Power distribution network line loss dynamic calculation method based on photovoltaic reverse transmission power correction
By identifying the reverse feed period and dynamically correcting the equivalent resistance, reversing the sign of the current, and calculating the root mean square current, the problem of large errors in existing line loss calculation methods under high-proportion photovoltaic access is solved, and high-precision line loss calculation is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-24
- Publication Date
- 2026-03-31
AI Technical Summary
Existing line loss calculation methods have large errors when there is a high proportion of distributed photovoltaic access. They cannot accurately calculate the bidirectional superimposed copper loss caused by photovoltaic backfeed power and the iron loss caused by voltage rise, and they cannot effectively deal with the additional losses caused by three-phase imbalance.
By collecting multi-source measurement data of the distribution network lines, identifying the reverse power period, constructing the reverse power correction coefficient, dynamically correcting the equivalent resistance, signifying the reverse current, and calculating the root mean square current, the spatiotemporal distribution of photovoltaic reverse power and three-phase unbalanced loss are accurately characterized, and dynamic calculation of line loss is realized.
It significantly improves the accuracy of line loss calculation, reducing it from 20%-40% to less than 5%, solving the error problem of existing algorithms under high-proportion photovoltaic access, supporting high-precision and interpretable dynamic line loss calculation, and meeting the needs of engineering deployment.
Smart Images
Figure CN121769830A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to a method for calculating power grid line losses. Background Technology
[0002] The existing line loss calculation system can be summarized as "three unidirectionalities": unidirectional power flow—power always flows from the substation bus to the load; unidirectional current—the sign of the current in each branch is always positive, without needing to consider direction reversal; and unidirectional losses—technical line losses only include copper losses and fixed iron losses, without the additional losses caused by "power backflow." Therefore, the existing line loss calculation system has certain limitations. Meanwhile, the industry continues to use simplified algorithms such as the "root mean square current method, equivalent resistance method, voltage drop method, and maximum load loss hour method." These methods could control errors within 3% in the era of "passive distribution networks," meeting the requirements for new energy line loss statistics and energy conservation. The ability to assess demand, with the "three new" challenges brought about by the high proportion of distributed photovoltaic, the "three new" are: new operating mode - during the daytime 08:00–16:00, the power factor at the feeder head drops sharply from above 0.9 to -0.2, resulting in "negative power factor" and "reverse metering"; new loss mechanism - the reverse power generates "bidirectional superposition" copper loss in the transformer winding and line impedance, and the iron loss increases exponentially with the voltage increase. The existing algorithm only considers the unidirectional RMS current, systematically underestimating 20%–40%; new three-phase imbalance - the proportion of single-phase household photovoltaic is greater than 70%, the negative sequence and zero sequence current imbalance increases from 5% to 30%, the neutral line current increases by 3 to 5 times, and the additional loss increases by 10%–25%; the existing algorithm includes (1) the equivalent capacity method, which simply equates the photovoltaic output to the load subtraction, and cannot depict the spatiotemporal distribution of the reverse power flow, and the error is still above 15%. (2) Forward-backward power flow method; The forward-backward power flow method requires complete, real-time, and accurate topology and measurement. However, the error rate of medium-voltage feeder ledgers is 8%–12%, and the error rate of low-voltage household transformer relationship is 15%–20%. The algorithm diverges in scenarios where communication is lacking, resulting in poor engineering deployability. (3) Artificial intelligence black box model; The artificial intelligence black box model uses LSTM, XGboost, etc. to directly map "feature → line loss". Although it has high accuracy in the short term, it lacks physical interpretation and fails to extrapolate in extreme backfeed scenarios (full power generation during holidays, sudden load drop). It also requires 2–3 years of massive samples, making it difficult to implement quickly in newly commissioned areas. (4) Static correction coefficient method; Some literature proposes "photovoltaic penetration rate coefficient" or "backfeed margin coefficient", but the coefficient is only related to the installed capacity and does not consider key variables such as access location, power factor, and three-phase imbalance, which cannot adapt to dynamic operation modes. Summary of the Invention
[0003] The purpose of this invention is to solve the problem of large calculation errors in existing dynamic line loss calculation methods, and to propose a dynamic calculation method for distribution network line losses based on photovoltaic backfeed power correction.
[0004] The method for dynamic calculation of distribution network line losses based on photovoltaic backfeed power correction, as described in this invention, includes the following steps:
[0005] Collect multi-source measurement data of power distribution network lines;
[0006] Based on the collected multi-source measurement data of the distribution network lines, the reverse transmission characteristics are identified to determine the reverse transmission time period of the distribution network lines;
[0007] Based on the reverse power period of the distribution network line, construct the power correction coefficient for the reverse power of the distribution network line;
[0008] The initial equivalent resistance of the distribution network line is dynamically corrected by using the power backfeed correction coefficient of the distribution network line, and the corrected equivalent resistance of the distribution network line is obtained.
[0009] The current sequence of the distribution network line is divided into two reverse periods according to the reverse period indicator, and the sign of the current in the reverse period is reversed. The root mean square current of the distribution network line in the reverse period is calculated.
[0010] The real-time line loss of the distribution network is calculated using the corrected equivalent resistance of the distribution network line and the root mean square current of the distribution network line during the reverse transmission period.
[0011] Furthermore, the multi-source measurement data of the distribution network lines includes the active power of the distribution network lines. Reactive power of distribution network lines Three-phase voltage of distribution network lines and the current of the distribution network lines .
[0012] Furthermore, the criterion for identifying the reverse power supply period of the distribution network is that the reverse power is greater than 0 within the sampling period and lasts for at least 2 sampling periods.
[0013] Furthermore, the expression for the power correction factor for backfeeding from the distribution network is as follows:
[0014]
[0015] in, This is a correction factor for the reverse power transmission of the distribution network lines; The proportion of photovoltaic grid connection locations; This is the ratio of photovoltaic capacity to the maximum downstream load. The real-time power factor is volts.
[0016] Furthermore, the expression for the corrected equivalent resistance of the distribution network line is as follows:
[0017]
[0018] in, The equivalent resistance of the distribution network lines after correction; The initial equivalent resistance of the distribution network lines; This is the penalty factor for three-phase imbalance.
[0019] Furthermore, the three-phase imbalance penalty factor The expression is:
[0020]
[0021] in, This refers to the negative sequence current imbalance. This represents the zero-sequence current imbalance.
[0022] Furthermore, the formula for calculating the root mean square current during the reverse transmission period of the distribution network line is as follows:
[0023]
[0024] in, This represents the root mean square current during the reverse transmission period of the distribution network line. This represents the number of sampling points within the current time period. The sampling sequence number; For the first The instantaneous current value after sign reversal at each sampling time.
[0025] Furthermore, the formula for calculating the real-time line loss of the distribution network is as follows:
[0026]
[0027] in, This refers to the real-time line loss of the distribution network lines. This refers to the no-load loss of the distribution network lines. This is used to calculate the number of hours in the time period.
[0028] Compared with the prior art, the present invention has the following advantages:
[0029] The present invention provides a dynamic calculation method for distribution network line losses based on photovoltaic backfeed power correction, which significantly improves the accuracy of line loss calculation. Existing line loss calculation methods only consider unidirectional RMS current, ignoring the "bidirectional superimposed copper loss" caused by photovoltaic backfeed power and the "exponential iron loss" caused by voltage rise. The present invention reduces the error from 20%-40% to an acceptable range (below 5%) by dynamically correcting the equivalent resistance. It introduces a negative sequence / zero sequence current imbalance penalty factor to accurately characterize the 10%-25% additional loss caused by a 3-5 times increase in neutral line current, filling the gap in existing line loss calculation methods. For the first time, it achieves high-precision, interpretable, and robust dynamic calculation of line losses in high-proportion photovoltaic distribution networks, filling the technical gaps in the industry in three major areas: quantification of backfeed power loss, three-phase imbalance correction, and lightweight engineering deployment. It directly promotes line loss management from "static estimation" to the era of "dynamic precision". Attached Figure Description
[0030] Figure 1 The flowchart below shows a method for dynamically calculating line losses in a distribution network based on photovoltaic backfeed power correction, as described in Specific Implementation Method 1. Detailed Implementation
[0031] Specific Implementation Method 1: Combination Figure 1 This embodiment describes a method for dynamically calculating distribution network line losses based on photovoltaic backfeed power correction, which includes the following steps:
[0032] Collect multi-source measurement data of power distribution network lines;
[0033] Based on the collected multi-source measurement data of the distribution network lines, the reverse transmission characteristics are identified to determine the reverse transmission time period of the distribution network lines;
[0034] Based on the reverse power period of the distribution network line, construct the power correction coefficient for the reverse power of the distribution network line;
[0035] The initial equivalent resistance of the distribution network line is dynamically corrected by using the power backfeed correction coefficient of the distribution network line, and the corrected equivalent resistance of the distribution network line is obtained.
[0036] The current sequence of the distribution network line is divided into two reverse periods according to the reverse period indicator, and the sign of the current in the reverse period is reversed. The root mean square current of the distribution network line in the reverse period is calculated.
[0037] The real-time line loss of the distribution network is calculated using the corrected equivalent resistance of the distribution network line and the root mean square current of the distribution network line during the reverse transmission period.
[0038] In this implementation, line loss calculation is broken down into a rigid sequence of "first identifying the reverse feed period → then correcting the resistance → finally bidirectional metering," completely plugging the structural loopholes of the traditional "unidirectional algorithm" in photovoltaic reverse feed scenarios. This reduces the line loss error of the entire feeder from an average of 28% to less than 5%. Using "the lightest measurement, the fewest modifications, and the lowest computing power," the spatiotemporal distribution of photovoltaic reverse feed power, bidirectional copper loss, voltage iron loss, and three-phase imbalance loss are quantified in one go. This allows the "unidirectional" line loss formula to be directly upgraded to a "bidirectional dynamic" version in the era of "high-proportion photovoltaics." The entire process only involves sampling point traversal and one square accumulation. The 15-minute rolling line loss calculation takes less than 30 ms and can be completed locally at the distribution transformer terminal, significantly reducing the computing pressure on the main station and laying the algorithmic foundation for future "minute-level early warning of distribution area line loss."
[0039] Specific Implementation Method Two: This implementation method further defines the dynamic calculation method for distribution network line losses based on photovoltaic backfeed power correction described in Specific Implementation Method One. In this implementation method, the multi-source measurement data of the distribution network lines includes the active power of the distribution network lines. Reactive power of distribution network lines Three-phase voltage of distribution network lines and the current of the distribution network lines .
[0040] In this embodiment, the active power of the distribution network lines is explicitly defined. Reactive power of distribution network lines Three-phase voltage of distribution network lines and the current of the distribution network lines Four types of core measurement data, unified data input format, solve the problem of missed detection of backfeed features caused by the chaotic data type of traditional algorithms (such as relying only on active power), and improve the reliability of subsequent backfeed period identification (such as avoiding misjudging reactive backfeed as active backfeed).
[0041] Specific Implementation Method 3: This implementation method further defines the dynamic calculation method for distribution network line loss based on photovoltaic backfeed power correction described in Specific Implementation Method 1. In this implementation method, the criterion for identifying the backfeed period of the distribution network line is that the reverse power is greater than 0 within the sampling period and lasts for at least 2 sampling periods.
[0042] In this embodiment, the above-mentioned criteria are enriched, which can accurately lock the reverse transmission period to resist interference, effectively filter instantaneous sampling noise (such as short-term jitter of current transformer), reduce the false judgment rate by more than 90% (compared with single-cycle criteria), and ensure that the reverse transmission correction only applies to the actual reverse transmission period, avoiding over-correction.
[0043] Specific Implementation Method Four: This implementation method further defines the dynamic calculation method for distribution network line losses based on photovoltaic backfeed power correction described in Specific Implementation Method One. In this implementation method, the expression for the distribution network line backfeed power correction coefficient is:
[0044]
[0045] in, This is a correction factor for the reverse power transmission of the distribution network lines; The proportion of photovoltaic grid connection locations; This is the ratio of photovoltaic capacity to the maximum downstream load. The real-time power factor is volts.
[0046] In this embodiment, physical quantity coupling modeling of backfeed loss is achieved for the first time, and the backfeed power correction coefficient is established. Photovoltaic connection location Capacity ratio Real-time power factor The three key variables are incorporated into the same function, breaking through the limitation that static coefficients are only related to installed capacity, and quantifying the hidden losses such as "the difference in line loss of up to 15% due to different access locations under the same penetration rate" (e.g., the reverse path impedance of end access is 3 times higher than that of head access).
[0047] Specific Implementation Method Five: This implementation method further defines the dynamic calculation method for distribution network line losses based on photovoltaic backfeed power correction described in Specific Implementation Method Four. In this implementation method, the expression for the equivalent resistance of the distribution network line after correction is:
[0048]
[0049] in, The equivalent resistance of the distribution network lines after correction; The initial equivalent resistance of the distribution network lines; This is the penalty factor for three-phase imbalance.
[0050] In this embodiment, the initial equivalent resistance of the distribution network lines is used. The linear correction formula embeds the backfeed loss into the equivalent resistance method framework without requiring topology reconstruction, reducing the calculation time from minutes to milliseconds in the forward-backward substitution method, thus meeting the real-time computing needs of embedded terminals (such as intelligent fusion terminals).
[0051] Specific Implementation Method Six: This implementation method further defines the dynamic calculation method for distribution network line losses based on photovoltaic backfeed power correction described in Specific Implementation Method Five. In this implementation method, the three-phase imbalance penalty factor... The expression is:
[0052]
[0053] in, This refers to the negative sequence current imbalance. This represents the zero-sequence current imbalance.
[0054] In this embodiment, the penalty factor First assignment of zero-order imbalance With a 4x weighting, it accurately reflects the quadratic loss effect of the neutral current; it solves the problem of underestimation of more than 25% caused by the existing algorithm ignoring zero-sequence loss (for example, when the proportion of single-phase photovoltaic in residential buildings is greater than 70%, the neutral wire diameter is only 50% of that of the phase wire, and the resistance is 4 times higher).
[0055] Specific Implementation Method Seven: This implementation method further defines the dynamic calculation method for distribution network line losses based on photovoltaic power feedback correction described in Specific Implementation Method Six. In this implementation method, the formula for calculating the root mean square current during the distribution network line feedback period is as follows:
[0056]
[0057] in, This represents the root mean square current during the reverse transmission period of the distribution network line. This represents the number of sampling points within the current time period. The sampling sequence number; For the first The instantaneous current value after sign reversal at each sampling time.
[0058] In this embodiment, the instantaneous value of the current during the reverse feed period is forcibly reversed (e.g., +50A → -50A) to ensure the root mean square current. It accurately reflects the superposition effect of bidirectional power flow and avoids the 30% loss omission caused by "current direction cancellation" in existing RMS calculations (such as when the phase difference between the back current and the load current is 180°, existing RMS only counts the difference).
[0059] Specific Implementation Method Eight: This implementation method further defines the dynamic calculation method for distribution network line losses based on photovoltaic backfeed power correction described in Specific Implementation Method Seven. In this implementation method, the formula for calculating the real-time line loss of the distribution network line is:
[0060]
[0061] in, This refers to the real-time line loss of the distribution network lines. This refers to the no-load loss of the distribution network lines. This is used to calculate the number of hours in the time period.
[0062] In this embodiment, the real-time line loss formula It strictly distinguishes between copper loss and iron loss, solving the problem that the "equivalent capacity method" mistakenly includes the increase in iron loss caused by the rise in back-feed voltage (such as a 17% increase in iron loss at 1.1 pu voltage) in copper loss, and supports accurate cost allocation under time-of-use pricing (such as the need to bear the iron loss premium during the photovoltaic back-feed period).
[0063] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.
Claims
1. A method for dynamically calculating line losses in distribution networks based on photovoltaic backfeed power correction, characterized in that, Includes the following steps: Collect multi-source measurement data of power distribution network lines; Based on the collected multi-source measurement data of the distribution network lines, the reverse transmission characteristics are identified to determine the reverse transmission time period of the distribution network lines; Based on the reverse power period of the distribution network line, construct the power correction coefficient for the reverse power of the distribution network line; The initial equivalent resistance of the distribution network line is dynamically corrected by using the power backfeed correction coefficient of the distribution network line, and the corrected equivalent resistance of the distribution network line is obtained. The current sequence of the distribution network line is divided into two reverse periods according to the reverse period indicator, and the sign of the current in the reverse period is reversed. The root mean square current of the distribution network line in the reverse period is calculated. The real-time line loss of the distribution network is calculated using the corrected equivalent resistance of the distribution network line and the root mean square current of the distribution network line during the reverse transmission period.
2. The method for dynamic calculation of distribution network line losses based on photovoltaic backfeed power correction according to claim 1, characterized in that, Multi-source measurement data of distribution network lines include the active power of distribution network lines. Reactive power of distribution network lines Three-phase voltage of distribution network lines and the current of the distribution network lines .
3. The method for dynamic calculation of distribution network line losses based on photovoltaic backfeed power correction according to claim 1, characterized in that, The criterion for identifying the reverse power supply period of the distribution network is that the reverse power is greater than 0 within the sampling period and lasts for at least 2 sampling periods.
4. The method for dynamic calculation of distribution network line losses based on photovoltaic backfeed power correction according to claim 1, characterized in that, The expression for the power correction factor for backfeeding from distribution network lines is: in, This is a correction factor for the reverse power transmission of the distribution network lines; The proportion of photovoltaic grid connection locations; This is the ratio of photovoltaic capacity to the maximum downstream load. The real-time power factor is volts.
5. The method for dynamic calculation of distribution network line losses based on photovoltaic backfeed power correction according to claim 4, characterized in that, The expression for the corrected equivalent resistance of the distribution network line is: in, The equivalent resistance of the distribution network lines after correction; The initial equivalent resistance of the distribution network lines; This is the three-phase imbalance penalty factor.
6. The method for dynamic calculation of distribution network line losses based on photovoltaic backfeed power correction according to claim 5, characterized in that, The three-phase imbalance penalty factor The expression is: in, This refers to the negative sequence current imbalance. This represents the zero-sequence current imbalance.
7. The method for dynamic calculation of distribution network line losses based on photovoltaic backfeed power correction according to claim 6, characterized in that, The formula for calculating the root mean square current during the reverse transmission period of the distribution network line is as follows: in, This represents the root mean square current during the reverse transmission period of the distribution network line. This represents the number of sampling points within the current time period. The sampling sequence number; For the first The instantaneous current value after sign reversal at each sampling time.
8. The method for dynamic calculation of distribution network line losses based on photovoltaic backfeed power correction according to claim 7, characterized in that, The formula for calculating the real-time line loss of the distribution network is as follows: in, This refers to the real-time line loss of the distribution network lines. This refers to the no-load loss of the distribution network lines. This is used to calculate the number of hours in the time period.