Distributed photovoltaic on-grid power transformer loss statistical method

Abstract

A kind of distributed photovoltaic on-grid power transformer loss statistical method, step 1: collect the time-sharing power generation data of representative day in a year at photovoltaic module inclination in project location;Step 2: the power generation function is constructed to represent the time distribution characteristics of photovoltaic power generation;Step 3: the intersection of the difference of user load curve and power generation function is the actual on-grid hours of the day, and the area surrounded by user load curve and power generation function is the surplus power on-grid power;Step 4, according to the set step, the on-grid hours of user load and surplus power on-grid proportion k are sequentially counted;Step 5, calculate the transformer loss value under each step;Step 6: according to the above calculation result, the transformer loss under different on-grid power proportion is obtained by interpolation method, and the loss query table is made.The present application introduces real photovoltaic power generation curve and user load characteristics, dynamically calculates loss, and avoids the huge deviation of fixed proportion method.

Description

Technical Field

[0001] This invention proposes a statistical method for transformer losses in distributed photovoltaic power plants with surplus electricity fed into the grid, belonging to the fields of renewable energy engineering and electrical engineering. Background Technology

[0002] In China, distributed photovoltaic (PV) systems largely utilize 380V low-voltage grid connection, employing a model of self-consumption with surplus power fed into the grid. When settling accounts between PV power plant investors, users, and the grid, it is necessary to clearly define the user's self-consumption electricity, the electricity fed into the grid, and the transformer losses.

[0003] In a typical distributed photovoltaic design scheme, as shown in the attached... Figure 2 As shown, metering point 1 is generally set at the photovoltaic grid connection point to count the total photovoltaic power generation W1; metering point 2 is set on the 10kV high-voltage side of the user transformer to count the actual photovoltaic power generation W2; and transformer loss power W3 during the surplus power grid connection period is generally not measured by a separate meter.

[0004] According to the principle of energy conservation, the actual electricity consumption of a user = W1 - W2 - W3, where W3 is the transformer loss. However, this loss W3 is not a fixed value, but consists of no-load loss (iron loss) and load loss (copper loss), both of which change dynamically in real time with the actual operating conditions of the transformer (especially grid connection time and load rate). Currently, most of the distributed photovoltaic projects already in operation do not have a dedicated bidirectional meter installed on the low-voltage side of the transformer to directly measure W3.

[0005] In the absence of direct measurement methods and without explicit contractual stipulations, the value of W3 often becomes the core focus of settlement disputes. In existing technology, the following two rough estimation methods are commonly used: 1. Fixed Proportion Allocation: For example, simply calculating W3 as a fixed percentage (e.g., 1%-3%) of W1 or W2. This method ignores the strong correlation between W3 and the actual load factor, and the error is extremely large when photovoltaic output fluctuates greatly and user loads are uneven.

[0006] 2: Estimation based on rated parameters: Based on the no-load loss P0 and load loss P on the transformer nameplate. k Theoretical calculations were performed. However, this method assumes that the transformer operates at full or half load throughout the day, which is seriously inconsistent with the intermittent and fluctuating power generation characteristics of distributed photovoltaic power and the peak power consumption pattern of users, resulting in distorted results.

[0007] Therefore, the industry needs a precise statistical method that does not require modification of existing metering systems, is low in cost, and can dynamically reflect transformer losses based on actual operating data, so as to provide an objective and fair basis for commercial settlement. Summary of the Invention

[0008] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method for statistical analysis of transformer losses in distributed photovoltaic power generation based on photovoltaic power generation characteristics and user load characteristics. This method constructs a mathematical model to correlate the dynamic loss W3, which is difficult to measure directly, with the easily obtainable ratio of total photovoltaic power generation W1 to the actual photovoltaic power generation W2, and generates a quick lookup table, thereby achieving fast and accurate loss lookup.

[0009] To achieve this objective, the present invention adopts the following technical solution: A method for statistical analysis of transformer losses in distributed photovoltaic power grid connection includes the following steps: Step 1: Collect the annual representative day's photovoltaic module tilt angle energy time-of-use power generation data (kWh) at the project location, which are time arrays {t1,t2....t23,t24} and photovoltaic power generation arrays {h1,h2....h23,h24}; Step 2: A continuous analytical function, namely the power generation function h=f(t), will be constructed to characterize the temporal distribution characteristics of photovoltaic power generation. Given that solar irradiance and power generation exhibit a unimodal, symmetrical, and smooth bell-shaped distribution under clear weather conditions, the data will be fitted using the nonlinear least squares method to obtain the fitted function for power generation h as follows: ; Where A is the peak value of the power generation {h1,h2....h23,h24}; μ is the time t corresponding to the peak value A; σ is the distribution width value; under typical illumination, the power generation curve is approximately σ≈2.65.

[0010] Step 3: Based on user load conditions, calculate the duration and percentage of surplus electricity fed into the grid. Under the self-consumption and surplus electricity fed into the grid model, the load curve of a typical industrial user is approximately a straight line. When the user's electricity consumption curve is lower than the power generation curve, surplus electricity will be fed into the grid. The difference between the intersection points of the user load curve and the power generation function f(t) is the actual number of hours fed into the grid for the day, and the area enclosed by the user load curve and the power generation function f(t) is the surplus electricity fed into the grid. By drawing a straight line with h = 0.9A, we can find the intersection point with the function f(t). t i and t j The remaining hours of electricity connected to the internet are... t j - t i Then the actual grid-connected electricity W2 can be expressed in the definite integral form of f(t): ; The surplus electricity fed into the grid ratio k = surplus electricity fed into the grid / total electricity generation = W2 / W1 * 100%. This ratio is based on known data measured by the user's daily meter. Step 4: Set the calculation step size to 0.1A. According to the set step size, count the number of hours the user is online and the proportion of surplus power used online in the range of 0.8A to 0.1A. Step 5: Calculate the transformer loss value W3 at each step size from h=0.8A to 0.1A. W3 is derived from the no-load loss W. 30 With load power loss W 3k Composition, no-load power loss W 30 The calculation method is as follows: ; The transformer load factor can be approximated as: ; The calculation method for no-load power loss W3 is as follows: ; ; Where P0 is the no-load loss (kW) of the user's transformer, P k P is the load loss (kW), β is the transformer load rate (%), and S is the transformer rated capacity (kVA). 0、 P k Both S and S are the nameplate values ​​of the user's transformer; Step 6: Generate a lookup table: Based on the above calculation results, the transformer losses under different on-grid power ratios were obtained through interpolation, and a loss lookup table was created.

[0011] The beneficial effects of this invention are: 1. High precision: By introducing real photovoltaic power generation curves and user load characteristics, losses are dynamically calculated, avoiding the huge deviations of the fixed ratio method.

[0012] 2. Low cost: Based entirely on existing metering data (W1, W2) and transformer nameplate parameters, no new hardware meters or modifications to the power station are required.

[0013] 3. Easy to use: The final result is a simple query table that can be used intuitively by users, investors, and power grid companies, greatly simplifying the settlement process.

[0014] 4. Fairness and impartiality: It provides objective and transparent calculation basis, effectively resolves commercial disputes caused by unclear losses, and promotes the healthy development of the distributed photovoltaic industry. Attached Figure Description

[0015] Figure 1 This is a flowchart of the calculation process of the present invention.

[0016] Figure 2This is a diagram of a typical distributed photovoltaic (PV) power generation system. Figure 3 This is a schematic diagram of the surplus electricity fed into the grid. Detailed Implementation

[0017] like Figure 1 As shown, a method for statistically analyzing transformer losses in distributed photovoltaic power grid connection mainly includes obtaining the local time-of-use generation curve, statistically analyzing the duration and proportion of surplus power grid connection, statistically analyzing the duration and proportion of grid-connected power, statistically analyzing the load values ​​of different users according to a step size, calculating transformer losses, and creating a loss lookup table. The main steps include: Step 1: Collect the time-of-use power generation data (kWh) of the photovoltaic modules at the project location on representative days throughout the year, which are time arrays {t1,t2....t23,t24} and photovoltaic power generation arrays {h1,h2....h23,h24}.

[0018] Step 2: A continuous analytical function, namely the power generation function h=f(t), will be constructed to characterize the temporal distribution characteristics of photovoltaic power generation. Given that solar irradiance and power generation exhibit a unimodal, symmetrical, and smooth bell-shaped distribution under clear weather conditions, the data will be fitted using the nonlinear least squares method to obtain the fitted function for power generation h as follows: ; Where A is the peak value of the power generation {h1,h2....h23,h24}; μ is the time t corresponding to the peak value A; σ is the distribution width value, which is approximately 2.65 under typical illumination power generation curves.

[0019] Step 3: Based on user load conditions, calculate the duration and percentage of surplus electricity sold to the grid. In a self-consumption, surplus electricity sold to the grid model, such as... Figure 3 As shown, the load curve of a typical industrial user is approximately a straight line. When the user's electricity consumption curve is lower than the power generation curve, surplus electricity will be fed into the grid. The difference between the intersection points of the user's load curve and the power generation function f(t) is the actual number of hours the electricity is fed into the grid that day. The area enclosed by the user's load curve and the power generation function f(t) is the amount of surplus electricity fed into the grid. Figure 3 The shaded area shown.

[0020] By drawing a straight line with h = 0.9A, we can find the intersection point with the function f(t). t i and t j The remaining hours of electricity connected to the internet are... t j - t i Then the actual grid-connected electricity W2 can be expressed in the definite integral form of f(t): ; The surplus electricity fed into the grid ratio k = surplus electricity fed into the grid / total electricity generation = W2 / W1 * 100%. This ratio is based on known data obtained from the user's daily meter readings.

[0021] Step 4: Set the calculation step size to 0.1A. According to the set step size, count the number of hours the user is online and the proportion of surplus power used online in the range of 0.8A to 0.1A. Step 5: Calculate the transformer loss value W3 at each step size from h=0.8A to 0.1A. W3 is derived from the no-load loss W. 30 With load power loss W 3k Composition, no-load power loss W 30 The calculation method is as follows: ; The transformer load factor can be approximated as: ; The calculation method for no-load power loss W3 is as follows: ; ; Where P0 is the no-load loss (kW) of the user's transformer, P k P is the load loss (kW), β is the transformer load rate (%), and S is the transformer rated capacity (kVA). 0、 P k Both S and S are the nameplate values ​​of the user's transformer, which can be queried according to the actual situation.

[0022] Step 6: Generate a lookup table. Based on the above calculation results, the transformer loss under different grid-connected electricity ratios was obtained through interpolation. A loss lookup table was created to allow users to query transformer loss values ​​under different grid-connected electricity ratios.

[0023] According to the query table, users can quickly find out the transformer loss based on different time dimensions and the ratio of the electricity generated by the user's meter to the total power generation, providing a basis for settlement of commercial metering.

[0024] Based on the user's actual power generation curve, this invention obtains the correspondence between the proportion of electricity generated into the grid and the amount of electricity lost through function fitting and step-by-step calculation, and creates a quick reference table for users to look up information quickly.

[0025] Engineering Case: Taking a distributed photovoltaic project as an example, with a rated capacity of S=1250kVA, no-load loss P0=1.3kW, and load loss Pk=9.13kW, the photovoltaic installed capacity is 800kW.

[0026] The time-of-use power generation data for a typical sunny day on the vernal equinox in the local area were obtained, and the Gaussian function was fitted as follows: f(t) = 489.6 * exp(-0.0712*(t-12)^2) (kWh / h).

[0027] The user load L is set from 0.1A to 0.9A (A=489.6) and calculated in steps.

[0028] Taking L=0.5A=244.8kW as an example, the intersection point ta ≈ 9.0h, tb ≈ 15.0h, W2≈ 1850kWh, W1≈ 3500kWh, so the surplus electricity fed into the grid is k ≈ 52.9%.

[0029] Calculate W3:W in this scenario. 30 = 1.3 * 6 = 7.8kWh; W 3k = 0.86 * 6 = 5.2 kWh; W3 = 7.8 + 5.2 = 13 kWh Repeat the above steps to complete the full table construction.

[0030] The resulting user query table is as follows:

[0031] If a user's electricity consumption on the internet is 5% of the total electricity consumed that day, the total transformer loss would be approximately 5.5 kWh.

Claims

1. A method for statistical analysis of transformer losses in distributed photovoltaic power grid connection, characterized in that: Includes the following steps: Step 1: Collect the annual representative day's photovoltaic module tilt angle energy time-of-use power generation data (kWh) at the project location, which are time arrays {t1,t2....t23,t24} and photovoltaic power generation arrays {h1,h2....h23,h24}; Step 2: A continuous analytical function, namely the power generation function h=f(t), will be constructed to characterize the temporal distribution characteristics of photovoltaic power generation. Given that solar irradiance and power generation exhibit a unimodal, symmetrical, and smooth bell-shaped distribution under clear weather conditions, the data will be fitted using the nonlinear least squares method to obtain the fitted function for power generation h as follows: ； Where A is the peak value of the power generation {h1,h2....h23,h24}; μ is the time t corresponding to the peak value A; and σ is the distribution width. Step 3: Based on user load conditions, calculate the duration and percentage of surplus electricity fed into the grid. Under the self-consumption and surplus electricity fed into the grid model, the load curve of a typical industrial user is approximately a straight line. When the user's electricity consumption curve is lower than the power generation curve, surplus electricity will be fed into the grid. The difference between the intersection points of the user load curve and the power generation function f(t) is the actual number of hours fed into the grid for the day, and the area enclosed by the user load curve and the power generation function f(t) is the surplus electricity fed into the grid. By drawing a straight line with h = 0.9A, we can find the intersection point with the function f(t). t i and t j The remaining hours of electricity connected to the internet are... t j - t i Then the actual grid-connected electricity W2 can be expressed in the definite integral form of f(t): ； The surplus electricity fed into the grid ratio k = surplus electricity fed into the grid / total electricity generation = W2 / W1 * 100%. This ratio is based on known data obtained from the user's daily meter readings; Step 4: Set the calculation step size to 0.1A. According to the set step size, count the number of hours the user is online and the proportion of surplus power used online in the range of 0.8A to 0.1A. Step 5: Calculate the transformer loss value W3 at each step size from h=0.8A to 0.1A. W3 is derived from the no-load loss W. 30 With load power loss W 3k Composition, no-load power loss W 30 The calculation method is as follows: ； The transformer load factor can be approximated as: ; The calculation method for no-load power loss W3 is as follows: ； ； Where P0 is the no-load loss (kW) of the user's transformer, P k P is the load loss (kW), β is the transformer load rate (%), and S is the transformer rated capacity (kVA). 0、 P k Both S and S are the nameplate values ​​of the user's transformer; Step 6: Generate a lookup table: Based on the above calculation results, the transformer losses under different on-grid power ratios were obtained through interpolation, and a loss lookup table was created.

2. The method for statistical analysis of transformer losses in distributed photovoltaic power generation according to claim 1, characterized in that: Under typical illumination, the power generation curve is approximately σ≈2.65.

Images