Photovoltaic bearing capacity improving method and system considering seasonal network optimization

Abstract

The invention discloses a photovoltaic bearing capacity improving method and system considering seasonal network optimization. The method comprises the following steps: acquiring a plurality of seasonal scene sets; constructing a power distribution network seasonal network reconstruction optimization model for photovoltaic bearing capacity improvement; for each season, based on the improved whale algorithm, generating a multi-group power distribution network switch switching-on and switching-off scheme in each season; solving the reconstruction optimization model in the target season, and determining respective total target function values of the multiple sets of power grid switch on-off schemes in the target season; and based on an improved whale algorithm, obtaining a current optimal population, determining a switch on-off scheme enabling an objective function value to be optimal in a target season, obtaining a network reconstruction optimization scheme of the power distribution network in the target season, and sequentially solving until network reconstruction optimization schemes of all seasons are obtained. According to the invention, the technical problem of low photovoltaic bearing capacity in a power distribution network system is solved, and the photovoltaic bearing capacity can be improved through a seasonal network optimization and reconstruction means in the operation stage of the power distribution network.

Description

Technical Field

[0001] The present invention relates to the technical field of power distribution networks, and in particular to a method and system for improving photovoltaic carrying capacity by considering seasonal network optimization. Background Art

[0002] The construction of a new distribution system has made it an inevitable trend to connect a high proportion of distributed photovoltaic power to the distribution network. The continuous increase in the scale of distributed photovoltaic installed capacity has highlighted the problems that restrict the access of distributed photovoltaic power in some areas, such as weak network structure, excessive voltage deviation, and limited reverse power. In order to ensure the coordinated development of distributed photovoltaic power and distribution network, it is urgent to take various effective measures to improve the carrying capacity of distribution network for distributed photovoltaic power. Distributed photovoltaic carrying capacity refers to the maximum distributed photovoltaic capacity that can be added to the distribution network under the premise of ensuring safe and stable operation of the system and the fact that various electrical quantity indicators do not exceed the limit. The network structure of the distribution network is complex, with a large number of contact switches and section switches in the system. Network optimization and reconstruction achieves the goal of adjusting the network topology and optimizing the system operation status by opening or closing each switch. For areas with weak carrying capacity of distributed photovoltaic access to the distribution network, the carrying capacity can be improved from a planning perspective by replacing the distribution equipment. However, for a small number of scenarios where the carrying capacity is limited due to reverse flow overload and voltage over-limit, the potential for carrying capacity improvement can be explored through network optimization and reconstruction during the operation stage of the distribution network, which will reduce the investment in blindly replacing distribution equipment to a certain extent.

[0003] In terms of the methods of network reconstruction of distribution networks, most of the current research aims to reduce network losses and balance loads through static network reconstruction of a certain time section or dynamic network reconstruction of multiple switches in a day. However, in the actual operation of the distribution network, it is not realistic to complete dynamic network reconstruction by operating switches multiple times in a day, or even to complete static network reconstruction by switching operations in each period, due to the requirements of safe operation of the power grid. Grid companies usually complete a round of switching operations in a quarter or even longer period to reconstruct the network and eliminate heavy overload. In terms of enhancing the distribution network's acceptance capacity for distributed photovoltaics through network reconstruction, existing research also mainly takes one day as a cycle, and maximizes the distributed photovoltaic absorption capacity through dynamic network optimization and reconstruction, which is less feasible in practice. In order to economically and feasibly improve the distributed photovoltaic carrying capacity of the distribution network, it is urgent to study the photovoltaic carrying capacity improvement method considering seasonal network reconstruction optimization, so as to provide practical and feasible technical means for promoting the high-quality development of distributed photovoltaics.

[0004] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0005] The embodiments of the present invention provide a method and system for improving photovoltaic carrying capacity taking into account seasonal network optimization, so as to at least solve the technical problem of low photovoltaic carrying capacity in the distribution network system.

[0006] According to one aspect of an embodiment of the present invention, a method for improving photovoltaic carrying capacity considering seasonal network optimization is provided, comprising: a first step: obtaining multiple seasonal scenario sets of photovoltaic distribution networks corresponding to multiple seasons, wherein the multiple seasonal scenario sets respectively include photovoltaic output data and load power data of the photovoltaic distribution network in the corresponding season; a second step: taking the maximum photovoltaic carrying capacity, the minimum seasonal average network loss, and the optimal seasonal load balancing under different probability levels as the goals, and taking the lower limit constraint of the photovoltaic carrying capacity probability level, the radial topology structure constraint of the distribution network, and the upper limit constraint of the number of switch operations as constraints, a seasonal network reconstruction optimization model of the photovoltaic distribution network is constructed; a third step: for each season, based on the improved whale algorithm, each Multiple distribution network switch separation and combination schemes under different seasons; the fourth step: input multiple seasonal scenario sets into the seasonal network reconstruction optimization model respectively, solve the seasonal network reconstruction optimization model under the target season, and determine the total objective function value of each of the multiple distribution network switch separation and combination schemes under the target season; the fifth step: based on the improved whale algorithm, according to the total objective function value of each of the multiple distribution network switch separation and combination schemes under the target season, obtain the current optimal switch separation and combination scheme, iteratively update the multiple distribution network switch separation and combination schemes, and after reaching the predetermined maximum number of iterations, determine the switch separation and combination scheme that makes the objective function value optimal in the target season, and obtain the distribution network reconstruction optimization plan under the target season, and solve it in sequence until the distribution network reconstruction optimization plan for all seasons is obtained.

[0007] Optionally, the total objective function is a normalized function of the first objective function, the second objective function and the third objective function, the first objective function is to maximize the photovoltaic carrying capacity under different probability levels, the second objective function is to minimize the seasonal average network loss, and the third objective function is to optimize the seasonal load balance.

[0008] Optionally, multiple seasonal scenario sets are respectively input into the seasonal network reconstruction optimization model, the seasonal network reconstruction optimization model is solved in the target season, and the total objective function value of each of the multiple distribution network switch separation and combination schemes in the target season is determined, including: using the following method to determine the value of the total objective function in the seasonal network reconstruction optimization model under the target distribution network switch separation and combination strategy, to determine the value of the total objective function corresponding to the multiple distribution network switch separation and combination schemes: under the target distribution network switch separation and combination scheme, determine the photovoltaic carrying capacity, network loss and load balance of each scenario in the target seasonal scenario set; sort the photovoltaic carrying capacity of the scenarios in the target seasonal scenario set to determine the photovoltaic carrying capacity under multiple percentiles; determine the probability evaluation model of the photovoltaic carrying capacity based on the photovoltaic carrying capacity under multiple percentiles, wherein the mathematical model of the probability evaluation model of the photovoltaic carrying capacity is shown in the following formula:

[0009] S ε,i =max{S α |Pr{S α ≤S M,i (t)}≥ε α}t∈Ω;

[0010] Where: ε α S is the percentile of the distributed photovoltaic access distribution network carrying capacity assessment; α To meet the probability level ε α Distributed photovoltaic capacity that can be connected, MW; S M,i (t) is the maximum accessible capacity of distributed photovoltaic on bus i in scenario t, MW, where t=1,2,...,N S , is the t-th sampled scene; Ω is the complete set of scenes; S ε,i The busbar i is at percentile ε α The maximum load capacity at the time, MW; according to the probability assessment model of photovoltaic load capacity, the specified ε is determined α The photovoltaic carrying capacity under the probability level; according to the average grid loss, average load balance and ε of the target seasonal scenario set α The photovoltaic carrying capacity under the probability level is used to determine the value of the total objective function under the target distribution network switch separation and combination scheme.

[0011] Optionally, the formula of the first objective function f1 is expressed as:

[0012]

[0013] Where: S α To meet the probability level ε α The distributed photovoltaic capacity that can be connected under S M,i (t) is the maximum accessible capacity of distributed photovoltaic on bus i under scenario t, including the power data of the target season, where t=1,2,...,NS , Ω is the scene set including the power data of the target season; N S The number of scenarios included in the power data for the target season; ε α N is the probability level of distributed photovoltaic access to the distribution network carrying capacity; PV is the total number of nodes whose photovoltaic carrying capacity needs to be improved.

[0014] Optionally, the formula of the second objective function f2 is expressed as:

[0015]

[0016] Where: N L is the total number of branches in the PV distribution network; n is any branch; t is any scenario included in the power data of the target season; N S is the number of scenarios included in the power data of the target season; the switch state variable on the nth branch is k n , used to indicate whether the switch is closed or open; r n is the resistance value of branch n; P n (t) is the active power of branch n in scenario t; Q n (t) is the reactive power of branch n in scenario t; V n (t) is the voltage at the end of branch n in scenario t.

[0017] Optionally, the formula of the third objective function f3 is expressed as:

[0018]

[0019] Where: N L is the total number of branches in the PV distribution network; n is any branch; t is any scenario included in the power data of the target season; N S is the number of scenarios included in the power data of the target season; the switch state variable on the nth branch is k n , used to indicate whether the switch is closed or open; P n (t) is the active power of branch n in scenario t; P n,max is the maximum transmission power allowed for branch n.

[0020] Optionally, the seasonal network reconfiguration optimization model includes the following constraints: a lower limit constraint on the probability level of photovoltaic carrying capacity, a radial topology constraint on the distribution network, an upper limit constraint on the number of switch operations, a power flow equation constraint, a node voltage constraint, and an upper limit constraint on the reverse power flow of a branch; wherein the lower limit constraint on the probability level of photovoltaic carrying capacity is expressed by the following formula:

[0021] ε α ≥ε min

[0022] Where: ε α is the probability level of distributed photovoltaic access to the distribution network carrying capacity; ε min is the minimum limit of the probability level of PV carrying capacity;

[0023] The radial topology constraint of the distribution network is expressed by the following formula:

[0024] g∈G

[0025] Where: g is the network structure of the distribution network; G is the set of all radial networks;

[0026] The upper limit constraint of the number of switch operations is expressed by the following formula:

[0027]

[0028] Where: k n,past is the switch state variable on the nth branch before optimization; k n,now is the switch state variable on the nth branch after optimization; k max It is the limit of the number of times the main switch is opened and closed;

[0029] The power flow equality constraint is expressed as follows:

[0030]

[0031] Where: P i (t) and Q i (t) are the active power and reactive power injected by node i in scenario t, MW and MVar respectively; U i (t) is the voltage amplitude of node i in scenario t, V; Ω i is the set of nodes connected to node i; G ij and B ij is the element of the admittance matrix; θ ij is the phase angle difference between nodes i and j;

[0032] The node voltage constraint is expressed by the following formula:

[0033] U i,min ≤U i (t)≤U i,max (t=1,2,3,...,Ns)

[0034] Where: U i (t) is the voltage amplitude of node i in scenario t, V; U i,max and U i,min are the upper and lower limits of the voltage at node i, V, which are 105% and 95% of the rated voltage respectively;

[0035] The upper limit constraint of the branch reverse flow is expressed by the following formula:

[0036] |P n (t)|≤P n,max (t=1,2,3,...,Ns)

[0037] Where: P n (t) is the reverse active power flow of branch n in scenario t; P n,max is the maximum transmission power limit of branch n, MW, which is taken as 80% of the maximum transmission power of the line and transformer equipment.

[0038] Optionally, multiple seasonal scenario sets of the photovoltaic distribution network corresponding to multiple seasons are obtained, including: obtaining the light intensity time series historical data and the load time series historical data of the photovoltaic distribution network; dividing the light intensity time series historical data and the load time series historical data into multiple groups according to the season; establishing a photovoltaic output beta distribution random model for each of the multiple seasons and a load power normal distribution random model for each of the multiple seasons according to the light intensity time series historical data and the load time series historical data corresponding to the multiple seasons; and sampling the photovoltaic output beta distribution random model for each of the multiple seasons and the load power normal distribution random model for each of the multiple seasons using the Latin hypercube sampling method to determine multiple seasonal scenario sets.

[0039] Optionally, based on the improved whale algorithm, according to the total objective function value of each of the multiple groups of distribution network switch separation and combination schemes, the current optimal population is obtained, the multiple groups of distribution network switch separation and combination schemes are iteratively updated, and after reaching a predetermined maximum number of iterations, the distribution network network reconstruction optimization scheme is determined, including: obtaining multiple groups of distribution network switch separation and combination schemes, setting the maximum number of iterations to M max , the initial number of iterations is m = 1, where the total number of the multiple distribution network switch separation and combination schemes is N; calculate the total objective function value of each switch separation and combination scheme in the multiple distribution network switch separation and combination schemes, update A and C according to the following formula, and generate a random number p on [0,1]:

[0040] A=2a·ra

[0041] C=2·r

[0042] Where A and C are vector coefficients, a decreases linearly from 2 to 0 during iteration; r is a random number between 0 and 1;

[0043] If the random number p ≥ 0.5, then according to Update the switch opening and closing scheme; if the random number p<0.5, determine whether the value of |A| is greater than 1. If |A|≥1, follow the formula Update the switch opening and closing scheme. If |A|<1, then follow the formula Update the switch separation and combination scheme, where m is the current iteration number; || represents the distance between two sets of vectors in the solution space, which is a non-negative number; b is a constant representing the spiral shape; l is a random number between -1 and 1; is the random position distance between the current switch separation scheme and the optimal switch separation scheme in the solution space, which is affected by the random number C; is the i-th switch opening and closing scheme of the m-th generation; is the optimal switch separation and combination scheme of the mth generation, which will be updated with iterations. is the random switch separation and combination scheme set; calculate the total objective function value of each updated switch separation and combination scheme, update the optimal objective function value and the optimal switch separation and combination scheme; determine whether the number of iterations m is equal to M max If not, re-update A and C and generate a random number p on [0,1] and the following steps until m equals M max Established, output the optimal objective function value and the optimal switch separation and combination scheme.

[0044] According to another aspect of an embodiment of the present invention, a photovoltaic carrying capacity enhancement system considering seasonal network optimization is also provided, including: an acquisition module, used to acquire multiple seasonal scenario sets of photovoltaic distribution networks corresponding to multiple seasons, wherein the multiple seasonal scenario sets respectively include photovoltaic output data and load power data of the photovoltaic distribution network in the corresponding season; a construction module, used to maximize photovoltaic carrying capacity, minimize seasonal average network loss, and optimize seasonal load balancing under different probability levels, and to construct a seasonal network reconstruction optimization model for the photovoltaic distribution network with the lower limit constraint of the photovoltaic carrying capacity probability level, the radial topology structure constraint of the distribution network, and the upper limit constraint of the number of switch operations as constraints; a switch separation and combination scheme generation module, used to generate a seasonal network reconstruction optimization model for the photovoltaic distribution network based on the improved whale algorithm for each season. Generate multiple groups of distribution network switch separation and combination schemes for each season; a reconstruction optimization module is used to input multiple seasonal scenario sets into the seasonal network reconstruction optimization model respectively, solve the seasonal network reconstruction optimization model in the target season, and determine the total objective function value of each of the multiple groups of distribution network switch separation and combination schemes in the target season; an optimal solution determination module is used to obtain the current optimal switch separation and combination scheme based on the improved whale algorithm according to the total objective function value of each of the multiple groups of distribution network switch separation and combination schemes in the target season, iteratively update the multiple groups of distribution network switch separation and combination schemes, and after reaching a predetermined maximum number of iterations, determine the switch separation and combination scheme that makes the objective function value optimal in the target season, obtain the distribution network network reconstruction optimization scheme in the target season, and solve the distribution network network reconstruction optimization schemes of all seasons in sequence.

[0045] In an embodiment of the present invention, multiple seasonal scenario sets of photovoltaic distribution networks corresponding to multiple seasons are obtained, wherein the multiple seasonal scenario sets respectively include photovoltaic output data and load power data of the photovoltaic distribution network in the corresponding season; with the maximum photovoltaic carrying capacity, the minimum seasonal average network loss and the best seasonal load balance under different probability levels as the goals, and with the lower limit constraint of the photovoltaic carrying capacity probability level, the radial topology structure constraint of the distribution network and the upper limit constraint of the number of switch operations as constraints, a seasonal network reconstruction optimization model of the photovoltaic distribution network is constructed; for each season, based on the improved whale algorithm, multiple groups of distribution network switch separation and combination schemes are generated in each season; the multiple seasonal scenario sets are respectively input into the seasonal network reconstruction optimization model, and the seasonal network reconstruction optimization model in the target season is constructed. The model is solved to determine the total objective function value of each of the multiple distribution network switch separation and combination schemes in the target season; based on the improved whale algorithm, the current optimal switch separation and combination scheme is obtained according to the total objective function value of each of the multiple distribution network switch separation and combination schemes in the target season, and the multiple distribution network switch separation and combination schemes are iteratively updated. After reaching the predetermined maximum number of iterations, the switch separation and combination scheme that makes the objective function value optimal in the target season is determined, and the distribution network reconstruction optimization scheme in the target season is obtained. The solution is performed in sequence until the distribution network reconstruction optimization scheme for all seasons is obtained, which achieves the purpose of determining the distribution network switch separation and combination optimization scheme according to the season, thereby achieving the technical effect of effectively improving the operation efficiency and reliability of the photovoltaic distribution network, and then solving the technical problem of low photovoltaic carrying capacity in the distribution network system. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0047] Figure 1 A hardware structure block diagram of a computer terminal for implementing a photovoltaic carrying capacity enhancement method considering seasonal network optimization is shown;

[0048] Figure 2 is a flow chart of a method for improving photovoltaic carrying capacity considering seasonal network optimization provided in accordance with an embodiment of the present invention;

[0049] Figure 3 is a flow chart of a Latin hypercube sampling process for generating a sample scenario according to an optional embodiment of the present invention;

[0050] Figure 4 is a schematic diagram of a node system loop provided according to an optional embodiment of the present invention;

[0051] Figure 5 is a flow chart of a whale optimization algorithm provided according to an optional embodiment of the present invention;

[0052] Figure 6 It is a flow chart of realizing active network reconstruction optimization of a distribution network by using an improved whale algorithm provided in an optional embodiment of the present invention;

[0053] Figure 7 It is a structural block diagram of a photovoltaic carrying capacity enhancement system considering seasonal network optimization provided according to an embodiment of the present invention. DETAILED DESCRIPTION

[0054] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme 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 only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0055] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0056] According to an embodiment of the present invention, a method embodiment for improving photovoltaic carrying capacity considering seasonal network optimization is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in an order different from that shown here.

[0057] The method embodiment provided in the first embodiment of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1 The hardware structure block diagram of a computer terminal for implementing a method for improving photovoltaic carrying capacity considering seasonal network optimization is shown. Figure 1As shown, the computer terminal 10 may include one or more (102a, 102b, ..., 102n are used to illustrate) processors (the processor may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 104 for storing data. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations shown.

[0058] It should be noted that the one or more processors and / or other data processing circuits described above may generally be referred to herein as "data processing circuits". The data processing circuits may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuit may be a single independent processing module, or may be incorporated in whole or in part into any of the other components in the computer terminal 10. As described in the embodiments of the present application, the data processing circuit acts as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0059] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the method for improving the photovoltaic carrying capacity considering seasonal network optimization in the embodiment of the present invention. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, the method for improving the photovoltaic carrying capacity considering seasonal network optimization of the above-mentioned application program is realized. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the computer terminal 10 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0060] The display may be, for example, a touch screen liquid crystal display (LCD) that enables a user to interact with a user interface of the computer terminal 10 .

[0061] Figure 2 is a flow chart of a method for improving photovoltaic carrying capacity considering seasonal network optimization provided in accordance with an embodiment of the present invention. Figure 2 As shown, the method comprises the following steps:

[0062] The first step is to obtain multiple seasonal scene sets of the photovoltaic distribution network corresponding to multiple seasons, wherein the multiple seasonal scene sets respectively include photovoltaic output data and load power data of the photovoltaic distribution network in the corresponding season.

[0063] The second step: With the goal of maximizing the photovoltaic carrying capacity, minimizing the seasonal average network loss, and optimizing the seasonal load balance under different probability levels, and taking the lower limit constraint of the photovoltaic carrying capacity probability level, the radial topology structure constraint of the distribution network, and the upper limit constraint of the number of switch operations as constraints, a seasonal network reconstruction optimization model of the photovoltaic distribution network is constructed.

[0064] Step 3: For each season, based on the improved whale algorithm, generate multiple groups of distribution network switch separation and combination schemes for each season.

[0065] Step 4: Input multiple seasonal scenario sets into the seasonal network reconstruction optimization model respectively, solve the seasonal network reconstruction optimization model in the target season, and determine the total objective function value of each of the multiple groups of distribution network switch separation and combination schemes in the target season.

[0066] Step 5: Based on the improved whale algorithm, the current optimal switching scheme is obtained according to the total objective function value of each of the multiple distribution network switching schemes in the target season, and the multiple distribution network switching schemes are iteratively updated. After reaching the predetermined maximum number of iterations, the switching scheme that makes the objective function value optimal in the target season is determined, and the distribution network reconstruction optimization scheme in the target season is obtained. The solutions are solved in sequence until the distribution network reconstruction optimization scheme for all seasons is obtained.

[0067] Through the above steps, the purpose of determining the optimal switch separation and combination optimization plan of the distribution network according to the season is achieved. The present invention solves the technical problem of low photovoltaic carrying capacity in the distribution network system, and can improve the photovoltaic carrying capacity by means of seasonal network optimization and reconstruction during the operation stage of the distribution network.

[0068] As an optional embodiment, multiple seasonal scene sets of a photovoltaic distribution network corresponding to multiple seasons are obtained, including: obtaining time series historical data of light intensity and time series historical data of load of the photovoltaic distribution network; dividing the time series historical data of light intensity and time series historical data of load into multiple groups according to seasons; establishing a photovoltaic output beta distribution random model for each of the multiple seasons and a load power normal distribution random model for each of the multiple seasons according to the time series historical data of light intensity and time series historical data of load corresponding to the multiple seasons; and sampling the photovoltaic output beta distribution random model for each of the multiple seasons and the load power normal distribution random model for each of the multiple seasons using a Latin hypercube sampling method to determine multiple seasonal scene sets.

[0069] Since photovoltaic and load are random, the present invention needs to study the output and load characteristics of distributed photovoltaic access to the distribution network, so a random model of distributed photovoltaic and load is established in a probabilistic statistical way. Considering that Monte Carlo sampling requires a high number of samplings to achieve high accuracy, the present invention adopts the Monte Carlo sampling method of Latin hypercube for sampling.

[0070] (1) Photovoltaic output random model

[0071] According to statistics of large sample photovoltaic historical data, the light intensity within a certain period of time is Beta distributed. Since the light intensity and photovoltaic output are positively correlated, the Beta distribution can be used to describe the probability density function of photovoltaic output, as shown in the following model:

[0072]

[0073] Where: P PV Photovoltaic output, MW; P N is the rated output of photovoltaic power, MW; a and b are the two shape parameters of Beta distribution; μ PV and δ PV are sample statistics P PV / P N The mean and variance of .

[0074] (2) Load random model

[0075] Load fluctuations and prediction errors will cause a certain degree of random error between the load prediction value and the actual load value, which shows that the load also has uncertainty. According to large sample historical data and similar studies, the uncertainty of the load presents a normal distribution, that is, the probability density function of the load active power is shown in the following model:

[0076]

[0077] Where: P L is the load active power, MW; μ L and σ L are the mean and variance of the active load respectively.

[0078] (1) Latin Hypercube Sampling

[0079] Latin Hypercube Sampling (LHS) can ensure that samples in each dimension are evenly distributed and reduce sample redundancy. Its sample points can be organized into a closed hypercube grid, that is, each sample point can have enough space to represent different sampling areas. The steps of sampling distributed photovoltaic output and load random model using Latin Hypercube Sampling are as follows:

[0080] 1) Divide the probability value [0,1] interval into N equal parts S equal subintervals, and set the sample number i=1.

[0081] 2) For the extraction of the i-th sample, first in [(i-1) / N S ,i / N S ] The subinterval generates a random number p i , its mathematical expression is:

[0082]

[0083] Where r is a random number uniformly distributed between 0 and 1.

[0084] 3) Get a random number p i Then, substitute the inverse function of the random model of distributed photovoltaic output and the inverse function of the random model of load (as shown in the following formula (4)) to obtain a sample value of distributed photovoltaic output and load power.

[0085] x i =F -1 (p i ) (4)

[0086] Where: x i is the sample value; F -1 is the inverse transformation function.

[0087] 4) Let i = i + 1 and repeat the above steps 2), 3), and 4) until i>N S , sampling is completed to obtain N S Distributed photovoltaic output, N S Load power sample values.

[0088] Figure 3 FIG. 1 is a flow chart of a Latin hypercube sampling process for generating a sample scenario according to an optional embodiment of the present invention. Figure 3 As shown in the figure, the process of generating sample scenes by Latin hypercube sampling method is as follows:

[0089] 1) Parameter estimation of photovoltaic and load power stochastic models. The actual light intensity time series historical data and load time series historical data are divided into four groups according to the four seasons of the year. The uncertainty parameters of the four groups of photovoltaic output Beta distribution and load normal distribution models with regional and seasonal characteristics are estimated by equations (1) and (2).

[0090] 2) Latin hypercube sampling. Determine the selected sample season and sample number, extract random variables, map the photovoltaic output and load power as sample values ​​through the inverse function of the distribution function, and sample at one-hour intervals to obtain the scene sample set of photovoltaic output and load power in the corresponding season.

[0091] 3) The number of samples is updated iteratively by season, and finally a scene sample set of distributed photovoltaic output and load power in four seasons is obtained.

[0092] As an optional embodiment, multiple seasonal scenario sets are respectively input into the seasonal network reconstruction optimization model, the seasonal network reconstruction optimization model is solved in the target season, and the total objective function value of each of the multiple groups of distribution network switch separation and combination schemes in the target season is determined, including: using the following method to determine the value of the total objective function in the seasonal network reconstruction optimization model under the target distribution network switch separation and combination strategy, to determine the value of the total objective function corresponding to the multiple groups of distribution network switch separation and combination schemes: under the target distribution network switch separation and combination scheme, determine the photovoltaic carrying capacity, network loss and load balance of each scenario in the target seasonal scenario set; sort the photovoltaic carrying capacity of the scenarios in the target seasonal scenario set to determine the photovoltaic carrying capacity under multiple percentiles; determine the probability evaluation model of the photovoltaic carrying capacity based on the photovoltaic carrying capacity under multiple percentiles; determine the specified ε based on the probability evaluation model of the photovoltaic carrying capacity α The photovoltaic carrying capacity under the probability level; according to the average grid loss, average load balance and ε of the target seasonal scenario set α The photovoltaic carrying capacity under the probability level is used to determine the value of the total objective function under the target distribution network switch separation and combination scheme.

[0093] According to the operation data of each device in the distribution network, the flow calculation of a certain time section can be performed to obtain the power (i.e. branch flow) flowing on each line or transformer device under the jurisdiction of the distribution network at that time section. When the output of the distributed power supply in the power supply area under the jurisdiction of a certain line or transformer device at that moment is greater than the load, the line or transformer device obtained by the flow calculation will have reverse power (reverse flow). The ratio of reverse power to the maximum rated power of the line or transformer device is called the reverse load rate of the distribution equipment. The calculation formula for the reverse load rate of the line and transformer is as follows:

[0094]

[0095] Where: P Net(t) is the reverse power of the line or transformer obtained by power flow calculation at time t, MW; when λ(t) < 0, it means that the photovoltaic absorption capacity in the power supply area of ​​the line or transformer is strong and there is no reverse power flow; when λ(t) > 0, it means that there is a reverse power flow in the line or transformer; S e is the maximum practical operating limit of the line or transformer. If the transformer is a three-winding transformer, then S e Take the actual operating limit on the side corresponding to the transformer power supply area, MW.

[0096] When the maximum distributed photovoltaic access capacity is obtained during thermal stability verification, the reverse load rate at this moment is used as the evaluation object to calculate the maximum accessible capacity in the power supply area corresponding to the line or transformer. The calculation formula is shown in (6), where when the reverse load rate is not less than the reverse transmission margin, the photovoltaic carrying capacity is 0.

[0097]

[0098] Where: S M,k (t) is the maximum additional capacity of distributed photovoltaic power generation in the power supply area of ​​the kth line or transformer at time t, MW; λ r is the reverse power margin allowed for distributed photovoltaic access to the grid line or transformer equipment. When the value is 0, the calculation result is the maximum photovoltaic carrying capacity when reverse power flow is not allowed; when the value is 0.8, the calculation result is the maximum photovoltaic carrying capacity when the thermal stability of the distribution equipment is not allowed to exceed the limit; η PV (t) is the actual power of distributed photovoltaic unit capacity at time t, and its value is related to the local photovoltaic power generation efficiency.

[0099] After the photovoltaic carrying capacity of the power supply area of ​​a certain line or transformer equipment k is obtained according to the above formula (6), the photovoltaic carrying capacity of a certain bus i can be obtained by the following formula (7).

[0100]

[0101] Where: S M,k (t) is the maximum additional capacity of distributed photovoltaic power generation in the power supply area of ​​line or transformer k at time t, that is, the photovoltaic carrying capacity assessment result of line or transformer equipment k; S' M,i (t) is the photovoltaic carrying capacity assessment result of the adjacent busbar i under the positive power flow of the line or transformer equipment k at time t; S M,i-1 (t) is the photovoltaic carrying capacity assessment result of bus i at the next level above i-1 under positive power flow at time t.

[0102] There may be a very small number of cases where the carrying capacity is limited in a large number of historical scenarios. Considering that the carrying capacity assessed based on the worst scenario in history will be too conservative, the present invention uses the probability statistics method to assess the photovoltaic carrying capacity. The probability statistics method provides a more powerful scientific basis for the evaluation of photovoltaic carrying capacity levels, which can exclude a few extreme cases of photovoltaic carrying capacity limitations, allowing distributed photovoltaics to be scientifically and reasonably connected to the distribution network and achieve high-quality development of distributed photovoltaics.

[0103] The present invention uses the statistical calculation idea of ​​quantiles to obtain the size of the photovoltaic carrying capacity of each percentile. Percentile statistics is a method of mathematical statistics, which sorts the sample statistical data and calculates the percentile of each data position. It can be written as: the carrying capacity values ​​in a group of m historical scenes are sorted by size, and the carrying capacity value at the p% position is called the photovoltaic carrying capacity under the pth percentile. This statistical method comprehensively considers all historical scene conditions, sorts the photovoltaic carrying capacity values ​​from large to small, and provides a carrying capacity statistical value based on probability distribution. By evaluating the carrying capacity of distributed photovoltaics at different percentiles, it is possible to objectively and scientifically quantify and analyze the actual data, avoid deviations in extreme cases, and improve the credibility of the carrying capacity assessment under the historical scene set. Its mathematical model is shown below:

[0104] S ε,i =max{S α |Pr{S α ≤S M,i (t)}≥ε α} t∈Ω (8)

[0105] Where: ε α S is the percentile of the distributed photovoltaic access distribution network carrying capacity assessment; α To meet the probability level ε α Distributed photovoltaic capacity that can be connected, MW; S M,i (t) is the maximum accessible capacity of distributed photovoltaic on bus i in scenario t, MW, where t=1,2,...,N S , is the t-th sampled scene; Ω is the complete set of scenes; S ε,i The busbar i is at percentile ε α Maximum load-bearing capacity at , MW.

[0106] For example, if there are 20 PV carrying capacity values ​​(in MW) in statistics, the order from largest to smallest is: 10, 9.8, 9.8, 9.5, 9.4, 9.4, 9.3, 9.1, 9.0, 9.0, 9.0, 8.8, 8.6, 8.6, 7.5, 6.5, 6.4, 6.3, 6.2, 2.1. When the carrying capacity percentile is set to 95%, the 20×95%=19th value 6.2 is the maximum carrying capacity when the percentile is 95%, that is, S ε,i,From formula (8), it can be known that the carrying capacity values ​​of 95% of the scenarios under the historical statistical scenario set are no less than 6.2MW.

[0107] The objective function and constraints of the seasonal network optimization reconstruction model for distribution network to improve photovoltaic carrying capacity include the following:

[0108] 1. The objective function is to maximize the photovoltaic carrying capacity, minimize the seasonal average network loss, and optimize the seasonal load balance under different probability levels.

[0109] (1) Maximum photovoltaic carrying capacity at different probability levels

[0110] Due to the randomness of load and photovoltaic output, the present invention considers the photovoltaic carrying capacity under different probability levels, and takes each distributed photovoltaic access bus as α The maximum total carrying capacity at the time is taken as an optimization target for the distribution network reconstruction. According to formula (8), the objective function formula is as follows:

[0111]

[0112] Where: S α To meet the probability level ε α The distributed photovoltaic capacity that can be connected under S M,i (t) is the maximum accessible capacity of distributed photovoltaic on bus i under scenario t, including the power data of the target season, where t=1,2,...,N S , Ω is the scene set including the power data of the target season; N S The number of scenarios included in the power data for the target season; ε α N is the probability level of distributed photovoltaic access to the distribution network carrying capacity; PV is the total number of nodes whose photovoltaic carrying capacity needs to be improved.

[0113] (2) Minimum seasonal average network loss

[0114] During the reconstruction of the distribution network, network loss is an important control target. The line loss of the distribution network is related to factors such as the state of the switch, active power, reactive power, etc. The objective function model for establishing the minimum network loss of the distribution network is as follows:

[0115]

[0116] Where: N L is the total number of branches in the PV distribution network; n is any branch; t is any scenario included in the power data of the target season; N S is the number of scenarios included in the power data of the target season; the switch state variable on the nth branch is k n, used to indicate whether the switch is closed or open; r n is the resistance value of branch n; P n (t) is the active power of branch n in scenario t; Q n (t) is the reactive power of branch n in scenario t; V n (t) is the voltage at the end of branch n in scenario t.

[0117] (3) Best seasonal load balance

[0118] Load balancing refers to the reasonable distribution of loads in the distribution network to achieve a relatively balanced load on the line to reduce overload operation of the line. The better the load balancing degree in the power system, the higher the power supply reliability. The optimal objective function model for establishing seasonal load balancing degree of the distribution network is as follows:

[0119]

[0120] Where: N L is the total number of branches in the PV distribution network; n is any branch; t is any scenario included in the power data of the target season; N S is the number of scenarios included in the power data of the target season; the switch state variable on the nth branch is k n , used to indicate whether the switch is closed or open; P n (t) is the active power of branch n in scenario t; P n,max is the maximum transmission power allowed for branch n.

[0121] 2. The seasonal network reconstruction optimization model includes the following constraints: lower limit constraint on the probability level of photovoltaic carrying capacity, radial topology constraint on the distribution network, upper limit constraint on the number of switch operations, power flow equation constraint, node voltage constraint, and upper limit constraint on branch reverse power flow.

[0122] (1) The lower limit constraint of the probability level of PV carrying capacity is expressed by the following formula:

[0123] ε α ≥ε min (12)

[0124] Where: ε α is the probability level of distributed photovoltaic access to the distribution network carrying capacity; ε min is the minimum limit of the probability level of PV carrying capacity;

[0125] (2) The radial topology constraint of the distribution network is expressed by the following formula:

[0126] g∈G(13)

[0127] Where: g is the network structure of the distribution network; G is the set of all radial networks;

[0128] (3) The upper limit constraint of the number of switch operations is expressed by the following formula:

[0129]

[0130] Where: k n,past is the switch state variable on the nth branch before optimization; k n,now is the switch state variable on the nth branch after optimization; k max It is the limit of the number of times the main switch is opened and closed;

[0131] (4) The power flow equality constraint is expressed by the following formula:

[0132]

[0133] Where: P i (t) and Q i (t) are the active power and reactive power injected by node i in scenario t, MW and MVar respectively; U i (t) is the voltage amplitude of node i in scenario t, V; Ω i is the set of nodes connected to node i; G ij and B ij is the element of the admittance matrix; θ ij is the phase angle difference between nodes i and j;

[0134] (5) The node voltage constraint is expressed by the following formula:

[0135] U i,min ≤U i (t)≤U i,max (t=1,2,3,...,Ns) (16)

[0136] Where: U i (t) is the voltage amplitude of node i in scenario t, V; U i,max and U i,min are the upper and lower limits of the voltage at node i, V, which are 105% and 95% of the rated voltage respectively;

[0137] (6) The upper limit constraint of the branch reverse flow is expressed by the following formula:

[0138] |P n (t)|≤P n,max (t=1,2,3,...,Ns) (17)

[0139] Where: P n (t) is the reverse active power flow of branch n in scenario t; P n,maxis the maximum transmission power limit of branch n, MW, which is taken as 80% of the maximum transmission power of the line and transformer equipment.

[0140] Optionally, the total objective function is a normalized function of the first objective function, the second objective function and the third objective function, the first objective function is to maximize the photovoltaic carrying capacity under different probability levels, the second objective function is to minimize the seasonal average network loss, and the third objective function is to optimize the seasonal load balance.

[0141] Optionally, based on the improved whale algorithm, according to the total objective function value of each of the multiple groups of distribution network switch separation and combination schemes, the current optimal population is obtained, the multiple groups of distribution network switch separation and combination schemes are iteratively updated, and after reaching a predetermined maximum number of iterations, the distribution network network reconstruction optimization scheme is determined, including: obtaining multiple groups of distribution network switch separation and combination schemes, setting the maximum number of iterations to M max , the initial number of iterations is m = 1, where the total number of the multiple distribution network switch separation and combination schemes is N; calculate the total objective function value of each switch separation and combination scheme in the multiple distribution network switch separation and combination schemes, update A and C according to the following formula, and generate a random number p on [0,1]:

[0142] A=2a·ra

[0143] C=2·r

[0144] Where A and C are vector coefficients, a decreases linearly from 2 to 0 during iteration; r is a random number between 0 and 1;

[0145] If the random number p ≥ 0.5, then according to Update the switch opening and closing scheme; if the random number p<0.5, determine whether the value of |A| is greater than 1. If |A|≥1, follow the formula Update the switch opening and closing scheme. If |A|<1, then follow the formula Update the switch separation and combination scheme, where m is the current iteration number; || represents the distance between two sets of vectors in the solution space, which is a non-negative number; b is a constant representing the spiral shape; l is a random number between -1 and 1; is the random position distance between the current switch separation scheme and the optimal switch separation scheme in the solution space, which is affected by the random number C; is the i-th switch opening and closing scheme of the m-th generation; is the optimal switch separation and combination scheme of the mth generation, which will be updated with iterations. is the random switch separation and combination scheme set; calculate the total objective function value of each updated switch separation and combination scheme, update the optimal objective function value and the optimal switch separation and combination scheme; determine whether the number of iterations m is equal to M maxIf yes, output the optimal objective function value and the optimal switch separation and combination scheme; if not, re-update A and C and generate a random number p on [0,1].

[0146] Optionally, in order to combine the network topology of the distribution network with the intelligent algorithm, the distribution network needs to be encoded. The present invention encodes the network switches of the distribution network, and uses the intelligent algorithm to update the switch combination to obtain an updated network topology, which can convert the network reconstruction problem into a mathematical problem of a permutation set. In order to increase the calculation speed and reduce the number of invalid solutions for isolated islands or ring networks, the present invention uses decimal encoding to improve the efficiency of the algorithm. In order to prevent the occurrence of isolated islands and ring networks in the final network, the present invention performs the following steps for decimal encoding:

[0147] 1) Label each branch in the distribution network, where the switch on the branch is marked as 0, and the switch on the branch is marked as 1. The number of tie switches in the divided basic loops is equal to the number of loops.

[0148] 2) After obtaining the distribution network system with l loops, list the basic loop matrix H according to the loops. The number of rows in the H matrix is ​​the number of basic loops, the number of columns is the number of the longest branches, and the elements behind the matrices of other loops are set to 0. The coding rule is to code from the leftmost branch in the distribution network to the rightmost branch, and from the topmost branch in the distribution network to the bottommost branch. The common branches that are initially coded are no longer coded in subsequent loops.

[0149] 3) Select a non-zero number in each row of the basic ring matrix H. When implementing the intelligent algorithm network reconstruction, in each iteration, it is set that only one branch in each ring can be disconnected, the selected branch coding state is 0, and the other branch coding states are 1. Using the above method, there will be no infeasible solution in the distribution network reconstruction result, which will also simplify the network reconstruction process and speed up the optimization.

[0150] Figure 4 is a schematic diagram of a node system loop provided according to an optional embodiment of the present invention, such as Figure 4 As shown in Figure 1, the IEEE 33-node system has 37 branches, that is, 37 switches. According to the traditional binary coding method, the coding length of each chromosome is 37, and the individuals in the population will have 2 37 = 137438953472 possibilities. However, due to the radial operation of the distribution network, the combination of switches is not arbitrary, and feasible solutions only account for a small part of them. Therefore, the use of binary coding will only produce a very small number of feasible solutions, making the optimization algorithm extremely inefficient. Improving the coding strategy and using the decimal coding method provided by the present invention will greatly shorten the coding length.

[0151] All the tie switches in the closed network will form loops, and the number of loops = the number of tie switches. If a tie switch is closed, a certain section switch in the loop must be opened, and the other switches in the loop must also be closed to keep the distribution network running radially. Figure 4 The IEEE33 node system has 5 tie switches, represented by dotted lines, which form 5 loops after closing. The numbers of the switches in the loops are shown in Table 1. Taking loop 1 as an example, it consists of branches 2, 3, 4, 5, 6, 7, 33, 20, 19, and 18, which are numbered 1-10. The chromosome code length is equal to the number of loops. Figure 3 The decimal code is the number of the disconnected branch in each loop, for example, Figure 3 All the tie switches 33, 34, 35, 36, 37 are coded as (74 5 4 5). This greatly shortens the length of the code. The number of possible individuals in the population is 10×4×6×8×8=15360, which is much smaller than the number of possible individuals in binary coding 2 37 =137438953472, and there is no infeasible solution.

[0152] Table 1 IEEE33 system decimal switch codes and corresponding ring switch numbers

[0153]

[0154] In summary, the present invention adopts the decimal encoding method, and the dimension to be solved is the number of ring networks, where the jth sub-chromosome is used To represent the encoding of the control variable:

[0155]

[0156] In the formula, The nth l The switch state of the branch is disconnected, that is, the nth l The first branch is coded as 0, and the other branches in the loop are coded as 1.

[0157] Optionally, the initial switch strategy can be updated by using the whale algorithm to obtain an updated switch strategy. The present invention uses the whale optimization algorithm (WOA) to formulate the separation and combination scheme of the distribution network tie switch and the section switch, that is, to find the optimal chromosome by using the whale algorithm. The whale optimization algorithm includes three ways to update the position, namely whale encirclement predation, whale bubble attack and search for prey.

[0158] 1) Whales surround and hunt

[0159] In the whale optimization algorithm of the present invention, the switch splitting scheme is regarded as an individual of the population, and the optimal individual is the switch splitting scheme when the photovoltaic carrying capacity is the largest. In each subsequent iteration, assuming that the current best candidate switch splitting scheme is the target prey, other switch splitting schemes will surround and shrink towards this position. The mathematical model is as follows:

[0160]

[0161] Where: A and C are vector coefficients; m is the current number of iterations; || represents the distance between two sets of vectors in the solution space, which is a non-negative number; is the random position distance between the current switch separation scheme and the optimal switch separation scheme in the solution space, which is affected by the random number C; is the i-th switch opening and closing scheme of the m-th generation; is the optimal switch separation and combination scheme of the mth generation, which will be updated with iterations.

[0162] The specific expressions of coefficients A and C are as follows:

[0163] A=2a·ra (21)

[0164] C=2·r (22)

[0165]

[0166] Where: M max is the maximum number of iterations; the convergence factor a decreases linearly from 2 to 0 during iteration; r is a random number between 0 and 1. Due to the selection of a and r, a non-integer solution vector is often obtained after iteration, while the code of the actual disconnection switch of the distribution network is an integer, which causes a contradiction. The present invention obtains the switch separation and combination scheme by rounding the solution vector to an integer.

[0167] 2) Whale Bubble Attack

[0168] Based on the whale's encirclement and predation, the humpback whale's behavior of making a bubble net is simulated. The solution will create a virtual "bubble net" to gradually narrow the search range and focus on the most promising switch splitting and closing scheme. The whale moves in a spiral shape and continuously searches for the current switch splitting and closing scheme. The expression for updating the spiral is as follows:

[0169]

[0170] Wherein: b is a constant representing the spiral shape; l is a random number between -1 and 1.

[0171] Whales can find the optimal switch splitting and closing scheme through Bubble Network, and can also use the surrounding contraction mechanism to approach the optimal switch splitting and closing scheme. In order to achieve the coordination of the two mechanisms, the corresponding threshold is set to realize the model construction, and its expression is as follows:

[0172]

[0173] Where: p is a random number between 0 and 1.

[0174] 3) Whales searching for prey

[0175] In the process of iterative update of the whale population position, in order to prevent the algorithm from reaching the local optimum, when |A|≥1, the switch splitting and combining scheme will no longer change with the optimal switch splitting and combining scheme. In this stage, the switch splitting and combining scheme will be forced to move to the randomly selected reference switch splitting and combining scheme. Update. So the mathematical model expression is as follows:

[0176]

[0177] Where: The random switching scheme is set; is the random position distance between the current switch splitting scheme and the random switch splitting scheme in the solution space, which is affected by the random number C.

[0178] Figure 5 is a flow chart of a whale optimization algorithm provided according to an optional embodiment of the present invention, such as Figure 5 As shown, the steps of implementing the whale optimization algorithm are summarized as follows:

[0179] ① Given the original switch splitting and closing scheme and some feasible switch splitting and closing schemes, the total number of switch splitting and closing schemes initially input is N, and the maximum number of iterations is set to M max , the initial number of iterations is m=1.

[0180] ② Calculate the fitness of each switch opening and closing scheme, update A and C according to equations (21) and (22), and generate a random number p on [0,1].

[0181] ③If the random number p≥0.5, update the switch separation and combination scheme according to formula (24) and enter ④; if the random number p<0.5, determine whether the value of |A| is greater than 1. If |A|≥1, update the switch separation and combination scheme according to formula (28) and enter ④; if |A|<1, update the switch separation and combination scheme according to formula (20) and enter ④.

[0182] ④ Calculate the fitness value of each switch separation and combination scheme, and update the optimal fitness value and the optimal switch separation and combination scheme.

[0183] ⑤ Determine whether the number of iterations m is equal to M max , if so, output the optimal value and the optimal switch separation and combination scheme; if not, return to ②.

[0184] In view of the shortcomings of the current whale optimization algorithm, such as being prone to falling into local optimal solutions and having low convergence accuracy, the present invention also proposes an improved whale optimization algorithm.

[0185] 1) Chaotic map initialization population

[0186] The above-mentioned whale optimization algorithm uses a random method to initialize the population. It is difficult for the whale population to be evenly distributed in the entire search space, which makes the algorithm search inefficient. The chaotic mapping method can more comprehensively find the switch separation and combination scheme, which has the characteristics of randomness and ergodicity. The present invention uses this method to realize the initialization of the whale population. The chaotic mapping initialization model is as follows:

[0187]

[0188] Where: x n It takes values ​​between 0 and 1; when the parameter μ is 1.99, the mapping is uniformly distributed.

[0189] 2) Nonlinear convergence factor and inertia weight

[0190] Since the algorithm is incomplete in the third step of searching for the optimal switch separation and combination scheme and has a slow convergence speed, the present invention modifies the convergence coefficient a to optimize the algorithm. On the basis of always keeping the convergence factor a decreasing with the increase of the number of iterations, a nonlinear convergence factor is proposed. The nonlinear convergence factor model is as follows:

[0191]

[0192] Where: m is the current iteration number, M max is the maximum number of iterations.

[0193] In order to make full use of the optimal solution during search, the following adaptive weight strategy model is proposed:

[0194]

[0195] Based on the nonlinear convergence factor and adaptive weight model proposed above, the algorithm position model formula can be obtained as follows:

[0196]

[0197] is the next switch opening and closing plan; m is the current iteration number; A is the coefficient; is the distance between the switch separation scheme and the optimal switch separation scheme; is the currently optimal switching scheme; b is a constant representing the spiral shape; l is a random number between -1 and 1.

[0198] The above formula makes the whale algorithm more powerful in global optimization. In the later stage of the whale algorithm with relatively large weight, the whale position is greatly affected by the optimal switch separation and combination scheme, and the algorithm has a strong local search effect.

[0199] In addition, since the algorithm has the disadvantage of being easily trapped in the local optimal solution, the Cauchy mutation is integrated into the whale algorithm, and the Cauchy mutation is performed on the current optimal switch separation and combination scheme. Its mathematical model is as follows:

[0200]

[0201] Where: is the switch separation and combination scheme after Cauchy mutation; cauchy(0,1) is the Cauchy operator.

[0202] This model performs a Cauchy mutation on the current optimal switch separation and combination scheme, which can make the whale algorithm escape from the state of local optimal solution, realize global search, and achieve the purpose of improving the global optimization capability.

[0203] (1) Network Coding Methods

[0204] In order to combine the network topology of the distribution network with the intelligent algorithm, the distribution network needs to be encoded. The present invention encodes the network switches of the distribution network, and uses the intelligent algorithm to update the switch combination to obtain an updated network topology, which can convert the network reconstruction problem into a mathematical problem of a permutation set. In order to increase the calculation speed and reduce the number of invalid solutions for isolated islands or ring networks, the present invention uses decimal encoding to improve the efficiency of the algorithm. In order to prevent the occurrence of isolated islands and ring networks in the final network, the present invention performs the following steps for decimal encoding:

[0205] 1) Label each branch in the distribution network, where the switch on the branch is marked as 0, and the switch on the branch is marked as 1. The number of tie switches in the divided basic loops is equal to the number of loops.

[0206] 2) After obtaining the distribution network system with l loops, list the basic loop matrix H according to the loops. The number of rows in the H matrix is ​​the number of basic loops, the number of columns is the number of the longest branches, and the elements behind the matrices of other loops are set to 0. The coding rule is to code from the leftmost branch in the distribution network to the rightmost branch, and from the topmost branch in the distribution network to the bottommost branch. The common branches that are initially coded are no longer coded in subsequent loops.

[0207] 3) Select a non-zero number in each row of the basic ring matrix H. When implementing the intelligent algorithm network reconstruction, in each iteration, it is set that only one branch in each ring can be disconnected, the selected branch coding state is 0, and the other branch coding states are 1. Using the above method, there will be no infeasible solution in the distribution network reconstruction result, which will also simplify the network reconstruction process and speed up the optimization.

[0208] (2) Improved multi-objective seasonal network reconstruction optimization method based on whale algorithm

[0209] The present invention uses the discrete optimization of the whale algorithm to achieve the optimization of network reconstruction. First, the three multi-objective problems of maximum photovoltaic carrying capacity under different probability levels, minimum seasonal average network loss, and optimal seasonal load balancing are converted into single objectives. Then, the switch state of the branch is set as the control variable of network reconstruction optimization by decimal coding, and the individual position of the whale and the distance between the whale and the target prey are encoded respectively. Finally, the specific steps of using the improved whale algorithm to achieve the optimization of distribution network reconstruction are proposed.

[0210] 1) Normalization of multiple objective functions

[0211] The present invention takes the maximum photovoltaic carrying capacity under different probability levels, the minimum seasonal average network loss, and the optimal seasonal load balance as the objective functions, and converts the three objective functions into a single objective solution by normalizing the multi-objective function. Since the dimensions of these three objective functions are different, and the carrying capacity and load balance affect each other, the present invention uses the range normalization method to unify the dimensions of each objective function. The mathematical model is as follows:

[0212]

[0213] Where: r imax The larger the index, the better. The invention represents the maximum objective function f1 of the normalized bearing capacity under different probability levels; r imin The smaller the index, the better. The present invention represents the objective functions f2 and f3 with the minimum normalized seasonal average network loss and the best seasonal load balance; f i An element of the objective function indicator.

[0214] Through the above processing, the multi-objective function can be converted into a single objective function:

[0215] F=-μ1r 1max +μ2r 2min +μ3r 3min (35)

[0216] Wherein: μ1, μ2, μ3 are the weight coefficients of the three objective functions of maximum carrying capacity, minimum seasonal average network loss and optimal seasonal load balance under different probability levels; F is the total objective function, the smaller F is, the better.

[0217] 2) Whale individual coding

[0218] The control variable of network reconstruction optimization is the switch state of the branch. Since the present invention adopts the decimal encoding method, the dimension to be solved is the number of ring networks. Set k l,nl The nth l The switch state of the branch is open. The position of the i-th whale is encoded as The distance between the i-th whale and the target prey is encoded as: Therefore, the individual position of the whale and the distance between the whale and the target prey can be recorded as:

[0219] As a specific embodiment, Figure 6 is a flow chart of realizing active network reconstruction optimization of distribution network by using the improved whale algorithm provided in an optional embodiment of the present invention, such as Figure 6 As shown in the figure, this scheme may include the following steps: ① Input the topological information of the distribution network, the information of each branch, the load of each node and the distributed photovoltaic access information, various constraint limits, etc. The distribution network is decimal-coded, the encoded basic loop matrix H is listed, and the sample season scenario N is obtained by Latin hypercube sampling. S (N1, N2, N3, N4), where 1, 2, 3, and 4 represent the four seasons of spring, summer, autumn, and winter respectively.

[0220] ② Set the initial scenario t = 1, calculate the power flow results under the original network topology, and obtain the network loss, load balance and photovoltaic carrying capacity under the scenario t = 1. Update the scenario t = t + 1, and calculate the network loss, load balance and photovoltaic carrying capacity under this scenario again until all scenarios N under the current network topology are obtained. S The grid loss, load balance and photovoltaic carrying capacity are calculated, and then the seasonal average grid loss, seasonal load balance and ε under the original network topology are obtained. α Initial value of PV carrying capacity under probability level.

[0221] ③ Encode the position of the individual whale, and take the state of the nth branch switch of the lth loop as the individual whale position, that is,

[0222] ④ Set the maximum number of iterations of the algorithm as the termination condition of the algorithm iteration, set the initial number of iterations to 1, and set the vector parameters, constants and random numbers of the whale algorithm. Determine whether it is a feasible solution. If it is not a feasible solution, repeat the second step.

[0223] ⑤ Calculate and obtain the topological structure of m=1 generations of feasible solutions, and set the initial scenario t=1.

[0224] ⑥ Calculate the power flow results, obtain the network loss, load balance and photovoltaic carrying capacity under scenario t, and determine whether t is less than N S If it is established, then t+1, calculate the power flow again; if it is not established, calculate the seasonal average network loss, seasonal load balance and statistics to obtain ε α Values ​​of PV carrying capacity at probability levels.

[0225] ⑦ Obtain the normalized fitness function F to obtain the optimal population and update the whale position. Determine whether it is the optimal solution, update the whale population, determine whether the optimal individual needs Cauchy mutation, and perform Cauchy mutation on the whale individual Then determine whether it is the maximum number of iterations. If not, update the number of iterations m=m+1 and proceed to step 5. If so, output the result of the improved whale algorithm network reconstruction optimization.

[0226] ⑧ Complete the improved whale algorithm network reconstruction optimization, and obtain the distribution network reconstruction optimization reconstruction plan for improving photovoltaic carrying capacity in each season.

[0227] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the described order of actions, because according to the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

[0228] Through the description of the above implementation methods, those skilled in the art can clearly understand that the photovoltaic carrying capacity improvement method considering seasonal network optimization according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, disk, CD), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0229] According to an embodiment of the present invention, a photovoltaic carrying capacity improvement system considering seasonal network optimization is also provided for implementing the photovoltaic carrying capacity improvement method considering seasonal network optimization. Figure 7is a structural block diagram of a photovoltaic carrying capacity enhancement system considering seasonal network optimization according to an embodiment of the present invention. Figure 7 As shown, the photovoltaic carrying capacity improvement system considering seasonal network optimization includes: an acquisition module, a construction module, a switch separation and combination scheme generation module, a reconstruction optimization module and an optimal scheme determination module. The photovoltaic carrying capacity improvement system considering seasonal network optimization is described below.

[0230] The acquisition module is used to acquire multiple seasonal scene sets of the photovoltaic distribution network corresponding to multiple seasons, wherein the multiple seasonal scene sets respectively include photovoltaic output data and load power data of the photovoltaic distribution network in the corresponding season.

[0231] The construction module is used to maximize the photovoltaic carrying capacity, minimize the seasonal average network loss, and optimize the seasonal load balance under different probability levels, and to construct a seasonal network reconstruction optimization model for the photovoltaic distribution network with the lower limit constraint of the photovoltaic carrying capacity probability level, the radial topology structure constraint of the distribution network, and the upper limit constraint of the number of switch operations as constraints.

[0232] The switch separation and combination scheme generation module is used to generate multiple groups of distribution network switch separation and combination schemes for each season based on the improved whale algorithm.

[0233] The reconstruction optimization module is used to input multiple seasonal scenario sets into the seasonal network reconstruction optimization model respectively, solve the seasonal network reconstruction optimization model in the target season, and determine the total objective function value of each of the multiple groups of distribution network switch separation and combination schemes in the target season.

[0234] The optimal solution determination module is used to obtain the current optimal switch separation and combination scheme based on the improved whale algorithm according to the total objective function value of each of the multiple distribution network switch separation and combination schemes in the target season, iteratively update the multiple distribution network switch separation and combination schemes, and after reaching a predetermined maximum number of iterations, determine the switch separation and combination scheme that makes the objective function value optimal in the target season, obtain the distribution network reconstruction optimization scheme in the target season, and solve the distribution network reconstruction optimization schemes for all seasons in turn.

[0235] It should be noted that the acquisition module, construction module, switch separation and combination scheme generation module, reconstruction optimization module and optimal scheme determination module correspond to steps 1 to 5 in the embodiment, and the examples and application scenarios implemented by the multiple modules and the corresponding steps are the same, but are not limited to the contents disclosed in the above embodiment. It should be noted that the above modules, as part of the device, can be run in the computer terminal 10 provided in the embodiment.

[0236] An embodiment of the present invention may provide a computer device. Optionally, in this embodiment, the computer device may be located in at least one network device among multiple network devices of a computer network. The computer device includes a memory and a processor.

[0237] Among them, the memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the photovoltaic carrying capacity improvement method and system considering seasonal network optimization in the embodiment of the present invention. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, realizing the above-mentioned photovoltaic carrying capacity improvement method considering seasonal network optimization. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include a memory remotely arranged relative to the processor, and these remote memories can be connected to the computer terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0238] The processor can call the information and application programs stored in the memory through the transmission device to perform the following steps: obtain multiple seasonal scene sets of the photovoltaic distribution network corresponding to multiple seasons, wherein the multiple seasonal scene sets respectively include the photovoltaic output data and load power data of the photovoltaic distribution network in the corresponding season; take the maximum photovoltaic carrying capacity, the minimum seasonal average network loss, and the best seasonal load balance under different probability levels as the goals, and take the lower limit constraint of the photovoltaic carrying capacity probability level, the radial topology structure constraint of the distribution network, and the upper limit constraint of the number of switch operations as constraints to construct a seasonal network reconstruction optimization model for the photovoltaic distribution network; for each season, based on the improved whale algorithm, generate multiple seasonal network reconstruction optimization models for each season. The switch separation and combination schemes of the distribution network are grouped; multiple seasonal scenario sets are respectively input into the seasonal network reconstruction optimization model, the seasonal network reconstruction optimization model is solved in the target season, and the total objective function value of each of the multiple distribution network switch separation and combination schemes in the target season is determined; based on the improved whale algorithm, according to the total objective function value of each of the multiple distribution network switch separation and combination schemes in the target season, the current optimal switch separation and combination scheme is obtained, and the multiple distribution network switch separation and combination schemes are iteratively updated. After reaching the predetermined maximum number of iterations, the switch separation and combination scheme that makes the objective function value optimal in the target season is determined, and the distribution network reconstruction optimization scheme in the target season is obtained, and the solutions are solved in sequence until the distribution network reconstruction optimization schemes of all seasons are obtained.

[0239] By adopting the embodiment of the present invention, a solution for improving photovoltaic carrying capacity taking into account seasonal network optimization is provided. By obtaining multiple seasonal scenario sets of photovoltaic distribution networks corresponding to multiple seasons, wherein the multiple seasonal scenario sets respectively include photovoltaic output data and load power data of the photovoltaic distribution network in the corresponding season; taking the maximum photovoltaic carrying capacity, the minimum seasonal average network loss, and the best seasonal load balance under different probability levels as the goals, and taking the lower limit constraint of the photovoltaic carrying capacity probability level, the radial topology structure constraint of the distribution network, and the upper limit constraint of the number of switch operations as constraints, a seasonal network reconstruction optimization model of the photovoltaic distribution network is constructed; for each season, based on the improved whale algorithm, multiple groups of distribution network switch separation and combination schemes are generated in each season; the multiple seasonal scenario sets are respectively input into the seasonal network reconstruction optimization model, and the seasonal network reconstruction optimization model in the target season is solved, Determine the total objective function value of each of the multiple distribution network switch separation and combination schemes in the target season; based on the improved whale algorithm, according to the total objective function value of each of the multiple distribution network switch separation and combination schemes in the target season, obtain the current optimal switch separation and combination scheme, iteratively update the multiple distribution network switch separation and combination schemes, and after reaching the predetermined maximum number of iterations, determine the switch separation and combination scheme that makes the objective function value optimal in the target season, and obtain the distribution network reconstruction optimization scheme in the target season, and solve it in sequence until the distribution network reconstruction optimization scheme for all seasons is obtained, thereby achieving the purpose of determining the distribution network's switch separation and combination optimization scheme according to the season, thereby achieving the technical effect of effectively improving the operating efficiency and reliability of the photovoltaic distribution network, and then solving the technical problem of low photovoltaic carrying capacity in the distribution network system.

[0240] A person of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a non-volatile storage medium, and the storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0241] The embodiment of the present invention further provides a non-volatile storage medium. Optionally, in this embodiment, the non-volatile storage medium can be used to store the program code executed by the photovoltaic carrying capacity improvement method considering seasonal network optimization provided by the above embodiment.

[0242] Optionally, in this embodiment, the non-volatile storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.

[0243] Optionally, in this embodiment, the non-volatile storage medium is configured to store program codes for executing the following steps: obtaining multiple seasonal scenario sets of the photovoltaic distribution network corresponding to multiple seasons, wherein the multiple seasonal scenario sets respectively include photovoltaic output data and load power data of the photovoltaic distribution network in the corresponding season; taking the maximum photovoltaic carrying capacity, the minimum seasonal average network loss, and the optimal seasonal load balancing degree under different probability levels as the goals, and taking the lower limit constraint of the photovoltaic carrying capacity probability level, the radial topology structure constraint of the distribution network, and the upper limit constraint of the number of switch operations as constraints, constructing a seasonal network reconstruction optimization model for the photovoltaic distribution network; for each season, based on the improved whale algorithm, generating a seasonal network reconstruction optimization model for each season. multiple groups of distribution network switch separation and combination schemes; input multiple seasonal scenario sets into the seasonal network reconstruction optimization model respectively, solve the seasonal network reconstruction optimization model in the target season, and determine the total objective function value of each of the multiple distribution network switch separation and combination schemes in the target season; based on the improved whale algorithm, according to the total objective function value of each of the multiple distribution network switch separation and combination schemes in the target season, obtain the current optimal switch separation and combination scheme, iteratively update the multiple distribution network switch separation and combination schemes, after reaching the predetermined maximum number of iterations, determine the switch separation and combination scheme that makes the objective function value optimal in the target season, and obtain the distribution network reconstruction optimization scheme in the target season, and solve it in sequence until the distribution network reconstruction optimization scheme for all seasons is obtained.

[0244] An embodiment of the present invention further provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the method for improving photovoltaic carrying capacity considering seasonal network optimization in various embodiments of the present application.

[0245] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0246] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0247] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0248] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0249] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0250] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a non-volatile storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0251] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for improving photovoltaic carrying capacity considering seasonal network optimization, characterized in that: include: Step 1: Acquire multiple seasonal scene sets of the photovoltaic distribution network corresponding to multiple seasons, wherein the multiple seasonal scene sets respectively include photovoltaic output data and load power data of the photovoltaic distribution network in the corresponding season; Step 2: With the goal of maximizing the photovoltaic carrying capacity, minimizing the seasonal average network loss, and optimizing the seasonal load balance under different probability levels, and taking the lower limit constraint of the photovoltaic carrying capacity probability level, the radial topology constraint of the distribution network, and the upper limit constraint of the number of switch operations as constraints, a seasonal network reconstruction optimization model of the photovoltaic distribution network is constructed; Step 3: For each season, based on the improved whale algorithm, multiple groups of distribution network switch separation and combination schemes are generated for each season; Step 4: input the multiple seasonal scenario sets into the seasonal network reconstruction optimization model respectively, solve the seasonal network reconstruction optimization model in the target season, and determine the total objective function value of each of the multiple groups of distribution network switch separation and combination schemes in the target season; Step 5: Based on the improved whale algorithm, according to the total objective function value of each of the multiple groups of distribution network switch separation and combination schemes in the target season, the current optimal switch separation and combination scheme is obtained, and the multiple groups of distribution network switch separation and combination schemes are iteratively updated. After reaching a predetermined maximum number of iterations, the switch separation and combination scheme that makes the objective function value optimal in the target season is determined, and the distribution network reconstruction optimization scheme in the target season is obtained. The solutions are solved in sequence until the distribution network reconstruction optimization scheme for all seasons is obtained.

2. The method according to claim 1, characterized in that The total objective function is a function obtained by normalizing the first objective function, the second objective function and the third objective function. The first objective function is to maximize the photovoltaic carrying capacity under different probability levels, the second objective function is to minimize the seasonal average network loss, and the third objective function is to optimize the seasonal load balance.

3. The method according to claim 1, characterized in that Inputting the multiple seasonal scenario sets into the seasonal network reconstruction optimization model respectively, solving the seasonal network reconstruction optimization model in the target season, and determining the total objective function value of each of the multiple groups of distribution network switch separation and combination schemes in the target season, including: The following method for determining the value of the total objective function in the seasonal network reconstruction optimization model under the target distribution network switch separation and combination strategy is used to determine the value of the total objective function corresponding to the multiple groups of distribution network switch separation and combination schemes: Under the target distribution network switch separation and combination scheme, determine the photovoltaic carrying capacity, network loss and load balance degree of each scenario in the target seasonal scenario set; Sorting the photovoltaic carrying capacity of the target seasonal scenario set scenarios to determine the photovoltaic carrying capacity at multiple percentiles; According to the photovoltaic carrying capacity under the multiple percentiles, a probability evaluation model of the photovoltaic carrying capacity is determined, wherein the mathematical model of the probability evaluation model of the photovoltaic carrying capacity is shown in the following formula: S ε,i =max{S α |Pr{S α ≤S M,i (t)}≥ε α }t∈Ω; Where: ε α S is the percentile of the distributed photovoltaic access distribution network carrying capacity assessment; α To meet the probability level ε α Distributed photovoltaic capacity that can be connected, MW; S M,i (t) is the maximum accessible capacity of distributed photovoltaic on bus i in scenario t, MW, where t=1,2,...,N S , is the t-th sampled scene; Ω is the complete set of scenes; S ε,i The busbar i is at percentile ε α Maximum bearing capacity at , MW; According to the probabilistic evaluation model of PV capacity, the specified ε α PV carrying capacity under probability level; According to the average network loss, average load balance degree and the ε α The photovoltaic carrying capacity under the probability level is used to determine the value of the total objective function under the target distribution network switch separation and combination scheme.

4. The method according to claim 2, characterized in that: The formula of the first objective function f1 is expressed as: Where: S α To meet the probability level ε α The distributed photovoltaic capacity that can be connected under S M,i (t) is the maximum accessible capacity of distributed photovoltaic on bus i under scenario t, including the power data of the target season, where t = 1, 2, ..., N S , Ω is the scene set included in the power data of the target season; N S The number of scenarios included in the power data for the target season; α The probability level of distributed photovoltaic access to the distribution network carrying capacity; N PV is the total number of nodes whose photovoltaic carrying capacity needs to be improved.

5. The method according to claim 2, characterized in that: The formula of the second objective function f2 is expressed as: Where: N L is the total number of branches in the photovoltaic distribution network; n is any branch; t is any scenario included in the power data of the target season; N S is the number of scenarios included in the power data of the target season; the switch state variable on the nth branch is k n , used to indicate whether the switch is closed or open; r n is the resistance value of branch n; P n (t) is the active power of branch n in scenario t; Q n (t) is the reactive power of branch n in scenario t; V n (t) is the voltage at the end of branch n in scenario t.

6. The method according to claim 2, characterized in that The formula of the third objective function f3 is expressed as: Where: N L is the total number of branches in the photovoltaic distribution network; n is any branch; t is any scenario included in the power data of the target season; N S is the number of scenarios included in the power data of the target season; the switch state variable on the nth branch is k n , used to indicate whether the switch is closed or open; P n (t) is the active power of branch n in scenario t; P n,max is the maximum transmission power allowed for branch n.

7. The method according to claim 2, characterized in that: The seasonal network reconfiguration optimization model includes the following constraints: lower limit constraint of photovoltaic carrying capacity probability level, radial topology constraint of distribution network, upper limit constraint of switch operation times, power flow equation constraint, node voltage constraint, and upper limit constraint of branch reverse power flow; The lower limit constraint of the probability level of photovoltaic carrying capacity is expressed by the following formula: e α ≥e min Where: ε α is the probability level of distributed photovoltaic access to the distribution network carrying capacity; ε min is the minimum limit of the probability level of PV carrying capacity; The radial topology constraint of the distribution network is expressed by the following formula: g∈G Where: g is the network structure of the distribution network; G is the set of all radial networks; The upper limit constraint of the number of switch operations is expressed by the following formula: Where: k n,past is the switch state variable on the nth branch before optimization; k n,now is the switch state variable on the nth branch after optimization; k max It is the limit of the number of times the main switch is opened and closed; The power flow equality constraint is expressed by the following formula: Where: P i (t) and Q i (t) are the active power and reactive power injected by node i in scenario t, MW and MVar respectively; U i (t) is the voltage amplitude of node i in scenario t, V; Ω i is the set of nodes connected to node i; G ij and B ij is the element of the admittance matrix; θ ij is the phase angle difference between nodes i and j; The node voltage constraint is expressed by the following formula: U i,min ≤U i (t)≤U i,max (t=1,2,3,...,Ns) Where: U i (t) is the voltage amplitude of node i in scenario t, V; U i,max and U i,min are the upper and lower limits of the voltage at node i, V, which are 105% and 95% of the rated voltage respectively; The upper limit constraint of the branch reverse power flow is expressed by the following formula: |P n (t)|≤P n,max (t=1,2,3,...,Ns) Where: P n (t) is the reverse active power flow of branch n in scenario t; P n,max is the maximum transmission power limit of branch n, MW, which is taken as 80% of the maximum transmission power of the line and transformer equipment.

8. The method according to claim 1, characterized in that: The step of obtaining a plurality of seasonal scene sets of the photovoltaic distribution network corresponding to a plurality of seasons comprises: Acquire the time series historical data of light intensity and load of the photovoltaic distribution network; Dividing the light intensity time series historical data and the load time series historical data into multiple groups according to seasons; According to the time series historical data of light intensity and the time series historical data of load corresponding to the multiple seasons, a beta distribution random model of photovoltaic output for each of the multiple seasons and a normal distribution random model of load power for each of the multiple seasons are established; The Latin hypercube sampling method is used to sample the photovoltaic output beta distribution random models of the multiple seasons and the load power normal distribution random models of the multiple seasons to determine the multiple seasonal scene sets.

9. The method according to claim 1, characterized in that: The improved whale algorithm is based on the total objective function value of each of the multiple groups of distribution network switch separation and combination schemes to obtain the current optimal population, iteratively update the multiple groups of distribution network switch separation and combination schemes, and after reaching a predetermined maximum number of iterations, determine the distribution network network reconstruction optimization plan, including: Get the multiple distribution network switch separation and combination schemes, and set the maximum number of iterations to M max , the initial number of iterations is m=1, wherein the total number of the multiple groups of distribution network switch separation and combination schemes is N; Calculate the total objective function value of each switch separation scheme in the multiple groups of distribution network switch separation schemes, update A and C according to the following formula, and generate a random number p on [0,1]: A=2a·ra C＝2·r Where A and C are vector coefficients, a decreases linearly from 2 to 0 during iteration; r is a random number between 0 and 1; If the random number p ≥ 0.5, then according to Update the switch opening and closing scheme; if the random number p < 0.5, determine whether the value of |A| is greater than 1. If |A| ≥ 1, then follow the formula Update the switch opening and closing scheme. If |A|<1, then follow the formula Update the switch separation and combination scheme, where m is the current iteration number; || represents the distance between two sets of vectors in the solution space, which is a non-negative number; b is a constant representing the spiral shape; l is a random number between -1 and 1; is the random position distance between the current switch separation scheme and the optimal switch separation scheme in the solution space, which is affected by the random number C; is the i-th switch opening and closing scheme of the m-th generation; is the optimal switch separation and combination scheme of the mth generation, which will be updated with iterations. The random switching scheme is set; Calculate the total objective function value of each updated switch separation and combination scheme, and update the optimal objective function value and the optimal switch separation and combination scheme; Determine whether the number of iterations m is equal to M max If not, repeat the steps of updating A and C and generating a random number p on [0,1] until m is equal to M. max Established, output the optimal objective function value and the optimal switch separation and combination scheme.

10. A photovoltaic capacity enhancement system considering seasonal network optimization, characterized in that: include: An acquisition module, used to acquire multiple seasonal scene sets of the photovoltaic distribution network corresponding to multiple seasons, wherein the multiple seasonal scene sets respectively include photovoltaic output data and load power data of the photovoltaic distribution network in the corresponding season; A construction module is used to construct a seasonal network reconstruction optimization model of the photovoltaic distribution network with the goals of maximizing photovoltaic carrying capacity, minimizing seasonal average network loss, and optimizing seasonal load balancing under different probability levels, and taking the lower limit constraint of photovoltaic carrying capacity probability level, the radial topology constraint of the distribution network, and the upper limit constraint of the number of switch operations as constraint conditions; The switch separation and combination scheme generation module is used to generate multiple sets of distribution network switch separation and combination schemes for each season based on the improved whale algorithm; A reconstruction optimization module, used to input the multiple seasonal scenario sets into the seasonal network reconstruction optimization model respectively, solve the seasonal network reconstruction optimization model in the target season, and determine the total objective function value of each of the multiple groups of distribution network switch separation and combination schemes in the target season; The optimal solution determination module is used to obtain the current optimal switch separation and combination scheme based on the improved whale algorithm according to the total objective function value of each of the multiple groups of distribution network switch separation and combination schemes in the target season, iteratively update the multiple groups of distribution network switch separation and combination schemes, and after reaching a predetermined maximum number of iterations, determine the switch separation and combination scheme that makes the objective function value optimal in the target season, obtain the distribution network network reconstruction optimization scheme in the target season, and solve the distribution network network reconstruction optimization schemes for all seasons in sequence.

Images