A transformer area line loss anomaly analysis method, system, device and medium based on source network load storage collaboration

Abstract

The present application relates to the technical field of power system distribution network operation monitoring and line loss analysis, and discloses a transformer area line loss anomaly analysis method, system, equipment and medium based on source-network-load-storage cooperation, which comprises the following steps: fusing multi-source heterogeneous data through time synchronization and nonlinear interpolation to construct a unified feature vector; adopting a line loss calculation model fused with a storage energy dynamic correction term to accurately calculate the actual loss of the transformer area; identifying line loss anomalies through a dynamic self-adaptive algorithm based on historical statistics and instantaneous change trend; and positioning anomalies to specific sides of source, network, load and storage through multi-source residual contribution degree analysis. The present application effectively improves the accuracy of complex distribution transformer area line loss calculation, the sensitivity and robustness of anomaly detection, and realizes the explainable and accurate positioning of anomaly sources.

Description

Technical Field

[0001] This invention relates to the field of technology, and in particular to a method, system, equipment and medium for analyzing abnormal line losses in transformer substations based on source-grid-load-storage coordination. Background Technology

[0002] With the continuous expansion of new energy access, electricity users' consumption behavior is becoming increasingly complex, and the operating environment of traditional distribution substations has also undergone significant changes. The collaborative operation of multiple entities—source, grid, load, and storage—has become the main trend in distribution system development. In this context, line loss is a key indicator reflecting the operational efficiency of the distribution network and the compliance of electricity consumption behavior. The accuracy of line loss calculation and the timely identification of anomalies directly determine the quality of distribution network operation and maintenance, as well as the strength of user-side risk control capabilities. However, current line loss analysis methods have many limitations. Most methods still rely solely on the difference in electricity consumption between transformer inlets and smart meters, without considering the interference of energy storage systems, distributed power sources, and dynamic loads on energy balance, leading to distorted line loss calculation results. Furthermore, inconsistencies in sampling frequencies, misaligned timestamps, and missing data are common problems in data acquisition systems, preventing unified time-series data fusion processing and affecting the accuracy of subsequent anomaly identification.

[0003] Currently, most line loss anomaly identification methods rely on fixed threshold judgments, which cannot adapt to dynamic changes in the operating environment. When faced with sudden issues such as electricity theft, equipment failure, and metering anomalies, they lack effective adaptive judgment mechanisms. After identifying line loss anomalies, most existing methods cannot further clarify the source of the anomaly, lacking refined tracing capabilities. They cannot determine whether the anomaly is caused by user electricity consumption behavior, distributed power output fluctuations, energy storage system strategy switching, or problems with the power distribution equipment itself, thus limiting the ability of power distribution companies to locate and manage line loss problems. Summary of the Invention

[0004] In view of the above-mentioned existing problems, the present invention provides a method, system, equipment and medium for anomaly analysis of transformer area line loss based on source-grid-load-storage coordination.

[0005] Therefore, the technical problem solved by this invention is that existing methods for analyzing line loss in distribution areas are difficult to adapt to the complex scenario of multi-side coordinated operation of sources, grids, loads, and storage, and cannot effectively handle problems such as time misalignment and inconsistent sampling frequencies caused by heterogeneous data from multiple sources. This results in large line loss calculation errors, inaccurate anomaly identification, high misjudgment rate, and unclear source tracing and positioning, which in turn restricts the distribution network's ability to accurately perceive and quickly respond to line loss anomalies.

[0006] To solve the above technical problems, the present invention provides the following technical solution: a method for analyzing line loss anomalies in distribution transformer areas based on source-grid-load-storage coordination, comprising: acquiring real-time operation measurement data of the source side, grid side, load side and storage side in the distribution transformer area, and performing time synchronization and fusion processing on the multi-source heterogeneous measurement data to construct a unified feature vector characterizing the overall operation status of the distribution transformer area; Based on the unified feature vector, an improved transformer substation line loss formula is used to calculate the corrected line loss, which reflects the actual loss of the transformer substation under the synergistic effect of the source, grid, load, and storage. Based on the corrected line loss sequence, a dynamic adaptive algorithm combining historical statistical characteristics and instantaneous change trends is used for real-time analysis and judgment to identify abnormal line loss events. In response to the detection of abnormal line loss, the residuals of each dimension of the unified feature vector and the baseline state are decomposed, and multi-source contribution analysis is performed to locate the source of the abnormal line loss.

[0007] As a preferred embodiment of the method for analyzing abnormal line loss in transformer substations based on the synergy of source, grid, load and storage described in this invention, the calculation of the corrected line loss amount reflecting the actual loss of the transformer substation under the synergistic effect of source, grid, load and storage includes calculation using an improved transformer substation line loss formula based on the unified feature vector at the current moment. The improved transformer substation line loss formula introduces an energy storage correction term and a nonlinear response factor by calculating the difference between incoming line energy and load power, thereby distinguishing and adjusting the energy flow caused by the charging and discharging behavior of the energy storage system.

[0008] As a preferred embodiment of the method for analyzing abnormal line loss in transformer substations based on source-grid-load-storage coordination described in this invention, the method for identifying abnormal line loss events includes determining the statistical analysis window at the current moment based on historical line loss data and calculating the statistical characteristic quantities within the window. The deviation of the statistical characteristic quantity and the instantaneous rate of change of the line loss quantity are calculated based on the current line loss quantity, and the anomaly comprehensive score at the current moment is calculated. Based on the statistical features and the deviation direction and magnitude of the current line loss, a judgment threshold is dynamically generated. The anomaly detection is completed by comparing the comprehensive anomaly score with the judgment threshold.

[0009] As a preferred embodiment of the source-grid-load-storage coordination-based transformer area line loss anomaly analysis method described in this invention, wherein: the calculation of the anomaly comprehensive score at the current moment includes calculating a first score component that reflects the degree to which the current line loss deviates from the historical statistical distribution; Calculate the second scoring component that reflects the drastic change in the current line loss; The first and second scoring components are fused to form a preliminary comprehensive score; Based on the direction in which the current line loss deviates from the historical average level, an exponential weighted mapping is introduced to adjust the weight of the preliminary comprehensive score. A dynamic adaptive threshold function is defined to obtain the final anomaly comprehensive score.

[0010] As a preferred embodiment of the source-grid-load-storage coordination-based transformer area line loss anomaly analysis method described in this invention, the multi-source contribution analysis includes comparing the operating parameter data in the unified feature vector with the corresponding historical benchmark values ​​and calculating the normalized residual. Based on the normalized residuals corresponding to each operating parameter data, the contribution weight of each operating parameter data to the current abnormal state is determined. The system side to which the operational parameter data with the largest contribution weight belongs is the source side of the anomaly.

[0011] As a preferred embodiment of the transformer area line loss anomaly analysis method based on source-grid-load-storage coordination described in this invention, the step of time synchronization and fusion processing of multi-source heterogeneous measurement data includes establishing a unified time series benchmark and sampling frequency. A nonlinear interpolation algorithm is used to map the raw data from the source, grid, load, and storage sides from different acquisition periods onto the same time series; Using the mapped data, construct the unified feature vector corresponding to each time step.

[0012] As a preferred embodiment of the method for analyzing abnormal line losses in transformer substations based on source-grid-load-storage coordination described in this invention, the operational measurement data includes the output power data of distributed power sources on the source side, the metering data of transformer incoming and outgoing lines on the grid side, the household electricity load data on the load side, and the charging and discharging power and state of charge data on the storage side. The unified feature vector is configured to synchronously integrate the operational measurement data to characterize the collaborative operational status of the transformer substations at a unified time.

[0013] This invention provides a transformer substation line loss anomaly analysis system based on source-grid-load-storage coordination.

[0014] As a preferred embodiment of the transformer area line loss anomaly analysis system based on source-grid-load-storage coordination described in this invention, it includes: a data synchronization and fusion module, a collaborative line loss calculation module, an intelligent anomaly detection module, and an anomaly tracing and analysis module; The data synchronization and fusion module is used to acquire multi-source heterogeneous measurement data, perform time synchronization and fusion, and construct a unified feature vector. The collaborative line loss calculation module is used to calculate the corrected line loss based on the unified feature vector and using the improved transformer area line loss formula. The intelligent anomaly detection module is used to perform real-time analysis and judgment based on the corrected line loss sequence using a dynamic adaptive algorithm to identify abnormal line loss events. The anomaly source analysis module is used to respond to detected anomalies by decomposing the residual between the unified feature vector and the baseline state to perform multi-source contribution analysis and locate the source of the anomaly.

[0015] The present invention provides a computer device, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of a method for analyzing abnormal line loss in transformer areas based on source-grid-load-storage coordination.

[0016] The present invention provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, it implements the steps of a method for analyzing line loss anomalies in transformer substations based on source-grid-load-storage coordination.

[0017] The beneficial effects of this invention are as follows: This invention uses an exponential weighted interpolation method to synchronize and align measurement data from different sampling frequencies (smart meters, energy storage, photovoltaic inverters, etc.), preserving the fluctuation characteristics of the original data while improving the response speed to key nodes, and greatly improving the time consistency of the operating data of the distribution area.

[0018] This processing method can eliminate data misalignment, missing data, and mismatch problems, laying the foundation for subsequent line loss calculation and anomaly identification. It constructs a dual offset scoring function to enhance the detection capability of sudden anomalies and achieve a unified response to stable offset anomalies and abrupt anomalies. Without the need to use a machine learning model, it ensures both the robustness and sensitivity of anomaly identification. The weighting of the offset direction is adjusted using the exponential Sigmoid function. A dynamic adaptive threshold is constructed to achieve adaptive adjustment of anomaly judgment. The residual path decomposition algorithm is used to make the anomaly source interpretable and improve the diagnostic accuracy. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a schematic flowchart of a method for analyzing line loss anomalies in transformer substations based on source-grid-load-storage coordination, provided as an embodiment of the present invention. Detailed Implementation

[0021] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0022] Example 1, referring to Figure 1 This is the first embodiment of the present invention, which provides a method for analyzing abnormal line losses in transformer substations based on source-grid-load-storage coordination, including: S1: Acquire real-time operational measurement data from the source side, grid side, load side, and storage side within the distribution transformer area, and perform time synchronization and fusion processing on the multi-source heterogeneous measurement data to construct a unified feature vector characterizing the overall operational status of the transformer area.

[0023] S2: Based on a unified feature vector, an improved transformer substation line loss formula is used to calculate the corrected line loss, which reflects the actual loss of the transformer substation under the synergistic effect of the source, grid, load, and storage.

[0024] S3: Based on the corrected line loss sequence, a dynamic adaptive algorithm combining historical statistical characteristics and instantaneous change trends is used for real-time analysis and judgment to identify abnormal line loss events.

[0025] S4: In response to the detection of abnormal line loss, decompose the residuals of each dimension of the unified feature vector with the baseline state, perform multi-source contribution analysis, and locate the source of the abnormal line loss.

[0026] It should be noted that the exponential weighted interpolation method was used to connect data from smart meters, energy storage, photovoltaic inverters and other devices with different sampling frequencies, maintaining the fluctuation characteristics of the original data, enhancing the data responsiveness of key nodes and greatly improving the time consistency of the operating data of the distribution area.

[0027] The energy direction function and power-law nonlinear response factor are used to adjust the energy storage behavior, ensuring the physical rationality and dynamic accuracy of the line loss calculation, which better characterizes the real physical loss and prevents interference from energy storage actions from misleading the line loss identification results.

[0028] The offset probability weighted anomaly detection algorithm using statistical inference introduces the historical window statistical mean and standard deviation to establish a dual offset score quantification. Combined with the rate of change of the current line loss and the line loss at the previous time, it can respond to both stable offset anomalies and abrupt anomalies.

[0029] A dynamic adaptive threshold is established to achieve adaptive adjustment of anomaly judgment. The residual path decomposition algorithm is used to make the anomaly source interpretable and improve the diagnostic accuracy.

[0030] Example 2 is an embodiment of the present invention. Based on the above embodiment, a method for analyzing line loss anomalies in transformer substations based on source-grid-load-storage coordination is provided.

[0031] Furthermore, in this embodiment, step S1 acquires real-time operational measurement data from the source side, grid side, load side, and storage side within the distribution transformer area, and performs time synchronization and fusion processing on the multi-source heterogeneous measurement data to construct a unified feature vector characterizing the overall operational status of the transformer area. Specific steps include S101-S103: S101: Data collection is achieved from all sides of the distribution area through multi-source acquisition terminals, including the instantaneous output power of the source side (such as photovoltaic inverters and wind turbines), the power metering data of the high and low voltage ports of the grid-side distribution transformer, the individual household power consumption of the smart meters on the user side, and the raw measurement data such as the charging and discharging amount of the energy storage system, the battery state of charge and operating mode signals.

[0032] S102: Due to the heterogeneous sampling periods of the aforementioned devices—some, such as smart meters, have a sampling period of 15 minutes, distribution transformers and energy storage systems have a sampling period of 1 minute, and photovoltaic inverters even reach the second level—direct splicing would lead to timing misalignment and information loss. Therefore, a unified time reference axis is constructed. The sampling period is determined by using the device with the most frequent sampling as the reference clock, i.e., the time step. Arbitrary variable Represents discrete time points in a unified time coordinate system. This is the start time of the analysis period. This marks the end of the current analysis period; and the original measurement data of all devices are nonlinearly interpolated and mapped using the introduced exponential weighted interpolation method.

[0033] Specifically, for any original measurement data (such as the energy storage charging and discharging capacity), at two original sampling points and Between these points, we need to find the time interval at any given moment. The estimated value By employing an exponentially weighted interpolation method, the response to nearby values ​​is enhanced, while the response to distant values ​​is weakened. The specific formula is as follows: in, This indicates the first value obtained after interpolation. Each measurement data at the target time The variable value; Indicates the first Each measurement data point at time point The original value at that location; Indicates the first Each measurement data point at time point The original value at that location; This is the interpolation attenuation parameter, in units of The larger the value, the more the interpolation biases towards closer data points. This is determined using existing Bayesian optimization methods, with a reference range of values. ; Interpolation is smoother than linear or spline interpolation, has stronger resistance to abrupt noise, and is beneficial for recovering continuous-time signals to form a complete time-series data tensor. After interpolation, a unified feature vector is formed, and all values ​​are synchronized to the same reference time point, ensuring temporal consistency for subsequent analysis.

[0034] S103: The improved transformer area line loss formula introduces an energy storage correction term and a nonlinear response factor by calculating the difference between the incoming line energy and the load power, so as to distinguish and adjust the energy flow caused by the charging and discharging behavior of the energy storage system.

[0035] In a unified feature vector Based on this, it is necessary to monitor the distribution radio area at every moment. To accurately calculate the actual line loss, the traditional line loss is defined as the sum of the transformer's incoming power energy and the user's total power meter readings. However, this does not consider the bidirectional energy flow disturbance caused by energy storage intervention. Therefore, an energy storage correction term is introduced to construct an improved formula for the line loss of the transformer substation: in, For the current moment The corrected line loss is represented at time [time value missing]. The line loss of the transformer substation is the unmeasurable loss remaining after subtracting the user's power and energy storage from the transformer's incoming power. This represents the total incoming power of the transformer, indicating the incoming energy on both the high-voltage and low-voltage sides of the transformer. It reflects the total input power of the distribution area and is derived from a unified feature vector. ; It is the first The metered electricity consumption of each user indicates the first individual users at any time Electricity consumption is taken from a unified feature vector. ; This refers to the number of users in the Taiwan region; It is an energy storage device at all times The net energy interaction value, representing the value at time t, is... The energy flow direction and magnitude of an energy storage system: positive values ​​represent charging (absorbing energy), and negative values ​​represent discharging (releasing energy), derived from a unified feature vector. ; This refers to the energy conversion efficiency of an energy storage system, representing the efficiency of electrical energy conversion during the charging and discharging process (considering the energy losses of batteries, power electronics, and other equipment). It is defined according to the equipment specifications, i.e., obtained from system configuration parameters or equipment technical documents; sign function. Provides energy flow direction judgment, so that the energy consumption of the energy storage during charging is regarded as the increase of load in the transformer area, and the energy provided by the energy storage during discharging should be deducted from the energy of the incoming line. +1 indicates charging, -1 indicates discharging, and 0 indicates no energy exchange. It is a nonlinear power response factor that reflects the nonlinear response characteristics of energy storage devices under different power levels or charging states. It makes a power-law adjustment and correction to the energy storage output efficiency and is set based on experience.

[0036] It should be noted that this step, by establishing a unified time reference axis and using exponential weighted interpolation, effectively solves the problem of time series data misalignment and missing data caused by different sampling frequencies of multiple devices from the source, network, load, and storage sides. It achieves high-fidelity synchronous fusion of multi-source heterogeneous data, providing a reliable data foundation with consistent time series for subsequent accurate analysis.

[0037] Furthermore, in this embodiment, step S2, based on a unified feature vector, uses an improved transformer substation line loss formula to calculate the corrected line loss amount, reflecting the actual loss of the transformer substation under the synergistic effect of the source, grid, load, and storage. The specific steps include: S201: Based on the corrected line loss, an anomaly detection algorithm based on statistical inference and weighted offset probability is used to detect anomalies. When an anomaly is detected in the line loss, a residual path decomposition algorithm based on multi-source tensors is introduced to trace the source of the anomaly and determine the source of the anomaly.

[0038] After line loss calculation is completed, a probability-weighted anomaly detection algorithm based on statistical inference is constructed to identify any anomalies. The core objective of this algorithm is to detect anomalies at any given time. By comparing the deviation of the current line loss value from the line loss distribution in the historical window, and considering the changing trend of the line loss mutation rate, multiple deviation factors are introduced to form a unified scoring index.

[0039] S202: The time window determined by autocorrelation function analysis based on historical line loss data obtained from existing databases. Inside, among which The sliding time window length, i.e., the window width, is first calculated using the historical average line loss. with standard deviation : in, Indicates the index of a time point within a time window; Indicates time The amount of line loss in the transformer area.

[0040] It should be noted that the anomaly detection framework based on statistical inference is adopted, and the historical statistical characteristics (mean and standard deviation) of line loss are calculated using a sliding time window. This provides an adaptive dynamic benchmark for anomaly judgment based on the recent operating status of the transformer area, avoiding the problem of poor environmental adaptability of the fixed threshold method.

[0041] Furthermore, in this embodiment, step S3, based on the corrected line loss sequence, employs a dynamic adaptive algorithm combining historical statistical characteristics and instantaneous trends for real-time analysis and judgment to identify abnormal line loss events. Specific steps include: S301: Calculate the first scoring component, which reflects the degree to which the current line loss deviates from the historical statistical distribution; calculate the second scoring component, which reflects the degree of drastic change in the current line loss; and fuse the first and second scoring components to form a preliminary comprehensive score.

[0042] Furthermore, a dual-offset scoring quantization is constructed. Its structure is as follows: in, It is the current moment. The baseline line loss anomaly score, i.e., the quantification of the double offset score; It is to prevent zero items, take ; It was the previous moment The line loss value; It is the rate of change response factor, which is set using fuzzy control methods; It should be noted that the first scoring component is... This is used to measure the degree of deviation of the current line loss value from its historical distribution; the second scoring component is... It reflects the rate of change of line loss; the sum of the two items reflects the current deviation of line loss from the overall amount.

[0043] S302: Based on the direction in which the current line loss deviates from the historical average level, an exponential weighted mapping is introduced to adjust the weighted preliminary comprehensive score, and a dynamic adaptive threshold function is defined to obtain the final anomaly comprehensive score.

[0044] To map the aforementioned double-offset score quantization to an anomaly risk score, an exponential weighted mapping is introduced: in, Indicates the current time Anomaly intensity scoring index; The slope factor of the Sigmoid function (adjusting response sensitivity) is set empirically; the Sigmoid function in the denominator controls the sensitivity to deviations in direction. If the current value is significantly higher than the mean (such as in cases of electricity theft or metering device malfunctions), the weighting is increased; if the deviation is small or the direction is not significant, the score will be weakened.

[0045] S303: Define a dynamic adaptive threshold function: in, It is the current moment. The linear decision threshold; The basic sensitivity coefficient is used to adjust the tolerance for line loss fluctuations. It is determined through a Bayesian search optimization method, and the reference value range is [value range missing]. ; The parameter for adjusting the nonlinear enhancement term controls the pulling effect of the deviation direction term on the threshold. It is determined empirically, with a reference range of values. The judgment rule is: like If the line loss is abnormal, it is determined that there is an abnormal line loss in the transformer area at the current moment; otherwise, it is determined that there is a normal line loss in the transformer area at the current moment.

[0046] Furthermore, in this embodiment, step S4, in response to detecting an anomaly in line loss, decomposes the residuals of each dimension of the unified feature vector with the baseline state, performs multi-source contribution analysis, and locates the source of the anomaly in line loss. Specific steps include: S401: Compare the data of each running parameter in the unified feature vector with the corresponding historical benchmark values ​​and calculate the normalized residual.

[0047] Once an anomaly is identified, a residual path decomposition algorithm based on multi-source tensors is introduced to achieve refined localization and trace the responsible party for the anomaly.

[0048] In the residual path decomposition algorithm based on multi-source tensors, the unified feature vector of the state at the current sampling time is... Element-level residuals will be calculated between the reference sequence (from an existing database, the mean sequence of normal operating days for the same period over the past 30 days) and the reference sequence. The residual vector is defined as any element Represented as: in, It is the first Each residual vector element; Indicates the current time The Types of measurement data (such as load, energy storage, power source output, etc.), i.e., interpolated measurement data. This indicates the value corresponding to the same measurement data in the reference sequence.

[0049] S402: Based on the normalized residuals corresponding to each operating parameter data, determine the contribution weight of each operating parameter data to the current abnormal state, and determine the system side to which the operating parameter data with the largest contribution weight belongs as the source side of the abnormality.

[0050] To standardize the measurement, a normalized deviation weight is further constructed: in, This represents the total number of measured data; the final weighting is based on the largest deviation. Define the current source of the anomaly. If the largest deviation occurs in the load-side data, it may indicate a sudden change in large user activity or illegal electricity use. If the residual error of the energy storage system is the largest, it indicates a switching of energy storage strategies. If the abnormal output of the source side is the main cause, it indicates power fluctuations of distributed power sources.

[0051] It should be noted that the multi-source tensor residual path decomposition algorithm is used to analyze the contribution of the discovered overall line loss anomalies to four specific operating parameters: source, grid, load, and storage. The advantage of this method is that it enables refined and interpretable localization of the anomaly source, distinguishing whether the anomaly is caused by user-side electricity consumption behavior, energy storage system strategies, distributed power source fluctuations, network-side equipment failures, etc., thereby improving operation and maintenance efficiency.

[0052] Example 3 is the third embodiment of the present invention, which differs from the previous two embodiments in that: This embodiment also provides a transformer area line loss anomaly analysis system based on source-grid-load-storage coordination, including: a data synchronization and fusion module, a collaborative line loss calculation module, an intelligent anomaly detection module, and an anomaly tracing and analysis module; The data synchronization and fusion module is used to acquire multi-source heterogeneous measurement data, perform time synchronization and fusion, and construct a unified feature vector. The collaborative line loss calculation module is used to calculate the corrected line loss based on the unified feature vector and using an improved transformer area line loss formula. The intelligent anomaly detection module is used to perform real-time analysis and judgment based on the corrected line loss sequence using a dynamic adaptive algorithm to identify abnormal line loss events. The anomaly source analysis module is used to respond to detected anomalies by decomposing the residual between the unified feature vector and the baseline state to perform multi-source contribution analysis and locate the source of the anomaly.

[0053] This embodiment also provides an electronic device, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the transformer area line loss anomaly analysis method based on source-grid-load-storage coordination proposed in the above embodiment.

[0054] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements a method for analyzing line loss anomalies in transformer substations based on source-grid-load-storage coordination, as proposed in the above embodiment.

[0055] The storage medium proposed in this embodiment belongs to the same inventive concept as the method for analyzing line loss anomalies in transformer areas based on source-grid-load-storage coordination proposed in the above embodiments. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0056] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0057] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for analyzing line loss anomalies in transformer substations based on source-grid-load-storage coordination, characterized in that: include, Real-time operational measurement data from the source side, grid side, load side, and storage side within the distribution transformer area are acquired, and time synchronization and fusion processing are performed on the multi-source heterogeneous measurement data to construct a unified feature vector characterizing the overall operational status of the transformer area. Based on the unified feature vector, an improved transformer substation line loss formula is used to calculate the corrected line loss, which reflects the actual loss of the transformer substation under the synergistic effect of the source, grid, load, and storage. Based on the corrected line loss sequence, a dynamic adaptive algorithm combining historical statistical characteristics and instantaneous change trends is used for real-time analysis and judgment to identify abnormal line loss events. In response to the detection of abnormal line loss, the residuals of each dimension of the unified feature vector and the baseline state are decomposed, and multi-source contribution analysis is performed to locate the source of the abnormal line loss.

2. The method for analyzing line loss anomalies in transformer substations based on source-grid-load-storage coordination as described in claim 1, characterized in that: The calculation yields a corrected line loss that reflects the actual losses in the transformer area under the synergistic effect of the source, grid, load, and storage. This includes calculating the line loss of the transformer area using an improved formula based on the unified feature vector at the current moment; The improved transformer substation line loss formula introduces an energy storage correction term and a nonlinear response factor by calculating the difference between incoming line energy and load power, thereby distinguishing and adjusting the energy flow caused by the charging and discharging behavior of the energy storage system.

3. The method for analyzing line loss anomalies in transformer substations based on source-grid-load-storage coordination as described in claim 2, characterized in that: The identification of abnormal line loss events includes determining the statistical analysis window for the current moment based on historical line loss data and calculating the statistical characteristic quantities within the window; The deviation of the statistical characteristic quantity and the instantaneous rate of change of the line loss quantity are calculated based on the current line loss quantity, and the anomaly comprehensive score at the current moment is calculated. Based on the statistical features and the deviation direction and magnitude of the current line loss, a judgment threshold is dynamically generated. The anomaly detection is completed by comparing the comprehensive anomaly score with the judgment threshold.

4. The method for analyzing line loss anomalies in transformer substations based on source-grid-load-storage coordination as described in claim 3, characterized in that: The calculation of the current moment's anomaly comprehensive score includes calculating the first score component, which reflects the degree to which the current line loss deviates from the historical statistical distribution; Calculate the second scoring component that reflects the drastic change in the current line loss; The first and second scoring components are fused to form a preliminary comprehensive score; Based on the direction in which the current line loss deviates from the historical average level, an exponential weighted mapping is introduced to adjust the weight of the preliminary comprehensive score. A dynamic adaptive threshold function is defined to obtain the final anomaly comprehensive score.

5. The method for analyzing line loss anomalies in transformer substations based on source-grid-load-storage coordination as described in claim 4, characterized in that: The multi-source contribution analysis includes comparing the data of each operating parameter in the unified feature vector with the corresponding historical benchmark value and calculating the normalized residual. Based on the normalized residuals corresponding to each operating parameter data, the contribution weight of each operating parameter data to the current abnormal state is determined. The system side to which the operational parameter data with the largest contribution weight belongs is the source side of the anomaly.

6. The method for analyzing line loss anomalies in transformer substations based on source-grid-load-storage coordination as described in claim 5, characterized in that: The time synchronization and fusion processing of multi-source heterogeneous measurement data includes establishing a unified time series benchmark and sampling frequency; A nonlinear interpolation algorithm is used to map the raw data from the source, grid, load, and storage sides from different acquisition periods onto the same time series; Using the mapped data, construct the unified feature vector corresponding to each time step.

7. The method for analyzing line loss anomalies in transformer substations based on source-grid-load-storage coordination as described in claim 6, characterized in that: The operational measurement data includes distributed power output data on the source side, transformer incoming and outgoing line metering data on the grid side, household electricity load data on the load side, and charging and discharging power and state of charge data on the storage side. The unified feature vector is configured to synchronously integrate the operational measurement data to characterize the collaborative operational status of the transformer substations at a unified time.

8. A transformer substation line loss anomaly analysis system based on source-grid-load-storage coordination, employing the method for transformer substation line loss anomaly analysis based on source-grid-load-storage coordination as described in any one of claims 1 to 7, characterized in that, include: Data synchronization and fusion module, collaborative line loss calculation module, intelligent anomaly detection module, and anomaly source tracing and analysis module; The data synchronization and fusion module is used to acquire multi-source heterogeneous measurement data, perform time synchronization and fusion, and construct a unified feature vector. The collaborative line loss calculation module is used to calculate the corrected line loss based on the unified feature vector and using the improved transformer area line loss formula. The intelligent anomaly detection module is used to perform real-time analysis and judgment based on the corrected line loss sequence using a dynamic adaptive algorithm to identify abnormal line loss events. The anomaly source analysis module is used to respond to detected anomalies by decomposing the residual between the unified feature vector and the baseline state to perform multi-source contribution analysis and locate the source of the anomaly.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method for analyzing abnormal line losses in transformer areas based on source-grid-load-storage coordination, as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for analyzing abnormal line losses in transformer areas based on source-grid-load-storage coordination as described in any one of claims 1 to 7.

Images