Distributed photovoltaic storage system with linear power access and its implementation method

Abstract

The present invention provides a distributed photovoltaic storage system with linear power access and its implementation method, including: Step S1: Utilizing a high-precision prediction module for renewable energy output to predict daily photovoltaic power generation and obtain a prediction result; Step S2: Using a linear power modulation module to modulate the daily power generation curve based on the prediction result, converting the steamed bun curve into linear grid-connected power. This system modulates the traditional "steamed bun curve" into a stable linear grid-connected power, dynamically compensating for it through an energy storage system. This reduces the impact of photovoltaic output fluctuations on the grid, improves the grid's ability to accommodate renewable energy, and meets the grid's requirements for stable power supply.

Description

Technical Field

[0001] The present invention relates to the field of smart grid and new energy dispatching and control technology, and in particular to a distributed photovoltaic storage system with direct power access to the grid and an implementation method thereof. Background Art

[0002] With the rapid global development of renewable energy generation, the installed capacity of renewable energy sources such as wind power and photovoltaics continues to grow. However, user photovoltaic installations are often located in rural areas. Due to the rudimentary power grid structure and low distribution network and line capacity in rural areas, large-scale photovoltaic deployments can easily cause reverse power overloads on low-voltage transformers in rural areas, impacting the lifespan of transformers and lines. Furthermore, peak photovoltaic power generation periods can lead to overvoltages and increased harmonics, posing a threat to the safe operation of distribution networks. This results in the classic "red zone" phenomenon, whereby power grids slow down the pace of photovoltaic grid integration and adjust distribution network structures to increase their capacity to absorb photovoltaic power. Furthermore, photovoltaic power generation peaks during midday hours, a characteristic that significantly deteriorates the power supply and demand relationship during daytime hours. In addition to leading to curtailed solar power, the rapidly growing penetration of distributed photovoltaic power will significantly lower midday on-grid electricity prices in the spot market, reducing the return on investment.

[0003] Using energy storage to adjust the processing characteristics of photovoltaics is considered an effective solution. The current mainstream design model is "self-generation for own use, with surplus power fed to the grid." This approach is often applied to distributed industrial and commercial projects. Self-use reduces the owner's transmission and distribution costs, while energy storage is used to collect low-priced photovoltaic power during the afternoon, discharging it during peak evening hours to achieve arbitrage. However, in practice, this model also carries significant risks. First, poor operational sustainability of the power purchasing companies that purchase electricity from distributed photovoltaic storage systems can easily result in photovoltaic storage system investors failing to achieve their expected returns. Second, in the context of the spot market, the market and price differentials between peak and valley electricity prices are erratic, making energy storage arbitrage opportunities highly challenging to the robustness of the profit model. Finally, in the surplus power fed to the grid model, the photovoltaic storage system still shifts the responsibility of energy consumption to the grid, resulting in serious grid-source conflicts. As the penetration rate of photovoltaic storage continues to rise, the system costs of this model will also gradually increase, impacting the sustainability of the photovoltaic storage system.

[0004] With the full marketization of renewable energy, the future profitability of new energy systems will increasingly depend on the state of the electricity market. The development of solar-plus-storage systems will ultimately hinge on their ability to compete with other power sources, such as thermal and hydropower. Therefore, a novel coordinated control strategy for solar-plus-storage systems is urgently needed. This strategy would allow solar-plus-storage systems to mimic the role of other power sources within the system, addressing the negative impact of midday solar power surges on the grid while also improving long-term profitability. This approach also relies heavily on solar power forecasting and power trading strategies. Summary of the Invention

[0005] In view of the defects in the prior art, the purpose of the present invention is to provide a distributed photovoltaic storage system with direct power access to the grid and an implementation method thereof.

[0006] A method for implementing a distributed photovoltaic storage system with linear power access to the grid provided by the present invention includes:

[0007] Step S1: using the new energy output high-precision prediction module to predict the photovoltaic daily power generation to obtain the prediction result;

[0008] Step S2: The linear power modulation module modulates the daily power generation curve according to the prediction results, and modulates the steamed bun curve into a linear grid-connected power;

[0009] The new energy output high-precision prediction module includes: using an empirical regression model to meet the preset requirements of photovoltaic power generation when the weather is clear As a benchmark, the photovoltaic daily power generation is predicted under the given conditions of the obscuration degree SDI and cloud proportion CCI to obtain the prediction results;

[0010] The linear power modulation module includes: counting photovoltaic high-power generation time based on the prediction result, and determining the linear grid-connected power level based on the photovoltaic high-power generation time.

[0011] Preferably, the step S1 includes:

[0012] Step S1.1: Periodically collect cloud image information covering the predicted area using an all-weather imager deployed at a renewable energy power station;

[0013] Step S1.2: pre-processing the collected cloud image information to obtain pre-processed cloud image information;

[0014] Step S1.3: Calculate the current position of the sun and the position at a preset time in the future and the movement trajectory of the clouds based on the preprocessed cloud image information, so as to predict whether the sun will be blocked at the preset time in the future;

[0015] Step S1.4: Calculate the average brightness of the sun area using the image data and compare it with the sunny day benchmark to obtain the degree of obstruction SDI;

[0016] Step S1.5: Segment the cloud layer using pixel thresholding or machine learning methods, and calculate the cloud fraction CCI;

[0017] ;

[0018] Step S1.6: Use the empirical regression model to meet the preset requirements of photovoltaic power generation in clear weather As a benchmark, under the given conditions of shading degree SDI and cloud proportion CCI, the power generation of photovoltaic power generation in the future preset time period is calculated and predicted;

[0019]

[0020] in, Indicates the power generation of the day; GHI indicates the predicted power generation of photovoltaic power generation in the future preset time period.

[0021] Preferably, the step S1.2 includes:

[0022] Step S1.2.1: Filter the cloud image information using a bilateral filtering method to obtain processed cloud image information;

[0023]

[0024] in, Is the output image at position Pixel value of is the input image at position Pixel value of Therefore The neighborhood window centered on is the standard deviation of the spatial domain, which controls the decay rate of the spatial weight; is the standard deviation of the value range, which controls the decay rate of pixel value weight; is a normalization constant used to ensure that the weights sum to 1;

[0025]

[0026] Step S1.2.2: Calculate the spatial weight and pixel value weight, and delete the pixels whose weight values ​​are less than the threshold to obtain the processed cloud image information;

[0027] Spatial weight is calculated based on the distance between pixels in space. The farther the distance, the smaller the weight.

[0028]

[0029] in, Pixels and pixels The Euclidean distance of Smoothing parameters that control the spatial extent; represents spatial weight;

[0030] The pixel value weight is calculated based on the difference in pixel values. The greater the difference in pixel values, the smaller the weight.

[0031]

[0032] in, is the difference between pixels; is a parameter that controls the sensitivity of pixel similarity; Represents pixel weight.

[0033] Preferably, step S2 includes:

[0034] Step S2.1: Counting the photovoltaic high power generation time based on the prediction results;

[0035] Step S2.2: Determine the straight-line grid power level based on the photovoltaic high-power generation time;

[0036]

[0037] in, To determine the level of power generation; is the photovoltaic installed power; To predict the period of high photovoltaic power generation; is the annual effective photovoltaic power generation hours;

[0038] Step S2.3: Perform linear power balance compensation based on the determined linear grid-connected power level.

[0039] Preferably, the step S2.3 includes:

[0040] When the photovoltaic output power is lower than the direct grid power level n%, it is photovoltaic low power output, and the energy storage output power is:

[0041]

[0042] in, Indicates the energy storage output power;

[0043] When the photovoltaic output power is higher than the direct grid power level by n%, it is photovoltaic high power output, and the energy storage absorption power is:

[0044]

[0045] in, Indicates the energy storage absorption power;

[0046] When the photovoltaic output power is at n% of the direct grid power level, the energy storage system neither charges nor discharges.

[0047] Preferably, the method further comprises:

[0048] Step S3: constructing a real-time node electricity price prediction model, and using the constructed real-time node electricity price prediction model to predict the electricity price level;

[0049] Step S4: Based on the predicted electricity price level and the modulated direct-line grid power, dynamic adjustment of the charging and discharging strategy is achieved.

[0050] Preferably, the real-time node electricity price prediction model includes:

[0051] Based on the unified load curve , output of non-market units , output of new energy units and outbound channel load Calculate bidding space ;

[0052] ;

[0053] Based on bidding space Calculate node energy prices;

[0054] Calculate the congestion price based on the shadow price and power flow sensitivity of the node-related branch / section security constraints;

[0055]

[0056] in, represents the blocking price, Shadow price representing the safety constraints of the node-related branches / sections; Indicates tidal current sensitivity;

[0057] The real-time node electricity price prediction model includes:

[0058] ；

[0059] in, Represents the node energy price.

[0060] Preferably, step S4 includes:

[0061] Based on the modulated linear grid power, when it is in a high electricity price period that meets the preset requirements, the output is increased; when it is in a low electricity price period that meets the preset requirements, the output is reduced;

[0062] Adjust the grid-connected power curve of the PV-storage system to match real-time electricity price fluctuations, including:

[0063] Based on the maximum output power of the energy storage system, the power output within different electricity price ranges is determined. During low electricity price periods that meet preset requirements, the energy storage system's on-grid power is reduced or no power is supplied to the grid at all. During high electricity price periods that meet preset requirements, the energy storage system's on-grid power is increased.

[0064] When photovoltaic power generation is sufficient, the grid-connected power is adjusted according to the real-time electricity price.

[0065] A distributed solar-storage system with linear power access to the grid provided by the present invention includes:

[0066] High-precision prediction module for renewable energy output: Using empirical regression model to meet the preset requirements of photovoltaic power generation in clear weather As a benchmark, the photovoltaic daily power generation is predicted under the given conditions of the obscuration degree SDI and cloud proportion CCI to obtain the prediction results;

[0067] Linear power modulation module: Based on the prediction results, the photovoltaic high-power generation time is counted, and the linear grid-connected power level is determined based on the photovoltaic high-power generation time.

[0068] Preferably, the new energy output high-precision prediction module includes:

[0069] Module M1.1: All-weather imagers deployed at renewable energy power generation sites periodically collect cloud image information covering the forecast area;

[0070] Module M1.2: pre-processing the collected cloud image information to obtain pre-processed cloud image information;

[0071] Module M1.3: Calculates the sun's current position and the cloud's movement trajectory at a preset time in the future based on preprocessed cloud image information, in order to predict whether the sun will be obscured at a preset time in the future.

[0072] Module M1.4: Use image data to calculate the average brightness of the sun area and compare it with the sunny day benchmark to obtain the degree of obstruction SDI;

[0073] Module M1.5: Segment cloud layers using pixel thresholding or machine learning methods and calculate cloud fraction (CCI);

[0074] ;

[0075] Module M1.6: Use empirical regression model to meet the preset requirements of photovoltaic power generation in clear weather As a benchmark, under the given conditions of shading degree SDI and cloud proportion CCI, the power generation of photovoltaic power generation in the future preset time period is calculated and predicted;

[0076]

[0077] in, Indicates the power generation of the day; GHI indicates the predicted power generation of photovoltaic power in the future preset time period;

[0078] The module M1.2 includes:

[0079] Module M1.2.1: Use bilateral filtering to filter the cloud image information to obtain the processed cloud image information;

[0080]

[0081] in, Is the output image at position Pixel value of is the input image at position Pixel value of Therefore The neighborhood window centered on is the standard deviation of the spatial domain, which controls the decay rate of the spatial weight; is the standard deviation of the value range, which controls the decay rate of pixel value weight; is a normalization constant used to ensure that the weights sum to 1;

[0082]

[0083] Module M1.2.2: Calculate spatial weights and pixel value weights, and delete pixels with weights less than the threshold to obtain processed cloud image information;

[0084] Spatial weight is calculated based on the distance between pixels in space. The farther the distance, the smaller the weight.

[0085]

[0086] in, Pixels and pixels The Euclidean distance of Smoothing parameters that control the spatial extent; represents spatial weight;

[0087] The pixel value weight is calculated based on the difference in pixel values. The greater the difference in pixel values, the smaller the weight.

[0088]

[0089] in, is the difference between pixels; is a parameter that controls the sensitivity of pixel similarity; represents pixel weight;

[0090] The linear power modulation module includes:

[0091] Module M2.1: Calculate the photovoltaic high-power generation time based on the prediction results;

[0092] Module M2.2: Determine the straight-line grid power level based on the photovoltaic high-power generation time;

[0093]

[0094] in, To determine the level of power generation; is the photovoltaic installed power; To predict the period of high photovoltaic power generation; is the annual effective photovoltaic power generation hours;

[0095] Module M2.3: Performs linear power balance compensation based on the determined linear grid power level;

[0096] The module M2.3 includes:

[0097] When the photovoltaic output power is lower than the direct grid power level n%, it is photovoltaic low power output, and the energy storage output power is:

[0098]

[0099] in, Indicates the energy storage output power;

[0100] When the photovoltaic output power is higher than the direct grid power level by n%, it is photovoltaic high power output, and the energy storage absorption power is:

[0101]

[0102] in, Indicates the energy storage absorption power;

[0103] When the photovoltaic output power is at n% of the direct grid power level, the energy storage system neither charges nor discharges.

[0104] Compared with the prior art, the present invention has the following beneficial effects:

[0105] 1. This invention uses an all-weather imager to collect cloud images, and combines bilateral filtering preprocessing, sun position calculation, and surface cloud coverage (SDI) and cloud concentration (CCI) analysis to accurately predict the daily photovoltaic power generation, thereby reducing prediction errors caused by weather fluctuations and providing a reliable data basis for subsequent power modulation.

[0106] 2. This invention modulates the traditional "steamed bun curve" into a stable linear grid-connected power, and dynamically compensates through the energy storage system, reducing the impact of photovoltaic output fluctuations on the power grid, improving the grid's ability to accept renewable energy, and meeting the grid's requirements for stable power supply;

[0107] 3. The present invention dynamically adjusts the energy storage charging and discharging strategy based on the prediction results, reduces the grid dispatching pressure, and improves the grid-connectedness of the photovoltaic storage system;

[0108] 4. After adopting linear power generation, the solar-storage system can use the linear power generation curve to quote in the medium- and long-term trading sequence. The quotation level is directly linked to the medium- and long-term trading price level of thermal power, which is conducive to stabilizing the return on investment in new energy, and is also conducive to the development of green energy substitution, further increasing the power grid's ability to absorb renewable energy.

[0109] 5. This invention proposes a control and management method for linear power generation using integrated photovoltaic and storage systems. In this mode, the photovoltaic and storage system can simulate the generation characteristics of a thermal power plant and participate in medium- and long-term transactions in the electricity market. This can also gradually approach the medium- and long-term transaction prices of thermal power. Furthermore, this technology can be combined with power trading and price forecasting technologies to precisely adjust the linear medium- and long-term contract curve, achieving higher levels of market-based arbitrage. BRIEF DESCRIPTION OF THE DRAWINGS

[0110] Other features, objects and advantages of the present invention will become more apparent upon reading the detailed description of non-limiting embodiments with reference to the following drawings:

[0111] Figure 1 Schematic diagram of a distributed photovoltaic storage system with linear power access to the grid. DETAILED DESCRIPTION

[0112] The present invention will be described in detail below with reference to specific embodiments. The following examples will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those skilled in the art, several changes and improvements can be made without departing from the scope of the present invention. These all fall within the scope of protection of the present invention.

[0113] Example 1

[0114] The present invention provides a distributed solar-to-storage system with linear power access. This system combines energy storage and photovoltaics to modulate the PV power curve into linear grid-connected power, enabling participation in electricity market transactions. The key concept of this invention is to enable the solar-to-storage system to mimic the output characteristics of thermal power through the combined control of photovoltaics and energy storage.

[0115] The distributed solar storage system with straight-line power access to the internet, such as Figure 1 As shown, it includes: a high-precision prediction module for new energy output, a linear power modulation module, a real-time node electricity price prediction module, and a trading strategy dynamic optimization module.

[0116] The high-precision prediction module for renewable energy output is the basic module for power generation curve modulation. In order to modulate the photovoltaic steamed bun line into a straight-line grid-connected power, it is necessary to perform high-precision measurement of the photovoltaic daily power generation, which will determine the level of the straight-line grid-connected power.

[0117] The linear power modulation module modulates the daily power generation curve according to the prediction results, converting the steamed bun line into a straight line to connect to the grid;

[0118] The real-time node electricity price prediction module provides basic support for trading strategies;

[0119] The trading strategy dynamic optimization module adjusts the positions of medium- and long-term transactions or day-ahead transactions under the execution of node electricity prices to optimize the profit performance of direct grid-connected power. If the market mechanism permits, non-direct grid-connected power can also be used to enhance profits.

[0120] The new energy output high-precision prediction module includes:

[0121] Module 1.1: All-weather imager data acquisition;

[0122] In this embodiment, cloud image information covering the predicted area is periodically collected by an all-weather imager deployed at a new energy power generation station.

[0123] The cloud image information includes:

[0124] Automatic capture: The imager automatically captures sky images at a preset frequency. These images, including visible light and infrared images, record information such as cloud shape, location, and thickness.

[0125] Data storage: The captured image data is automatically stored in the device's local memory. At the same time, the data is transmitted to a server or cloud platform via the network for subsequent processing and analysis.

[0126] Module 1.2: Due to the existence of external interference factors, the noise in the cloud image information is pre-processed to obtain the processed cloud image information;

[0127] In this embodiment, the cloud image information is first preprocessed using a bilateral filtering method, including:

[0128] The formula for bilateral filtering can be expressed as:

[0129]

[0130] in: Is the output image at position Pixel value of is the input image at position The pixel value of . Therefore The neighborhood window centered on . is the standard deviation of the spatial domain, which controls the decay rate of the spatial weight It is the standard deviation of the value range, which controls the decay rate of pixel value weight. is a normalization constant used to ensure that the weights sum to 1:

[0131]

[0132] Then, by calculating the spatial weight and pixel value weight, the pixels with weight values ​​less than the threshold are deleted to improve the algorithm's recognition of cloud edges.

[0133] The spatial weight includes:

[0134] The spatial weight is calculated based on the distance between pixels in space. The farther the distance, the smaller the weight, ensuring that only neighboring pixels have a greater impact on the central pixel.

[0135]

[0136] in, Represents pixels and pixels Euclidean distance (spatial distance); represents the smoothing parameter that controls the spatial range;

[0137] The pixel value weights include:

[0138] The pixel value weight is calculated based on the difference in pixel values. The greater the difference in pixel values, the smaller the weight, ensuring that only pixels with values ​​close to the center pixel have a greater impact on the center pixel.

[0139]

[0140] in, Represents the difference between pixels (color or grayscale difference); Represents the parameter that controls the sensitivity of pixel similarity;

[0141] The processed cloud image information can be used to calculate the current and future positions of the sun and the movement trajectory of the clouds, so as to predict whether the sun will be blocked in the next few minutes.

[0142] The average brightness (weighted average) of the sun area is calculated using image data and compared with the sunny day benchmark to obtain the degree of obstruction SDI;

[0143] Segment the cloud layer using pixel thresholding or machine learning methods (such as CNN) and calculate the cloud fraction (CCI):

[0144]

[0145] An empirical regression model is used to meet the preset requirements of photovoltaic power generation in clear weather. As a benchmark, under the given conditions of shading degree SDI and cloud proportion CCI, the power generation of photovoltaic power generation in the future preset time period is calculated and predicted;

[0146]

[0147] Specifically, the linear power modulation module includes: When a new energy power station participates in spot power market transactions, the power curve generated by the generating unit determines its electricity sales revenue. Ordinary photovoltaic power lines or "abnormal power lines" for self-use cannot achieve high returns in the power market, and conventional arbitrage models cannot guarantee stable returns. However, linear grid-connected power can bring the generation curve of the photovoltaic storage system closer to that of thermal power, facilitating the future participation of photovoltaic storage systems in medium- and long-term power transactions.

[0148] This module aims to determine the linear grid-connected power level based on the new energy power prediction module, and provide a control strategy to ensure that the output power remains near the linear level when the photovoltaic power is near the critical value of the grid-connected power level;

[0149] More specifically, the linear power modulation module includes:

[0150] Module 2.1: Calculate the photovoltaic high-power generation time based on the prediction results;

[0151] Conduct actual statistics on photovoltaic power generation, observe periods of high photovoltaic power generation in different seasons for different installed powers, inclination angles, available hours, etc., conduct statistics and obtain available data.

[0152]

[0153] The high-power generation period is when the hourly power generation accounts for more than 4%. During the daily operation of the photovoltaic storage system, the photovoltaic high-power generation period needs to be adjusted according to the power generation forecast;

[0154] Module 2.2: Determine the power level of the direct-line grid;

[0155] The formula for determining the online power level is:

[0156]

[0157] The design level of energy storage installed capacity is:

[0158]

[0159] in, To predict the period of high photovoltaic power generation; is the annual effective photovoltaic power generation hours; To determine the level of power generation; is the discharge depth of the energy storage system; is the maximum power of energy storage; is the installed photovoltaic power.

[0160] Module 2.3: Perform linear power balance compensation;

[0161] By simulating human thinking, the fuzzy control algorithm converts the difference between photovoltaic output power and load power, the state of charge of the energy storage battery, and the precise values ​​of the energy storage system's charge and discharge power into linguistic variables in a fuzzy set, such as "large," "medium," and "small." It also defines photovoltaic output power as 5% below the direct-line grid power level as low, and photovoltaic output power as 5% above the direct-line grid power level as high. When photovoltaic output power is low, the energy storage output power is:

[0162]

[0163] When the photovoltaic power output is high, the energy storage absorbs power, and its magnitude is:

[0164]

[0165] When the photovoltaic output power is at 5% of the direct-line grid power level, the energy storage system neither charges nor discharges.

[0166] During actual operation, the photovoltaic output power status is first determined, and then the energy storage system is activated to charge and discharge, thereby achieving the effect of outputting linear power.

[0167] Specifically, the real-time node electricity price prediction module includes:

[0168] Under normal circumstances, a linear grid-connected solar-storage system can participate in medium- and long-term transactions to sell a linear power curve, generating revenue significantly higher than the natural grid-connected revenue of photovoltaic power generation. Due to the volatility and forecast errors of photovoltaic power generation, trading risk management is necessary to further stabilize revenue performance through power trading. Therefore, a real-time node electricity price forecasting module is crucial.

[0169] More specifically, the real-time node electricity price prediction module includes: using node energy prices to reflect the supply and demand relationship of the system, using extreme congestion prices to reflect the impact of local node grid lines on node electricity prices, and combining the two to predict the real-time average price of the node.

[0170] Calculate the market bidding space and calculate the node energy price based on the market bidding space;

[0171]

[0172] in, Indicates the unified load curve; Indicates the output of non-market units; Indicates the output of new energy units; Indicates the load of the outgoing channel.

[0173] As shown in the above formula, the bidding space It represents the bid position of thermal power units participating in the market bidding, at which the power generation and power consumption are fully matched for matching.

[0174] If node electricity prices are subject to channel constraints or congestion caused by adjustments to system operating conditions or maintenance of related data centers, they will cause significant changes in node electricity prices. The congestion price represents the congestion price of the effective constraints and reflects the grid congestion situation at the node's location. When the power flow of a transmission line or section reaches its transmission limit, congestion occurs, requiring adjustments to power generation or load to maintain safe grid operation. The additional cost of these adjustments is the congestion cost, which is reflected in the electricity price as the congestion price.

[0175] The congestion price is equal to the sum of the shadow price of the node-related branch / section security constraint and the power flow sensitivity. The specific calculation formula is:

[0176]

[0177] in, is the shadow price, which refers to the sensitivity of the system objective function (such as power generation cost) to the marginal change of the constraint conditions under the constraint conditions; It refers to the degree of influence of node load changes on transmission line power flow.

[0178] Based on the above results, the final prediction model of real-time node electricity price is:

[0179]

[0180] Specifically, the trading strategy dynamic optimization module includes:

[0181] Based on the previous calculation results, this module further constructs an optimized control mechanism for the energy storage system for the spot period of operation, realizing the dynamic update of charging and discharging strategies and improving profits.

[0182] Based on the current electricity price level, PV power forecast and energy storage SOC, the linear grid power level is optimized. Under a given electricity price trend, the linear grid power level is optimized. The specific steps are:

[0183] Adjustment strategy determination: On a linear basis, output should be appropriately increased during periods of high electricity prices and reduced during periods of low electricity prices.

[0184] SOC adjustment: This involves adjusting the grid-connected power curve of the PV-storage system to match real-time electricity price fluctuations. Specific steps include:

[0185] Determine maximum power output: Based on the maximum output power of the energy storage system (limited by battery capacity, power converter capabilities, etc.), determine the power output within different electricity price ranges.

[0186] When electricity prices are low, the energy storage system's grid power should be reduced, or even not supplied to the grid at all, but instead charged.

[0187] When electricity prices are high, the grid-connected power of the energy storage system should be increased, and the stored electricity should be discharged during the period with the highest electricity prices to optimize profits.

[0188] Adjusting PV output power: When PV generation is sufficient, the grid-connected power can be adjusted based on the real-time electricity price. If PV generation is high but the electricity price is low, the energy can be stored. Conversely, when the electricity price is high, the grid-connected power can be increased as much as possible.

[0189] Those skilled in the art will appreciate that, in addition to implementing the system, device, and various modules provided by the present invention in purely computer-readable program code, it is entirely possible to implement the same program in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, embedded microcontrollers, and the like by logically programming the method steps. Therefore, the system, device, and various modules provided by the present invention can be considered a hardware component, and the modules included therein for implementing various programs can also be considered structures within the hardware component; the modules for implementing various functions can also be considered both software programs for implementing the method and structures within the hardware component.

[0190] The above describes specific embodiments of the present invention. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art may make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. The embodiments of this application and the features in the embodiments may be combined with each other in any manner unless there is a conflict.

Claims

1. A method for implementing a distributed solar storage system with linear power access to the grid, characterized in that: include: Step S1: using the new energy output high-precision prediction module to predict the photovoltaic daily power generation to obtain the prediction result; Step S2: The linear power modulation module modulates the daily power generation curve according to the prediction results, and modulates the steamed bun curve into a linear grid-connected power; The new energy output high-precision prediction module includes: using an empirical regression model to meet the preset requirements of photovoltaic power generation when the weather is clear As a benchmark, the photovoltaic daily power generation is predicted under the given conditions of the obscuration degree SDI and cloud proportion CCI to obtain the prediction results; The linear power modulation module includes: counting photovoltaic high-power generation time based on the prediction results, and determining the linear grid-connected power level based on the photovoltaic high-power generation time; The step S1 comprises: Step S1.1: Periodically collect cloud image information covering the predicted area using an all-weather imager deployed at a renewable energy power station; Step S1.2: pre-processing the collected cloud image information to obtain pre-processed cloud image information; Step S1.3: Calculate the current position of the sun and the position at a preset time in the future and the movement trajectory of the clouds based on the preprocessed cloud image information, so as to predict whether the sun will be blocked at the preset time in the future; Step S1.4: Calculate the average brightness of the sun area using the image data and compare it with the sunny day benchmark to obtain the degree of obstruction SDI; Step S1.5: Segment the cloud layer using pixel thresholding or machine learning methods, and calculate the cloud fraction CCI; ； Step S1.6: Use the empirical regression model to meet the preset requirements of photovoltaic power generation in clear weather As a benchmark, under the given conditions of shading degree SDI and cloud proportion CCI, the power generation of photovoltaic power generation in the future preset time period is calculated and predicted; in, Indicates the power generation of the day; GHI indicates the predicted power generation of photovoltaic power in the future preset time period; The step S2 comprises: Step S2.1: Counting the photovoltaic high power generation time based on the prediction results; Step S2.2: Determine the straight-line grid power level based on the photovoltaic high-power generation time; in, To determine the level of power generation; is the photovoltaic installed power; To predict the period of high photovoltaic power generation; is the annual effective photovoltaic power generation hours; Step S2.3: performing linear power balance compensation based on the determined linear grid power level; The step S2.3 includes: When the photovoltaic output power is lower than the direct grid power level n%, it is photovoltaic low power output, and the energy storage output power is: in, Indicates the energy storage output power; When the photovoltaic output power is higher than the direct grid power level by n%, it is photovoltaic high power output, and the energy storage absorption power is: in, Indicates the energy storage absorption power; When the photovoltaic output power is at n% of the direct grid power level, the energy storage system neither charges nor discharges.

2. The method for implementing a distributed solar-storage system with direct-line power access to the grid according to claim 1, characterized in that: The step S1.2 includes: Step S1.2.1: Filter the cloud image information using a bilateral filtering method to obtain processed cloud image information; in, Is the output image at position Pixel value of is the input image at position Pixel value of Therefore The neighborhood window centered on is the standard deviation of the spatial domain, which controls the decay rate of the spatial weight; is the standard deviation of the value range, which controls the decay rate of pixel value weight; is a normalization constant used to ensure that the weights sum to 1; Step S1.2.2: Calculate the spatial weight and pixel value weight, and delete the pixels whose weight values ​​are less than the threshold to obtain the processed cloud image information; Spatial weight is calculated based on the distance between pixels in space. The farther the distance, the smaller the weight. in, Pixels and pixels The Euclidean distance of Smoothing parameters that control the spatial extent; represents spatial weight; The pixel value weight is calculated based on the difference in pixel values. The greater the difference in pixel values, the smaller the weight. in, is the difference between pixels; is a parameter that controls the sensitivity of pixel similarity; Represents pixel weight.

3. The method for implementing a distributed photovoltaic storage system with linear power access to the grid according to claim 1, characterized in that: The method further comprises: Step S3: constructing a real-time node electricity price prediction model, and using the constructed real-time node electricity price prediction model to predict the electricity price level; Step S4: Based on the predicted electricity price level and the modulated direct-line grid power, dynamic adjustment of the charging and discharging strategy is achieved.

4. The method for implementing a distributed solar-storage system with direct-line power access to the grid according to claim 3, characterized in that: The real-time node electricity price prediction model includes: Based on the unified load curve , output of non-market units , output of new energy units and outbound channel load Calculate bidding space ; ； Based on bidding space Calculate node energy prices; Calculate the congestion price based on the shadow price and power flow sensitivity of the node-related branch / section security constraints; in, represents the blocking price, Shadow price representing the safety constraints of the node-related branches / sections; Indicates tidal current sensitivity; The real-time node electricity price prediction model includes: ； in, Represents the node energy price.

5. The method for implementing a distributed solar-storage system with direct-line power access to the grid according to claim 3, characterized in that: The step S4 comprises: Based on the modulated linear grid power, when it is in a high electricity price period that meets the preset requirements, the output is increased; when it is in a low electricity price period that meets the preset requirements, the output is reduced; Adjust the grid-connected power curve of the PV-storage system to match real-time electricity price fluctuations, including: Based on the maximum output power of the energy storage system, the power output within different electricity price ranges is determined. During low electricity price periods that meet preset requirements, the energy storage system's on-grid power is reduced or no power is supplied to the grid at all. During high electricity price periods that meet preset requirements, the energy storage system's on-grid power is increased. When photovoltaic power generation is sufficient, the grid-connected power is adjusted according to the real-time electricity price.

6. A distributed solar storage system with linear power access to the grid, characterized in that: include: High-precision prediction module for renewable energy output: Using empirical regression model to meet the preset requirements of photovoltaic power generation in clear weather As a benchmark, the photovoltaic daily power generation is predicted under the given conditions of the obscuration degree SDI and cloud proportion CCI to obtain the prediction results; Linear power modulation module: Calculates the photovoltaic high-power generation time based on the prediction results, and determines the linear grid-connected power level based on the photovoltaic high-power generation time; The new energy output high-precision prediction module includes: Module M1.1: All-weather imagers deployed at renewable energy power generation sites periodically collect cloud image information covering the forecast area; Module M1.2: pre-processing the collected cloud image information to obtain pre-processed cloud image information; Module M1.3: Calculates the sun's current position and the cloud's movement trajectory at a preset time in the future based on preprocessed cloud image information, in order to predict whether the sun will be obscured at a preset time in the future. Module M1.4: Use image data to calculate the average brightness of the sun area and compare it with the sunny day benchmark to obtain the degree of obstruction SDI; Module M1.5: Segment cloud layers using pixel thresholding or machine learning methods and calculate cloud fraction (CCI); ； Module M1.6: Use empirical regression model to meet the preset requirements of photovoltaic power generation in clear weather As a benchmark, under the given conditions of shading degree SDI and cloud proportion CCI, the power generation of photovoltaic power generation in the future preset time period is calculated and predicted; in, Indicates the power generation of the day; GHI indicates the predicted power generation of photovoltaic power in the future preset time period; The module M1.2 includes: Module M1.2.1: Use bilateral filtering to filter the cloud image information to obtain the processed cloud image information; in, Is the output image at position Pixel value of is the input image at position Pixel value of Therefore The neighborhood window centered on is the standard deviation of the spatial domain, which controls the decay rate of the spatial weight; is the standard deviation of the value range, which controls the decay rate of pixel value weight; is a normalization constant used to ensure that the weights sum to 1; Module M1.2.2: Calculate spatial weights and pixel value weights, and delete pixels with weights less than the threshold to obtain processed cloud image information; Spatial weight is calculated based on the distance between pixels in space. The farther the distance, the smaller the weight. in, Pixels and pixels The Euclidean distance of Smoothing parameters that control the spatial extent; represents spatial weight; The pixel value weight is calculated based on the difference in pixel values. The greater the difference in pixel values, the smaller the weight. in, is the difference between pixels; is a parameter that controls the sensitivity of pixel similarity; represents pixel weight; The linear power modulation module includes: Module M2.1: Calculate the photovoltaic high-power generation time based on the prediction results; Module M2.2: Determine the straight-line grid power level based on the photovoltaic high-power generation time; in, To determine the level of power generation; is the photovoltaic installed power; To predict the period of high photovoltaic power generation; is the annual effective photovoltaic power generation hours; Module M2.3: Performs linear power balance compensation based on the determined linear grid power level; The module M2.3 includes: When the photovoltaic output power is lower than the direct grid power level n%, it is photovoltaic low power output, and the energy storage should output power, the size of which is: in, Indicates the energy storage output power; When the photovoltaic output power is higher than the direct grid power level by n%, it is photovoltaic high power output, and the energy storage should absorb power, the size of which is: in, Indicates the energy storage absorption power; When the photovoltaic output power is at n% of the direct grid power level, the energy storage system neither charges nor discharges.

Images