Micro-grid planning method and system based on improved normal boundary intersection method

Abstract

The application discloses a micro-grid planning method and system based on an improved normal boundary intersection method, and the method comprises the following steps: constructing a target function of a micro-grid planning model considering ecological, climate and human settlement influences; constructing constraint conditions of the micro-grid planning model considering environmental benefit constraints; constructing a compact form of the micro-grid planning model; based on the improved normal boundary intersection method of Euclidean distance correction, the compact form of the micro-grid planning model is solved to obtain a Pareto frontier; and an optimal compromise solution is selected from the Pareto frontier to obtain a micro-grid expansion planning scheme. The application comprehensively considers ecological, climate and human settlement influences, quantitatively models environmental benefits, and improves the environmental adaptability of the micro-grid planning. In view of the nonlinearity of the model caused by the modeling of the environmental benefits, the improved normal boundary intersection method based on the Euclidean distance correction is designed to solve the problem, the uniformity of the solution set of the Pareto frontier is improved, the solution set is prevented from being gathered or having a vacancy area, and a candidate scheme set with higher coverage is provided for decision makers.

Description

Technical Field

[0001] This invention belongs to the field of power systems and relates to microgrid planning technology, specifically to a microgrid planning method and system based on an improved normal boundary intersection method. Background Technology

[0002] Microgrids, as an important vehicle for achieving a green and low-carbon energy transition, are being widely promoted and applied. Microgrids integrate various power sources, load types, and energy storage devices, playing a crucial role in enhancing power supply reliability and promoting the consumption of renewable energy. Therefore, constructing optimal planning models for nano-microgrid clusters has become a key research direction.

[0003] In planning models, common planning objectives include improving the economic efficiency, environmental benefits, and reliability of power systems. Dufo et al., based on historical information on wind speed, solar irradiance, and load, established a two-objective optimization problem considering the minimization of costs and emissions for stand-alone wind-solar-diesel-storage (SLS) systems. Diab et al., for stand-alone SLS microgrids, established optimal configuration models for SLS microgrids based on different control strategies. Khalilpour et al., for grid-connected SLS microgrids, studied the multi-period planning problem for photovoltaic (PV) and energy storage based on annual PV and load curves. Hoppmann et al., for grid-connected SLS microgrids, used annual operation simulation to conduct a techno-economic analysis of PV and energy storage investment issues.

[0004] Furthermore, for solving multi-objective programming models, the normal boundary cross method is often used because standard multi-objective particle swarm optimization and multi-objective genetic algorithms are prone to randomness and getting trapped in local optima. Duan Ziyue et al., addressing potential unstable operating conditions in flexible low-frequency offshore wind power systems, employed the normal boundary cross method and fuzzy membership function theory to solve a multi-objective system stability optimization problem considering operational efficiency and stability margin. Tan Zhukui et al. used an improved generalized normal boundary cross method to solve the Pareto front of a multi-objective optimization scheduling model for an electro-pneumatic interconnected system, providing dispatchers with diverse decision-making solutions.

[0005] However, the above studies still have the following shortcomings: (1) At the modeling level, the modeling of environmental benefits simply focuses on indicators such as carbon emissions, without fully quantifying the comprehensive impact of ecological, climatic, and human settlement factors on microgrid planning. (2) At the solution level, the existing normal boundary intersection method has the problem of uneven solution set distribution, making it difficult to obtain a uniform Pareto front solution set. Summary of the Invention

[0006] Purpose of the invention: In order to overcome the shortcomings of the existing technology, a microgrid planning method and system based on the improved normal boundary intersection method is provided.

[0007] Technical Solution: To achieve the above objectives, this invention provides a microgrid planning method based on an improved normal boundary intersection method, comprising the following steps:

[0008] S1: Objective function for constructing a microgrid planning model that takes into account the impacts of ecology, climate, and human settlements;

[0009] S2: Constraints for constructing a microgrid planning model that takes into account environmental benefits;

[0010] S3: Based on the objective function constructed in step S1 and the constraints constructed in step S2, construct a compact form of the microgrid planning model;

[0011] S4: For compact microgrid planning models, the Pareto front is obtained by solving the improved normal boundary crossing method based on Euclidean distance correction.

[0012] S5: Select the optimal compromise solution from the Pareto front to obtain the microgrid expansion planning scheme.

[0013] Furthermore, the objective functions of the microgrid planning model that takes into account the ecological, climatic, and human settlement impacts in step S1 include an economic cost objective function, an environmental benefit objective function, and a reliability objective function.

[0014] Furthermore, the establishment of the economic cost objective function in step S1 includes:

[0015] Microgrid planning aims to achieve total economic minimization, that is, minimizing the sum of planning and operating costs:

[0016]

[0017] In the formula: The objective function is the economic cost. For planning costs; For the total set of scenes; For the total time period set; variable subscript Indicates the first Scenario; variable subscript Indicates time period ; Cost of purchasing electricity for microgrids; For gas turbine fuel costs; For adjustable load incentive costs; Cost of charging and discharging energy storage.

[0018] Regarding planning costs :

[0019]

[0020] In the formula: Cost of expanding per unit energy storage capacity; Expanding energy storage capacity; Cost of expanding the capacity of a single new energy generating unit; Expanding the capacity of new energy generating units; For candidate energy storage sets; A collection of candidate new energy generating units;

[0021] Regarding operating costs:

[0022]

[0023]

[0024]

[0025]

[0026] In the formula: The electricity purchase price; For the power purchased; This refers to the fuel cost coefficient for gas turbines. This refers to the output power of the gas turbine. This is the load incentive cost coefficient; For flexible load power; This represents the load demand value under no-load excitation. , These are the cost coefficients for energy storage charging and discharging, respectively. , These are energy storage charging efficiency and energy storage discharging efficiency, respectively. and These are the energy storage charging power and the energy storage discharging power, respectively.

[0027] Among them, adjustable load incentive cost For nonlinear terms containing absolute values, the Big M method is used for linearization:

[0028]

[0029]

[0030]

[0031]

[0032]

[0033] in, For the introduction of continuous auxiliary variables; The 0-1 auxiliary variables are introduced; It is a very large positive number.

[0034] Furthermore, the establishment of the environmental benefit objective function in step S1 includes:

[0035] Based on carbon emissions, three dimensions—ecological impact, climate impact, and human settlement impact—are further introduced to construct an environmental benefit objective function:

[0036]

[0037] in, The objective function is the environmental benefit function; The environmental benefit objective function in existing microgrid planning; , , These are the weighting factors for ecological impact, climate impact, and human settlement impact, respectively. Ecological impact factors; Climate influencing factors; Factors influencing human settlements.

[0038] Furthermore, the establishment of the reliability objective function in step S1 includes:

[0039] In microgrid planning, reliability aims to minimize wind and solar power curtailment and load shedding, as expressed below:

[0040]

[0041] in, The reliability objective function; This indicates the amount of wind and solar power curtailed. Indicates the load shedding amount; The penalty cost per unit for curtailing wind and solar power; Penalty cost per unit of load shedding; A collection of new energy generating units; For load sets.

[0042] Furthermore, the constraints in step S2 include external power purchase constraints, gas turbine constraints, flexible load constraints, energy storage constraints, planning constraints, and environmental benefit constraints, specifically:

[0043] External power purchase constraints: These include upper and lower limits on external power purchases.

[0044]

[0045] in, This represents the maximum power purchase capacity of the microgrid.

[0046] Gas turbine constraints: For gas turbines, these include upper and lower output limits and ramping constraints.

[0047]

[0048]

[0049]

[0050] in, , These are the upper and lower limits of the active power of the gas turbine, respectively. This represents the maximum uphill climbing power of the gas turbine. This represents the maximum downhill ramp power of the gas turbine.

[0051] Adjustable load constraints: For adjustable loads, this includes total load constraints and load shifting constraints.

[0052]

[0053]

[0054] in, , These represent the maximum and minimum electricity demand, respectively.

[0055] Energy storage constraints: The regulation capability of energy storage is modeled as follows:

[0056]

[0057]

[0058] in, and They are respectively Time and The remaining energy stored at the end of the time interval; , These are the energy storage charging and discharging efficiencies, respectively. , These are the energy storage charging and discharging power, respectively; Net charging power for energy storage;

[0059] Energy storage regulation constraints include upper and lower limits for charge and discharge power, and upper and lower limits for energy storage capacity.

[0060]

[0061]

[0062]

[0063]

[0064] in, , These are the upper and lower limits of the energy storage capacity of the energy storage battery, respectively. , These are the upper limits for charging and discharging power of the energy storage battery, respectively. , These represent the charging and discharging states of the energy storage battery, which are variables of 0 and 1, respectively.

[0065] Planning constraints: For planned energy storage:

[0066]

[0067]

[0068]

[0069] in, The planning energy storage uses 0-1 auxiliary variables, where 0 indicates no expansion and 1 indicates expansion; To plan for maximum energy storage capacity;

[0070] Regarding the planning of new energy sources:

[0071]

[0072] in, The planning of new energy sources uses 0-1 auxiliary variables, where 0 indicates no expansion and 1 indicates expansion; To plan for the maximum capacity of new energy sources;

[0073] Environmental benefit constraints include ecological red line constraints, ecological buffer zone distance constraints, and human settlement suitability constraints, specifically:

[0074] Ecological red line constraints:

[0075]

[0076] in, Indicates the first Whether or not a region will build new energy power units is a 0 / 1 variable. This represents the set of areas covered by the ecological red line;

[0077] Ecological buffer zone distance constraints:

[0078]

[0079] in, Indicates the construction area and ecologically sensitive areas The distance between them; This is a collection of areas suitable for planning new energy sources. This is a collection of ecologically sensitive areas; Indicates the distance of the ecological buffer zone;

[0080] Human settlement suitability constraints:

[0081]

[0082] in, For the construction of new energy areas with residential areas The distance between them; It is a collection of residential areas; This indicates the distance suitable for human habitation.

[0083] Furthermore, the compact form of the microgrid planning model in step S3 is as follows:

[0084]

[0085] in, For decision variables; superscript Indicates transpose; , These are the first and second coefficient matrices of the economic cost objective function, respectively. The environmental benefit objective function is in a nonlinear form. , These are the first and second coefficient matrices of the reliability objective function, respectively. , These are the coefficient matrix and constant term of the inequality constraints, respectively. , These are the coefficient matrix and constant term of the inequality constraints, respectively. , Decision variables The lower and upper bound matrices.

[0086] Further, step S4 includes:

[0087] A1: Determine the boundary points and auxiliary surfaces of the Pareto front;

[0088] For the three sets of objective functions in the multi-objective optimization problem, each objective function is optimized separately to obtain the corresponding optimal solution. The three sets of optimal solutions are denoted as Pareto front boundary points.

[0089] Let the optimal solution corresponding to the economic cost objective function be... At this point, the objective function value is , , The optimal solution corresponding to the environmental benefit objective function is: At this point, the objective function value is , , The optimal solution corresponding to the reliability objective function optimization is: At this point, the objective function value is , , ;

[0090] Denote the boundary points of the Pareto front. , , The defined triangular plane is the Pareto front auxiliary plane;

[0091] A2: Normalize the three objective functions in the model so that their values ​​are all between [0,1], thus removing the influence of dimensions and orders of magnitude on the uniformity of the solution.

[0092]

[0093]

[0094]

[0095] in, This is the normalized economic cost objective function; , These represent the maximum and minimum values ​​of the economic cost objective function, respectively. This is the normalized objective function for environmental benefits; , These represent the maximum and minimum values ​​of the environmental benefit objective function, respectively. This is the normalized reliability objective function; , These represent the maximum and minimum values ​​of the reliability objective function, respectively.

[0096] After normalization, the Pareto front boundary points become the normalized Pareto front boundary points, and the Pareto front auxiliary surface becomes the normalized Pareto front auxiliary surface.

[0097] A3: Obtain the normalized Pareto front point set;

[0098] On the normalized Pareto front auxiliary surface, select a set of uniformly distributed points. , represented as the normalized Pareto front boundary point , , Three-point linear combination:

[0099]

[0100]

[0101] in, , , These are the boundary points of the normalized Pareto front on the auxiliary surface of the normalized Pareto front. , , Weighting coefficients;

[0102] Passing through each point Draw a perpendicular line to the normalized Pareto front auxiliary surface, calculate the following optimization model, and obtain the point set projected onto the normalized Pareto front surface. :

[0103]

[0104] in, express and The distance;

[0105] A4: Adaptive adjustment of point set uniformity based on Euclidean distance correction, including the following steps:

[0106] A4-1: Calculate the set of adjacent points on the normalized Pareto front surface. , Euclidean distance between :

[0107]

[0108] in, express The values ​​of the axis coordinates; express The values ​​of the axis coordinates; express The values ​​of the axis coordinates;

[0109] A4-2: Calculate the average Euclidean distance between all adjacent point sets. :

[0110]

[0111] in, Indicates the total number of points in the set;

[0112] A4-3: Setting the adjustment threshold ;

[0113] like On the normalized Pareto front auxiliary surface, the point The three-dimensional coordinate position is updated as follows:

[0114]

[0115] in, This is a dynamically settable parameter with a value range between [0,1].

[0116] After the location is updated Draw a perpendicular line to the normalized Pareto front auxiliary surface to obtain the set of points projected onto the normalized Pareto front surface. Recalculate at this point. :like Then increase ;like Then decrease ;like Then fix the current Location;

[0117] like On the normalized Pareto front auxiliary surface, the points The three-dimensional coordinate position is updated as follows:

[0118]

[0119] After the location is updated Draw a perpendicular line to the normalized Pareto front auxiliary surface to obtain the set of points projected onto the normalized Pareto front surface. Recalculate at this point. :like Then decrease ;like Then increase ;like Then fix the current Location;

[0120] like This indicates the point. With point If the spacing is within a reasonable range, no adjustment will be made.

[0121] Furthermore, the specific method for selecting the optimal compromise solution from the Pareto front in step S5 includes:

[0122] For the normalized economic cost objective function Normalized environmental benefit objective function Normalized reliability objective function The satisfaction level can be calculated using the following formulas:

[0123]

[0124] in, Indicates the first The satisfaction of a normalized objective function takes values ​​between [0,1]. For the first A normalized objective function, For the first The maximum value of the normalized objective function. For the first The minimum value of a normalized objective function;

[0125] After calculating the satisfaction levels of each normalized objective function, the overall satisfaction level of the solution is determined by the following formula:

[0126]

[0127] in, The number of normalized objective functions. This represents the overall satisfaction score.

[0128] Among all solutions on the Pareto front, the solution with the highest overall satisfaction is the optimal compromise solution.

[0129] This invention also provides a microgrid planning system based on an improved normal boundary intersection method, comprising:

[0130] The objective function construction module is used to construct the objective function of the microgrid planning model that takes into account the impacts of ecology, climate, and human settlements.

[0131] The constraint construction module is used to construct the constraint conditions for a microgrid planning model that takes into account environmental benefit constraints.

[0132] The model optimization module is used to build a compact form of the microgrid planning model;

[0133] The model solving module is used to solve the Pareto front by using the improved normal boundary crossing method based on Euclidean distance correction. The optimal compromise solution is selected from the Pareto front to obtain the microgrid expansion planning scheme.

[0134] Beneficial effects: Compared with the prior art, the present invention has the following advantages:

[0135] (1) This invention integrates ecological, climatic and human settlement impacts, quantifies and models environmental benefits, and embeds them into the objective function and constraints, thereby improving the environmental adaptability of microgrid planning;

[0136] (2) In view of the nonlinearity of the model caused by environmental benefit modeling, the present invention designs an improved normal boundary intersection method based on Euclidean distance correction to solve the Pareto front, which improves the uniformity of the solution set of the Pareto front, avoids the occurrence of clustered or empty regions in the solution set, and provides decision-makers with a more comprehensive set of candidate solutions, making it easier to make more reasonable decisions. Attached Figure Description

[0137] Figure 1 This is a schematic flowchart of the method of the present invention;

[0138] Figure 2 A schematic diagram of the Pareto front boundary points and Pareto front auxiliary surfaces;

[0139] Figure 3 This is a schematic diagram of the normalized Pareto front boundary points and the normalized Pareto front auxiliary surface;

[0140] Figure 4 A schematic diagram of the points projected onto the Pareto front surface set;

[0141] Figure 5 A diagram illustrating the Pareto front solution obtained by the method of this invention;

[0142] Figure 6 The diagram illustrates the Pareto front solutions obtained by three different algorithms. Detailed Implementation

[0143] The present invention will be further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. After reading this invention, any modifications of the invention in various equivalent forms by those skilled in the art will fall within the scope defined by the appended claims.

[0144] Example 1:

[0145] like Figure 1 As shown, this embodiment provides a microgrid planning method based on the improved normal boundary intersection method, including the following steps:

[0146] S1: Objective function for constructing a microgrid planning model that takes into account the impacts of ecology, climate, and human settlements;

[0147] S2: Constraints for constructing a microgrid planning model that takes into account environmental benefits;

[0148] S3: Based on the objective function constructed in step S1 and the constraints constructed in step S2, construct a compact form of the microgrid planning model;

[0149] S4: For compact microgrid planning models, the Pareto front is obtained by solving the improved normal boundary crossing method based on Euclidean distance correction.

[0150] S5: Select the optimal compromise solution from the Pareto front to obtain the microgrid expansion planning scheme.

[0151] The objective functions of the microgrid planning model that takes into account the ecological, climatic, and human settlement impacts in step S1 include the economic cost objective function, the environmental benefit objective function, and the reliability objective function;

[0152] The establishment of the economic cost objective function includes:

[0153] Microgrid planning aims to achieve total economic minimization, that is, minimizing the sum of planning and operating costs:

[0154]

[0155] In the formula: The objective function is the economic cost. For planning costs; For the total set of scenes; For the total time period set; variable subscript Indicates the first Scenario; variable subscript Indicates time period ; Cost of purchasing electricity for microgrids; For gas turbine fuel costs; For adjustable load incentive costs; Cost of charging and discharging energy storage.

[0156] Regarding planning costs :

[0157]

[0158] In the formula: Cost of expanding per unit energy storage capacity; Expanding energy storage capacity; Cost of expanding the capacity of a single new energy generating unit; Expanding the capacity of new energy generating units; For candidate energy storage sets; A collection of candidate new energy generating units;

[0159] Regarding operating costs:

[0160]

[0161]

[0162]

[0163]

[0164] In the formula: The electricity purchase price; For the power purchased; This refers to the fuel cost coefficient for gas turbines. This refers to the output power of the gas turbine. This is the load incentive cost coefficient; For flexible load power; This represents the load demand value under no-load excitation. , These are the cost coefficients for energy storage charging and discharging, respectively. , These are energy storage charging efficiency and energy storage discharging efficiency, respectively. and These are the energy storage charging power and the energy storage discharging power, respectively.

[0165] Among them, adjustable load incentive cost For nonlinear terms containing absolute values, the Big M method is used for linearization:

[0166]

[0167]

[0168]

[0169]

[0170]

[0171] in, For the introduction of continuous auxiliary variables; The 0-1 auxiliary variables are introduced; It is a very large positive number.

[0172] The establishment of the environmental benefit objective function includes:

[0173] In existing microgrid planning, environmental benefits are typically measured by carbon emissions, i.e.:

[0174]

[0175] in, The environmental benefit objective function in existing microgrid planning; The carbon emission cost of gas turbines; For carbon price; Carbon emission factors for gas turbine power generation; This refers to the output power of the gas turbine.

[0176] The aforementioned modeling methods fail to fully reflect the nonlinear disturbances of distributed energy construction to regional ecosystems. Therefore, this embodiment, in addition to carbon emissions, further introduces three dimensions: ecological impact, climate impact, and human settlement impact, to construct an environmental benefit objective function:

[0177]

[0178] in, The objective function is the environmental benefit function; The environmental benefit objective function in existing microgrid planning; , , These are the weighting factors for ecological impact, climate impact, and human settlement impact, respectively. Ecological impact factors; Climate influencing factors; Factors influencing human settlements.

[0179] The modeling methods for each factor are described below:

[0180] First, ecological impact factors:

[0181] Ecological impact factors are used to characterize the degree of disturbance of new energy units to the ecosystem, and mainly consider the following three aspects:

[0182] 1. The disturbance of birds by the wind turbine ( High-speed rotating blades of wind turbines can easily cause birds to collide with and die, especially in areas such as migratory bird flyways and wetland reserves. The quantitative modeling is as follows:

[0183]

[0184] in, For wind power installed capacity, The distance between the wind turbine and the bird habitat. , Formulas The first coefficient and the second coefficient.

[0185] The above function represents the variation of wind turbine collision risk to birds with wind power capacity and distance. (Regarding...) As installed capacity increases, the risk of bird collisions rises; however, considering that high-capacity wind turbines cause fewer bird deaths per unit capacity, therefore... This is to reflect the diminishing marginal risk of increased capacity; Regarding The exponential decay is used, indicating that the farther the wind turbine is from bird habitat, the lower the collision risk. In this embodiment, the ecological impact factor... item, Take 0.18, Take 0.20.

[0186] 2. Damage to surface vegetation and animal habitats caused by photovoltaic power generation units ( The extensive use of land for photovoltaic installations has led to fragmented plant communities and restricted animal activity. Its quantitative modeling is as follows:

[0187]

[0188] in, For photovoltaic installed capacity, The distance between photovoltaic units and ecologically sensitive areas. , Formulas The first coefficient and the second coefficient.

[0189] The above function describes the impact of new energy unit planning on vegetation destruction and animal habitat loss. It is stated that as the installed capacity of new energy sources increases, the land area required and construction disruptions increase. However, further increases in capacity will utilize more developed land, thus mitigating marginal losses. Furthermore, This indicates that the farther the generating unit is from ecologically sensitive areas, the smaller its ecological impact. In this embodiment, the ecological impact factors... item, Take 0.09, Take 0.10.

[0190] 3. The disturbance of soil and water bodies by new energy units ( The planning and construction of new energy power units alters the hydrogeomorphic structure, increasing the risk of soil erosion. Its quantitative modeling is as follows:

[0191]

[0192] in, The distance between new energy generating units and water- and soil-sensitive areas. , Formulas The first coefficient and the second coefficient.

[0193] The above function represents the impact of engineering construction on soil structure and water pollution risks. It is stated that the larger the installed capacity of new energy sources, the greater the required construction intensity, thus increasing the degree of soil and water disturbance. However, as the scale expands, the marginal effect of subsequent capacity increases on new disturbances decreases. Furthermore, This indicates that if construction is located far from water- and soil-sensitive areas, its environmental impact will be significantly reduced. In this embodiment, the ecological impact factors... item, Take 0.10, Take 0.15.

[0194] In summary, ecological impact factors can be modeled as follows:

[0195]

[0196] in, , , These are the first, second, and third weighting coefficients of the ecological impact factor, which can be set according to factors such as the type of unit, the degree of overlap between the site selection area and the protected area. In this embodiment, they are taken as 0.3, 0.4, and 0.3, respectively.

[0197] Second, climate influencing factors:

[0198] Climate impact factors are used to describe the extent to which new energy facilities alter the local microclimate, and mainly consider the following two aspects:

[0199] 1. Turbulence effect caused by the fan and temperature disturbance ( Wind turbines disturb the atmospheric boundary layer, leading to a decrease in surface wind speed and nighttime surface warming, which affects the ecological balance.

[0200]

[0201] in, for The distance to the boundary is calculated. The distance from the wind turbine. , Formulas The first coefficient and the second coefficient.

[0202] The above function reflects the impact of large-scale wind farm operation on local atmospheric turbulence. This indicates that wind turbine operation causes increased turbulence and decreased wind speed in the wake region, and the intensity of this effect increases with increasing installed capacity, but the additional reduction in wind speed with each additional unit of capacity becomes relatively smaller. Furthermore, This indicates that the turbulence effect attenuates with distance. In this embodiment, the climate influence factor... item, Take 2000m, Take 0.25, Take 0.05.

[0203] 2. Changes in the ground albedo of photovoltaic panels ( Large-scale photovoltaic deployment reduces surface reflectivity and alters heat distribution, potentially triggering the "heat island effect."

[0204]

[0205] in, for The distance to the boundary is calculated. Distance from photovoltaic power plants, , Formulas The first coefficient and the second coefficient.

[0206] The above function describes the environmental impact of photovoltaic power plants altering the land surface albedo. This indicates that as photovoltaic (PV) installed capacity increases, the proportion of land covered rises, leading to a greater decrease in average albedo. However, once the density of PV installations reaches a certain level, the albedo reduction effect tends to saturate. Furthermore, This is used to represent the spatial limitation of the impact, which is mainly confined to the vicinity of the site and has almost no impact on distant areas. In this embodiment, the climate impact factor... item, Take 2000m, Take 0.20, Take 0.05.

[0207] In summary, climate influencing factors can be modeled as follows:

[0208]

[0209] in, , These are the first and second weighting coefficients of the climate impact factor, respectively. They can be set according to data such as the scale of new energy units and regional meteorological sensitivity. In this embodiment, they are set to 0.5 and 0.5, respectively.

[0210] Third, factors affecting human settlements:

[0211] The human settlement impact factor is used to quantify the impact of new energy facilities on residents' living environment and psychological well-being, and mainly includes:

[0212] 1. Noise pollution ( The operation of the fan generates low-frequency noise, which can affect residents' health and sleep.

[0213]

[0214] in, This refers to the distance between the wind turbine and residential areas. , , Formulas The first coefficient, the second coefficient, and the third coefficient.

[0215] The above function describes the attenuation of wind turbine operating noise with the scale of the installed capacity and the distance. This indicates that wind turbine noise increases with the increase of installed capacity, but due to the logarithmic effect of the noise superposition of multiple wind turbines, the contribution of each additional wind turbine to the total noise decreases. This reflects the approximately exponential decay of noise with increasing distance during propagation. In this embodiment, the human settlement impact factor... item, Take 0.90, Take 0.35, Take 0.20.

[0216] 2. Visual pollution ( Tall wind towers and dense photovoltaic arrays damage the natural landscape and reduce resident satisfaction.

[0217]

[0218] in, The distance between the new energy generating unit and residential areas. , Formulas The first coefficient and the second coefficient.

[0219] The above function is used to measure the degree of impact of wind turbines or photovoltaic units on the visual landscape. This indicates that as installed capacity increases and the land area occupied increases, the degree of occupation and damage to the landscape intensifies until the visual impact approaches a certain upper limit. This reflects the attenuation of visual impact with observer distance. In this embodiment, the human settlement impact factor... item, Take 0.25, Take 0.15.

[0220] In summary, the human settlement impact factors can be modeled as follows:

[0221]

[0222] in: , These are the first and second weighting coefficients of the human settlement impact factor, respectively. They can be set based on factors such as the distance between the generator unit and residential areas and noise simulation results. In this embodiment, they are set to 0.5 and 0.5, respectively.

[0223] Based on carbon emissions, and combined with ecological impact factors, climate impact factors, and human settlement impact factors, the environmental benefit objective function proposed in this invention reflects that: when planning microgrids, ecological protection, climate regulation, and human settlement friendliness should be coordinated to achieve greater environmental adaptability.

[0224] The establishment of the reliability objective function includes:

[0225] In microgrid planning, reliability aims to minimize wind and solar power curtailment and load shedding, as expressed below:

[0226]

[0227] in, The reliability objective function; This indicates the amount of wind and solar power curtailed. Indicates the load shedding amount; The penalty cost per unit for curtailing wind and solar power; Penalty cost per unit of load shedding; A collection of new energy generating units; For load sets.

[0228] The constraints in step S2 include external power purchase constraints, gas turbine constraints, flexible load constraints, energy storage constraints, planning constraints, and environmental benefit constraints, specifically:

[0229] External power purchase constraints: These include upper and lower limits on external power purchases.

[0230]

[0231] in, This represents the maximum power purchase capacity of the microgrid.

[0232] Gas turbine constraints: For gas turbines, these include upper and lower output limits and ramping constraints.

[0233]

[0234]

[0235]

[0236] in, , These are the upper and lower limits of the active power of the gas turbine, respectively. This represents the maximum uphill climbing power of the gas turbine. This represents the maximum downhill ramp power of the gas turbine.

[0237] Adjustable load constraints: For adjustable loads, this includes total load constraints and load shifting constraints.

[0238]

[0239]

[0240] in, , These represent the maximum and minimum electricity demand, respectively.

[0241] Energy storage constraints: The regulation capability of energy storage is modeled as follows:

[0242]

[0243]

[0244] in, and They are respectively Time and The remaining energy stored at the end of the time interval; , These are the energy storage charging and discharging efficiencies, respectively. , These are the energy storage charging and discharging power, respectively; Net charging power for energy storage;

[0245] Energy storage regulation constraints include upper and lower limits for charge and discharge power, and upper and lower limits for energy storage capacity.

[0246]

[0247]

[0248]

[0249]

[0250] in, , These are the upper and lower limits of the energy storage capacity of the energy storage battery, respectively. , These are the upper limits for charging and discharging power of the energy storage battery, respectively. , These represent the charging and discharging states of the energy storage battery, which are variables of 0 and 1, respectively.

[0251] Planning constraints: For planned energy storage:

[0252]

[0253]

[0254]

[0255] in, The planning energy storage uses 0-1 auxiliary variables, where 0 indicates no expansion and 1 indicates expansion; To plan for maximum energy storage capacity;

[0256] Regarding the planning of new energy sources:

[0257]

[0258] in, The planning of new energy sources uses 0-1 auxiliary variables, where 0 indicates no expansion and 1 indicates expansion; To plan for the maximum capacity of new energy sources;

[0259] Environmental benefit constraints include ecological red line constraints, ecological buffer zone distance constraints, and human settlement suitability constraints, specifically:

[0260] Ecological red line constraints:

[0261]

[0262] in, Indicates the first Whether or not a region will build new energy power units is a 0 / 1 variable. This represents the set of areas covered by the ecological red line;

[0263] Ecological buffer zone distance constraints:

[0264]

[0265] in, Indicates the construction area and ecologically sensitive areas The distance between them is characterized using Euclidean distance; This is a collection of areas suitable for planning new energy sources. This is a collection of ecologically sensitive areas; This indicates the distance of the ecological buffer zone, which can be set to 300-500m; in this embodiment, it is set to 400m.

[0266] Human settlement suitability constraints:

[0267]

[0268] in, For the construction of new energy areas with residential areas The distance between them is characterized using Euclidean distance; It is a collection of residential areas; The distance representing the suitability of the living environment can be set from 300m to 1000m, and is set to 500m in this embodiment.

[0269] The compact form of the microgrid planning model in step S3 is as follows:

[0270]

[0271] in, For decision variables; superscript Indicates transpose; , These are the first and second coefficient matrices of the economic cost objective function, respectively. The environmental benefit objective function is in a nonlinear form. , These are the first and second coefficient matrices of the reliability objective function, respectively. , These are the coefficient matrix and constant term of the inequality constraints, respectively. , These are the coefficient matrix and constant term of the inequality constraints, respectively. , Decision variables The lower and upper bound matrices.

[0272] As can be seen from the compact form of the above model, the constructed model is a nonlinear multi-objective optimization problem due to the consideration of environmental benefits.

[0273] Step S4 includes:

[0274] A1: Determine the boundary points and auxiliary surfaces of the Pareto front;

[0275] For the three sets of objective functions in the multi-objective optimization problem, each objective function is optimized separately to obtain the corresponding optimal solution. The three sets of optimal solutions are denoted as Pareto front boundary points.

[0276] like Figure 2 As shown, let the optimal solution corresponding to the optimization of the economic cost objective function be... At this point, the objective function value is , , The optimal solution corresponding to the environmental benefit objective function is: At this point, the objective function value is , , The optimal solution corresponding to the reliability objective function optimization is: At this point, the objective function value is , , ;

[0277] Denote the boundary points of the Pareto front. , , The defined triangular plane is the Pareto front auxiliary plane;

[0278] A2: As Figure 3As shown, the three objective functions in the model are normalized so that their values ​​are all between [0,1], thus removing the influence of dimensions and orders of magnitude on the uniformity of the solution:

[0279]

[0280]

[0281]

[0282] in, This is the normalized economic cost objective function; , These represent the maximum and minimum values ​​of the economic cost objective function, respectively. This is the normalized objective function for environmental benefits; , These represent the maximum and minimum values ​​of the environmental benefit objective function, respectively. This is the normalized reliability objective function; , These represent the maximum and minimum values ​​of the reliability objective function, respectively.

[0283] After normalization, the Pareto front boundary points become the normalized Pareto front boundary points, and the Pareto front auxiliary surface becomes the normalized Pareto front auxiliary surface.

[0284] A3: Obtain the normalized Pareto front point set;

[0285] like Figure 4 As shown, a set of uniformly distributed points are selected on the normalized Pareto front auxiliary surface. , represented as the normalized Pareto front boundary point , , Three-point linear combination:

[0286]

[0287]

[0288] in, , , These are the boundary points of the normalized Pareto front on the auxiliary surface of the normalized Pareto front. , , Weighting coefficients;

[0289] Passing through each point Draw a perpendicular line to the normalized Pareto front auxiliary surface, calculate the following optimization model, and obtain the point set projected onto the normalized Pareto front surface. :

[0290]

[0291] in, express and Distance is indicated;

[0292] A4: Adaptive Adjustment of Point Set Uniformity Based on Euclidean Distance Correction

[0293] Because of the introduction of mathematical modeling for environmental benefits, the normalized Pareto front exhibits nonlinear properties, i.e., it is a curved surface. Although points The auxiliary surface of the normalized Pareto front is uniformly selected, but due to the nonlinear properties of the normalized Pareto front, the point set projected onto the normalized Pareto front surface... The distribution is uneven. This conclusion can be drawn from... Figure 4 The points selected on the normalized Pareto front auxiliary surface are reflected in the following: The points are uniformly distributed, but projected onto the normalized Pareto front surface. It is a non-uniform distribution;

[0294] Step A4 includes the following steps:

[0295] A4-1: Calculate the set of adjacent points on the normalized Pareto front surface. , Euclidean distance between :

[0296]

[0297] in, express The values ​​of the axis coordinates; express The values ​​of the axis coordinates; express The values ​​of the axis coordinates;

[0298] A4-2: Calculate the average Euclidean distance between all adjacent point sets. :

[0299]

[0300] in, Indicates the total number of points in the set;

[0301] A4-3: Setting the adjustment threshold ;

[0302] like This indicates the point. With point The spacing is too small; therefore, on the normalized Pareto front auxiliary surface, the points are... The three-dimensional coordinate position is updated as follows:

[0303]

[0304] in, This is a dynamically settable parameter with a value range between [0,1].

[0305] After the location is updated Draw a perpendicular line to the normalized Pareto front auxiliary surface to obtain the set of points projected onto the normalized Pareto front surface. Recalculate at this point. :like Then increase appropriately. ;like Then reduce appropriately. ;like Then fix the current Location;

[0306] like This indicates the point. With point The spacing is too large; therefore, on the normalized Pareto front auxiliary surface, the points are... The three-dimensional coordinate position is updated as follows:

[0307]

[0308] After the location is updated Draw a perpendicular line to the normalized Pareto front auxiliary surface to obtain the set of points projected onto the normalized Pareto front surface. Recalculate at this point. :like Then reduce appropriately. ;like Then increase appropriately. ;like Then fix the current Location;

[0309] like This indicates the point. With point If the spacing is within a reasonable range, no adjustment will be made.

[0310] In summary, by adaptively adjusting the point set uniformity based on Euclidean distance correction, the obtained normalized Pareto front surface shows a more uniform point set, and the Euclidean distances between adjacent point sets are all within the range of... This improves the uniformity of the solution results by bridging the range between the two values.

[0311] The specific methods for selecting the optimal compromise solution from the Pareto front in step S5 include:

[0312] For the normalized economic cost objective function Normalized environmental benefit objective function Normalized reliability objective function The satisfaction level can be calculated using the following formulas:

[0313]

[0314] in, Indicates the first The satisfaction of a normalized objective function takes values ​​between [0,1]. For the first A normalized objective function, For the first The maximum value of the normalized objective function. For the first The minimum value of a normalized objective function;

[0315] After calculating the satisfaction levels of each normalized objective function, the overall satisfaction level of the solution is determined by the following formula:

[0316]

[0317] in, The number of normalized objective functions. This represents the overall satisfaction score.

[0318] Among all solutions on the Pareto front, the solution with the highest overall satisfaction is the optimal compromise solution. Furthermore, microgrid planning expansion schemes include the expansion results of energy storage and new energy units.

[0319] Example 2:

[0320] Based on the method of Embodiment 1, this embodiment provides a microgrid planning system based on the improved normal boundary intersection method, including:

[0321] The objective function construction module is used to construct the objective function of the microgrid planning model that takes into account the impacts of ecology, climate, and human settlements.

[0322] The constraint construction module is used to construct the constraint conditions for a microgrid planning model that takes into account environmental benefit constraints.

[0323] The model optimization module is used to build a compact form of the microgrid planning model;

[0324] The model solving module is used to solve the Pareto front by using the improved normal boundary crossing method based on Euclidean distance correction. The optimal compromise solution is selected from the Pareto front to obtain the microgrid expansion planning scheme.

[0325] Example 3:

[0326] This embodiment uses a structure where three microgrids are connected to the same distribution network to verify the effectiveness of the proposed model and solution algorithm. The capacity parameter settings for each device are shown in Table 1.

[0327] Table 1 Capacity Parameter Settings

[0328]

[0329] Other parameter settings for the microgrid planning model are shown in Table 2:

[0330] Table 2 Other parameter settings

[0331]

[0332] For the renewable energy-load scenario, this invention uses Latin hypercube sampling to generate 100 daily scenarios, and then uses the K-means algorithm to reduce the number of scenarios, ultimately obtaining 15 typical daily scenarios for use in the microgrid planning model. Correspondingly, in the total scenario set... The document contains 15 typical daily scenarios; within each typical daily scenario, the total time period is set. It contains 24 moments.

[0333] First, the improved normal boundary intersection method proposed in this invention is used to solve the problem, resulting in 20 Pareto solutions on the normalized Pareto front, such as... Figure 5 As shown.

[0334] Depend on Figure 5 It can be seen that, on the one hand, the Pareto front set obtained by using the improved normal boundary intersection method can achieve a more uniformly distributed solution set, resulting in better quality. On the other hand, observing the Pareto front reveals that as the value of one objective function increases, the values ​​of the other two objective functions decrease accordingly. This indicates that there is a certain conflict between the three objective functions—economic cost, environmental benefits, and reliability—in microgrid planning, requiring trade-offs among them.

[0335] Secondly, for the 20 solutions in the normalized Pareto front, three compromise solutions P1, P2, and P3 were selected, with P2 being the optimal compromise solution. The results are shown in Table 3.

[0336] Table 3 Optimization results of different compromise solutions

[0337]

[0338] As can be seen from Table 3, P1 performs better in terms of environmental benefits, but has the highest economic cost and the worst reliability; P3 has the lowest economic cost and good reliability, but performs weaker in terms of environmental benefits; P2, as the solution with the strongest balance, achieves good results in all three types of objective functions, demonstrating a reasonable trade-off performance.

[0339] The extended programming solutions corresponding to the optimal compromise solution P2 are shown in Table 4:

[0340] Table 4. Planning results corresponding to P2

[0341]

[0342] In contrast, in environmental benefit modeling, only carbon emissions + ecological factors + climate factors (comparison scheme 1), carbon emissions + ecological factors + human settlement factors (comparison scheme 2), and carbon emissions + climate factors + human settlement factors (comparison scheme 3) were considered, with other conditions remaining unchanged. The comparison of the planning scheme results is shown in Table 5:

[0343] Table 5 Comparison of Planning Schemes Obtained from Modeling Different Environmental Benefits

[0344]

[0345] Comparing the results in Tables 4 and 5, it can be seen that compared to the planning scheme of this invention, which comprehensively considers ecological, climatic, and human settlement factors, the planned capacity of new energy units (photovoltaics and wind turbines) in all three comparative schemes has increased. This indicates that this invention, by comprehensively considering various environmental benefit indicators, can take into account the limitations of ecological buffer zones and human settlement adaptability on the installed capacity of new energy, thereby rationally planning new energy installations and improving the environmental adaptability of microgrid planning.

[0346] To verify the effectiveness and rationality of the improved normal boundary intersection method proposed in this invention, the following comparative algorithm is designed to compare and analyze the Pareto front of the multi-objective optimization problem:

[0347] Algorithm 1: The improved normal boundary intersection method proposed in this invention.

[0348] Algorithm 2: Existing normal boundary intersection method.

[0349] Algorithm 3: NSGA-II algorithm.

[0350] Under a unified model and parameter setting, based on the three different algorithms mentioned above, 20 different Pareto front solutions were obtained in the normalized space, such as... Figure 6 As shown.

[0351] Furthermore, the relevant statistical indicators of the 20 Pareto front solutions for the three algorithms are shown in Table 6:

[0352] Table 6. Statistical Indicators of Pareto Front Solutions for Three Algorithms

[0353]

[0354] according to Figure 6 Table 6 shows that the standard deviation of Algorithm 1 is only 0.012, and the coefficient of variation is only 0.014, indicating that its solution set is concentrated and more stable. Compared with Algorithms 2 and 3, the solution set generated by Algorithm 1 is evenly distributed across the entire Pareto front, avoiding local clustering or sparse regions. Therefore, the uniformity adaptive adjustment mechanism based on Euclidean distance proposed in this invention effectively overcomes the problem of uneven solution set distribution that traditional algorithms easily encounter in nonlinear Pareto fronts. The method of this invention is particularly suitable for microgrid planning problems with high requirements for solution set quality and diversity, providing decision-makers with a more valuable solution set.

Claims

1. A microgrid planning method based on an improved normal boundary intersection method, characterized in that, The method comprises the following steps: S1: constructing a target function of a micro-grid planning model considering ecological, climatic and human settlement impacts; S2: constructing constraint conditions of the micro-grid planning model considering environmental benefit constraints; S3: constructing a compact form of the micro-grid planning model according to the target function constructed in step S1 and the constraint conditions constructed in step S2; S4: solving the compact form of the micro-grid planning model to obtain a Pareto front based on an improved normal boundary intersection method with Euclidean distance correction; S5: selecting an optimal compromise solution from the Pareto front to obtain a micro-grid expansion planning scheme; The target function of the micro-grid planning model considering ecological, climatic and human settlement impacts in step S1 comprises an economic cost target function, an environmental benefit target function and a reliability target function; The establishment of the economic cost target function in step S1 comprises: The micro-grid planning aims to minimize the total economic cost, that is, the sum of the planning and operation costs is minimized: wherein: f1is an economic cost objective function; C inv is the planning cost; S is the total set of scenarios; T is the total set of time periods; variable subscript s denotes the s-th scenario; variable subscript t denotes time period t; C br,s,t is the microgrid electricity purchase cost; C gas,s,t is the gas turbine fuel cost; C l,s,t is the adjustable load incentive cost; C se,s,t is the energy storage charge and discharge cost; For planning cost C inv : In the formula: a se The unit energy storage capacity expansion cost; x se The energy storage expansion capacity; a r The unit new energy unit capacity expansion cost; x r The new energy unit expansion capacity; Ω se+ The candidate energy storage set; Ω r+ The candidate new energy unit set; For the operation cost: C br,s,t = a br P br,s,t C gas,s,t = a gas P gas,s,t In the formula: a br is the electricity purchase price; P br,s,t is the electricity purchase power; a gas is the gas turbine fuel cost coefficient; P gas,s,t is the gas turbine output power; a l is the load incentive cost coefficient; P l,s,t is the flexible load power; is the load demand value without load incentive; a se,c , a se,d are the energy storage charging and discharging cost coefficients respectively; η se,in , η se,d are the energy storage charging efficiency and energy storage discharging efficiency respectively; P se,d,s,t and P se,c,s,t are the energy storage charging power and energy storage discharging power respectively; where the adjustable load incentive cost For the nonlinear term with absolute value, the large M method is used for linearization: u∈{0,1} Wherein, z is a continuous auxiliary variable introduced; u is a 0-1 auxiliary variable introduced; M is a normal number; The establishment of the environmental benefit target function in step S1 comprises: On the basis of carbon emissions, three dimensions of ecological impact, climatic impact and human settlement impact are further introduced to construct the environmental benefit target function: min f2 = f 2,0 + ω1EI + ω2CI + ω3HI wherein f2 is an environmental benefit objective function; f 2,0 is an environmental benefit objective function in the existing microgrid planning; ω1, ω2, and ω3 are weight factors of ecological impact, climate impact, and human settlement impact, respectively; EI is an ecological impact factor; CI is a climate impact factor; and HI is a human settlement impact factor. The establishment of the reliability target function in step S1 comprises: In the micro-grid planning, the reliability aims to minimize the degree of wind and light abandonment and the degree of load shedding, which is expressed as follows: wherein f3 is a reliability objective function; represents the amount of abandoned wind and light; represents the amount of load shedding;c r is the unit abandoned wind and light penalty cost;c d is the unit load shedding penalty cost;Ω r is a new energy unit set;Ω d is a load set; The compact form of the micro-grid planning model in step S3 is: Wherein, x is the decision variable; superscript T represents transposition; a, b are respectively the first coefficient matrix, the second coefficient matrix of the economic cost objective function; F(x) is the nonlinear form of the environmental benefit objective function; c, d are respectively the first coefficient matrix, the second coefficient matrix of the reliability objective function; A ieq , b ieq are respectively the coefficient matrix, the constant term of the inequality constraint; A eq , b eq are respectively the coefficient matrix, the constant term of the inequality constraint; L bound , U bound are respectively the lower, upper bound matrix of the decision variable x.

2. The microgrid planning method based on the improved line boundary intersection method according to claim 1, wherein, The constraint conditions in step S2 comprise external power purchase constraints, gas turbine constraints, flexible load constraints, energy storage constraints, planning constraints and environmental benefit constraints, and specifically are as follows: The external power purchase constraints comprise upper and lower constraints for power purchase: 0 < P br,s,t ≤ P br,max Pmax= Pgrid+ Pload+ Pstorage+ Pgeneration br,max Pmax= Pgrid+ Pload+ Pstorage+ Pgeneration The gas turbine constraints comprise output upper and lower constraints and climbing constraints for the gas turbine: P gas,min ≤P gas,s,t ≤P gas,max where P gas,max , P gas,min are the upper and lower active power limits of the gas turbine, respectively; is the maximum up-ramp power of the gas turbine; is the maximum down-ramp power of the gas turbine; The adjustable load constraints comprise total load constraints and load translatable constraints for the adjustable load: wherein, respectively the maximum and minimum load electricity demand; The energy storage constraints are modeled as follows: P se,s,t = P se,c,s,t - P se,d,s,t Wherein, E se,s,t and E se,s,t-1 are the remaining energy of the energy storage at time t and at the end of time t-1, respectively; η se,in , η se,d are the charging and discharging efficiency of the energy storage, respectively; P se,c,s,t , P se,d,s,t are the charging and discharging power of the energy storage, respectively; P se,s,t is the net charging power of the energy storage. The energy storage adjustment constraints comprise upper and lower constraints for charging and discharging power and upper and lower constraints for storage capacity: wherein, respectively the upper and lower energy storage limits of the energy storage battery; respectively the upper and lower charge and discharge power limits of the energy storage battery; respectively the charge and discharge state 0 / 1 variables of the energy storage battery; The planning constraints are as follows for the planning energy storage: x se ≤u se x se,max wherein u se is a 0-1 auxiliary variable for planning energy storage, 0 means not to expand, 1 means to expand; x se,max is the maximum capacity of the planned energy storage; The planning new energy constraints are as follows: x r ≤u r x r,max wherein u r is a 0-1 auxiliary variable for planning new energy, 0 represents no expansion, and 1 represents expansion; x r,max is the maximum capacity of planning new energy; The environmental benefit constraints comprise ecological red line constraints, ecological buffer zone distance constraints and human settlement environment suitability constraints, and specifically are as follows: The ecological red line constraints are as follows: wherein, x i is a 0 / 1 variable indicating whether a new energy unit is built in the i-th region, E redline represents a set of ecological red line covered regions; The ecological buffer zone distance constraints are as follows: wherein d i,j represents the distance between the construction area i and the ecological sensitive area j; P is the set of new energy planning areas; H is the set of ecological sensitive areas; D eco represents the distance of the ecological buffer zone; The human settlement environment suitability constraints are as follows: where d i,k is the distance between new energy construction region i and residential area k; R is the residential area set; D resid represents the human settlement environment suitability distance.

3. The microgrid planning method based on the improved line boundary intersection method according to claim 2, characterized in that, The step S4 comprises: A1: determining a Pareto front boundary point and a Pareto front auxiliary surface; For the three groups of target functions in the multi-objective optimization problem, the respective optimal solutions are obtained by separately optimizing each target function, and the three groups of optimal solutions are denoted as a Pareto front boundary point; The optimal solution corresponding to the optimization of the economic cost objective function is The objective function value at this time is The optimal solution corresponding to the optimization of the environmental benefit objective function is The objective function value at this time is The optimal solution corresponding to the optimization of the reliability objective function is The objective function value at this time is points on the pareto frontier the determined triangle plane is a pareto frontier auxiliary plane; A2: normalizing the three target functions in the model so that the target functions take values between 0 and 1, and removing the influence of dimensions and orders of magnitude on the uniformity of the solutions: Wherein, f1'(x) is the normalized economic cost objective function; f 1,max , f 1,min are the maximum and minimum values of the economic cost objective function respectively; f2'(x) is the normalized environmental benefit objective function; f 2,max , f 2,min are the maximum and minimum values of the environmental benefit objective function respectively; f3'(x) is the normalized reliability objective function; f 3,max , f 3,min are the maximum and minimum values of the reliability objective function respectively. After normalization, the Pareto front boundary point becomes a normalized Pareto front boundary point, and the Pareto front auxiliary surface becomes a normalized Pareto front auxiliary surface; A3: obtaining a normalized Pareto front surface point set; On the normalized Pareto frontier auxiliary plane, a set of uniformly distributed points p i are selected, denoted as normalized Pareto frontier boundary points Three-point linear combination: w1+w2+w3=1 where w 1,i , w 2,i , w 3,i are the weight coefficients of the normalized Pareto frontier boundary points on the normalized Pareto frontier auxiliary surface, respectively. passing through each point p i , the normal line of the normalized Pareto front auxiliary plane, the following optimization model is calculated to obtain the projection to the normalized Pareto front point set p i ′: where d i represents p i the distance from p i ' ; A4: Adaptive adjustment of point set uniformity based on Euclidean distance correction, including the following steps: A4-1: Calculate the Euclidean distance d(i,i+1) between the adjacent point sets p on the normalized Pareto front face i , p i+1 between the adjacent point sets p on the normalized Pareto front face Wherein, x(·) represents the value of x-axis coordinate; y(·) represents the value of y-axis coordinate; z(·) represents the value of z-axis coordinate; A4-2: Calculate the average of the Euclidean distances between all neighboring point sets Wherein, I represents the total number of point sets; A4-3: Set adjustment threshold ε>0; If On the normalized Pareto front auxiliary plane, the three-dimensional coordinate position of point p i is updated as follows: Wherein, λ is a parameter that can be dynamically set, and the value range is between [0, 1]; For the updated p i , the perpendicular of the normalized Pareto front auxiliary plane is obtained, and the point set p i ′ projected on the normalized Pareto front plane is obtained; at this time, d(i, i+1) is recalculated: if d(i, i+1) > 0, then λ is increased; if d(i, i+1) < 0, then λ is decreased; if d(i, i+1) = 0, then the current p ′ position is fixed. i ′ position is fixed.​​ If The three-dimensional coordinate position of the point p i on the normalized Pareto front auxiliary plane is updated as follows: For p after updating position i Draw a perpendicular line to the normalized Pareto front auxiliary surface to obtain the point set p′ projected onto the normalized Pareto front surface. i At this point, d(i,i+1) is recalculated. ： like Then decrease λ; if Then increase λ; if Then fix the current p i Location; If then it is indicated that the point p' i is within a reasonable range of the point p' i+1 and no adjustment is made.

4. The microgrid planning method based on the improved line boundary intersection method according to claim 3, wherein, The specific method of selecting the optimal compromise solution from the Pareto frontier in step S5 includes: For the normalized economic cost objective function f1'(x), the normalized environmental benefit objective function f2'(x) and the normalized reliability objective function f3'(x), their satisfaction degrees are solved by the following formula respectively: wherein mem(i) represents the satisfaction degree of the i-th normalized objective function, and takes a value in the range of [0, 1]; f i f(i) is the i-th normalized objective function, i max f(i) is the i-th normalized objective function, i min f(i) is the i-th normalized objective function, After calculating the satisfaction degrees of each normalized objective function, the comprehensive satisfaction degree of the solution is determined by the following formula: Wherein, m is the number of normalized objective functions, and summem is the value of comprehensive satisfaction degree. Among all the solutions on the Pareto frontier, the solution with the maximum comprehensive satisfaction degree is the optimal compromise solution.

5. A microgrid planning system based on an improved line boundary crossing method, characterized by, The system for implementing the method of claim 1 comprises: A target function construction module for constructing a target function of a microgrid planning model considering ecological, climate and human settlement impacts; A constraint condition construction module for constructing constraint conditions of the microgrid planning model considering environmental benefit constraints; A model optimization module for constructing a compact form of the microgrid planning model; A model solving module for solving the Pareto frontier based on the improved normal boundary intersection method with Euclidean distance correction, selecting the optimal compromise solution from the Pareto frontier, and obtaining the microgrid expansion planning scheme.

Images