Method and device for analyzing influence of distributed photovoltaic power generation on line loss of distribution network

Abstract

The application discloses a kind of distributed photovoltaic power generation to the analysis method and device of influence of line loss of distribution network, belong to power grid technical field.The method includes: based on photovoltaic power generation historical data, through kernel density estimation method and Joe estimation function to establish joint probability distribution function, to accurately depict the space-time correlation and uncertainty of photovoltaic output;Through the sampling and inverse transform of joint probability distribution function to generate typical daily curve;Based on typical daily curve, introduce energy storage adjustment strategy based on loss response, use multi-objective adaptive evolution algorithm based on heuristic mutation to dynamically optimize energy storage charging and discharging behavior, construct distribution network model;The active power loss of each line and element of distribution network is calculated, considering the influence of integrated power of grid connection point after photovoltaic access on line loss, to obtain the result of network loss calculation.The application can effectively and quickly analyze the line loss of distribution system, and real-time energy storage adjustment can reduce the adverse effects of line loss caused by fluctuation of distributed generation access.

Description

Technical Field

[0001] This invention belongs to the field of power grid technology, specifically relating to a method and apparatus for analyzing the impact of distributed photovoltaic power generation on distribution network line losses. Background Technology

[0002] With the continuous improvement of global environmental awareness and the rapid iteration of new energy technologies, distributed photovoltaic (PV) power generation, as a clean, efficient, and renewable energy technology, is gaining unprecedented widespread application and promotion in the power grid sector. Guided by the national "dual-carbon" strategy, the development of distributed power sources is showing a vigorous development trend, becoming not only an important path for energy transformation but also a key measure to achieve carbon neutrality. The large-scale integration of distributed PV is fundamentally changing the network structure of traditional distribution networks, gradually evolving it from a single radial power supply mode into a complex network system with multiple power sources and nodes. This structural transformation means that the distribution network is no longer a passive system that transmits energy in one direction, but a smart grid with bidirectional energy flow and dynamic adjustment capabilities. The intermittent and fluctuating output of distributed PV brings significant randomness and uncertainty to the net load of the distribution system. This poses a greater challenge to the operational stability of the distribution network, causing it to face a series of technical problems such as voltage fluctuations, voltage over-limits, three-phase voltage imbalance, harmonic pollution, and increased line losses.

[0003] With the rapid advancement of energy storage technology, energy storage systems, characterized by rapid regulation and flexible energy management, have emerged, providing a more economical, safe, and efficient new paradigm for the operation and control of distributed photovoltaic (PV) grid integration. Benefiting from technological progress and economies of scale, the cost of energy storage systems continues to decline, and their application scope is rapidly expanding. From its initial role in power peak shaving and frequency regulation to its widespread application in distributed generation and microgrid construction, energy storage has become a crucial link connecting new energy sources, the power grid, and users. Although academia and industry have conducted multi-faceted research on the impact of PV grid integration, deeply coupling distributed PV with energy storage systems and systematically assessing their impact on distribution network line losses remains a research area urgently requiring breakthroughs.

[0004] Energy storage systems, with their unique advantages of rapid power regulation and flexible energy management, can significantly improve the operating characteristics of distribution networks through proactive charging and discharging strategies. They can not only optimize load balance and voltage distribution on lines, but also improve the economics of integrating distributed photovoltaic (PV) power generation into the grid. Therefore, in-depth research into the impact of distributed PV power generation, considering energy storage regulation, on distribution network line losses is not only of significant theoretical importance, but will also provide crucial technical support for the construction of future smart grids. Summary of the Invention

[0005] The present invention aims to at least partially solve one of the technical problems in the aforementioned related technologies.

[0006] Therefore, the purpose of this invention is to provide a method and apparatus for analyzing the impact of distributed photovoltaic power generation on distribution network line losses, which can effectively and quickly analyze the line losses of the distribution system.

[0007] To solve the above-mentioned technical problems, the present invention is implemented as follows:

[0008] This invention provides a method for analyzing the impact of distributed photovoltaic power generation on distribution network line losses. The method includes:

[0009] Based on historical data of photovoltaic power generation, a joint probability distribution function is established using kernel density estimation and Joe estimation function to accurately characterize the spatiotemporal correlation and uncertainty of photovoltaic output; through sampling and inverse transformation of the joint probability distribution function, a typical daily curve considering the correlation and uncertainty of photovoltaic output is generated.

[0010] Based on typical daily curves, an energy storage regulation strategy based on loss response is introduced. A multi-objective adaptive evolutionary algorithm based on heuristic mutation is used to dynamically optimize the charging and discharging behavior of energy storage, with the goal of minimizing line losses, and a distribution network model is constructed.

[0011] The active power loss of each line and component in the distribution network is calculated, and the impact of the comprehensive power at the grid connection point after photovoltaic access on the line loss is considered to obtain the network loss calculation results.

[0012] Furthermore, the method for analyzing the impact of distributed photovoltaic power generation on distribution network line losses according to the present invention may also have the following additional technical features:

[0013] In some of these implementations, the kernel density estimation method uses a Gaussian kernel function for density estimation.

[0014] In some of these implementations, the active power loss is calculated using the root mean square current method, which calculates the active power loss of each line and component of the distribution network based on 24-hour hourly load data.

[0015] In some of these implementations, the combined power at the grid connection point after photovoltaic access is a coupling of load power and photovoltaic output.

[0016] In some implementations, the multi-objective adaptive evolutionary algorithm based on heuristic mutation includes:

[0017] Initialization: An initial population is generated using a heuristic mapping to ensure that the population is uniformly and purposefully distributed in the search space; and the positions of individuals in the initial population are initialized.

[0018] Evaluation and screening: The initial population is sorted by quality and density;

[0019] Iterative updates: The sorted optimal solution guides the global search, expanding the solution space coverage; adaptive boundary optimization is used for local fine-grained search to improve the quality of the solution.

[0020] Iteration Termination: After reaching the maximum number of iterations, the optimal solution set is output for energy storage charging and discharging strategy optimization.

[0021] In some of these implementations, the specific content of the evaluation and screening includes:

[0022] Calculate the individual objective function value and select the superior solution based on the ranking of merits;

[0023] The dominant solutions are density-sorted, and individuals with uniform distribution and good performance are retained.

[0024] In some implementations, iterative updates include an exploration phase and a development phase;

[0025] The exploration phase involves a global search: guided by the current optimal solution, before updating... N / 2 individual positions enhance global convergence; N Population size;

[0026] The development phase involves local search: an adaptive boundary is introduced to perform a fine-grained local search, which improves the accuracy of the solution.

[0027] In some of these implementations, the energy storage regulation strategy based on loss response includes:

[0028] Using line loss responsiveness as the core basis for energy storage regulation, the dynamic response characteristics of system loss to node power fluctuations are quantified, and the net power of grid connection points is optimized through energy storage charging and discharging strategies to reduce line losses.

[0029] Construct a constraint model for the energy storage system to ensure the safety and effectiveness of energy storage regulation.

[0030] In some of these implementations, the energy storage system constraint model includes power balance constraints, power limitation constraints, and power security constraints.

[0031] This invention also provides an analysis device for the impact of distributed photovoltaic power generation on distribution network line losses, including a processor and a memory. The memory stores a software program, and when the processor runs the software program, it can implement the content of the analysis method for the impact of distributed photovoltaic power generation on distribution network line losses as described above.

[0032] Compared with the prior art, the present invention has at least the following beneficial effects:

[0033] In this embodiment of the invention, the method for analyzing the impact of distributed photovoltaic (PV) power generation on distribution network line losses considers the distribution network with distributed PV access. It analyzes the impact of factors such as PV configuration, climate conditions, and load type on distribution network line losses. The proposed energy storage regulation method, based on line loss response, utilizes a heuristic mutation-based multi-objective adaptive evolutionary algorithm to achieve energy storage regulation, which can reduce distribution network line losses and improve the economic efficiency of distribution network operation. The calculation results show that the access of distributed PV is beneficial to reducing network line losses, and the degree of improvement is correlated with PV capacity, PV installation location, and switching time. Therefore, optimizing PV configuration will help reduce distribution network line losses and optimize the economic benefits of distribution network operation. Under the action of energy storage regulation, distribution network line losses under different conditions are further reduced. Furthermore, the energy storage regulation amount obtained by this method is calculated based on the line loss response, without the need for power flow calculation, thus facilitating real-time application in actual distribution network operation. In addition, future research will focus on optimizing line losses and further studying strategies for PV-storage synergistic configuration optimization considering energy storage conditions.

[0034] The apparatus for analyzing the impact of distributed photovoltaic (PV) power generation on distribution network line losses of the present invention includes the method for analyzing the impact of distributed PV power generation on distribution network line losses, and therefore possesses at least all the features and advantages of the method for analyzing the impact of distributed PV power generation on distribution network line losses, which will not be repeated here. Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0035] Figure 1 This is a flowchart of an embodiment of the present invention for analyzing the impact of distributed photovoltaic power generation on distribution network line losses;

[0036] Figure 2 This is a topology diagram of an IEEE 33-node distribution system considering distributed photovoltaics, as disclosed in one embodiment of the present invention.

[0037] Figure 3 This is a graph showing the actual annual power output data of a distributed photovoltaic system according to an embodiment of the present invention.

[0038] Figure 4 This is a distributed photovoltaic scenario sampling diagram disclosed in one embodiment of the present invention;

[0039] Figure 5 This is a comparative graph showing the improvement in distribution network line loss under different photovoltaic penetration rates, as disclosed in an embodiment of the present invention.

[0040] Figure 6 This is a comparison diagram showing the impact of changes in photovoltaic output at different locations on line loss improvement, as disclosed in an embodiment of the present invention.

[0041] Figure 7 This is a comparison chart showing the impact of different switching times on line loss improvement in a single embodiment of the present invention.

[0042] Figure 8 This is a comparison diagram of line loss in a typical scenario for switching scheme 1 and scheme 3 disclosed in an embodiment of the present invention;

[0043] Figure 9 This is a typical distributed photovoltaic scenario diagram in different climates disclosed in an embodiment of the present invention;

[0044] Figure 10 This is a comparison chart of line loss improvement under different climatic conditions disclosed in an embodiment of the present invention;

[0045] Figure 11 This is a line graph showing the normalized power values ​​of different types of loads disclosed in one embodiment of the present invention;

[0046] Figure 12 This is a comparison chart of losses in distributed photovoltaic power grids under different loads, as disclosed in one embodiment of the present invention.

[0047] Figure 13 This is a comparison chart of industrial load and typical distributed photovoltaic power output in one embodiment of the present invention;

[0048] Figure 14 This is a comparison chart of losses in a distributed photovoltaic power grid under industrial load, as disclosed in one embodiment of the present invention.

[0049] Figure 15 This is a comparison chart of losses in a distributed photovoltaic power grid under residential load, as disclosed in one embodiment of the present invention. Detailed Implementation

[0050] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings and specific examples and application scenarios.

[0052] In some embodiments of the present invention, an analysis method is provided for the impact of distributed photovoltaic power generation considering energy storage regulation on distribution network line losses: First, the nonlinear spatiotemporal correlation of photovoltaic output is accurately described using the joint probability distribution function, and the scenario method is used to accurately characterize the simulated distributed photovoltaic output, providing data support for subsequent analysis; Second, the root mean square current method, which is more in line with practical applications, is adopted for the line loss calculation model, improving the practicality of the line loss analysis results and mechanisms.

[0053] To achieve efficient and accurate distribution network line loss analysis, this invention proposes an energy storage regulation optimization algorithm based on the responsibility of node complex power changes. This method significantly improves computational efficiency by quantifying the dynamic response characteristics of system losses to node power fluctuations, providing a reliable theoretical basis for real-time regulation. In simulation experiments on the IEEE 33-node distribution system, the system comprehensively evaluated multiple factors, including distributed photovoltaic capacity configuration, spatial distribution, operating status, and regional climate and load characteristics. The system revealed the mechanism by which grid-connected photovoltaics affect distribution network line losses and empirically demonstrated the significant effect of energy storage systems in improving line losses.

[0054] In some embodiments of the present invention, a systematic calculation method is proposed for analyzing the impact of distributed photovoltaic power generation on distribution network line losses. The overall process is as follows: Figure 1 As shown in the diagram, firstly, based on historical photovoltaic (PV) power generation data (including voltage V, current I, and power P), a probability density function for PV output is constructed using kernel density estimation. A joint probability distribution model is then established using the Joe estimation function to accurately characterize the spatiotemporal correlation and uncertainty of PV output. This function effectively captures the positive and negative correlations between variables, thus more realistically reflecting the output characteristics of distributed PV. Secondly, through sampling and inverse transformation of the joint probability distribution function, typical daily curves considering the correlation and randomness of PV output are generated, providing data support for subsequent line loss analysis. Finally, using a daily scenario as the calculation period, the actual operating state of the distribution network is simulated, and an energy storage regulation strategy based on loss response is introduced. Through a multi-objective adaptive evolutionary algorithm based on heuristic mutation, the charging and discharging behavior of energy storage is dynamically optimized to minimize line losses, achieving efficient and economical operation of the distribution network. This method not only fully considers the volatility and spatial correlation of PV output but also effectively improves the stability and economy of the distribution network through energy storage regulation, providing a new technical path for line loss analysis in the context of distributed PV integration.

[0055] The losses in a distribution network are the cumulative sum of power losses of all electrical components in the network over time. However, in actual operation, the power losses of electrical components change dynamically over time, making it difficult to describe precisely with an explicit expression. Therefore, a calculation model for line losses is designed using mathematical statistics methods combined with certain assumptions.

[0056] Assume the equivalent resistance of the power grid line is The equivalent current of the line is Then the active power loss of this line is:

[0057] (1)

[0058] in, This is a correction factor to reflect subtle differences in actual operation.

[0059] To obtain a more comprehensive picture of energy loss, it is necessary to integrate the instantaneous energy loss over 24 hours:

[0060] (2)

[0061] In practical applications, current As an instantaneous variable, its precise measurement presents technical challenges. Approximate equivalence can be achieved by dividing the observation time into small time intervals. A common method in engineering practice is to utilize typical 24-hour hourly load data, assuming that the current remains relatively stable within each hour. The total loss of the distribution network can be expressed as the sum of the equivalent resistive losses of all lines and devices:

[0062] (3)

[0063] In the formula: and These are the equivalent resistances of different components such as lines and transformers; and These refer to the total number of lines and the total number of transformers, respectively. The calculation period is in hours. , These are the correction factors for each component; and These are the equivalent currents of the corresponding components.

[0064] To obtain these equivalent currents, statistical calculations can be performed based on hourly power data over a 24-hour period.

[0065] (4)

[0066] In the formula: and These represent the active and reactive power at time t, respectively. This represents the voltage at the corresponding moment.

[0067] In the practice of energy conservation and loss reduction in power distribution networks, traditional technical approaches mainly revolve around several key dimensions:

[0068] (1) Optimize the power supply radius: By reducing the length of low-voltage lines, the power loss caused by line impedance during current transmission is reduced.

[0069] (2) Reactive power compensation strategy: Adjust the power factor to reduce the adverse effects of reactive power on system losses.

[0070] (3) Load balancing management: By balancing the three-phase load, the negative impact of three-phase imbalance on system losses is reduced.

[0071] (4) Transformer capacity configuration: Select transformer capacity reasonably and optimize transformer operating efficiency.

[0072] With the widespread integration of distributed photovoltaic (PV) power generation, the current and voltage distribution of the power distribution network has become more complex. A PV power plant can be considered a controllable load source with time-varying characteristics. At the grid connection point, existing loads can be comprehensively coupled with PV output:

[0073] (5)

[0074] In the formula, For overall power, and These are the load power and photovoltaic output, respectively.

[0075] From the perspective of distribution network loss, a specific time segment is selected to compare the line losses before and after photovoltaic access, as shown in equation (6):

[0076] (6)

[0077] In the formula, , These represent the total number of system nodes and the number of photovoltaic connections, respectively. and They are nodes Load power and node voltage, and These refer to the photovoltaic output and voltage of a photovoltaic system.

[0078] The change in loss can be expressed as:

[0079] (7)

[0080] As can be seen from the above formula, the system impact of photovoltaic (PV) grid connection exhibits complex nonlinear characteristics, and the location and capacity of PV grid connection have varying degrees of influence on line losses:

[0081] (1) Access capacity: Reasonable selection of photovoltaic capacity can effectively reduce system network loss.

[0082] (2) Penetration critical point: When the photovoltaic penetration rate is too high, it may cause problems such as reverse power flow, power backfeed, increased grid line loss, and voltage over-limit risk.

[0083] (3) Sensitivity of access location: When the access location is close to the power supply side, the voltage support for the end load is limited and the line loss improvement effect is weak; when the access location is close to the end of the line, the voltage support effect is significant and the network loss improvement effect is more obvious.

[0084] This invention analyzes multiple factors affecting losses, including photovoltaic grid connection capacity, photovoltaic grid connection location, photovoltaic grid commissioning and decommissioning conditions, distribution network topology, regional climate characteristics, and load type diversity.

[0085] In the context of distributed photovoltaic (PV) power generation being integrated into the distribution network, this invention proposes a method for assessing line losses. By introducing a line loss improvement rate index, the impact of distributed PV on distribution network losses is quantified, as shown in equation (8):

[0086] (8)

[0087] in, This represents the improvement rate of line loss; the smaller the value, the better the improvement in line loss.

[0088] Energy storage technology plays a crucial role in modern power distribution networks, offering unique advantages such as fast response, high energy conversion efficiency, long lifespan, and low geographical requirements. These characteristics enable energy storage systems to flexibly adapt to grid demands and achieve efficient energy management. Specifically, energy storage systems can absorb and charge electricity during off-peak hours and release it during peak hours, effectively achieving peak shaving and valley filling.

[0089] For 10kV medium-voltage distribution networks, this invention focuses on battery energy storage systems (ESS) and selects to deploy energy storage on the network side of the distributed photovoltaic (DG) grid connection point to effectively adjust the uncertainty of photovoltaic output.

[0090] To suppress line losses, this invention introduces loss response as the core calculation basis for energy storage regulation. For radial distribution networks, the system power loss variation caused by multiple nodes, i.e., line losses, can be approximated as:

[0091] (9)

[0092] In the formula, Indicates the resistance of the circuit; This indicates the change in power at the network point; and The lines are respectively The conjugate value of the power change at the first node and the conjugate value of the power change amount; For the line The conjugate value of the first node voltage; This is the conjugate value of the grid-connected node voltage. To extract the real part of the complex number, ensure that the calculation result is the active power loss.

[0093] Net power change at grid connection point for:

[0094] (10)

[0095] In the formula, For changes in the power of the energy storage system; Changes in photovoltaic power generation output; This refers to changes in load power.

[0096] Furthermore, the overall power grid loss can be obtained by superimposing the effects of all line losses, as shown below:

[0097] (11)

[0098] In the formula, It represents the collection of the entire power grid lines.

[0099] Loss Response Reflecting the degree of impact of line power changes on system losses:

[0100] (12)

[0101] To ensure reduced system losses, energy storage regulation must meet the following requirements:

[0102] (13)

[0103] in, Indicates energy storage point With the line The equivalent impedance between them.

[0104] Energy storage system constraint model:

[0105] (1) Power balance constraint:

[0106] (14)

[0107] in, express Remaining battery level at any given time and express The charging or discharging power of the energy storage battery at any given time; and These represent the battery's charging efficiency and discharging efficiency, respectively.

[0108] (2) Power limitation constraints:

[0109] (15)

[0110] in, and This indicates the maximum allowable charging and discharging power of the energy storage battery.

[0111] (3) Power safety constraints:

[0112]

[0113] In the formula, and Indicates the remaining power of the energy storage The lower limit and upper limit.

[0114] To accurately achieve dynamic distribution network optimization through energy storage regulation, this study proposes a multi-objective adaptive evolutionary algorithm based on heuristic mutation. This algorithm effectively balances global exploration and local exploitation through an intelligent search strategy, providing an efficient solution for complex distribution network optimization problems.

[0115] An improved heuristic mapping method is used to generate the initial population, and its mapping equation is defined as:

[0116] (16)

[0117] In the formula, \alpha \in \left [ {0,1} \right ] Heuristic parameters; This is the mapping value for the current iteration.

[0118] Location of evolutionary individual generation:

[0119] (17)

[0120] In the formula, The vector generated by the heuristic mapping function, and They have the same dimension; and To optimize the upper and lower bounds of variables, Population size.

[0121] Individual location representation:

[0122] (18)

[0123] in, For the first The individual in the first The position of the dimension From a population perspective.

[0124] Fitness ranking is a key technique in multi-objective optimization, used to effectively filter the fitness levels of different individuals. For two solutions... , Their superiority-inferiority relationship is defined as: if Outperforms on all objective functions Then it is called Superior :

[0125] (19)

[0126] in, The p-th objective function value.

[0127] Density sorting evaluates the distribution characteristics of the solution set, and the calculation process consists of three steps:

[0128] First, calculate the search space density:

[0129] (20)

[0130] In the formula, and It is the i-th solution. Adjacent values ​​of a dimension; and For the first The maximum and minimum values ​​of the dimension.

[0131] Secondly, calculate the target spatial density:

[0132] (twenty one)

[0133] In the formula, and For the first The solution is at the th solution. The neighboring values ​​of the objective function; and Then the upper and lower bounds of the corresponding function value.

[0134] Finally, calculate the overall density:

[0135] (twenty two)

[0136] Individual updates during the exploration phase are as follows:

[0137] (twenty three)

[0138] In the formula, To update the location during the exploration phase; Current optimal solution; This is the scaling factor.

[0139] Individual updates during development phase :

[0140] (twenty four)

[0141] Adaptive boundary:

[0142] (25)

[0143] in, , This represents the current iteration number. This represents the maximum number of iterations.

[0144] The proposed multi-objective adaptive evolutionary algorithm based on heuristic mutation mainly consists of population initialization, a heuristic mutation exploration phase, an development phase, and a solution evaluation mechanism based on superiority ranking and density ranking. Unlike traditional random initialization, this algorithm introduces a heuristic mapping (H) to generate the initial population, enabling a more even and purposeful distribution of individuals in the search space, laying a solid foundation for subsequent evolution. During population evolution, a superiority ranking mechanism identifies solutions with relative advantages, and individuals exhibiting excellent performance across various indicators are selected based on rigorous comparisons of multiple objective functions. Furthermore, a comprehensive density ranking is used in the superior solution set. As a secondary selection criterion, it considers not only the distribution of the search space but also the distribution characteristics of the target space, comprehensively measuring the diversity and representativeness of the solutions. This mechanism ensures that the evaluation of a solution depends not only on the performance of the objective function but also on its good distribution characteristics in the solution space. The algorithm designs a dynamic balance mechanism in two phases: the exploration phase focuses on global search to guide the population to converge toward the optimal solution region; the development phase focuses on detailed local search to improve the accuracy of the solution. Through an adaptive scaling factor, the search strategy is dynamically adjusted to effectively balance global exploration and local development.

[0145] Using the aforementioned scheme of this invention, the IEEE 33 bus system was selected as the simulation platform to explore the multidimensional impact of distributed photovoltaic power generation on power distribution system losses in the Matlab environment. The system topology is as follows: Figure 2 As shown, the bus voltage is set to 1.02 pu, and the reference voltage is 10kV. The photovoltaic power generation system is connected from nodes 18, 22, and 33, and the annual measured output data of grid-connected photovoltaic power from a certain industrial park is used as the research sample. The raw data is then normalized (e.g., ...). Figure 3As shown), combined with the scene generation method innovatively proposed in this invention, 100 typical photovoltaic power output scenarios are finally obtained (such as...). Figure 4 (As shown).

[0146] The following section primarily analyzes the impact mechanism of distributed photovoltaic (PV) grid integration on system losses. PV penetration rate, defined as the ratio of PV installed capacity to distribution transformer capacity, is a crucial parameter for quantifying the scale of distributed PV integration. By introducing the PV penetration rate index, the impact of different PV power generation capacities on the operating characteristics of the distribution network can be systematically assessed. The mathematical expression can be described as:

[0147] (26)

[0148] in, Representative node Photovoltaic installed capacity at the location; Represents a node Distribution transformer capacity.

[0149] This invention focuses on a systematic analysis of the regulatory effect of photovoltaic power generation on distribution network losses from three dimensions: photovoltaic penetration rate, grid connection location, and switching time.

[0150] First, using generated scenario data, the changes in line loss were tested under five different photovoltaic penetration rates (10%, 20%, 40%, 60%, 80%, and 100%). The research results (e.g.) Figure 5 The study revealed a nonlinear pattern: as photovoltaic penetration increases, line losses initially decrease and then increase. This indicates that moderately increasing photovoltaic output is beneficial for reducing line losses, but excessive penetration may lead to power backflow, increasing network current carrying capacity and consequently causing increased losses. Introducing energy storage for regulation not only further improves network losses but also effectively suppresses the increase in line losses caused by surplus photovoltaic output.

[0151] Secondly, by analyzing the changes in photovoltaic output at three different nodes (18, 22, and 33), the impact of grid connection location on line loss was studied (e.g., Figure 6 The results show that: Node 18 is located at the end of the network, and its photovoltaic capacity change has the most significant impact on network loss, with the network loss improvement rate changing from less than 1 to greater than 1; the increase in photovoltaic output of nodes 22 and 33 is generally beneficial to reducing network loss; energy storage regulation helps to reduce network loss to some extent, but due to safety constraints, its ability to reduce loss is limited in the case of photovoltaic output backfeed.

[0152] Finally, the impact of three different photovoltaic switching time schemes (6:00-22:00, 8:00-20:00, 10:00-18:00) on distribution network losses was studied (e.g., Figure 7The results showed that Scheme 1 and Scheme 2 had similar degrees of improvement in grid loss; Scheme 3 had a significantly weaker effect on grid loss improvement; after introducing energy storage regulation, the grid loss of Scheme 3 was improved to a certain extent; excessively shortening the photovoltaic switching time would significantly reduce the grid loss improvement effect.

[0153] Figure 8 The study further demonstrates the loss variations of Scheme 1 and Scheme 3 under typical daily load and photovoltaic output scenarios when the photovoltaic capacity factor is 0.8. The comparative results show that Scheme 3 suffers from problems of late deployment and premature shutdown; the effect of photovoltaic power in improving grid losses is weakened due to improper deployment and shutdown timing; and energy storage has a certain loss reduction effect when photovoltaic output is insufficient.

[0154] Due to the uncertain impact of climate change on distributed photovoltaic power generation, this invention constructs a system by dividing annual photovoltaic output data into quarterly segments and using K-means clustering. The study identified typical scenarios for each quarter and analyzed the probability of occurrence for each scenario. The results (such as...) Figure 9 The data clearly shows the significant differences in photovoltaic output between seasons: photovoltaic output in summer and autumn is significantly higher than that in winter and spring, while photovoltaic output in winter is at its lowest level.

[0155] Based on typical scenarios, this invention further evaluates the loss improvement rate of photovoltaic power output connected to the distribution network. By weighted calculations for each typical scenario, a seasonal comprehensive evaluation result is obtained (e.g., ...). Figure 10 The results showed that distributed photovoltaic (PV) grid connection in summer and autumn had the most significant effect on improving line losses, while the improvement in power output in winter was relatively the smallest. Under different climatic conditions, the energy storage system always played a key regulatory role, and could continuously and effectively improve the line loss improvement effect.

[0156] The diversity of load types adds further complexity to distribution network loss analysis. This invention examines four typical load types—industrial, agricultural, commercial, and residential—and analyzes their differential impact on distribution network losses. While the power of distribution network loads approximately follows a normal distribution, the expected power and standard deviation of different loads vary significantly across different time periods. Figure 11 The normalized curves of typical daily active power expectation values ​​for four load types are presented, revealing the diversity of load characteristics.

[0157] By assessing losses in three scenarios—no PV grid connection, distributed PV grid connection, and energy storage regulation—the study reveals the significant impact of load type on network losses (e.g., ...). Figure 12 The highest losses occur under industrial loads, while the lowest losses occur under agricultural loads. Figure 13 A comparison of typical daily scenarios for industrial load and photovoltaic (PV) power clearly demonstrates the significant differences between peak PV output and peak load periods.

[0158] like Figure 14 As shown, when there is a severe mismatch between photovoltaic (PV) output and load timing characteristics, the loss reduction effect of PV will be greatly reduced. Take residential load as an example (e.g., Figure 15 During midday when power output is low, high-capacity photovoltaic (PV) output can easily lead to a large surplus of power, causing backfeeding and increasing network losses. This phenomenon highlights the importance of coordinating PV power generation systems with loads. Against this backdrop, energy storage systems demonstrate their significant value as a flexible resource, dynamically balancing PV output based on real-time network conditions, effectively reducing network losses and improving overall economic efficiency.

[0159] For any part of this invention not described in detail, please refer to the prior art or the art known to those skilled in the art.

[0160] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of the present invention.

Claims

1. A distributed photovoltaic power generation impact on power distribution network line loss analysis method, characterized in that, The method includes the following contents: Based on the historical data of photovoltaic power generation, the joint probability distribution function is established by kernel density estimation method and Joe estimation function to accurately describe the spatio-temporal correlation and uncertainty of photovoltaic output; the typical day curve considering the correlation and uncertainty of photovoltaic output is generated by sampling and inverse transformation of the joint probability distribution function; Based on the typical day curve, the energy storage adjustment strategy based on loss response degree is introduced, and the charging and discharging behavior of energy storage is dynamically optimized by using the multi-objective adaptive evolutionary algorithm based on heuristic mutation to minimize the line loss, and the distribution network model is constructed; The active power loss of each line and element of the distribution network is calculated, and the influence of the integrated power at the grid connection point after the photovoltaic access on the line loss is considered to obtain the line loss calculation result; The multi-objective adaptive evolutionary algorithm based on heuristic mutation includes the following contents: Initialization: the initial population is generated by using heuristic mapping to ensure that the population is uniformly and purposefully distributed in the search space; and the position of the individual in the initial population is initialized; Evaluation and screening: the initial population is sorted and density sorted; Iterative update: the optimal solution after sorting is used to guide the global search to expand the solution space coverage; the adaptive boundary is used for local fine search to improve the quality of the solution; Iteration termination: after reaching the maximum number of iterations, the optimal solution set is output for energy storage charging and discharging strategy optimization; The energy storage adjustment strategy based on loss response degree includes the following contents: The line loss response degree is used as the core basis for energy storage adjustment, the dynamic response characteristics of system loss to node power fluctuation are quantified, the grid point net power is optimized by energy storage charging and discharging strategy, and the line loss is reduced; A constraint model of the energy storage system is constructed to ensure the safety and effectiveness of the energy storage adjustment. 2.The method of claim 1, wherein, The kernel density estimation method uses Gaussian kernel function for density estimation. 3.The method of claim 1, wherein, The root mean square current method is used for active power loss calculation, and based on the 24-hour integral point load data, the active power loss of each line and element of the distribution network is calculated. 4.The method of claim 1, wherein, The integrated power at the grid connection point after the photovoltaic access is the coupling of load power and photovoltaic output.

5. The method of claim 1, wherein the method further comprises: The specific contents of evaluation and screening include: The individual objective function value is calculated, and the advantage solution is screened based on the superior-inferior sorting; The density sorting is performed on the advantage solution to retain the individuals with uniform distribution and good performance. 6.The method of claim 1, wherein, The iterative update includes the exploration stage and the development stage; The exploration stage performs global search: the current optimal solution is used as the guide to update the position of the first N / 2 individuals to enhance the global convergence; N is the population size; The development stage performs local search: the adaptive boundary is introduced for local fine search to improve the accuracy of the solution. 7.The method of claim 1, wherein, The constraint model of the energy storage system includes power balance constraint, power limit constraint and power safety constraint.

8. A distributed photovoltaic power generation distribution network line loss influence analysis device, comprising a processor and a memory, the memory has a software program stored thereon, characterized in that, When the processor runs the software program, the contents of the distributed photovoltaic power generation influence analysis method for distribution network line loss of any one of claims 1-7 can be realized.

Images