Optimal design method and device for optimal absorption photovoltaic power station

Abstract

The invention discloses an optimal consumption photovoltaic power station optimization design method and device, and the method comprises the following steps: (1), collecting load data through an intelligent electric meter, carrying out the preprocessing of the load data, and carrying out the curve analysis, and obtaining the load characteristics of a load curve; (2) under the condition of the load curve characteristics obtained in the step (1), in combination with a photovoltaic output model, performing optimization design on the photovoltaic power station by adopting a single component arrangement mode algorithm based on a PSO algorithm to obtain a designed component arrangement mode; and (3) based on the characteristics of the load curve in the step (1), repeating the step (2) for multiple times, designing the comprehensive power station, and assisting with PSO algorithm acceleration to obtain the optimal design of the comprehensive power station and complete the absorption target. According to the method, only the photovoltaic power station is designed through an algorithm strategy under the condition that an energy storage device and a scheduling algorithm are not used, and the operation efficiency of a power station design algorithm is improved through a PSO algorithm. Therefore, the investment cost of the power station in the early stage can be greatly reduced, the cost and difficulty of later operation and maintenance can be reduced, and an effective and economical solution is provided for sustainable construction and operation of the photovoltaic power station.

Description

Technical Field

[0001] The invention discloses a photovoltaic power station optimization design method and device for optimal consumption, belonging to the field of photovoltaic power station design. Background Art

[0002] As global demand for energy grows and the goal of achieving carbon neutrality is advanced, solar photovoltaics play an increasingly critical role in the energy structure. However, due to the discontinuity and instability of solar power generation, it poses challenges to the dispatch and stable operation of the power grid. In areas where solar power generation is more developed, when power generation exceeds electricity demand, the excess electricity is difficult to be effectively utilized, and sometimes even leads to abnormal electricity market prices (such as "negative electricity prices"), which poses a threat to the economy and sustainability of the power grid. Therefore, researchers at home and abroad have conducted a lot of technical research on the problem of solar power generation absorption, in order to find optimization strategies to improve the absorption capacity of photovoltaic power stations.

[0003] At present, the main research direction for solving the problem of absorption is to use energy storage systems and scheduling schemes to alleviate the volatility of solar power generation. For example, Wan Yuexiang et al. proposed a method for energy storage planning for distributed photovoltaic medium-voltage distribution networks, and realized the automated planning of energy storage by in-depth research on energy storage technology. Jin Li et al. proposed a "generation-energy storage-load" coordinated scheduling strategy, which takes into account the demand response of the load side and electrochemical energy storage system at different time scales to achieve coordinated operation of load and energy storage and improve the absorption capacity of power stations. Zhang Ruixue et al. discussed an accurate load forecasting technology combining weather forecast and deep learning, predicted power demand and solar power generation through algorithms, and designed a dynamic reconstruction strategy for distribution networks to optimize the absorption effect of solar energy. Chen Bin et al. proposed a solar energy absorption optimization scheme suitable for AC / DC hybrid distribution stations. By establishing a joint output model of thermal power, solar energy and power equipment, and adopting an energy interaction control method, the absorption performance of photovoltaic power stations was effectively improved.

[0004] Although many studies have proposed various strategies to solve the problem of PV power station consumption, there are still some shortcomings, mainly including: 1) Most of the current research focuses on the adjustment of energy scheduling strategies, and these adjustments often require the configuration of energy storage facilities, which have relatively high costs; 2) The research content fails to fully consider the actual environment of PV power station construction, resulting in greater difficulty in its implementation. Summary of the invention

[0005] The present invention adopts an algorithm strategy to design photovoltaic power stations without the help of energy storage devices and scheduling algorithms, and accelerates the operation efficiency of the power station design algorithm through the PSO algorithm (particle swarm algorithm). This can not only greatly reduce the initial investment cost of the power station, but also reduce the cost and difficulty of later operation and maintenance, providing an effective and economical solution for the sustainable construction and operation of photovoltaic power stations.

[0006] The technical solution of the present invention is as follows:

[0007] A photovoltaic power station optimization design method for optimal consumption includes the following steps:

[0008] (1) Collect load data through smart meters, pre-process the load data, and perform curve analysis to obtain the load characteristics of the load curve;

[0009] (2) Under the load curve characteristic conditions obtained in step (1), combined with the photovoltaic output model, a single component layout algorithm based on the PSO algorithm is used to optimize the design of the photovoltaic power station to obtain a designed component layout;

[0010] (3) Based on the characteristics of the load curve in step (1), step (2) is repeated multiple times, and a comprehensive power plant design is performed, assisted by the PSO algorithm to accelerate the optimization design of the comprehensive power plant and achieve the consumption target.

[0011] Preferably, the specific steps of the above step (1) are as follows:

[0012] (1-1) Load data collection: smart meters are used to continuously monitor data over a period of time. Data is recorded in hours, minutes or seconds to capture fluctuations in load changes.

[0013] (1-2) In order to ensure the accuracy of the data, the collected data needs to be preprocessed. The preprocessing steps include outlier cleaning and missing value filling. Outlier cleaning requires identifying and removing unreasonable load data, such as instantaneous abnormal mutation data. Missing value filling mainly fills in missing data points to ensure their continuity. Interpolation or averaging methods can be used for filling;

[0014] (1-3) After completing data collection and preprocessing, the basic features of the load curve need to be extracted. The main extracted features include time features and frequency features. Time features include the time distribution of peaks and troughs and the specific time and value of occurrence. This feature can reflect the user's electricity demand in a specific period of time. The load peak period usually occurs in the morning and afternoon, while the load trough period usually occurs in the late night or early morning. The frequency characteristics mainly include the periodic components in the load curve, which can extract the daily, weekly and monthly changes in the load, reflect the user's electricity consumption habits, and provide a reference for periodic power dispatch and load management;

[0015] (1-4) Finally, fluctuation analysis is to be carried out. Fluctuation analysis is to study the changing trend of the load curve on a time scale. Its main purpose is to analyze the amplitude of changes around peaks and troughs. The measurement standard for fluctuation analysis is the Pearson correlation coefficient, which is used to measure the degree of correlation between two variables and is usually represented by the symbol r. Its value is between -1 and 1. When r = 1, the two variables are completely positively correlated, that is, there is a perfect linear relationship between the two variables, and one of them increases as the other increases; when r = -1, the two variables are completely negatively correlated, that is, one of them decreases as the other increases; when r = 0: the two variables are not correlated, that is, one of the variables is independent of the increase or decrease of the other variable. The specific mathematical model of the Pearson correlation coefficient is:

[0016]

[0017] Where: X i is the i-th sample in variable X; X is the average value of sample X; Y i is the i-th sample in variable Y; Y is the mean of sample Y.

[0018] Preferably, the specific steps of the above step (2) are as follows:

[0019] (2-1) The inclination and azimuth of the components in the power plant design are used as the solution domain of the PSO algorithm, and the layout of the power plant components is represented by the particle position.

[0020] (2-2) Construct the objective function of the single component layout algorithm based on the PSO algorithm, introduce the photovoltaic power station output model and the Pearson coefficient to construct the objective function of the algorithm, and its mathematical expression is:

[0021] f(x)=A·abs(T energy -T grid )-Br

[0022] Where: A and B are empirical coefficients, A defaults to 2, B defaults to 1; T energyis the peak moment of the photovoltaic power station output model; T grid is the peak moment of the load curve; r is the Pierce coefficient of the two curves;

[0023] (2-3) Based on the objective function and solution domain, the fitness function of the PSO algorithm is constructed to optimize the PSO algorithm for the single component layout method;

[0024] (2-4) Then the output of the photovoltaic power station is calculated to obtain the hourly power generation data for the whole year. Through aggregation operation, the hourly transformation data of the day is obtained and compared with the load curve for analysis.

[0025] Preferably, the specific steps of the above step (3) are as follows:

[0026] (3-1), according to the load curve characteristics, repeat step (2) multiple times to obtain a power station design scheme for a single component layout under different load characteristics;

[0027] (3-2), the number of components of the power plant design schemes with single component layout under different load characteristics is defined as the solution space, where the particle position represents the number of components of each power plant design;

[0028] (3-3) Construct the objective function of the comprehensive power station optimization algorithm based on the PSO algorithm, and its mathematical expression is:

[0029]

[0030] Where: p represents the time; T MAX is the maximum moment; N is the number of characteristics of the load curve, E pq,ac is the power generation of the qth optimal arrangement at the pth time moment; E p,grid is the load power at the pth moment, x q is the number of components required under the qth optimal layout;

[0031] (3-4) According to the power plant design plan of the single component layout under each load characteristic and the corresponding component quantity, the comprehensive design of the power plant is carried out. This result can achieve the effect of power consumption, and its result can achieve the effect of consumption.

[0032] The photovoltaic output model steps in the above steps (2) and (3) are as follows, and the flow chart is as follows: Figure 2 As shown,

[0033] (1) The horizontal irradiance model is connected to the Meteonorm meteorological database. The horizontal irradiance data obtained;

[0034] (2) The horizontal irradiance data obtained by the model is further combined with the arrangement of components to calculate the irradiance on the inclined surface;

[0035] The inclined surface irradiance model in this algorithm is an anisotropic model, which consists of three parts: direct irradiation, scattered irradiation, and ground reflected irradiation, which can be expressed as:

[0036] G tilt =G tilt,b +G tilt,d +G tilt,r

[0037] G tilt,b is the direct irradiance on the inclined surface, and its calculation formula is:

[0038] G tilt,b =G b R b

[0039] Where: G b is the direct irradiance on the horizontal plane; R b It is the ratio of direct irradiance on the inclined plane to that on the horizontal plane, which can be expressed as:

[0040]

[0041] Where: θ Z is the solar zenith angle; θ is the solar incidence angle, and the specific formula is:

[0042] θ=arccos(cosθ Z cosβ+sinθ Z sinβcos(γ s -γ))

[0043] Where: β is the tilt angle of the component tilt surface; γ is the tilt azimuth, with the south direction as 0°; γ s ——Solar azimuth.

[0044] G tilt,d is the scattered radiation on the inclined surface, and its calculation formula is:

[0045]

[0046] Where: G d is the scattered irradiance on the horizontal plane; F1 is the circumstellar coefficient; F2 is the horizontal brightness coefficient; a is max{0, cosθ Z}; b is max{cos85°, cosθ Z}

[0047] G tilt,r is the ground reflected radiation, and its calculation formula is:

[0048]

[0049] Where: G is the total horizontal irradiance; is the ground reflectivity.

[0050] (3) Calculate the output of the PV module by combining the obtained irradiance with the performance of the PV module itself;

[0051] The battery cell of the module can be equivalent to a single diode circuit model. The equivalent circuit diagram is as follows: Figure 3 As shown:

[0052] The characteristic relationship equation of the output current and voltage of the photovoltaic cell simulated by the ideal single diode circuit model, that is, the IV characteristic equation can be expressed as:

[0053]

[0054] Where: I L is the photoinduced dark current; I o is the saturation current of the equivalent diode; R sh is the parallel resistance; Rs is the series resistance; T is the temperature of the battery; N s is the number of solar cells connected in series; γ is the diode ideal factor; k is the Boltzmann constant (1.38×10-23J / k); q is the charge (1.6×10-9C); α is the modified ideal factor based on the component.

[0055] The key to solving the IV characteristic equation is to obtain I L ,I o ,α,R s , R sh The five parameters, including the power factor, are used to obtain the output power characteristics of the module. These parameters can be calculated based on the nameplate data under standard test conditions of the photovoltaic module. Under different environmental conditions, these parameters need to be corrected. The correction formula is as follows:

[0056]

[0057]

[0058] Where: T C is the component operating temperature; E g is the energy band gap of the material; μ is the short-circuit current temperature coefficient; G is the irradiance on the component surface.

[0059] (4) Through the formula

[0060]

[0061] Using the above formula and Monte Carlo iteration method, the DC side power generation of the photovoltaic module can be calculated. Through the inverter model, the photovoltaic output power on the AC side can be obtained. The inverter model used in this article is the SUNDIA model, which is a linearized model proposed by Sandia National Laboratories in the United States. The AC output power calculation formula of the inverter is as follows:

[0062]

[0063] Where: P ac is the AC output power; P dc is the input DC power; P aco is the maximum output power under standard working conditions; P dco P is the input DC power under standard working conditions; so V is the inverter startup DC power, that is, the inverter self-consumption DC power; dc is the DC input voltage; V dco is the input DC voltage under standard working conditions; C0-C3 are the empirical coefficients of the model, which are usually set to 0, but can be obtained through experimental data to improve the accuracy of the model.

[0064] The particle swarm algorithm steps in the above steps (2) and (3) are as follows:

[0065] The particle swarm optimization (PSO) algorithm is an optimization technique that simulates swarm intelligence and is inspired by the foraging behavior of birds. In this algorithm, the potential solutions to the optimization problem are regarded as "particles" in the search space, each particle represents a possible solution, and the best solution in the group is likened to the food that the flock of birds is looking for. The execution steps of PSO are as follows:

[0066] 1) Initialization When the particle swarm algorithm is started, the position and velocity of the particles are randomly initialized.

[0067] 2) Evaluate the fitness of particles. Fitness is an indicator for evaluating the quality of particle solutions. It guides the search direction and path of the particle swarm during the optimization process. The fitness evaluation criteria depend on the specific optimization problem and objective function. For the objective function f(x), the fitness function Fit(x) can be defined as:

[0068]

[0069] 3) Update the personal best position of the particle. Compare the fitness value calculated in the current iteration with the best fitness value of the particle in history. If the current value is better, update the personal best position of the particle, which is recorded as pbest.

[0070] 4) Update the best position of the particle swarm. Compare the fitness value in the current iteration with the best fitness value in the history of the entire particle swarm. If the current value is better, update the best position of the entire particle swarm, which is recorded as gbest.

[0071] 5) Iteratively update the speed and position of the particle. For a particle, its position is represented by X and its speed is represented by V. In the tth iteration, the update formula of the speed and position is as follows:

[0072]

[0073] Where: m represents the position of the first dimension in the particle search space, n represents the position of the second dimension; ω is the inertia weight, which affects the continuity of the particle's previous velocity; c1 is the individual learning factor, c2 is the group learning factor, and the value is usually 2; r1 is the individual random factor, r2 is the group random factor, which is a random number between [0,1] to increase randomness; is the individual optimal position of the particle; Best location for the group.

[0074] For the inertia weight ω, a larger value can make the particle explore more widely in the search space, but may cause the particle to miss the optimal area or become unstable; a smaller value can enhance the local search ability of the particle, but may cause it to fall into the local optimum too early, limiting the comprehensive exploration of the search space. Therefore, the inertia weight needs to strike a balance between global search and local search, which can be achieved by dynamically adjusting the value of ω. This paper adopts the linear decreasing weight (LDW) strategy to adjust ω, that is, starting from a larger initial value w max Decreases linearly to a smaller final value w min This method helps to expand the search range in the early stage of optimization to find the global optimal solution, and narrow the search range in the later stage of optimization to accurately find the optimal solution. For the tth iteration, the formula of the LDW strategy is as follows:

[0075] w t =(w max -w min )(t max -t) / t max +w min

[0076] Where: t max is the maximum number of iterations.

[0077] The present invention also discloses a photovoltaic power station optimization design device for optimal consumption, comprising a processor and a memory, wherein the memory is used to store a computer program, and the processor is used to execute the computer program to perform any of the methods described above.

[0078] The beneficial effects of the present invention are:

[0079] The present invention simulates the optimization scheme and compares and analyzes the absorption ratio of the traditional power station. The results show that the photovoltaic power station design scheme optimized by the algorithm has a certain absorption capacity for the load, which can greatly improve the self-generation and self-use capacity of the photovoltaic power station; at the same time, the economic analysis of the scheme shows that the optimization scheme has higher economic benefits than the traditional layout scheme in a negative electricity price environment. This paper only adopts the design of photovoltaic power stations to achieve self-absorption of power stations. Without the support of intelligent scheduling and energy storage technologies, it only redesigns the power station to improve the absorption ratio of the power station. This can not only greatly reduce the initial investment cost of the power station, but also reduce the cost and difficulty of later operation and maintenance, and provide an effective and economical solution for the sustainable construction and operation of photovoltaic power stations. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 is a flow chart of the design method of the present invention;

[0081] Figure 2 is a flow chart of the PV output model;

[0082] Figure 3 is the equivalent circuit diagram;

[0083] Figure 4 It is the flow chart of particle swarm algorithm;

[0084] Figure 5 is a load curve diagram in the embodiment;

[0085] Figure 6 It is the optimized layout scheme based on the load curve in the embodiment, the final design layout of the power station, and the specific consumption effect diagram;

[0086] Figure 7 In the embodiment, the maximum absorption ratio is taken as the goal, and the traditional component layout method and the optimized layout method are respectively used to optimize the design of the power station, and the absorption evaluation is carried out, and the specific absorption effect diagram is shown;

[0087] Figure 8 It is a comparison chart of the traditional layout absorption ratio, the optimized layout absorption ratio and the standard absorption ratio;

[0088] Fig. 9 It is a graph of the optimized layout economic curve and the traditional layout economic curve. DETAILED DESCRIPTION

[0089] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0090] A photovoltaic power station optimization design method for optimal consumption, such as Figure 1 As shown, the following steps are included:

[0091] (1) Collect load data through smart meters, pre-process the load data, and perform curve analysis to obtain the load characteristics of the load curve;

[0092] (2) Under the load curve characteristic conditions obtained in step (1), combined with the photovoltaic output model, a single component layout algorithm based on the PSO algorithm is used to optimize the design of the photovoltaic power station to obtain a designed component layout;

[0093] (3) Based on the characteristics of the load curve in step (1), step (2) is repeated multiple times, and a comprehensive power plant design is performed, assisted by the PSO algorithm to accelerate the optimization design of the comprehensive power plant and achieve the consumption target.

[0094] Preferably, the specific steps of the above step (1) are as follows:

[0095] (1-1) Load data collection: smart meters are used to continuously monitor data over a period of time. Data is recorded in hours, minutes or seconds to capture fluctuations in load changes.

[0096] (1-2) The steps of preprocessing load data include outlier cleaning and missing value filling. Outlier cleaning requires identifying and removing unreasonable load data; missing value filling is to fill the missing data points to ensure their continuity, using interpolation or mean method to fill;

[0097] (1-3) After completing data collection and preprocessing, the basic characteristics of the load curve are extracted. The extracted characteristics include time characteristics and frequency characteristics. The time characteristics include the time distribution of peaks and troughs and the specific time and value of occurrence. The frequency characteristics include the periodic components in the load curve, and the daily, weekly and monthly change patterns of the load are extracted.

[0098] (1-4) Finally, the fluctuation analysis is carried out. The fluctuation analysis is to study the changing trend of the load curve on the time scale. Its main purpose is to analyze the variation amplitude around the peak and the trough;

[0099] The measurement standard of fluctuation analysis is the Pearson correlation coefficient, which is used to measure the correlation between two variables. It is represented by the symbol r. The specific mathematical model of the Pearson correlation coefficient is:

[0100]

[0101] Where: X i is the i-th sample in variable X; X is the average value of sample X; Y iis the i-th sample in variable Y; is the mean value of sample Y.

[0102] Preferably, the specific steps of the above step (2) are as follows:

[0103] (2-1) The inclination and azimuth of the components in the power plant design are used as the solution domain of the PSO algorithm, and the layout of the power plant components is represented by the particle position.

[0104] (2-2) Construct the objective function of the single component layout algorithm based on the PSO algorithm, introduce the photovoltaic power station output model and the Pearson coefficient to construct the objective function of the algorithm, and its mathematical expression is:

[0105] f(x)=A·abs(T energy -T grid )-Br

[0106] Where: A and B are empirical coefficients, A defaults to 2, B defaults to 1; T energy is the peak moment of the photovoltaic power station output model; T grid is the peak moment of the load curve; r is the Pierce coefficient of the two curves;

[0107] (2-3) Based on the objective function and solution domain, the fitness function of the PSO algorithm is constructed to optimize the PSO algorithm for the single component layout method;

[0108] (2-4) Then the output of the photovoltaic power station is calculated to obtain the hourly power generation data for the whole year. Through aggregation operation, the hourly transformation data of the day is obtained and compared with the load curve for analysis.

[0109] Preferably, the specific steps of the above step (3) are as follows:

[0110] (3-1), according to the load curve characteristics, repeat step (2) multiple times to obtain a power station design scheme for a single component layout under different load characteristics;

[0111] (3-2), the number of components in the power plant design scheme with a single component arrangement under different load characteristics is defined as the solution domain, and the number of components in each power plant design is represented by the particle position;

[0112] (3-3) Construct the objective function of the comprehensive power station optimization algorithm based on the PSO algorithm, and its mathematical expression is:

[0113]

[0114] Where: p represents the time; T MAX is the maximum moment; N is the number of characteristics of the load curve, E pq,acis the power generation of the qth optimal arrangement at the pth time moment; E p,grid is the load power at the pth moment, x q is the number of components required under the qth optimal layout;

[0115] (3-4) Based on the single component layout mode under each load characteristic and the corresponding component quantity under this mode, a comprehensive design of the power station is carried out, and the result can achieve the effect of absorption.

[0116] Example:

[0117] The power load data of a factory was collected through smart meters, and data cleaning and missing value processing were performed to ensure the integrity and accuracy of the data. Since the structure and component layout of a photovoltaic power station are fixed once it is built, the optimal design of the power station must not only ensure high accuracy, but also maintain good absorption capacity in long-term operation. By aggregating the hourly load data throughout the year and simplifying it into hourly change data within a day, its change trend can be analyzed. The change trend is as follows: Figure 5 shown.

[0118] The analysis results show that the load curve of the plant presents a "double peak" shape, reaching peaks at 10 am and 3 pm respectively. Faced with this load characteristic, the traditional single component layout method is difficult to achieve efficient consumption. Therefore, this study adopts a design combining multiple layout methods to achieve a balance between load and photovoltaic power generation.

[0119] Considering the bimodal characteristics of the load curve, this study selected two component layouts that adapt to different load characteristics. Using the power plant absorption design optimization based on the PSO algorithm, the optimal design strategy of the power plant was obtained to achieve effective absorption of the load curve. Specifically, two component layouts with an azimuth of about -85.1° and an inclination of 84° and an azimuth of about 131.2° and an inclination of 56° were adopted, and they were arranged in a ratio of 1:1.7.

[0120] Since the photovoltaic power station cannot generate electricity at night, the consumption period is set from 7 am to 7 pm. Based on the optimized layout plan of the load curve, the specific consumption effect of the power station is as follows: Figure 6 shown.

[0121] With the maximum absorption ratio as the goal, the power station is optimally designed using traditional component layout and optimized layout, and absorption evaluation is carried out. The specific absorption effect is as follows: Figure 7 As shown:

[0122] As can be seen from the figure, the power generation curve of the optimized power station is consistent with the power demand during the period from 10 am to 4 pm, and the absorption effect is significant; in other periods, due to the intermittent nature of photovoltaic power generation, the power generation cannot fully meet the demand. In contrast, the power demand gap of the traditional power station is larger before 11 am and after 2 pm. In addition, the power generated by the traditional power station during the noon period cannot be fully absorbed, resulting in power loss, affecting economic benefits and possibly causing damage to equipment.

[0123] In order to intuitively compare the absorption capacity between the optimized and traditional power stations, it is necessary to calculate the absorption ratio of the power generation curve, and the formula is:

[0124]

[0125] Where: E gen E is the amount of electricity generated for self-use; grid The power consumption of the load.

[0126] This formula quantifies the ability of a photovoltaic power station to generate electricity for its own use, and reflects the absorption effect of different layout schemes. In addition, in order to comprehensively evaluate the matching degree between the power generation curve and the power demand, this paper also introduces the mean absolute error (MAE), root mean square error (RMSE) and mean absolute percentage error (MAPE) as evaluation indicators. The specific mathematical formula is:

[0127]

[0128] Where: n is the total number of time; r max is the maximum absorption ratio, which is 100%.

[0129] Table 1 Comprehensive Error of Consumption Ratio

[0130]

[0131] It can be seen from the figure that the power generation curve of the optimized layout is more consistent with the standard line than the traditional layout in the intervals of 7:00 to 10:00 in the morning and 4:00 to 7:00 in the afternoon, showing a better absorption ratio; and in the time period from 10:00 in the morning to 4:00 in the afternoon, its absorption ratio is basically maintained near the standard line, and the absorption effect is better than the traditional layout.

[0132] The comprehensive error of the absorption ratio in Table 1 shows that the degree of agreement between the optimized layout and the standard line is significantly better than that of the traditional layout, which leads to the conclusion that the power station layout after optimized design can significantly enhance the absorption capacity of the power station.

[0133] In order to cope with the negative electricity price situation, this study deeply analyzed the economic feasibility of the power system under the negative electricity price environment. Negative electricity prices usually occur when the power supply significantly exceeds the demand, causing the electricity price to fall below zero. With the increase in the use of renewable energy, the energy regulation pressure of the power grid increases, and the problem of negative electricity prices becomes more frequent. Therefore, evaluating and optimizing the economic feasibility under the negative electricity price scenario is crucial to the robust operation and sustainable development of the power system. According to the survey, negative electricity prices often occur between 12 noon and 2 pm, with an average electricity price of about 27 euros per megawatt-hour and a maximum of 130 euros per megawatt-hour. In this simulation environment, the negative electricity price is set at 0.27 yuan per kWh, and the purchase price is based on the electricity price of the Jiangsu power grid: 8 am to 12 noon and 5 pm to 10 pm are peak periods, with an electricity price of 1.227 yuan / kWh, 12 noon to 5 pm is flat peak period, with an electricity price of 0.734 yuan / kWh, and midnight to 8 am is the low period, with an electricity price of 0.331 yuan / kWh; the grid-connected electricity price is 0.391 yuan / kWh from 10 am to 12 noon and from 2 pm to 4 pm. After calculating the purchase price and the grid-connected electricity price, the comparison results were obtained. .

[0134] according to Fig. 9 The analysis results show that the optimized power station design is significantly better than the traditional design in terms of overall economic benefits. Especially when negative electricity prices are becoming increasingly prominent, the advantages of the optimized design are more prominent, showing its excellent performance in improving economic benefits and coping with market fluctuations.

[0135] 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 technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A photovoltaic power station optimization design method for optimal consumption, characterized in that: The steps include: (1) Collect load data through smart meters, pre-process the load data, and perform curve analysis to obtain the load characteristics of the load curve; (2) Under the load curve characteristic conditions obtained in step (1), combined with the photovoltaic output model, a single component layout algorithm based on the PSO algorithm is used to optimize the design of the photovoltaic power station to obtain a designed component layout; (3) Based on the characteristics of the load curve in step (1), step (2) is repeated multiple times, and a comprehensive power plant design is performed, assisted by the PSO algorithm to accelerate the optimization design of the comprehensive power plant and achieve the consumption target.

2. The optimal design method for photovoltaic power station with optimal consumption according to claim 1, characterized in that: The specific steps of step (1) are as follows: (1-1) Load data collection: smart meters are used to continuously monitor data over a period of time. Data is recorded in hours, minutes or seconds to capture fluctuations in load changes. (1-2) Preprocess the load data, including outlier cleaning and missing value filling. Outlier cleaning requires identifying and removing unreasonable load data; missing value filling is to fill in missing data points to ensure their continuity, using interpolation or mean method to fill in; (1-3) After completing data collection and preprocessing, the basic characteristics of the load curve are extracted. The extracted characteristics include time characteristics and frequency characteristics. The time characteristics include the time distribution of peaks and valleys and the specific time and value of occurrence. Frequency characteristics include the periodic components in the load curve, and extract the daily, weekly and monthly variation patterns of the load; (1-4) Finally, the fluctuation analysis is carried out. The fluctuation analysis is to study the changing trend of the load curve on the time scale. Its main purpose is to analyze the variation amplitude around the peak and the trough; The measurement standard of fluctuation analysis is the Pearson correlation coefficient, which is used to measure the correlation between two variables. It is represented by the symbol r. The specific mathematical model of the Pearson correlation coefficient is: Where: X i is the i-th sample in variable X; is the mean value of sample X; Y i is the i-th sample in variable Y; is the mean value of sample Y.

3. The photovoltaic power station optimization design method for optimal consumption according to claim 2 is characterized in that: The specific steps of step (2) are as follows: (2-1) The inclination and azimuth of the components in the power plant design are used as the solution domain of the PSO algorithm, and the layout of the power plant components is represented by the particle position. (2-2) Construct the objective function of the single component layout algorithm based on the PSO algorithm, introduce the photovoltaic power station output model and the Pearson coefficient to construct the objective function of the algorithm, and its mathematical expression is: f(x)=A·abs(T energy -T grid )-Br Where: A and B are empirical coefficients, A defaults to 2, B defaults to 1; T energy is the peak moment of the photovoltaic power station output model; T grid is the peak moment of the load curve; r is the Pierce coefficient of the two curves; (2-3) Based on the objective function and solution domain, the fitness function of the PSO algorithm is constructed to optimize the PSO algorithm for the single component layout method; (2-4) Then the output of the photovoltaic power station is calculated to obtain the hourly power generation data for the whole year. Through aggregation operation, the hourly transformation data of the day is obtained and compared with the load curve for analysis.

4. The photovoltaic power station optimization design method for optimal consumption according to claim 3 is characterized in that: The specific steps of step (3) are as follows: (3-1), according to the load curve characteristics, repeat step (2) multiple times to obtain a power station design scheme for a single component layout under different load characteristics; (3-2), the number of components of the power plant design schemes with single component layout under different load characteristics is defined as the solution space, where the particle position represents the number of components of each power plant design; (3-3) Construct the objective function of the comprehensive power station optimization algorithm based on the PSO algorithm, and its mathematical expression is: Where: p represents the time; T MAX is the maximum moment; N is the number of characteristics of the load curve, E pq,ac is the power generation of the qth optimal arrangement at the pth time moment; E p,grid is the load power at the pth moment, x q is the number of components required under the qth optimal layout; (3-4) According to the power plant design plan of the single component layout under each load characteristic and the corresponding component quantity, the comprehensive design of the power plant is carried out. This result can achieve the effect of power consumption, and its result can achieve the effect of consumption.

5. A storage medium for optimal design of photovoltaic power station with optimal consumption, used for storing computer programs, characterized in that: The computer program executes the method as described in any one of claims 1 to 4.

6. A photovoltaic power station optimization design device for optimal consumption, comprising a processor and the storage medium according to claim 5, characterized in that: The processor is used to execute the computer program to perform the method as described in any one of claims 1 to 4.

Images