A power grid line loss reduction optimization method and system for a power distribution area

Abstract

The application relates to the technical field of power grid line loss optimization, in particular to a power grid line loss reduction optimization method and system for a power distribution area, which comprises the following steps: collecting three-phase voltage and three-phase current used by power grid users in real time; determining a first characteristic parameter of a three-phase line by the high-frequency energy proportion of each phase current, the deviation between a measured fundamental wave time sequence signal and a standard fundamental wave time sequence signal, and the energy mutation degree caused by high-frequency harmonics; obtaining a phase deviation degree of the three-phase line; obtaining a second characteristic parameter of the three-phase line based on the phase deviation degree of the three-phase line and the first characteristic parameter, adjusting a smoothing coefficient in a smoothing algorithm, and predicting the line loss rate of the next moment of the current moment to judge the matching degree of the line loss and electricity stealing line loss of the current moment, and reducing the line loss of the power grid. Therefore, the line loss reduction effect of the power grid is improved.

Description

Technical Field

[0001] This application relates to the field of power grid line loss optimization technology, specifically to a method and system for optimizing power grid line loss reduction in a distribution substation. Background Technology

[0002] Line loss refers to the energy loss that occurs during the transmission of electrical energy from the power grid to the user, occurring at various stages such as transmission, transformation, distribution, and consumption. Achieving efficient reduction of power grid line loss is crucial for ensuring the efficient and economical operation of the power system. The prerequisite for efficient loss reduction is the timely detection of line loss anomalies. Analysis of line loss data reveals abnormal fluctuations and helps determine if there are anomalies such as electricity theft. Furthermore, to ensure early fault detection, predictive algorithms are typically used to forecast line loss anomalies. Therefore, the accuracy of line loss anomaly prediction directly affects the effectiveness of loss reduction optimization.

[0003] Smoothing algorithms are commonly used in power distribution network line loss reduction for prediction. For example, the Exponentially Weighted Moving Average (EWMA) algorithm essentially applies exponentially decaying weights to historical data to predict future line losses. However, this algorithm uses a fixed-size smoothing parameter for decay, which is effective for predicting slowly increasing line loss anomalies, but relatively poor for sudden, abnormal increases in line loss, such as electricity theft. If the decay weights are small, the prediction results still heavily rely on historical line loss data, smoothing the next prediction. For instance, if a user suddenly steals electricity, causing a sharp increase in line loss, the predicted line loss may not accurately reflect the actual situation, easily influenced by historical data, reducing the degree of anomaly and affecting the overall line loss reduction effect. Summary of the Invention

[0004] To address the aforementioned technical problems, the purpose of this application is to provide a method and system for optimizing power grid line loss reduction in distribution substations. The specific technical solution adopted is as follows: In a first aspect, embodiments of this application provide a method for optimizing power grid line loss reduction in a distribution substation, the method comprising the following steps: Real-time acquisition of three-phase voltage and three-phase current used by power grid users; The first characteristic parameter of the three-phase line is determined by the proportion of high-frequency energy in each phase current, the deviation between the measured fundamental frequency timing signal and the standard fundamental frequency timing signal, and the degree of energy change caused by high-frequency harmonics. The phase angles corresponding to the fundamental signals of each phase voltage are obtained, the phase deviations of each phase voltage from the standard phase angle are determined, and the phase deviations of each phase current from the standard phase angle are fused to obtain the phase deviation of the three-phase line. Based on the phase deviation of the three-phase line and the first characteristic parameter, the second characteristic parameter of the three-phase line is obtained. The smoothing coefficient in the smoothing algorithm is adjusted to predict the line loss rate of the next moment at the current moment, so as to determine the matching degree between the line loss at the current moment and the theft of power lines, and to reduce the power grid line loss.

[0005] In one embodiment, the process of acquiring the standard fundamental time-series signal is as follows: Based on the preset fundamental frequency of the power grid, the time-domain signal of each phase current is transformed in the frequency domain to obtain the amplitude and phase of the fundamental wave. Combined with the preset fundamental frequency, the standard fundamental wave time sequence signal corresponding to each phase current is reconstructed.

[0006] In one embodiment, determining the first characteristic parameter of the three-phase line includes: For each phase current, calculate the metric distance between the measured fundamental time series signal and the standard fundamental time series signal; obtain the time-frequency diagram of the time series current signal, calculate the energy deviation of each frequency at adjacent moments, and take the ratio of the maximum value of the energy deviation of each frequency to the energy of the first moment of its corresponding adjacent moment as the energy mutation rate of each frequency. Based on the magnitude of the energy mutation rate at all frequencies in the high-frequency band, and combined with the metric distance and the high-frequency energy ratio, the first characteristic parameter of the three-phase line is obtained.

[0007] In one embodiment, further determination of the first feature parameter includes: The maximum value of the energy mutation rate of all frequencies in the high-frequency band is taken as the global energy mutation rate of each phase current. The average value of the high-frequency energy ratio of the three-phase current, the average value of the measurement distance, and the average value of the global energy mutation rate are combined to obtain the first characteristic parameter of the three-phase line.

[0008] In one embodiment, determining the standard phase angle includes: The standard phase angle of phase A voltage is defined as the phase angle of phase A voltage, the standard phase angle of phase B voltage is the difference between the phase angle of phase A voltage and 120°, and the standard phase angle of phase C voltage is the difference between the phase angle of phase A voltage and 240°.

[0009] In one embodiment, determining the phase deviation of the three-phase line includes: Calculate the phase deviation of each phase voltage from the standard phase angle. The specific calculation method is as follows: In the formula, This represents the phase deviation of the phase angle of the k-phase voltage from the standard phase angle. Let be the phase angle of the k-phase voltage. Let be the standard phase angle of the k-phase voltage, and min() be the minimum value function; where k∈A, B, C; Calculate the ratio of the root mean square value of the phase deviation corresponding to all phase voltages to the maximum phase deviation, and record it as the first ratio. Accordingly, obtain the first ratio corresponding to all phase currents. The phase deviation of the three-phase line is obtained by weighted summing of the first ratio of all phase voltages and the first ratio of all phase currents, wherein the weight of the first ratio of all phase currents is greater than the weight of the first ratio of all phase voltages.

[0010] In one embodiment, the second characteristic parameter is the product of the phase deviation of the three-phase line and the first characteristic parameter.

[0011] In one embodiment, the smoothing coefficient in the adjustment smoothing algorithm is expressed as: In the formula, The smoothing coefficient is adjusted by the smoothing algorithm. This is the second characteristic parameter of a three-phase circuit. The preset maximum value for the smoothing coefficient. This is the preset minimum value for the smoothing coefficient. This is the preset adjustment coefficient.

[0012] In one embodiment, determining the degree of matching between the current line loss and the loss from wire theft, and then reducing power grid line loss, includes: The line loss type at the current moment is obtained based on a deep learning algorithm. If the line loss at the current moment is the loss due to electricity theft, the line loss rate of the current collection period is used as the input of a smoothing algorithm to predict the line loss rate at the next moment. If the error between the line loss rate at the next moment and the current moment is within a preset range, it is determined that the current user has engaged in electricity theft and the current user's electricity consumption is corrected. Otherwise, the current user's electricity consumption is not corrected.

[0013] Secondly, embodiments of this application also provide a power grid line loss reduction and optimization system for a distribution substation, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of any of the methods described above.

[0014] This application has at least the following beneficial effects: This application, by real-time acquisition of three-phase voltage and current used by power grid users, combined with the proportion of high-frequency energy, fundamental signal deviation, and the degree of abrupt changes in high-frequency harmonic energy, can accurately assess the characteristic parameters of three-phase lines. The high-precision real-time data makes the prediction of line losses more accurate, thereby improving the overall operating efficiency of the power grid. Furthermore, by analyzing the phase deviation of three-phase current and voltage, abnormal states in the power grid can be detected in a timely manner, especially abnormal line losses caused by electricity theft. Monitoring the phase angle deviation between voltage and current helps distinguish between normal power consumption and abnormal electricity theft, thereby effectively preventing and reducing theft. The application addresses the issue of power loss. Furthermore, by adjusting the smoothing coefficient in the smoothing algorithm in real time, it enables dynamic prediction and real-time optimization of the power grid line loss rate. It allows for flexible adjustment of the prediction algorithm based on current current and voltage fluctuations and the power grid's operating conditions, avoiding the error accumulation and real-time response lag issues caused by fixed parameters in traditional methods. This application accurately identifies the types of power grid line losses, improves the optimization effect of power grid line loss reduction, helps reduce electricity theft, significantly reduces unnecessary power losses, lowers power grid maintenance and operation costs, comprehensively improves the intelligent management level of the power grid, and enhances the stability and efficiency of the power grid. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating the steps of a method for optimizing power grid line loss reduction in a distribution substation, as provided in one embodiment of this application. Detailed Implementation

[0016] The following, in conjunction with the accompanying drawings, details the specific scheme of the power grid line loss reduction and optimization method and system provided in this application for a distribution substation.

[0017] Please see Figure 1 The diagram illustrates a flowchart of a method for optimizing power grid line loss reduction in a distribution substation according to an embodiment of this application. The method includes the following steps: S1 collects the three-phase voltage and three-phase current used by power grid users in real time and performs preprocessing.

[0018] This embodiment reads smart meter data from each user in the power grid using a communication protocol. The communication protocol can be DL / T 645-2007 or Modbus, etc., and this embodiment does not impose any special restrictions. The smart meter data includes the user ID, the user's three-phase voltage and current, electricity consumption, and power consumption. This embodiment collects smart meter data at a frequency of 5kHz for a 48-hour period, obtaining the time-series three-phase voltage and current signals within one collection period. The implementer can set the collection frequency and period according to actual conditions; this embodiment does not impose any restrictions on this.

[0019] To address the potential for missing or outlier values ​​in the collected voltage, current, and other data, this embodiment uses 3... The principle is to identify outliers, and then use linear interpolation to both correct outliers and fill in missing values, thus completing the preprocessing of three-phase voltage and three-phase current data. 3 Both the principle and the linear interpolation method are existing well-known techniques, and the specific process will not be elaborated here.

[0020] S2 determines the first characteristic parameter of the three-phase line by the proportion of high-frequency energy in each phase current, the deviation between the measured fundamental frequency timing signal and the standard fundamental frequency timing signal, and the degree of energy mutation caused by high-frequency harmonics.

[0021] When no electricity theft occurs in the power grid, the harmonics in the frequency domain signals of voltage and current are mainly concentrated near the low-frequency fundamental wave, with no high-frequency harmonics. Since the power grid frequency is 50Hz, the harmonics are mainly concentrated in integer multiples of the power frequency, and the energy concentration at low frequencies is also relatively high. If users illegally steal electricity through technical means, such as emitting high-frequency harmonics into smart meters, causing the fundamental waveform to deviate from the sine wave, a large number of high-frequency harmonics will also appear, increasing the proportion of frequency energy in the high-frequency band. At the same time, the abnormal line loss caused by electricity theft is sudden, and the energy at the same frequency may change abruptly at different times. Therefore, this embodiment measures the degree of line loss caused by user electricity theft based on the degree of harmonic anomaly in the frequency domain signals of voltage and current. Since current is more susceptible to harmonics than voltage, the more harmonics there are, the more obvious the abnormal changes in current are. This embodiment analyzes the degree of harmonic anomaly in the frequency domain signals of three-phase current.

[0022] Specifically, this embodiment takes the time-series three-phase current signal within a single acquisition cycle as an example for analysis, and analyzes the time-series signal of the three-phase current. , , As input, the corresponding current frequency domain signal is output using a Fast Fourier Transform. In this embodiment, the first phase current timing signal is used. For example, the current timing signal The corresponding current frequency domain signal is used as input. Due to the wiring of the power grid itself, there may be multiple harmonics at low frequencies. However, the harmonics emitted by electricity theft are mainly at high frequencies and have a greater impact on the current frequency domain signal. Therefore, in this embodiment, frequencies above 400Hz in the current frequency domain signal are set as the high-frequency band. The total energy value of the high-frequency band is calculated, along with the total energy of the entire current frequency domain signal. The ratio of high-frequency energy to total energy is then calculated to obtain the first phase current timing signal. High-frequency energy ratio The Fast Fourier Transform (FFT) is a well-known technique, and its specific process will not be elaborated upon here.

[0023] Since the power grid frequency is 50Hz, this embodiment determines 50Hz as the fundamental frequency, starting from the first phase current timing signal. The amplitude and phase of the fundamental wave are obtained from the current frequency domain signal after Fast Fourier Transform (FFT). Using the obtained fundamental frequency (50Hz), amplitude, and phase, a standard fundamental wave timing signal, i.e., a standard sine waveform, is reconstructed. Furthermore, the first phase current timing signal... The current frequency domain signal obtained after Fast Fourier Transform is truncated with a bandwidth of 50Hz to account for power grid frequency fluctuations. Therefore, it is retained. By setting the frequency domain points within the specified range to 0 and setting the rest to 0, a frequency domain signal containing only 50Hz and its local range is obtained. Then, an inverse Fourier transform is used to convert the obtained frequency domain signal to the time domain, yielding the actual fundamental frequency time series signal. Finally, the metric distance between the actual fundamental frequency time series signal and the standard fundamental frequency time series signal is calculated. This represents the degree of distortion of the actual fundamental wave signal. In this embodiment, the distance measurement is calculated using DTW distance. Implementers can choose other feasible distance measurement methods, such as Euclidean distance or Manhattan distance. Obtaining the corresponding amplitude, phase, fundamental wave signal reconstruction, inverse Fourier transform, and DTW distance calculation through fast Fourier transform are all existing well-known technologies, and the specific process will not be described in detail.

[0024] The first phase current timing signal As input, a short-time Fourier transform is used to obtain the corresponding time-frequency graph, where the horizontal axis represents time, the vertical axis represents frequency, and the corresponding coordinate value represents the energy value. Taking the energy at frequency f at time t as an example, the energy value corresponding to frequency f at time t, i.e., coordinate (t, f), is denoted as... Where time t starts from the second moment and frequency f starts from 400Hz (high frequency band). Calculation The energy value corresponding to the frequency f at the previous time (t-1 time) is energy deviation The specific calculation method is as follows: Furthermore, the energy deviation of frequency f at all times is calculated sequentially, and the maximum value among all energy deviations corresponding to frequency f is obtained. Let f represent the energy deviation of frequency f between time i and time i-1, and calculate the maximum value. The ratio of the energy of frequency f at time i-1 to the energy of frequency f is used as the energy mutation rate of frequency f. The larger the energy mutation rate, the more abnormal the energy mutation rate caused by high-frequency harmonics. It should be noted that when the energy of frequency f at time i-1 is 0, the energy mutation rate of frequency f is directly set to 0.

[0025] It should be noted that the energy deviation is calculated using the absolute value of the difference in this embodiment. Implementers may choose other calculation methods to measure the energy deviation, such as the square of the difference, the ratio, etc. This embodiment does not impose any restrictions on this.

[0026] The maximum value of the energy mutation rate across all frequencies in the high-frequency band is selected as the global energy mutation rate of the first-phase current. The short-time Fourier transform is a well-known technique, and its specific process will not be elaborated upon. Using the same calculation method as for the global energy mutation rate of the first-phase current, the global energy mutation rates of each phase current can be obtained.

[0027] Based on the above analysis, the first characteristic parameter of the three-phase line is calculated to characterize the degree of influence of high-frequency harmonics emitted by electricity theft on the three-phase line. Specifically: The first characteristic parameter of the three-phase line is obtained by combining the average high-frequency energy proportion of the three-phase current, the average metric distance, and the average global energy mutation rate. Here, "combining" means combining multiple variables, which can be calculated using methods such as addition, multiplication, or a combination of addition and multiplication.

[0028] In this embodiment, the expression for the first characteristic parameter of the three-phase line is: In the formula, This represents the first characteristic parameter of a three-phase circuit. The average proportion of high-frequency energy in the three-phase current represents the high-frequency energy characteristics of the three-phase line. The mean of the measured distance of the three-phase current represents the degree of difference between the fundamental signal of the three-phase line and the standard fundamental signal. This represents the average global energy mutation rate of the three-phase current. The larger the value of the first characteristic parameter of the three-phase line, the greater the degree to which the three-phase line is affected by high-frequency harmonics emitted during electricity theft.

[0029] It should be understood that the high-frequency harmonics emitted during electricity theft have a significant impact on the current data in three-phase lines. Therefore, this embodiment mainly analyzes the current timing signal in three-phase lines. High-frequency harmonics increase the high-frequency energy distribution and cause the fundamental signal to deviate from the standard fundamental waveform. During high-frequency harmonic emission, energy abrupt changes will occur in the high-frequency band, i.e., the energy abrupt change rate is high and exceeds the normal abrupt change rate range. The calculated mean value... , , All three are positively correlated with the degree of influence of high-frequency harmonics. Therefore, the result of multiplying the three together represents the first characteristic parameter of the three-phase line.

[0030] S3, obtain the phase angle corresponding to the fundamental signal of each phase voltage, determine the phase deviation of each phase voltage from the standard phase angle, and fuse the phase deviation of each phase current from the standard phase angle to obtain the phase deviation of the three-phase line.

[0031] Further analysis reveals that during normal operation of the three-phase line, the phases of the voltages and currents in each phase of the A, B, and C phases differ by 120°. When users steal electricity by emitting high-frequency harmonics into the smart meter, the voltage and current data become distorted. The more harmonics emitted, the greater the impact of the high-frequency harmonics emitted during electricity theft on the three-phase line, resulting in greater distortion and a greater deviation in the phase angles of the voltage and current in the three-phase line. This indicates a greater degree of abnormal line loss.

[0032] Specifically, taking the voltage timing signal of a single acquisition cycle in a three-phase line as an example, the timing signals of the three-phase voltages are expressed as follows: , , , obtain , , The fundamental wave signal and its corresponding phase angle , , Furthermore, taking the phase angle of phase A voltage as an example... Construct a standard symmetric phase angle for the initial phase , denoted as the standard phase angle, where, This represents the standard phase angle of phase k, where k can take values ​​A, B, or C. When k = A, When k=B, When k=C, Based on the calculated phase angle of the three-phase line voltage , , Compared with standard phase angle To determine the phase deviation of each phase voltage from the standard phase angle, since the phase angle has a 360° cycle, the specific calculation method is as follows: In the formula, This represents the phase deviation of the phase angle of the k-phase voltage from the standard phase angle. Let be the phase angle of the k-phase voltage. Let be the standard phase angle of the k-phase voltage, and min() be the function to take the minimum value.

[0033] Furthermore, the maximum value of the phase deviation between the phase angle of all phase voltages and the standard phase angle is obtained. The root mean square value of the phase deviation between the phase angle of all phase voltages and the standard phase angle is calculated as the ratio of the maximum phase deviation, denoted as the first ratio. The first ratio corresponding to all phase currents is obtained using the same calculation method as for the first ratio of all phase voltages. In particular, when the maximum phase deviation is zero, the first ratio is directly set to zero.

[0034] The phase deviation of the three-phase line is obtained by weighted summing of the first ratio of all phase voltages and the first ratio of all phase currents. Since harmonics have a greater impact on current, the weight of the first ratio of all phase currents is greater than the weight of the first ratio of all phase voltages. In this embodiment, the weight of the first ratio of all phase currents is set to 0.6, and the weight of the first ratio of all phase voltages is set to 0.4. Implementers can set these weights according to their actual circumstances; this embodiment does not impose any restrictions.

[0035] S4. Based on the phase deviation of the three-phase line and the first characteristic parameter, the second characteristic parameter of the three-phase line is obtained. The smoothing coefficient in the smoothing algorithm is adjusted to predict the line loss rate of the next moment at the current moment, so as to determine the matching degree between the line loss at the current moment and the theft of power lines, and to reduce the power grid line loss.

[0036] The second characteristic parameter of the three-phase line is calculated to characterize the degree of distortion of high-frequency harmonics emitted by electricity theft in the three-phase line. Specifically: The product of the phase deviation of the three-phase line and the first characteristic parameter is used as the second characteristic parameter of the three-phase line. The larger the second characteristic parameter of the three-phase line, the higher the degree of distortion of the three-phase line by high-frequency harmonics emitted by electricity theft.

[0037] It should be understood that the more high-frequency harmonics emitted during electricity theft, the greater the impact of these harmonics on the three-phase circuit, resulting in greater distortion. This means that the first characteristic parameter of the three-phase circuit is larger, and the phase angle deviation of the voltage and current in the three-phase circuit is greater, which is positively correlated with the second characteristic parameter.

[0038] When line loss detection technology is used to detect abnormal line loss due to electricity theft for the first time in a power grid loss reduction project in a distribution substation, in order to formulate a more accurate loss reduction plan, it is usually necessary to observe the abnormal line loss rate over a period of time to avoid temporary false anomalies caused by environmental factors. The exponentially weighted moving average algorithm is usually used to predict the line loss rate at the next moment. However, because the exponentially weighted moving average algorithm uses a fixed parameter for attenuation, the predicted value will be smoothed by the historical normal line loss rate, making it significantly lower than the detected line loss rate, which can mislead decision-making and make it difficult to determine whether the abnormal line loss is really caused by user electricity theft. Therefore, this embodiment adjusts the smoothing coefficient according to the voltage and current distortion of the three-phase line to make the predicted line loss rate more accurate.

[0039] Based on the above analysis, the smoothing coefficient of the exponentially weighted moving average algorithm is adjusted to characterize the degree of dependence on historical line loss rate data when predicting line loss rate. The specific expression is as follows: In the formula, The smoothing coefficient is adjusted by the smoothing algorithm. This is the second characteristic parameter of a three-phase circuit. The preset maximum value for the smoothing coefficient. This is the preset minimum value for the smoothing coefficient. The smoothing algorithm used in this embodiment is an exponentially weighted moving average algorithm, with a preset adjustment coefficient. Implementers can choose other feasible smoothing algorithms, such as exponential smoothing. The smoothing coefficient is set to range from 0.1 to 0.8. , Adjustment coefficient The value range is 0.5 to 1. In this embodiment... To avoid excessively adjusting the smoothing coefficient and relying too heavily on historical line loss rates, which makes the system susceptible to noise.

[0040] Furthermore, this embodiment trains a deep learning model using electricity consumption data collected from historical instances of electricity theft, resulting in a binary classification deep learning model capable of detecting such behavior. The electricity consumption data includes the amount of electricity supplied by the power grid, the electricity used by the user's meter, and the line loss rate. This embodiment uses a CNN-LSTM deep learning model, which is a well-known and existing technique. Implementers can choose other feasible deep learning models, such as ResNet-LSTM, and this embodiment does not impose any restrictions on this choice.

[0041] The electricity consumption data of the user in the current collection period is used as the input to the trained CNN-LSTM model, and the output is the line loss type of the user at the current moment. The line loss type includes theft-related line loss and non-theft-related line loss. If the line loss at the current moment is theft-related line loss, the line loss rate of the current collection period is used as the input to the exponentially weighted moving average algorithm, and the line loss rate of the next moment is output. If the error between the line loss rate of the next moment and the current moment is within a preset range, it is determined that the current user has engaged in electricity theft, and the current user's electricity consumption is corrected; otherwise, the current user's electricity consumption is not corrected. In this embodiment, the preset range is ±5%, and the implementer can set it according to the actual situation. This embodiment does not impose any restrictions on this.

[0042] If the abnormal line loss rate is caused by user electricity theft, the power to the smart meters of the users involved in the theft will be cut off in time to reduce line loss. However, high-frequency harmonic interference can cause electromagnetic coupling in the voltage and current sampling coils, leading to signal distortion. This causes the meter's chip to misjudge the amplitude and phase of the fundamental wave, resulting in a lower calculated fundamental active power, meaning the meter reading is lower than the actual electricity consumption. Therefore, this embodiment calculates the fundamental active power based on the amplitude and phase of the fundamental waves of the three-phase voltage and current. Where U and I represent the frequency domain signal amplitudes of voltage and current, respectively. This represents the phase difference between the fundamental voltage and fundamental current under the same phase. Therefore, the smart meter's electricity consumption can be corrected based on the fundamental active power and the actual power recorded by the smart meter. The corrected user electricity consumption is then calculated. for ,in, This indicates the actual measured value of electricity consumption by the smart meter. This indicates the actual recorded power value of the smart meter. Specifically, when the actual recorded power value of the smart meter... When it is zero, the fundamental active power is used directly. The user's electricity consumption was estimated and corrected based on the electricity usage time. .

[0043] This embodiment addresses technical electricity theft by automatically cutting off power to the electricity meter and correcting the meter's power consumption to promptly compensate for losses, thus optimizing line loss reduction. The calculation of the fundamental active power is a well-known existing technique, and the specific process will not be elaborated further.

[0044] Based on the same inventive concept as the above method, this application embodiment also provides a power grid line loss reduction and optimization system for a distribution substation, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above-described methods for power grid line loss reduction and optimization in a distribution substation.

Claims

1. A method for optimizing power grid line loss reduction in a distribution substation, characterized in that, The method includes the following steps: Real-time acquisition of three-phase voltage and three-phase current used by power grid users; The first characteristic parameter of the three-phase line is determined by the proportion of high-frequency energy in each phase current, the deviation between the measured fundamental frequency timing signal and the standard fundamental frequency timing signal, and the degree of energy change caused by high-frequency harmonics. The phase angles corresponding to the fundamental signals of each phase voltage are obtained, the phase deviations of each phase voltage from the standard phase angle are determined, and the phase deviations of each phase current from the standard phase angle are fused to obtain the phase deviation of the three-phase line. Based on the phase deviation of the three-phase line and the first characteristic parameter, the second characteristic parameter of the three-phase line is obtained. The smoothing coefficient in the smoothing algorithm is adjusted to predict the line loss rate of the next moment at the current moment, so as to determine the matching degree between the line loss at the current moment and the theft of power lines, and to reduce the power grid line loss.

2. The method for optimizing power grid line loss reduction in a distribution substation as described in claim 1, characterized in that, The process of acquiring the standard fundamental time-series signal is as follows: Based on the preset fundamental frequency of the power grid, the time-domain signal of each phase current is transformed in the frequency domain to obtain the amplitude and phase of the fundamental wave. Combined with the preset fundamental frequency, the standard fundamental wave time sequence signal corresponding to each phase current is reconstructed.

3. The method for optimizing power grid line loss reduction in a distribution substation as described in claim 1, characterized in that, The determination of the first characteristic parameter of the three-phase line includes: For each phase current, calculate the metric distance between the measured fundamental time series signal and the standard fundamental time series signal; obtain the time-frequency diagram of the time series current signal, calculate the energy deviation of each frequency at adjacent moments, and take the ratio of the maximum value of the energy deviation of each frequency to the energy of the first moment of its corresponding adjacent moment as the energy mutation rate of each frequency. Based on the magnitude of the energy mutation rate at all frequencies in the high-frequency band, and combined with the metric distance and the high-frequency energy ratio, the first characteristic parameter of the three-phase line is obtained.

4. The method for optimizing power grid line loss reduction in a distribution substation as described in claim 3, characterized in that, Further determination of the first feature parameter includes: The maximum value of the energy mutation rate of all frequencies in the high-frequency band is taken as the global energy mutation rate of each phase current. The average value of the high-frequency energy ratio of the three-phase current, the average value of the measurement distance, and the average value of the global energy mutation rate are combined to obtain the first characteristic parameter of the three-phase line.

5. The method for optimizing power grid line loss reduction in a distribution substation as described in claim 1, characterized in that, The determination of the standard phase angle includes: The standard phase angle of phase A voltage is defined as the phase angle of phase A voltage, the standard phase angle of phase B voltage is the difference between the phase angle of phase A voltage and 120°, and the standard phase angle of phase C voltage is the difference between the phase angle of phase A voltage and 240°.

6. The method for optimizing power grid line loss reduction in a distribution substation as described in claim 1, characterized in that, The determination of the phase deviation of the three-phase line includes: Calculate the phase deviation of each phase voltage from the standard phase angle. The specific calculation method is as follows: In the formula, This represents the phase deviation of the phase angle of the k-phase voltage from the standard phase angle. Let be the phase angle of the k-phase voltage. Let be the standard phase angle of the k-phase voltage, and min() be the minimum value function; where k∈A, B, C; Calculate the ratio of the root mean square value of the phase deviation corresponding to all phase voltages to the maximum phase deviation, and record it as the first ratio. Accordingly, obtain the first ratio corresponding to all phase currents. The phase deviation of the three-phase line is obtained by weighted summing of the first ratio of all phase voltages and the first ratio of all phase currents, wherein the weight of the first ratio of all phase currents is greater than the weight of the first ratio of all phase voltages.

7. The method for optimizing power grid line loss reduction in a distribution substation as described in claim 1, characterized in that, The second characteristic parameter is the product of the phase deviation of the three-phase line and the first characteristic parameter.

8. The method for optimizing power grid line loss reduction in a distribution substation as described in claim 1, characterized in that, The smoothing coefficient in the adjustment smoothing algorithm is expressed as follows: In the formula, The smoothing coefficient is adjusted by the smoothing algorithm. This is the second characteristic parameter of a three-phase circuit. The preset maximum value for the smoothing coefficient. This is the preset minimum value for the smoothing coefficient. This is the preset adjustment coefficient.

9. The method for optimizing power grid line loss reduction in a distribution substation as described in claim 1, characterized in that, The determination of the degree of matching between the current line loss and the loss from stolen electricity, and the subsequent reduction of power grid line loss, includes: The line loss type at the current moment is obtained based on a deep learning algorithm. If the line loss at the current moment is the loss due to electricity theft, the line loss rate of the current collection period is used as the input of a smoothing algorithm to predict the line loss rate at the next moment. If the error between the line loss rate at the next moment and the current moment is within a preset range, it is determined that the current user has engaged in electricity theft and the current user's electricity consumption is corrected. Otherwise, the current user's electricity consumption is not corrected.

10. A power grid line loss reduction and optimization system for a distribution substation, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1-9.

Images