters of the digital twin model.

2. The multi-objective scheduling method for AC-DC hybrid microgrid based on digital twin according to claim 1, characterized in that: In S1, the digital twin model of the AC-DC hybrid microgrid includes a wind power generation model, a photovoltaic power generation model, a solar thermal power generation model, and an energy storage device model; dynamically integrating relevant grid data, including wind speed, irradiance, load demand, and time-of-use electricity price data.

3. The multi-objective scheduling method for the AC-DC hybrid microgrid based on digital twin according to claim 2, wherein: The wind power generation model is: ， Among them, v t is the instantaneous wind speed, v in , v r and v out are the cut-in wind speed, rated wind speed and cut-out wind speed respectively, and P r is the rated power of the wind turbine; The photovoltaic power generation model is: ， Among them, P PV is the output power of the photovoltaic panel, G t is the solar irradiance, G STC is the irradiance under standard test conditions, is the temperature coefficient -0.0047, T STC is the surface temperature of the photovoltaic panel under standard test conditions of 25 °C, P STC is the output power of the photovoltaic panel under standard measurement.

4. The multi-objective scheduling method for AC-DC hybrid microgrid based on digital twin according to claim 2, wherein: The solar thermal power generation model is: Collector part: ， Among them, is the heat collection power of solar thermal power generation, is the heat collection efficiency of the solar thermal power station, is the mirror field area, and GI is the solar irradiance; Power generation part: ， Among them, is the output power of solar thermal power generation, is the heat collection power of solar thermal power generation, is the power generation efficiency of solar thermal power generation; The energy storage device model, including the models of the battery and the heat storage device, is: ， where SOC (t+1) and SOC (t) are the state of charge of the battery at times t+1 and t, and are the charge and discharge powers of the battery respectively, and are the charge and discharge efficiencies of the battery respectively; ， Among them, S HS (t + 1) and S HS (t) are the capacities of the heat storage device at times t + 1 and t, and are the charging and discharging powers of the heat storage device respectively, and are the charging and discharging efficiencies of the heat storage device respectively.

5. The multi-objective scheduling method for AC-DC hybrid microgrid based on digital twin according to claim 1, characterized in that: In S2, the comprehensive total cost model of the small AC-DC hybrid microgrid system includes power generation costs, energy storage operation and maintenance costs, and grid interaction costs, while taking into account the impact of the charge state SOC cycle interval on the lifespan, and using an exponential function to quantify the health status: ， Among them, —— Operating cost; Z —— Number of subnets; N —— Total number of time periods; s1 —— Operating cost per unit time of wind power generation unit, yuan / h; s2 —— Operating cost per unit time of photovoltaic power generation unit, yuan / h; s3 —— Operating cost per unit time of solar thermal power generation unit, yuan / h; s4 —— Operating cost per unit time of energy storage device, yuan / h; δ —— Unit time interval, h; c(t) —— Purchase and sale electricity price of large power grid at time t, yuan / kWh; P load (t) —— Grid power consumption load at time t, kW; f2 —— Change range of charge state of energy storage device, %; —— Charge state of energy storage device at time t; —— Average charge state of energy storage device, %.

6. The multi-objective scheduling method for AC-DC hybrid microgrid based on digital twin according to claim 1, wherein: In S3, the constraint conditions specifically include: The operation of the AC-DC hybrid microgrid system needs to satisfy the power constraints within the AC-DC hybrid microgrid: ， Among them, —— The power of the i-th subnet at time t, kW; —— The photovoltaic power of the i-th subnet at time t, kW; —— The wind power generation power of the i-th subnet at time t, kW —— The solar thermal power generation power of the i-th subnet at time t, kW; —— The power of the energy storage device in the i-th subnet at time t, kW; —— The load power of the i-th subnet at time t, kW; The operation of the AC-DC hybrid microgrid system needs to satisfy the power constraints between the AC-DC hybrid microgrids: ， The operation of the AC-DC hybrid microgrid system needs to satisfy the charge state constraints of the energy storage device and the charge and discharge power constraints of the energy storage device: ， Wherein: —— The charge state of the i-th subnet at the initial moment, %; —— The charge state of the i-th subnet at the termination moment, %; —— The minimum charge amount of the energy storage device, %; —— The maximum charge amount of the energy storage device, %; The operation of the AC-DC hybrid microgrid system needs to satisfy the tie-line power constraints: ， In the formula: —— Lower limit value, kW; —— Upper limit value, kW; —— Power of subnet i at time t, kW; The operation of the AC-DC hybrid microgrid system needs to satisfy the external power purchase and sale constraints: ， In the formula: —— External power selling upper limit, kW; —— External power purchase upper limit, kW; —— External power purchase quantity at time t, kW.

7. The multi-objective scheduling method for AC-DC hybrid microgrid based on digital twin according to claim 1, wherein: In S4, based on the system model, objective function, and constraint conditions proposed in S1-S3, establish a mathematical model of multi-objective optimal scheduling of the system, obtain the initial load data of the microgrid, obtain the annual meteorological historical data and time-of-use electricity price data, and obtain the basic parameters such as the operation efficiency, rated power, and start-stop time of the equipment.

8. The multi-objective scheduling method for AC-DC hybrid microgrid based on digital twin according to claim 1, characterized in that: In S5, design an improved multi-objective particle swarm optimization algorithm and improve the archive maintenance strategy; introduce a displacement-based density estimation ISDE index to screen non-dominated solutions with both convergence and distribution; Introduce a fuzzy decision-making method, quantify the satisfaction degree of each objective through a membership function, and select the optimal scheduling plan.

9. The multi-objective scheduling method for AC-DC hybrid microgrid based on digital twin according to claim 1, wherein: In S6, the energy storage charge and discharge strategy is: When the electricity price is at the peak period, give priority to selling electricity to earn income. When the power generation of the AC-DC hybrid microgrid is higher than the local load, the energy storage device discharges as much as possible within the allowable range; when the power generation of the AC-DC hybrid microgrid is lower than the local load, the energy storage device discharges to meet the electricity demand. When the energy storage device cannot meet the demand, consider purchasing electricity from the grid; When the electricity price is in the valley period, the power purchase strategy is given priority. Within the range of the charge state of the energy storage device, power is preferentially purchased from the large power grid. At this time, the purchased power simultaneously meets the charging demand of the energy storage device and the local load gap. When the charge state of the energy storage device reaches the upper limit, the energy storage ends. At this time, the purchased power only meets the local load gap. When the electricity price is in the flat stage, priority is given to maintaining the charge state of the energy storage device. When the power generation of the AC-DC hybrid microgrid is higher than the local load, the energy storage device is preferentially charged. When the local load exceeds the generating capacity of the AC-DC hybrid microgrid, the energy storage device fills the gap.

10. A multi-objective scheduling system for an AC-DC hybrid microgrid based on digital twin, which is applied to the multi-objective scheduling method for an AC-DC hybrid microgrid based on digital twin according to any one of claims 1-9 above, and is characterized in that: It includes a power system and a digital twin platform. The power system is used to provide power and is composed of three energy utilization forms: photovoltaic power generation, solar thermal power generation, and wind power generation, as well as energy storage devices. The digital twin platform is used for data collection, providing an interaction interface, and constructing a digital twin model. Based on the multi-objective particle swarm optimization algorithm of the improved file maintenance strategy and fuzzy decision-making method, the output situation of the power system is optimized in real-time scheduling.

Images