Port distributed power supply planning method and system based on improved Pedal optimization algorithm

Abstract

The invention discloses a port distributed power supply planning method and system based on an improved Pedal optimization algorithm, and the method comprises the steps: constructing a photovoltaic power generation model and a wind power generation model, and converting the wind speed and illumination intensity of a typical wind and light time sequence scene in each season into the time sequence output power of a distributed power supply; constructing a distributed power supply planning model based on the annual comprehensive cost objective function, the annual total voltage deviation objective function and constraint conditions of the distributed power supply multi-objective planning model; performing performance improvement on the Pedal optimization algorithm to obtain an improved Pedal optimization algorithm; and solving the distributed power supply planning model based on an improved Pedal optimization algorithm to obtain an optimal planning result of the port distributed power supply. According to the method, the overall economic benefit of the power system is remarkably improved, the operation cost is reduced by optimizing resource allocation, effective utilization of green energy is promoted, and the active loss of the system is successfully reduced by optimizing the layout and operation strategy of the distributed power supply.

Description

Technical Field

[0001] The invention belongs to the technical field of distributed power supply planning, and relates to a port distributed power supply planning method and system based on an improved gazelle optimization algorithm. Background Art

[0002] Distributed power generation is an important form of renewable energy generation. Distributed power is a regional power generation device close to the load side. Distributed power has become an important part of the port energy supply system and a priority development direction. However, the uncontrollable output of distributed power has undoubtedly added great difficulty to the control and coordination of the port power grid. Reasonable planning of the location and capacity of distributed power access to the port power grid is an important factor in ensuring the stable and reliable operation of the power system and improving the quality of power. Research on distributed power planning has important theoretical and practical significance for the development and construction of new power systems. The solution process of traditional planning algorithms is relatively complex, the solution speed is too slow, and it is easy to fall into local optimality. Summary of the invention

[0003] The purpose of the present invention is to solve the problem in the prior art that the capacity and access location of distributed power sources cannot be reasonably determined, which leads to the destruction of power grid stability and affects the normal operation of the power grid, and to provide a port distributed power source planning method and system based on an improved gazelle optimization algorithm.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions:

[0005] The port distributed power planning method based on the improved gazelle optimization algorithm includes:

[0006] Based on the operating parameters and conditions of photovoltaic equipment and the operating parameters and wind power of wind equipment in typical port wind and solar time sequence scenarios, photovoltaic power generation model and wind power generation model are constructed respectively;

[0007] Based on the photovoltaic power generation model and wind power generation model, the wind speed and light intensity of typical wind and solar time sequence scenes in each season are converted into the time sequence output power of distributed power sources;

[0008] Based on the fixed investment cost, operating cost, electricity purchase cost from the superior power grid, network loss cost and compensation cost of distributed power generation, determine the annual comprehensive cost objective function;

[0009] Based on the voltage per unit value of each node at each time in the unit scenario, the annual total voltage deviation objective function is obtained;

[0010] Based on the constraints of the distributed power multi-objective planning model, a distributed power planning model is constructed with the annual comprehensive cost objective function, the annual total voltage deviation objective function and the distributed power operation cost as the optimization objectives;

[0011] Improve the performance of the Gazelle optimization algorithm and obtain an improved Gazelle optimization algorithm;

[0012] The distributed power generation planning model is solved based on the improved gazelle optimization algorithm by inputting the distributed power generation's sequential output power, grid parameters and load-related data to obtain the optimal planning result of the port's distributed power generation.

[0013] A further improvement of the present invention is:

[0014] Furthermore, based on the photovoltaic equipment operating parameters and conditions of the typical port wind and solar time sequence scenario and the wind equipment operating parameters and wind power, a photovoltaic power generation model and a wind power generation model are constructed respectively, specifically:

[0015] Construct a photovoltaic power generation model, specifically: Beta distribution is used to model the uncertainty of photovoltaic power generation, and the probability density function is:

[0016]

[0017] Among them, α1 and α2 are the shape parameters of the Beta distribution, which depend on the expectation and variance of the light intensity; S and S max are the actual light intensity and the maximum light intensity respectively;

[0018] The relationship between the active output of photovoltaic power generation equipment and light intensity is:

[0019]

[0020] Among them, P pv and is the active output and rated power of photovoltaic, and the probability density function of photovoltaic active output power is obtained as follows:

[0021]

[0022] Construct a wind power generation model. Specifically, assume that the wind turbine generator set is under constant power control and adopts Weibull distribution. The probability density function is as follows:

[0023]

[0024] Where u represents the actual wind speed; k and β are the shape parameter and scale parameter of the Weibull distribution respectively; the output of the wind turbine is:

[0025]

[0026] Among them, u1, u2 and u e Respectively represent the cut-in wind speed, cut-out wind speed and rated wind speed; P wt is the active power output of the wind turbine at different wind speeds; is the rated power of the fan.

[0027] Furthermore, based on the photovoltaic power generation model and the wind power generation model, the wind speed and light intensity of the typical wind-solar time sequence scenes in each season are converted into the time sequence output power of the distributed power source, specifically:

[0028] Based on the K-means clustering algorithm, the data of wind speed and light intensity of the port for one year were processed to obtain the typical time series scenes of wind speed and light intensity in each season;

[0029] The wind speed and light intensity data of typical wind and solar time series scenarios in each season are input into the established wind power and photovoltaic time series output model, and the output is the wind power and photovoltaic time series output power corresponding to each scenario.

[0030] Furthermore, based on the fixed investment cost, operating cost, power purchase cost from the superior power grid, network loss cost, and compensation cost of the distributed power generation, the annual comprehensive cost objective function is determined as follows:

[0031] F1=f total =f l +f om +f lo +f en +f dr (6)

[0032] Among them, f l is the fixed investment cost; f om is the operating cost; lo is the network loss cost; f en The cost of purchasing electricity from the upper power grid; dr To compensate for costs.

[0033] Furthermore, based on the voltage per unit value of each node at each time in the unit scenario, the annual total voltage deviation objective function is obtained, which is specifically:

[0034]

[0035] Among them, V s,i,t is the voltage per unit value at the i-th node at time t in scenario s;

[0036] The annual comprehensive cost objective function and the annual total voltage deviation objective function are weighted respectively, and the correction coefficient Q is introduced C and Q v The weighted function is modified to obtain a modified objective function;

[0037] The annual comprehensive cost objective function and the annual total voltage deviation objective function are weighted respectively, as follows:

[0038] minF=wF1+(1-w)F2 (8)

[0039] Introducing correction factor Q C and Q v After correcting formula (8), the objective function is:

[0040] minF=wF1 / Q C +(1-w)F2 / Q V (9)

[0041] Among them, the correction factor Q C and Q v The calculation expression is:

[0042]

[0043] Among them, C ni,max and C n0 are the annual comprehensive costs of the i-th node after it is connected to the distributed power source with the largest capacity and when it is not connected to the distributed power source, V tvd,ni and V tvd,n0 are the total voltage deviations of the i-th node after it is connected to the distributed power source with the maximum capacity and when it is not connected to the distributed power source, is the upper limit of the capacity of distributed generation connected to node i, n DG is the number of nodes connected to the distributed generation.

[0044] Furthermore, the constraints of the distributed power generation multi-objective planning model are as follows:

[0045] (1) Installed capacity constraints of distributed generation

[0046]

[0047] In the formula, Indicates the maximum capacity of distributed generation allowed to be installed at the node i to be planned; and They represent the maximum allowable installed capacity of wind power and photovoltaic power at the node i to be planned;

[0048] (2) Node voltage limitation

[0049]

[0050] Where U i is the voltage amplitude at node i; and are the upper and lower limits of the node voltage amplitude respectively;

[0051] (3) Power balance constraints

[0052]

[0053] Where P i.p and Q i.s They represent the active power and reactive power injected into node i in scenario s respectively; U i.s and U j.s Respectively represent the voltage amplitude of node i and node j in scenario s; θ ij.s is the phase angle difference between nodes i and j in scene s;

[0054] (4) Branch capacity constraints

[0055]

[0056] Where S j and denote the apparent power and maximum value of line j respectively;

[0057] (5) Distributed power penetration rate constraints

[0058]

[0059] Where P z is the total active load of the distribution network; ζ is the maximum penetration rate of distributed generation allowed for distribution network planning;

[0060] (6) Distributed power generation operation constraints

[0061]

[0062] The types of distributed power sources are wind power and photovoltaic power. is the upper limit of active power output installed in node i scenario s; p i,DG and are the active output removal ratio and its maximum value of the distributed generation installed at node i respectively;

[0063] (7) Load transfer constraints

[0064]

[0065] Where P i,s,dr is the transferable load of node i in scenario s; is its maximum value.

[0066] Furthermore, the performance of the Gazelle optimization algorithm is improved to obtain an improved Gazelle optimization algorithm, specifically:

[0067] Improved population position initialization strategy;

[0068] The position of gazelle population X in the search space is described as:

[0069]

[0070] In the formula, n represents the number of gazelle populations, d represents the dimension of the optimization variable, and x i,j represents the position of the i-th gazelle individual in the j-dimension;

[0071] The individual with the smallest fitness value in the gazelle population is selected as the strongest gazelle and saved in the optimal gazelle matrix. The optimal gazelle matrix Elite is described as:

[0072]

[0073] In the formula, x topi,j represents the search position of the strongest gazelle in the current gazelle population;

[0074] The Tent mapping initialization strategy is introduced to improve the uniformity of the distribution of initial gazelle individuals in the solution space; the Tent mapping is described as follows:

[0075]

[0076] In the formula, x t is the generated chaotic sequence, n is the number of gazelle populations to be initialized, τ is the adjustment parameter of the chaotic sequence, 0<τ<1;

[0077] Introducing logarithmic inertia weights;

[0078] Development phase: This process assumes that the position of the gazelle individual follows the Brownian motion mode. The position update process of the gazelle individual in the development phase is described as follows:

[0079] x i.p1 =x i +v0·r2×R b (Elite-R b × i ) (twenty one)

[0080] In the formula, x i.p1 and x i They represent the positions of the i-th gazelle individual before and after the update in the development stage, v0 is the foraging speed of the gazelle, r2 is a random number in the range of [0,1], Elite represents the optimal gazelle individual position, R b is the Brownian motion position vector, which can be obtained by the following formula;

[0081]

[0082] In the formula, R b (x,μ,σ 2 ) represents the probability distribution function of the position of the gazelle at this stage, μ=0, σ 2 =1;

[0083] Exploration phase: The position update process of the gazelle in this phase is described as follows:

[0084] x i.p2 =x i +v·μ g ·r3×R l (Elite-R l × i ) (twenty three)

[0085] x i.p2' =x i +v·μ g ·c f ×R b (Elite-R l × i ) (twenty four)

[0086]

[0087] In the formula, x i.p2 and x i.p2' is the updated position of the i-th gazelle individual in the exploration phase, v is the maximum speed of the gazelle individual, μ g is the directional characteristic variable, R l is a Lévy flight step that conforms to the Lévy distribution, r3 is a random number in the range of [0, 1], c f is the cumulative effect of the predator, m and M represent the current iteration number and the maximum iteration number respectively, R l is the Lévy flight step length, α is the control index, u and v are two random variables that obey the normal distribution, and Γ is a function of one variable;

[0088] Introducing the logarithmic inertia weight w in the position update process of the gazelle individual in the development stage helps the algorithm maintain a certain inertia during the iteration process, thereby better balancing the exploration and development links; and adding the Gaussian coefficient of variation κ in the exploration stage can increase the diversity of gazelle individuals in the population and improve the global search ability of the algorithm; where the logarithmic inertia weight w is calculated by the following formula:

[0089] w=(m / M)×(lgw max / lgw min )-lgw max (29)

[0090] Where w is the logarithmic inertia weight, w max and w min is the maximum and minimum value of the inertia weight in the number of iterations, m is the current number of iterations, and M is the maximum number of iterations;

[0091] The improved position update formula of gazelle individuals in the development stage is as follows:

[0092]

[0093] The improved position update formula of gazelle individuals in the exploration phase is as follows:

[0094]

[0095] Differential Evolution Strategy

[0096] Escape phase: Whether the gazelle can successfully escape from the predator is described as:

[0097]

[0098] In the formula, u b and l b Respectively represent the upper and lower limits of the gazelle’s location, x r1 、x r2 are the positions of two random optimal gazelle individuals in the optimal gazelle matrix Elite, r4, r5 and r6 represent random numbers in the range [0,1], psrs is the escape rate, and Q is a binary vector consisting of random numbers in the range [0,1];

[0099] Introduce the differential mutation strategy into the position update of gazelle individuals; the differential evolution strategy includes three stages: mutation, crossover, and selection. The specific operations are as follows:

[0100] First, after the gazelle population completes the position update operation and calculates the corresponding fitness value, three gazelle individuals x are randomly selected from the current population. r3 、x r4 and x r5 , and x will be r3 、x r4 and x r5 Arrange them, and the fitness values ​​of the three are f r3 、f r4 and f r5 ; Finally, the mutant gazelle individual x is generated m The position of , thus completing the variation;

[0101] x m =x r3 +ζ(x r4 -x r5 ) (35)

[0102]

[0103] In the formula, ζ is the mutation rate, x m is the position of the mutant gazelle individual, ζu and l Indicates the upper and lower bounds of the mutation rate;

[0104] After the mutation operation is completed, the cross coefficient CF is introduced and a random number R between 0 and 1 is randomly generated and compared with CF. If R>CF, the individual position of the gazelle after mutation is adopted; if not, the individual position of the gazelle before mutation is kept unchanged; the calculation method of the cross coefficient CF is as follows:

[0105]

[0106] The crossover operation can be described as:

[0107]

[0108] Calculate the fitness values ​​of the new and old gazelle individuals before and after the mutation crossover, compare the fitness of the new and old gazelle individuals to achieve the survival of the fittest, and the selection operation is specifically implemented by the following formula:

[0109]

[0110] In the formula, f(x c ) and f(x i ) represent the fitness values ​​of gazelle individuals before and after mutation crossover.

[0111] Furthermore, the distributed power generation planning model is solved based on the improved gazelle optimization algorithm by inputting the time-series output power, grid parameters and load-related data of the distributed power generation, and the optimal planning result of the port distributed power generation is obtained, which is as follows:

[0112] The sum of the annual comprehensive cost objective function and the annual total voltage deviation objective function is used as the upper optimization objective to determine the installation type, location and capacity of distributed power sources; the distributed power source operation cost in the distributed power source multi-objective planning model is used as the lower optimization objective; the lower optimization is the optimization sub-problem of the distribution network in each scenario;

[0113] Set the parameters of the improved gazelle optimization algorithm, including the number of individuals in the population, the dimension, and the maximum number of iterations. The sum of the annual comprehensive cost objective function and the annual total voltage deviation objective function obtained after correction is used as the fitness function of the parameters of the improved gazelle optimization algorithm.

[0114] Based on the improved gazelle algorithm, the initial gazelle population, i.e. the initial planning scheme, is generated, including: the type, location and capacity of distributed power sources. Then the upper model passes the planning scheme of distributed power sources to the operation sub-problems of the lower model.

[0115] After receiving the planning scheme, the lower-level model calculates the total operating cost of each scenario after the distributed power generation planning based on the solver;

[0116] The operation result is passed to the upper model as a variable to calculate the upper objective function;

[0117] Determine whether the maximum number of iterations has been reached; if the maximum number of model iterations has not been reached, update the planning scheme of the gazelle population, i.e., the distributed power source, and continue to iterate; otherwise, output the optimal planning scheme.

[0118] The port distributed power planning system based on the improved gazelle optimization algorithm includes:

[0119] A first construction module, which constructs a photovoltaic power generation model and a wind power generation model based on photovoltaic equipment operating parameters and conditions and wind equipment operating parameters and wind power in a typical port wind and light time sequence scenario;

[0120] A conversion module, which converts the wind speed and light intensity of typical wind-solar time sequence scenes in each season into a time sequence output power of a distributed power source based on a photovoltaic power generation model and a wind power generation model;

[0121] A determination module, wherein the determination module determines an annual comprehensive cost objective function based on the fixed investment cost, operating cost, power purchase cost from the superior power grid, network loss cost, and compensation cost of the distributed power source;

[0122] An acquisition module, wherein the acquisition module acquires an annual total voltage deviation objective function based on the voltage per unit value of each node at each time in a unit scenario;

[0123] A second construction module, wherein the second construction module constructs a distributed power planning model with annual comprehensive cost objective function, annual total voltage deviation objective function and distributed power operation cost as optimization objectives based on the constraints of the distributed power multi-objective planning model;

[0124] An improvement module, wherein the improvement module improves the performance of the Gazelle optimization algorithm to obtain an improved Gazelle optimization algorithm;

[0125] A solution module is provided, wherein the solution module inputs the time-series output power, grid parameters and load-related data of the distributed power source, solves the distributed power source planning model based on the improved gazelle optimization algorithm, and obtains the optimal planning result of the port distributed power source.

[0126] Compared with the prior art, the present invention has the following beneficial effects:

[0127] The present invention firstly establishes a port distributed power planning model, including constructing the objective function, constraint conditions and conversion model structure of the port distributed power planning model; secondly, it proposes a method for solving the port distributed power planning model based on the improved gazelle optimization algorithm. The present invention significantly improves the overall economic benefits of the power system, reduces operating costs by optimizing resource allocation, and promotes the effective use of green energy. Moreover, by optimizing the layout and operation strategy of distributed power sources, the active power loss of the system is successfully reduced and the energy utilization efficiency is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0128] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0129] Figure 1 It is a flow chart of a method for planning distributed power sources in a port based on an improved gazelle optimization algorithm of the present invention;

[0130] Figure 2 It is a structural schematic diagram of a port distributed power supply planning system based on an improved gazelle optimization algorithm of the present invention;

[0131] Figure 3 It is a general flow chart of the port distributed power planning method based on the improved gazelle optimization algorithm of the present invention;

[0132] Figure 4(a) is a four-season curve diagram after clustering the light intensity of a port in one year;

[0133] Figure 4(b) is a four-season curve diagram after clustering the wind speed of a port in one year;

[0134] Figure 5 It is the Pareto distribution diagram of weights selected by sensitivity modified weight method;

[0135] Figure 6 It is the model architecture diagram of the port distributed power planning model;

[0136] Figure 7 It is an algorithm flow chart of the improved gazelle optimization algorithm proposed by the present invention;

[0137] Figure 8 It is the IEEE69 node system structure diagram;

[0138] Fig. 9 is the load curve before and after the implementation of demand response;

[0139] Fig.10It is a comparison diagram of the total voltage deviation distribution of each node in one day in the summer scenario before and after the installation of distributed power generation in the port. DETAILED DESCRIPTION

[0140] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0141] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0142] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0143] In the description of the embodiments of the present invention, it should be noted that if the terms "upper", "lower", "horizontal", "inner", etc. indicate an orientation or positional relationship based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship in which the product of the invention is usually placed when in use, it is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. In addition, the terms "first", "second", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0144] In addition, if the term "horizontal" appears, it does not mean that the component must be absolutely horizontal, but can be slightly tilted. For example, "horizontal" only means that its direction is more horizontal than "vertical", which does not mean that the structure must be completely horizontal, but can be slightly tilted.

[0145] In the description of the embodiments of the present invention, it is also necessary to explain that, unless otherwise clearly specified and limited, the terms "set", "install", "connect", and "connect" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal connection of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0146] The present invention is further described in detail below in conjunction with the accompanying drawings:

[0147] See also Figure 1 The present invention discloses a port distributed power planning method based on an improved gazelle optimization algorithm, comprising:

[0148] S101, based on the photovoltaic equipment operating parameters and conditions of the typical port wind and solar time sequence scenario and the wind equipment operating parameters and wind power, respectively construct a photovoltaic power generation model and a wind power generation model;

[0149] Construct a photovoltaic power generation model, specifically: Beta distribution is used to model the uncertainty of photovoltaic power generation, and the probability density function is:

[0150]

[0151] Among them, α1 and α2 are the shape parameters of the Beta distribution, which depend on the expectation and variance of the light intensity; S and S max are the actual light intensity and the maximum light intensity respectively;

[0152] The relationship between the active output of photovoltaic power generation equipment and light intensity is:

[0153]

[0154] Among them, P pv and is the active output and rated power of photovoltaic, and the probability density function of photovoltaic active output power is obtained as follows:

[0155]

[0156] Construct a wind power generation model. Specifically, assume that the wind turbine generator set is under constant power control and adopts Weibull distribution. The probability density function is as follows:

[0157]

[0158] Where u represents the actual wind speed; k and β are the shape parameter and scale parameter of the Weibull distribution respectively; the output of the wind turbine is:

[0159]

[0160] Among them, u1, u2 and u e Respectively represent the cut-in wind speed, cut-out wind speed and rated wind speed; P wt is the active power output of the wind turbine at different wind speeds; is the rated power of the fan.

[0161] S102, based on the photovoltaic power generation model and the wind power generation model, converting the wind speed and light intensity of the typical wind-solar time sequence scenes in each season into the time sequence output power of the distributed power source;

[0162] Based on the K-means clustering algorithm, the data of wind speed and light intensity of the port for one year were processed to obtain the typical time series scenes of wind speed and light intensity in each season;

[0163] The wind speed and light intensity data of typical wind and solar time series scenarios in each season are input into the photovoltaic power generation model and the wind power generation model, and the output is the time series output power of wind power and photovoltaic power corresponding to each scenario.

[0164] S103, determining an annual comprehensive cost objective function based on the fixed investment cost, operating cost, power purchase cost from the superior power grid, network loss cost, and compensation cost of the distributed power generation;

[0165] The optimization variables of the model are the capacity and location of the installed distributed generation, which are represented by the following vector:

[0166]

[0167] Determine the annual comprehensive cost objective function. The annual comprehensive cost F1 is expressed as follows:

[0168] F1=f total =f l +f om +f lo +f en +f dr (7)

[0169] Among them, f l is the fixed investment cost of distributed power generation; f om is the operating cost of distributed power generation; lo is the network loss cost; f en The cost of purchasing electricity from the upper power grid; dr To compensate for costs.

[0170] (1) Fixed investment cost of distributed power generation

[0171]

[0172] In the formula, λ wt and λ pv is the coefficient of converting the present value of wind power and photovoltaic power into equivalent annual value; n n The number of nodes to be planned for distributed power generation; and is the rated capacity installed at node i; and is the investment cost per unit capacity; a is the discount rate; y is the useful life of the distributed power source.

[0173] (2) Distributed power generation operating costs

[0174]

[0175] In the formula, and The operating cost per unit of electricity generated by wind power and photovoltaic power; and is the actual active power output of wind power and photovoltaic installed at node i in scenario s; n s is the number of scenarios run by simulation; t s Indicates the total number of running days of scenario s.

[0176] (3) Electricity purchase costs from the upper-level power grid

[0177]

[0178] Where P en.s,t and d s,t They are the active power and real-time electricity price purchased from the superior power grid at time t in scenario s, respectively.

[0179] (4) Network loss costs

[0180] Convert network loss into economic indicators to reduce energy waste. The conversion formula is as follows:

[0181]

[0182] In the formula, c lo P is the conversion cost of active network loss per unit of electricity; l.s Represents the total active network loss in scenario s.

[0183] (5) Compensation costs incurred by implementing demand response

[0184] The transferable load measures guide users to transfer part of the transferable load from the system peak load period to the valley or other non-peak load period through economic incentives. The calculation formula is as follows:

[0185]

[0186] Where P j,S,dr represents the transferable load power at load node j under scenario s; c dr It represents the compensation cost of transferable load per unit of electricity.

[0187] S104, obtaining an annual total voltage deviation objective function based on the voltage per unit value of each node at each time in the unit scenario;

[0188]

[0189] Among them, V s,i,t is the voltage per unit value at the i-th node at time t in scenario s;

[0190] The annual comprehensive cost objective function and the annual total voltage deviation objective function are weighted respectively, and the correction coefficient Q is introduced C and Q v The weighted function is modified to obtain a modified objective function;

[0191] The annual comprehensive cost objective function and the annual total voltage deviation objective function are weighted respectively, as follows:

[0192] minF=wF1+(1-w)F2 (16)

[0193] Introducing correction factor Q C and Q v After correcting formula (16), the objective function is:

[0194] minF=wF1 / Q C +(1-w)F2 / Q V (17)

[0195] Among them, the correction factor Q C and Q v The calculation expression is:

[0196]

[0197] Among them, C ni,max and C n0 are the annual comprehensive costs of the i-th node after it is connected to the distributed power source with the largest capacity and when it is not connected to the distributed power source, V tvd,ni and V tvd,n0 are the total voltage deviations of the i-th node after it is connected to the distributed power source with the maximum capacity and when it is not connected to the distributed power source, is the upper limit of the capacity of distributed generation connected to node i, n DG is the number of nodes connected to the distributed generation.

[0198] S105, based on the constraints of the distributed power multi-objective planning model, construct a distributed power planning model with annual comprehensive cost objective function, annual total voltage deviation objective function and distributed power operation cost as optimization objectives;

[0199] The constraints based on the distributed generation multi-objective planning model are as follows:

[0200] (1) Installed capacity constraints of distributed generation

[0201]

[0202] In the formula, Indicates the maximum capacity of distributed generation allowed to be installed at the node i to be planned; and They represent the maximum allowable installed capacity of wind power and photovoltaic power at the node i to be planned;

[0203] (2) Node voltage limitation

[0204]

[0205] Where U i is the voltage amplitude at node i; and are the upper and lower limits of the node voltage amplitude respectively;

[0206] (3) Power balance constraints

[0207]

[0208] Where P i.p and Q i.s They represent the active power and reactive power injected into node i in scenario s respectively; U i.s and U j.s Respectively represent the voltage amplitude of node i and node j in scenario s; θ ij.s is the phase angle difference between nodes i and j in scene s;

[0209] (4) Branch capacity constraints

[0210]

[0211] Where S j and denote the apparent power and maximum value of line j respectively;

[0212] (5) Distributed power penetration rate constraints

[0213]

[0214] Where P z is the total active load of the distribution network; ζ is the maximum penetration rate of distributed generation allowed for distribution network planning;

[0215] (6) Distributed power generation operation constraints

[0216]

[0217] The types of distributed power sources are wind power and photovoltaic power. is the upper limit of active power output installed in node i scenario s; p i,DG and are the active output removal ratio and its maximum value of the distributed generation installed at node i respectively;

[0218] (7) Load transfer constraints

[0219]

[0220] Where P i,s,dr is the transferable load of node i in scenario s; is its maximum value.

[0221] S106, improving the performance of the Gazelle optimization algorithm to obtain an improved Gazelle optimization algorithm;

[0222] S106.1, population position initialization strategy improvement;

[0223] The position of gazelle population X in the search space is described as:

[0224]

[0225] In the formula, n represents the number of gazelle populations, d represents the dimension of the optimization variable, and x i,j represents the position of the i-th gazelle individual in the j-dimension;

[0226] The individual with the smallest fitness value in the gazelle population is selected as the strongest gazelle and saved in the optimal gazelle matrix. The optimal gazelle matrix Elite is described as:

[0227]

[0228] In the formula, x topi,j represents the search position of the strongest gazelle in the current gazelle population;

[0229] The Tent mapping initialization strategy is introduced to improve the uniformity of the distribution of initial gazelle individuals in the solution space; the Tent mapping is described as follows:

[0230]

[0231] In the formula, x t is the generated chaotic sequence, n is the number of gazelle populations to be initialized, τ is the adjustment parameter of the chaotic sequence, 0<τ<1;

[0232] S106.2, introduce logarithmic inertia weight;

[0233] Development phase: This process assumes that the position of the gazelle individual follows the Brownian motion mode. The position update process of the gazelle individual in the development phase is described as follows:

[0234] x i.p1 =x i +v0·r2×Rb (Elite-R b × i ) (29)

[0235] In the formula, x i.p1 and x i They represent the positions of the i-th gazelle individual before and after the update in the development stage, v0 is the foraging speed of the gazelle, r2 is a random number in the range of [0,1], Elite represents the optimal gazelle individual position, R b is the Brownian motion position vector, which can be obtained by the following formula;

[0236]

[0237] In the formula, R b (x,μ,σ 2 ) represents the probability distribution function of the position of the gazelle at this stage, μ=0, σ 2 =1;

[0238] Exploration phase: The position update process of the gazelle in this phase is described as follows:

[0239] x i.p2 =x i +v·μ g ·r3×R l (Elite-R l × i ) (31)

[0240] x i.p2' =x i +v·μ g ·c f ×R b (Elite-R l × i ) (32)

[0241]

[0242]

[0243] In the formula, x i.p2 and x i.p2' is the updated position of the i-th gazelle individual in the exploration phase, v is the maximum speed of the gazelle individual, μ g is the directional characteristic variable, R l is a Lévy flight step that conforms to the Lévy distribution, r3 is a random number in the range [0,1], c f is the cumulative effect of the predator, m and M represent the current iteration number and the maximum iteration number respectively, R lis the Lévy flight step length, α is the control index, u and v are two random variables that obey the normal distribution, and Γ is a function of one variable;

[0244] Introducing the logarithmic inertia weight w in the position update process of the gazelle individual in the development stage helps the algorithm maintain a certain inertia during the iteration process, thereby better balancing the exploration and development links; and adding the Gaussian coefficient of variation κ in the exploration stage can increase the diversity of gazelle individuals in the population and improve the global search ability of the algorithm; where the logarithmic inertia weight w is calculated by the following formula:

[0245] w=(m / M)×(lgw max / lgw min )-lgw max (37)

[0246] Where w is the logarithmic inertia weight, w max and w min is the maximum and minimum value of the inertia weight in the number of iterations, m is the current number of iterations, and M is the maximum number of iterations;

[0247] The improved position update formula of gazelle individuals in the development stage is as follows:

[0248]

[0249] The improved position update formula of gazelle individuals in the exploration phase is as follows:

[0250]

[0251] S106.3. Differential Evolution Strategy

[0252] Escape phase: Whether the gazelle can successfully escape from the predator is described as:

[0253]

[0254] In the formula, u b and l b Respectively represent the upper and lower limits of the gazelle’s location, x r1 、x r2 are the positions of two random optimal gazelle individuals in the optimal gazelle matrix Elite, r4, r5 and r6 represent random numbers in the range [0,1], psrs is the escape rate, and Q is a binary vector consisting of random numbers in the range [0,1];

[0255] Introduce the differential mutation strategy into the position update of gazelle individuals; the differential evolution strategy includes three stages: mutation, crossover, and selection. The specific operations are as follows:

[0256] First, after the gazelle population completes the position update operation and calculates the corresponding fitness value, three gazelle individuals x are randomly selected from the current population. r3 、x r4 and x r5 , and x will be r3 、x r4 and x r5 Arrange them, and the fitness values ​​of the three are f r3 、f r4 and f r5 ; Finally, the mutant gazelle individual x is generated m The position of , thus completing the variation;

[0257] x m =x r3 +ζ(x r4 -x r5 ) (43)

[0258]

[0259] In the formula, ζ is the mutation rate, x m is the position of the mutant gazelle individual, ζ u and l Indicates the upper and lower bounds of the mutation rate;

[0260] After the mutation operation is completed, the cross coefficient CF is introduced and a random number R between 0 and 1 is randomly generated and compared with CF. If R>CF, the individual position of the gazelle after mutation is adopted; if not, the individual position of the gazelle before mutation is kept unchanged; the calculation method of the cross coefficient CF is as follows:

[0261]

[0262] The crossover operation can be described as:

[0263]

[0264] Calculate the fitness values ​​of the new and old gazelle individuals before and after the mutation crossover, compare the fitness of the new and old gazelle individuals to achieve the survival of the fittest, and the selection operation is specifically implemented by the following formula:

[0265]

[0266] In the formula, f(x c ) and f(x i ) represent the fitness values ​​of gazelle individuals before and after mutation crossover.

[0267] S107, inputting the distributed power generation's time-series output power, grid parameters, load and other related data, solving the distributed power generation planning model based on the improved gazelle optimization algorithm, and obtaining the optimal planning result of the port distributed power generation.

[0268] The sum of the annual comprehensive cost objective function and the annual total voltage deviation objective function is used as the upper optimization objective to determine the installation type, location and capacity of distributed power sources; the distributed power source operation cost in the distributed power source multi-objective planning model is used as the lower optimization objective; the lower optimization is the optimization sub-problem of the distribution network in each scenario;

[0269] Set the parameters of the improved gazelle optimization algorithm, including the number of individuals in the population, the dimension, and the maximum number of iterations. The sum of the annual comprehensive cost objective function and the annual total voltage deviation objective function obtained after correction is used as the fitness function of the parameters of the improved gazelle optimization algorithm.

[0270] Based on the improved gazelle algorithm, the initial gazelle population, i.e. the initial planning scheme, is generated, including: the type, location and capacity of distributed power sources. Then the upper model passes the planning scheme of distributed power sources to the operation sub-problems of the lower model.

[0271] After receiving the planning scheme, the lower-level model calculates the total operating cost of each scenario after the distributed power generation planning based on the solver;

[0272] The operation result is passed to the upper model as a variable to calculate the upper objective function;

[0273] Determine whether the maximum number of iterations has been reached; if the maximum number of model iterations has not been reached, update the planning scheme of the gazelle population, i.e., the distributed power source, and continue to iterate; otherwise, output the optimal planning scheme.

[0274] See also Figure 2 The present invention discloses a port distributed power planning system based on an improved gazelle optimization algorithm, comprising:

[0275] A first construction module, which constructs a photovoltaic power generation model and a wind power generation model based on photovoltaic equipment operating parameters and conditions and wind equipment operating parameters and wind power in a typical port wind and light time sequence scenario;

[0276] A conversion module, which converts the wind speed and light intensity of typical wind-solar time sequence scenes in each season into a time sequence output power of a distributed power source based on a photovoltaic power generation model and a wind power generation model;

[0277] A determination module, wherein the determination module determines an annual comprehensive cost objective function based on the fixed investment cost, operating cost, power purchase cost from the superior power grid, network loss cost, and compensation cost of the distributed power source;

[0278] An acquisition module, wherein the acquisition module acquires an annual total voltage deviation objective function based on the voltage per unit value of each node at each time in a unit scenario;

[0279] A second construction module, wherein the second construction module constructs a distributed power planning model with annual comprehensive cost objective function, annual total voltage deviation objective function and distributed power operation cost as optimization objectives based on the constraints of the distributed power multi-objective planning model;

[0280] An improvement module, wherein the improvement module improves the performance of the Gazelle optimization algorithm to obtain an improved Gazelle optimization algorithm;

[0281] A solution module is provided, wherein the solution module inputs the time-series output power, grid parameters and load-related data of the distributed power source, solves the distributed power source planning model based on the improved gazelle optimization algorithm, and obtains the optimal planning result of the port distributed power source.

[0282] Example:

[0283] See also Figure 3 , Figure 3 A port distributed power planning method based on an improved gazelle optimization algorithm is disclosed.

[0284] Step 1: Establish the output model of wind power and photovoltaic power, collect the data of wind speed and light intensity of a port for one year and use K-means clustering algorithm to obtain the typical scenes of wind speed and light intensity in each season as shown in Figure 4(a) and Figure 4(b) respectively. The typical wind speed and light intensity scenes in each season can be converted into the corresponding time-series output power of distributed power sources through the wind and photovoltaic output model;

[0285] Step 1.1, establish a photovoltaic power generation model according to equations (1)-(3);

[0286] Step 1.2, establish a wind power generation model according to equations (4) and (5);

[0287] Step 1.3, collect the data of wind speed and light intensity in a certain area for one year, and use K-means clustering algorithm to obtain the typical time series scenes of wind speed and light intensity in each season;

[0288] Step 1.4, input the wind speed and light intensity data of typical wind and solar time sequence scenes in each season into the established wind power and photovoltaic time sequence output model, and output the wind power and photovoltaic time sequence output power corresponding to each scene;

[0289] Step 2, propose and apply the sensitivity correction weight method to establish a distributed power generation planning model with annual total cost and power quality as optimization objectives;

[0290] Step 2.1, determine the annual comprehensive cost objective function according to equations (7)-(14);

[0291] Step 2.2, determine the annual total voltage deviation objective function according to formula (15);

[0292] Step 2.3, use the sensitivity modified weight method to obtain a new objective function according to equations (16)-(18), and set different weight values ​​to obtain the Pareto solution set as follows: Figure 5 As shown, the optimal value of weight w is 0.5;

[0293] Step 2.4, determine the constraints of the distributed generation multi-objective planning model according to equations (19)-(25);

[0294] Step 3: Convert the established port distributed power planning model into a double-layer structure as follows: Figure 6 As shown;

[0295] Step 4, improve the performance of the Gazelle Optimization Algorithm to obtain an improved Gazelle Optimization Algorithm. The flow chart of the improved Gazelle Optimization Algorithm is as follows: Figure 7 As shown;

[0296] Step 4.1, improve the population position initialization strategy according to formula (28);

[0297] Step 4.2, introduce the dynamic adaptive factor, whose expression is (37). The position update formulas of the improved gazelle optimization algorithm in the development phase and exploration phase are shown in equations (38)-(40);

[0298] Step 4.3, introduce the differential evolution strategy into the position update of gazelle individuals in the escape phase of the gazelle optimization algorithm. The expression of the differential evolution strategy is (43)-(47);

[0299] The IEEE69 node system is selected to simulate the port power distribution system architecture to test the distributed power planning method based on the improved gazelle optimization algorithm proposed in the present invention. Each node can represent the operation area to be connected to the distributed power supply. The IEEE69 structure diagram is shown in FIG. Figure 8 As shown, the total system load and reference voltage are 5000kW and 12.66kV respectively, the system parameters are default parameters, the types of distributed power sources to be selected are wind power and photovoltaic, the candidate access nodes for wind power access system are set to 41, 53, 61 and 64, the system maximum installed wind power total capacity is set to 1000kW, the candidate access nodes for photovoltaic access to IEEE69 system are 6, 69, 21 and 59, and the maximum installed distributed power capacity of each node is 800kW. The upper limit of the total installed distributed power capacity of the system is set to 2250kW. This study implements time-of-use electricity prices and transferable load measures for all nodes of the system;

[0300] Step 5, input the distributed power generation time sequence output power, grid parameters and load and other related data and set the parameters of the improved gazelle optimization algorithm. Apply the improved gazelle optimization algorithm to solve the distributed power generation planning model and obtain the optimal planning result of the distributed power generation;

[0301] Step 5.1, input the wind speed, light intensity, load data of a port area and the simulated port power distribution system data and set the parameters of the improved gazelle optimization algorithm. The number of individuals, dimension and maximum number of iterations of the population are set to 50, 8 and 100 respectively, and formula (18) is used as the fitness function of the algorithm;

[0302] Step 5.2, the upper layer determines to apply the improved gazelle algorithm to generate the initial gazelle population, i.e., the initial planning scheme (type, location and capacity of distributed power sources), and then the upper layer model passes the planning scheme of distributed power sources to the operation sub-problems of the lower layer;

[0303] Step 5.3: After receiving the planning scheme, the lower-level model uses the CPLEX solver to calculate the total operating cost C of each scenario after the distributed power planning. om ;

[0304] Step 5.4, the operation result is passed to the upper model as a variable to calculate the upper objective function;

[0305] Step 5.5, determine whether the maximum number of iterations is 100. If the maximum number of model iterations is not reached, update the planning scheme of the gazelle population, i.e., the distributed power generation, and continue to iterate. Otherwise, output the optimal planning scheme.

[0306] In addition, in an embodiment of the present invention, in order to better demonstrate the good performance of the proposed improved gazelle optimization algorithm, several classic methods such as particle swarm optimization algorithm (PSO), gray wolf optimization algorithm (GWO), whale optimization algorithm (WOA), and dwarf mongoose optimization algorithm (DMO) and arithmetic optimization algorithm (AOA) used in the latest literature are selected for comparison with the IGOA algorithm. The number of individuals in all algorithm populations is set to 50, and the number of iterations is 100 times. Table 1 shows the solution results of the model constructed by the present invention using different algorithms. It can be seen from Table 1 that the objective function value obtained by the improved gazelle optimization algorithm is the smallest, and the results obtained include the minimum annual comprehensive cost and total voltage deviation. The proposed improved gazelle optimization algorithm has the most superior solution performance.

[0307] Table 1 Model running results of different algorithms

[0308]

[0309]

[0310] The time-of-use electricity price and transferable load considered in the present invention enable users to voluntarily change their electricity consumption plans through price and economic incentives respectively, thereby reducing electricity consumption during peak periods of the port power grid and increasing electricity consumption during valley periods, playing a role in peak shaving and valley filling, and improving the safety margin and stability of the power supply of the power grid. Fig. 9 The load curves before and after the implementation of demand response. It can be clearly seen from the figure that after the implementation of demand response, the maximum value of the system load is reduced, the power supply safety margin is improved, the peak-to-valley difference of the system load curve is significantly reduced, and the system stability is significantly improved.

[0311] In the embodiment of the present invention, in order to demonstrate the effectiveness of the constructed port distributed power planning model, the annual comprehensive cost and total voltage deviation of distributed power access are taken as optimization targets, thereby improving the economy and power quality of the system. The smaller the annual comprehensive cost and total voltage deviation, the better the planning scheme obtained by the proposed algorithm and planning model.

[0312] Three different cases are set up for comparative experiments to verify the effectiveness of the proposed distributed power generation planning method.

[0313] Case 1: No distributed generation is installed and no demand response is considered.

[0314] Case 2: Using the proposed model and algorithm to install distributed generation in the system but not implementing demand response.

[0315] Case 3: Use the proposed model and algorithm to install distributed generation in the system and implement demand response.

[0316] The results under the three scenarios are shown in Table 2:

[0317] Table 2 Results under three scenarios

[0318]

[0319] It can be seen from Table 2 that case 1 is the operation result before the system is connected to distributed power sources. The values ​​of annual comprehensive cost and total voltage deviation are the maximum values ​​compared with other cases, and the economy and power quality of the system are the worst. Compared with case 1, case 2 installed distributed power sources but did not consider demand response. Its annual comprehensive cost was reduced by 25.67%, and the total voltage deviation was reduced by 33.90%. Compared with the initial case 1, the annual comprehensive cost of case 3, which installed distributed power sources and considered demand response, was reduced by 24.61%, and the total voltage deviation was reduced by 34.93%. Case 3 implemented load transfer measures and required compensation for power users. Compared with case 2, case 3 increased the compensation cost and annual comprehensive cost, but reduced the total voltage deviation, and improved the power quality and reliability of the system. The results show that the proposed algorithm and model can effectively determine the reasonable scheme for the access of distributed power sources to the system, and significantly reduce the annual comprehensive cost and total voltage deviation of the system.

[0320] Fig.10 The total voltage deviation distribution of each node in the summer scene in two scenarios is compared within one day. In the original system case 1, the overall voltage distribution fluctuates too much, and the sum of the voltage deviations of nodes 60 to 64 even exceeds 1.8pu, and the power quality is very low. After the distributed power supply is installed in case 2 and case 3, the node voltage deviation is significantly reduced, among which the power quality of nodes 5 to 26 and 49 to 70 is improved most significantly, the number of nodes exceeding the voltage limit is reduced, and the system stability is improved. Considering the demand response, case 3 is compared with case 2. The node voltage deviation value is reduced, the overall voltage distribution of the system is more stable, and the power quality is greatly improved. In summary, the reasonable installation of distributed power supplies in the system can improve the node voltage, improve the power quality and stability of the system, and the implementation of demand response can have a beneficial impact on the planning results.

[0321] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A port distributed power planning method based on the improved gazelle optimization algorithm is characterized by: include: Based on the operating parameters and conditions of photovoltaic equipment and the operating parameters and wind power of wind equipment in typical port wind and solar time sequence scenarios, photovoltaic power generation model and wind power generation model are constructed respectively; Based on the photovoltaic power generation model and wind power generation model, the wind speed and light intensity of typical wind and solar time sequence scenes in each season are converted into the time sequence output power of distributed power sources; Based on the fixed investment cost, operating cost, electricity purchase cost from the superior power grid, network loss cost and compensation cost of distributed power generation, determine the annual comprehensive cost objective function; Based on the voltage per unit value of each node at each time in the unit scenario, the annual total voltage deviation objective function is obtained; Based on the constraints of the distributed power multi-objective planning model, a distributed power planning model is constructed with the annual comprehensive cost objective function, the annual total voltage deviation objective function and the distributed power operation cost as the optimization objectives; Improve the performance of the Gazelle optimization algorithm and obtain an improved Gazelle optimization algorithm; The distributed power generation planning model is solved based on the improved gazelle optimization algorithm by inputting the distributed power generation's sequential output power, grid parameters and load-related data to obtain the optimal planning result of the port's distributed power generation.

2. The port distributed power planning method based on the improved gazelle optimization algorithm according to claim 1 is characterized in that: The photovoltaic power generation model and the wind power generation model are constructed based on the photovoltaic equipment operating parameters and conditions and the wind equipment operating parameters and wind power of the typical port wind and solar time sequence scene, respectively, specifically: Construct a photovoltaic power generation model, specifically: Beta distribution is used to model the uncertainty of photovoltaic power generation, and the probability density function is: Among them, α1 and α2 are the shape parameters of the Beta distribution, which depend on the expectation and variance of the light intensity; S and S max are the actual light intensity and the maximum light intensity respectively; The relationship between the active output of photovoltaic power generation equipment and light intensity is: Among them, P pv and is the active output and rated power of photovoltaic, and the probability density function of photovoltaic active output power is obtained as follows: Construct a wind power generation model. Specifically, assume that the wind turbine generator set is under constant power control and adopts Weibull distribution. The probability density function is as follows: Where u represents the actual wind speed; k and β are the shape parameter and scale parameter of the Weibull distribution respectively; the output of the wind turbine is: Among them, u1, u2 and u e Respectively represent the cut-in wind speed, cut-out wind speed and rated wind speed; P wt is the active power output of the wind turbine at different wind speeds; is the rated power of the fan.

3. The port distributed power planning method based on the improved gazelle optimization algorithm according to claim 2 is characterized in that: The method converts the wind speed and light intensity of typical wind-solar time sequence scenes in each season into the time sequence output power of the distributed power source based on the photovoltaic power generation model and the wind power generation model, specifically: Based on the K-means clustering algorithm, the data of wind speed and light intensity of the port for one year were processed to obtain the typical time series scenes of wind speed and light intensity in each season; The wind speed and light intensity data of typical wind and solar time series scenarios in each season are input into the established wind power and photovoltaic time series output model, and the output is the wind power and photovoltaic time series output power corresponding to each scenario.

4. The port distributed power planning method based on the improved gazelle optimization algorithm according to claim 3 is characterized in that: The annual comprehensive cost objective function is determined based on the fixed investment cost, operating cost, power purchase cost from the superior power grid, network loss cost, and compensation cost of the distributed power generation, which is specifically: F1=f total =f l +f om +f lo +f en +f dr (6) Among them, f l is the fixed investment cost; f om is the operating cost; lo is the network loss cost; f en The cost of purchasing electricity from the upper power grid; dr To compensate for costs.

5. The port distributed power planning method based on the improved gazelle optimization algorithm according to claim 4 is characterized in that: The annual total voltage deviation objective function is obtained based on the voltage per unit value of each node at each time in the unit scenario, which is specifically: Among them, V s,i,t is the voltage per unit value at the i-th node at time t in scenario s; The annual comprehensive cost objective function and the annual total voltage deviation objective function are weighted respectively, and the correction coefficient Q is introduced C and Q v The weighted function is modified to obtain a modified objective function; The annual comprehensive cost objective function and the annual total voltage deviation objective function are weighted respectively, as follows: min F=wF1+(1-w)F2 (8) Introducing correction factor Q C and Q v After correcting formula (8), the objective function is: min F=wF1 / Q C +(1-w)F2 / Q V (9) Among them, the correction factor Q C and Q v The calculation expression is: Among them, C ni,max and C n0 are the annual comprehensive costs of the i-th node after it is connected to the distributed power source with the largest capacity and when it is not connected to the distributed power source, V tvd,ni and V tvd,n0 are the total voltage deviations of the i-th node after it is connected to the distributed power source with the maximum capacity and when it is not connected to the distributed power source, is the upper limit of the capacity of distributed generation connected to node i, n DG is the number of nodes connected to the distributed generation.

6. The port distributed power planning method based on the improved gazelle optimization algorithm according to claim 5 is characterized in that: The constraints based on the distributed power generation multi-objective planning model are specifically: (1) Installed capacity constraints of distributed generation In the formula, Indicates the maximum capacity of distributed generation allowed to be installed at the node i to be planned; and They represent the maximum allowable installed capacity of wind power and photovoltaic power at the node i to be planned; (2) Node voltage limitation Where U i is the voltage amplitude at node i; and are the upper and lower limits of the node voltage amplitude respectively; (3) Power balance constraints Where P i.p and Q i.s They represent the active power and reactive power injected into node i in scenario s respectively; U i.s and U j.s Respectively represent the voltage amplitudes of nodes i and j in scenario s; θ ij.s is the phase angle difference between nodes i and j in scene s; (4) Branch capacity constraints Where S j and denote the apparent power and maximum value of line j respectively; (5) Distributed power penetration rate constraints Where P z is the total active load of the distribution network; ζ is the maximum penetration rate of distributed generation allowed for distribution network planning; (6) Distributed power generation operation constraints The types of distributed power sources are wind power and photovoltaic power. is the upper limit of active power output installed in node i scenario s; p i,DG and are the active output removal ratio and its maximum value of the distributed generation installed at node i respectively; (7) Load transfer constraints Where P i,s,dr is the transferable load of node i in scenario s; is its maximum value.

7. The port distributed power planning method based on the improved gazelle optimization algorithm according to claim 6 is characterized in that: The performance of the Gazelle optimization algorithm is improved to obtain an improved Gazelle optimization algorithm, specifically: 7.

1. Improvement of population position initialization strategy; The position of gazelle population X in the search space is described as: In the formula, n represents the number of gazelle populations, d represents the dimension of the optimization variable, and x i,j represents the position of the i-th gazelle individual in the j-dimension; The individual with the smallest fitness value in the gazelle population is selected as the strongest gazelle and saved in the optimal gazelle matrix. The optimal gazelle matrix Elite is described as: In the formula, x topi,j Indicates the search position of the strongest gazelle in the current gazelle population; The Tent mapping initialization strategy is introduced to improve the uniformity of the distribution of initial gazelle individuals in the solution space; the Tent mapping is described as follows: In the formula, x t is the generated chaotic sequence, n is the number of gazelle populations to be initialized, τ is the adjustment parameter of the chaotic sequence, 0<τ<1; 7.

2. Introducing logarithmic inertia weight; Development phase: This process assumes that the position of the gazelle individual follows the Brownian motion mode. The position update process of the gazelle individual in the development phase is described as follows: x i.p1 =x i +v0·r2×R b (Elite-R b ×x i ) (21) In the formula, x i.p1 and x i They represent the positions of the i-th gazelle individual before and after the update in the development stage, v0 is the foraging speed of the gazelle, r2 is a random number in the range of [0,1], Elite represents the optimal gazelle individual position, R b is the Brownian motion position vector, which can be obtained by the following formula; In the formula, R b (x,μ,σ 2 ) represents the probability distribution function of the position of the gazelle at this stage, μ=0, σ 2 =1; Exploration phase: The position update process of the gazelle in this phase is described as follows: x i.p2 =x i +v·μ g ·r3×R l (Elite-R l ×x i ) (23) x i.p2' =x i +v·μ g ·c f ×R b (Elite-R l ×x i ) (24) In the formula, x i.p2 and x i.p2' is the updated position of the i-th gazelle individual in the exploration phase, v is the maximum speed of the gazelle individual, μ g is the directional characteristic variable, R l is a Lévy flight step that conforms to the Lévy distribution, r3 is a random number in the range of [0, 1], c f is the cumulative effect of the predator, m and M represent the current iteration number and the maximum iteration number respectively, R l is the Lévy flight step length, α is the control index, u and v are two random variables that obey the normal distribution, and Γ is a function of one variable; Introducing the logarithmic inertia weight w in the position update process of the gazelle individual in the development stage helps the algorithm maintain a certain inertia during the iteration process, thereby better balancing the exploration and development links; and adding the Gaussian coefficient of variation κ in the exploration stage can increase the diversity of gazelle individuals in the population and improve the global search ability of the algorithm; where the logarithmic inertia weight w is calculated by the following formula: w=(m / M)×(lg w max / lg w min )-lg w max (29) Where w is the logarithmic inertia weight, w max and w min is the maximum and minimum value of the inertia weight in the number of iterations, m is the current number of iterations, and M is the maximum number of iterations; The improved position update formula of gazelle individuals in the development stage is as follows: The improved position update formula of gazelle individuals in the exploration phase is as follows: 7.3 Differential Evolution Strategy Escape phase: Whether the gazelle can successfully escape from the predator is described as: In the formula, u b and l b Respectively represent the upper and lower limits of the gazelle’s location, x r1 、x r2 are the positions of two random optimal gazelle individuals in the optimal gazelle matrix Elite, r4, r5 and r6 represent random numbers in the range [0,1], psrs is the escape rate, and Q is a binary vector consisting of random numbers in the range [0,1]; Introduce the differential mutation strategy into the position update of gazelle individuals; the differential evolution strategy includes three stages: mutation, crossover, and selection. The specific operations are as follows: First, after the gazelle population completes the position update operation and calculates the corresponding fitness value, three gazelle individuals x are randomly selected from the current population. r3 、x r4 and x r5 , and x will be r3 、x r4 and x r5 Arrange them, and the fitness values ​​of the three are f r3 、f r4 and f r5 ; Finally, the mutant gazelle individual x is generated m The position of , thus completing the variation; x m =x r3 +ζ(x r4 -x r5 ) (35) In the formula, ζ is the mutation rate, x m is the position of the mutant gazelle individual, ζ u and l Indicates the upper and lower bounds of the mutation rate; After the mutation operation is completed, the cross coefficient CF is introduced and a random number R between 0 and 1 is randomly generated and compared with CF. If R>CF, the individual position of the gazelle after mutation is adopted; if not, the individual position of the gazelle before mutation is kept unchanged; the calculation method of the cross coefficient CF is as follows: The crossover operation can be described as: Calculate the fitness values ​​of the new and old gazelle individuals before and after the mutation crossover, compare the fitness of the new and old gazelle individuals to achieve the survival of the fittest, and the selection operation is specifically implemented by the following formula: In the formula, f(x c ) and f(x i ) represent the fitness values ​​of gazelle individuals before and after mutation crossover.

8. The port distributed power planning method based on the improved gazelle optimization algorithm according to claim 7 is characterized in that: The input distributed power source time output power, grid parameters and load related data are used to solve the distributed power source planning model based on the improved gazelle optimization algorithm to obtain the optimal planning result of the port distributed power source, which is specifically: The sum of the annual comprehensive cost objective function and the annual total voltage deviation objective function is used as the upper optimization objective to determine the installation type, location and capacity of distributed power sources; the distributed power generation operation cost in the distributed power generation multi-objective planning model is used as the lower optimization objective; The optimization at the lower level is the optimization sub-problem of the distribution network in each scenario; Set the parameters of the improved gazelle optimization algorithm, including the number of individuals in the population, the dimension, and the maximum number of iterations. The sum of the annual comprehensive cost objective function and the annual total voltage deviation objective function obtained after correction is used as the fitness function of the parameters of the improved gazelle optimization algorithm. Based on the improved gazelle algorithm, the initial gazelle population, i.e. the initial planning scheme, is generated, including: the type, location and capacity of distributed power sources. Then the upper model passes the planning scheme of distributed power sources to the operation sub-problems of the lower model. After receiving the planning scheme, the lower-level model calculates the total operating cost of each scenario after the distributed power generation planning based on the solver; The operation result is passed to the upper model as a variable to calculate the upper objective function; Determine whether the maximum number of iterations has been reached; if the maximum number of model iterations has not been reached, update the planning scheme of the gazelle population, i.e., the distributed power source, and continue to iterate; otherwise, output the optimal planning scheme.

9. The port distributed power planning system based on the improved gazelle optimization algorithm is characterized by: include: A first construction module, which constructs a photovoltaic power generation model and a wind power generation model based on photovoltaic equipment operating parameters and conditions and wind equipment operating parameters and wind power in a typical port wind and light time sequence scenario; A conversion module, which converts the wind speed and light intensity of typical wind-solar time sequence scenes in each season into a time sequence output power of a distributed power source based on a photovoltaic power generation model and a wind power generation model; A determination module, wherein the determination module determines an annual comprehensive cost objective function based on the fixed investment cost, operating cost, power purchase cost from the superior power grid, network loss cost, and compensation cost of the distributed power source; An acquisition module, wherein the acquisition module acquires an annual total voltage deviation objective function based on the voltage per unit value of each node at each time in a unit scenario; A second construction module, wherein the second construction module constructs a distributed power planning model with annual comprehensive cost objective function, annual total voltage deviation objective function and distributed power operation cost as optimization objectives based on the constraints of the distributed power multi-objective planning model; An improvement module, wherein the improvement module improves the performance of the Gazelle optimization algorithm to obtain an improved Gazelle optimization algorithm; A solution module is provided, wherein the solution module inputs the time-series output power, grid parameters and load-related data of the distributed power source, solves the distributed power source planning model based on the improved gazelle optimization algorithm, and obtains the optimal planning result of the port distributed power source.

Images