A transformer area electricity stealing analysis system based on sectional line loss calculation

Abstract

The application discloses a kind of based on sectional line loss calculation's transformer area electricity larceny analysis system, belong to distribution network line loss detection and electricity larceny inspection technical field.For the technical problem that traditional distribution network line loss calculation exists data input and output strong uncertainty, dynamic and static line loss is calculated using unified solidification formula, without targeted calibration mechanism, leading to low line loss calculation precision, electricity larceny identification accuracy is poor, by configuration service subsystem, multi-node reusable energy storage unit and voltage sensor, independently divided dynamic line loss calculation module and static line loss calculation module are built-in service subsystem.The application completes dynamic line loss deviation calculation and parameter correction by the cross-node standardization power transmission task of reusable energy storage unit, and realizes the comparison calibration of static line loss by the line voltage data collected by voltage sensor in real time.Greatly improve sectional line loss calculation precision and model working condition adaptability.

Description

Technical Field

[0001] This invention belongs to the field of power system monitoring technology, specifically relating to a distribution area electricity theft analysis system based on segmented line loss calculation. Background Technology

[0002] Distribution network line loss data is the core basis for identifying electricity theft in the power grid, and accurate segmented line loss calculation is key to achieving accurate electricity theft assessment. Existing line loss calculation architectures for distribution network electricity theft detection have systemic flaws. They lack a unified data processing and calculation service platform, and the input and output of various operational data such as current and operating conditions in segmented transmission lines are uncertain, leading to unstable basic data for line loss calculation and significant deviations in calculation results. Furthermore, traditional line loss calculation models do not differentiate between dynamic and static line losses, using a uniform, fixed formula to calculate all line loss parameters. This fails to adapt to the characteristics and patterns of different loss types in the lines, and the model parameters cannot be dynamically adjusted according to the actual operating conditions of the power grid, resulting in serious computational limitations.

[0003] On the one hand, existing technologies lack a reusable node energy storage transmission calibration mechanism and have no dedicated unit to complete the standardized power transmission calibration task for segmented lines. As a loss parameter that changes with the power transmission operating conditions, dynamic line loss cannot rely on standardized and characteristic transmission scenarios to obtain accurate theoretical benchmarks, and cannot specifically correct the dynamic line loss calculation parameters. Therefore, the uncertainty problem of dynamic line loss calculation cannot be eliminated.

[0004] On the other hand, existing technologies, which target static line losses that fluctuate little over time, do not rely on transformers to configure a dedicated voltage sensing and acquisition architecture. They cannot obtain accurate voltage data of segmented lines in real time and rely solely on fixed theoretical parameters for static line loss calculation. This cannot match the slight changes in static losses caused by line equipment aging and environmental changes, and lacks a real-time calibration and correction mechanism for static line loss parameters. Summary of the Invention

[0005] In view of this, the purpose of this invention is to provide a transformer substation electricity theft analysis system based on segmented line loss calculation.

[0006] To solve the above-mentioned technical problems, the technical solution of the present invention is: a transformer substation electricity theft analysis system based on segmented line loss calculation, comprising: a service subsystem, several reusable energy storage units deployed at different transmission nodes of the distribution network, and several voltage sensors configured on segmented low-voltage distribution lines; the service subsystem is internally configured with a line loss calculation model and an electricity theft analysis model, the service subsystem is used to collect current operation data and grid operating condition data of segmented low-voltage distribution lines to generate power supply characteristic information, the line loss calculation model is used to calculate the corresponding segmented line loss based on the power supply characteristic information, and the electricity theft analysis model inputs the corresponding segmented line loss into the power supply topology network to output the corresponding electricity theft information; the line loss calculation model is internally divided into a dynamic line loss calculation module and a static line loss calculation module; the reusable energy storage units are configured with a local node power supply mode and an inter-node transmission mode, and the power supply characteristics and power consumption characteristics of each reusable energy storage unit are... The system uses pre-set known parameters. When the reusable energy storage unit operates in cross-node transmission mode, it performs segmented power transmission tasks. The service subsystem, based on the pre-set known power supply characteristics, the power consumption characteristics of the reusable energy storage unit at the receiving end, and the collected segmented current data, generates the theoretical and actual dynamic line losses of the segmented low-voltage distribution lines through a dynamic line loss calculation module. The dynamic line loss calculation module compares the theoretical and actual dynamic line losses to generate line loss deviations, and corrects the dynamic line loss calculation parameters based on the line loss deviations. The voltage sensor is used to collect the voltage data of the corresponding segmented low-voltage distribution lines in real time. The service subsystem inputs the real-time voltage data into the static line loss calculation module, which calculates the actual static line loss of the segmented low-voltage distribution lines based on the real-time voltage data. The static line loss calculation module compares the actual static line loss with the pre-set theoretical static line loss and corrects the static line loss calculation parameters based on the comparison results.

[0007] By centrally coordinating various power grid operation data through the service subsystem, the technical drawbacks of uncertain input and output data in traditional line loss calculation are effectively addressed. By separating the line loss calculation model into dynamic and static modules, dynamic line loss calibration is completed based on standardized transmission operations using reusable energy storage units, while static line loss calibration is completed based on real-time voltage data. This achieves precise correction of the differences between the two types of line loss parameters, overcoming the shortcomings of traditional unified calculation models that cannot adapt to different line loss characteristics. It also enhances the adaptive capability of the line loss model under operating conditions, laying a solid data foundation for accurate detection of electricity theft.

[0008] Furthermore, the reusable energy storage unit incorporates a built-in adaptive update module for characteristic parameters, which is communicatively connected to the service subsystem. This module periodically receives real-time operating parameters from the distribution network and iteratively updates its pre-stored power supply and consumption characteristic parameters based on these parameters, achieving dynamic self-calibration of node characteristic parameters. Compared to the traditional method of fixing characteristic parameters in energy storage units, this solution enables dynamic self-calibration and updating of the power supply and consumption characteristic parameters of the energy storage unit. It can iteratively optimize characteristic parameters in accordance with the real-time operating conditions of the distribution network, eliminating the problem of fixed parameters being out of sync with actual operating conditions. This ensures the authenticity and adaptability of the energy storage unit calibration scenario, providing a precise benchmark that closely matches real-time operating conditions for calculating dynamic line loss theoretical values ​​and comparing line loss deviations.

[0009] Furthermore, when the reusable energy storage unit performs cross-node segmented power transmission tasks, it incorporates an impedance adaptive matching module. This module collects the line impedance parameters of the current segmented low-voltage distribution lines in real time and adaptively adjusts its own transmission output impedance based on these parameters, ensuring that the transmission conditions of the transmitting and receiving energy storage units are in a standardized matching state. This added line impedance adaptive matching mechanism can sense and adapt to impedance fluctuations in the segmented low-voltage distribution lines in real time, actively adjusting the transmission output state of the energy storage unit to ensure that cross-node transmission calibration operations are always in a standardized matching state. This effectively avoids transmission condition disturbances caused by line impedance fluctuations, eliminates line loss deviations caused by non-theft operating conditions, and comprehensively guarantees the standardization and calculation accuracy of the dynamic line loss calibration process.

[0010] Furthermore, each voltage sensor incorporates a built-in clock drift self-calibration unit, which is bound to the global clock signal of the service subsystem. This unit compares the deviation between the local acquisition clock and the global clock in real time, compensates for the clock deviation, and corrects the acquisition timing to achieve synchronous acquisition of voltage data across all line segments. This solves the data deviation problem caused by clock drift and timing asynchrony in traditional multi-node voltage acquisition. By binding the voltage sensor to the global clock to achieve clock drift self-calibration, synchronous acquisition of voltage data across all line segments is completed, avoiding static line loss calculation errors caused by timing misalignment. This ensures the uniformity and comparability of static line loss acquisition data for each line segment, providing accurate raw data support for static line loss parameter correction.

[0011] Furthermore, the data preprocessing module of the service subsystem incorporates a timing misalignment correction algorithm. This module calculates timing deviations for current, voltage, and operating condition data from different acquisition cycles, and compensates for and aligns the misalignments of multi-source heterogeneous data based on these deviations, outputting standardized power supply characteristic information with fully coupled timing. Addressing the timing misalignment and inconsistent dimensionality issues in multi-source heterogeneous data of the distribution network, a timing misalignment correction algorithm is added to compensate and align various types of operating data. This effectively solves the data matching deviation problem caused by different acquisition devices and acquisition cycles, eliminates computational interference caused by differences in data dimensionality, outputs highly coupled standardized data, and improves the computational accuracy and operational reliability of the line loss calculation model from the underlying data dimension.

[0012] Furthermore, the dynamic line loss calculation module incorporates a deviation source tracing and correction mechanism. This mechanism correlates the load fluctuation, transmission distance, and power transmission power condition factors corresponding to the line loss deviation. Based on the weight ratio of different condition factors, it specifically matches the corresponding dynamic parameter correction coefficients to achieve targeted correction of the line loss deviation. This abandons the traditional, generalized overall correction mode, accurately locating the core causes of the deviation through the line loss deviation source tracing mechanism, and completing targeted parameter correction by combining the weight ratio of various power grid condition factors. This effectively avoids the problem of decreased model adaptability caused by blind correction, allowing the dynamic line loss model parameter adjustment to accurately match the actual causes of line losses, significantly improving the targeting and scientific rationality of dynamic line loss correction.

[0013] Furthermore, the static line loss calculation module is integrated with an environmental parameter compensation mechanism. This module is used to access ambient temperature and humidity around the line, as well as equipment aging time-series parameters, coupling these environmental and equipment aging parameters to the theoretical static line loss calculation formula. This overcomes the shortcomings of traditional static line loss theoretical parameters being fixed and not adapted to the effects of environment and equipment aging, by coupling environmental and equipment aging time-series parameters to the static line loss theoretical calculation formula. This enables dynamic compensation and updating of the theoretical static line loss parameters, ensuring that the theoretical calculation benchmark closely matches the actual loss characteristics of the line, effectively reducing the deviation between the theoretical and actual static line loss values, and improving the accuracy of static line loss calibration.

[0014] Furthermore, the line loss calculation model incorporates a bidirectional linkage calibration unit for dynamic and static line loss. This unit uses the corrected static line loss parameters as the baseline threshold for dynamic line loss calculation, while simultaneously using the long-term mean deviation of dynamic line loss to iteratively correct the static line loss benchmark value. This breaks the limitations of traditional independent and fragmented correction of dynamic and static line loss, constructing a bidirectional linkage closed-loop calibration mechanism. By using the corrected static line loss parameters as the baseline for dynamic line loss calculation, and iteratively optimizing the static line loss benchmark value based on the long-term mean deviation of dynamic line loss, the model achieves collaborative iterative optimization of the two types of line loss parameters, effectively improving the overall adaptability and global calculation accuracy of the line loss model.

[0015] Furthermore, the service subsystem incorporates a multi-path cross-calibration mechanism. This mechanism controls energy storage units within the same topology calibration group to switch between different transmission paths and transmission power to complete multiple rounds of calibration. It cross-verifies dynamic line loss deviations under multiple paths, eliminating random errors from single transmission calibrations. This multi-path cross-calibration mechanism overcomes the limitations of traditional single transmission calibration, which relies on a single operating condition and high randomness. By switching between multiple transmission paths and transmission power to complete multiple rounds of dynamic line loss calibration, and cross-verifying line loss deviations under multiple operating conditions, it effectively eliminates random errors caused by fluctuations in single transmission conditions and external random interference, improving the stability and reliability of dynamic line loss correction results and strengthening the model's anti-interference capability. Furthermore, the data preprocessing module incorporates an anomaly type identification and differentiated repair mechanism. This mechanism distinguishes between data anomalies caused by equipment malfunctions and data distortions caused by electricity theft, employing differentiated processing methods such as data gap masking and anomaly feature marking for each type of data. This addresses the shortcomings of traditional methods that uniformly remove data anomalies without distinguishing their causes, enabling accurate identification of data anomalies caused by equipment malfunctions and data distortions caused by electricity theft. The differentiated processing strategy for the two types of data effectively masks the interference of fault data on model correction while fully preserving the characteristics of electricity theft anomaly data, effectively avoiding model miscorrection and misjudgment of electricity theft, thus balancing model calibration accuracy and electricity theft identification accuracy.

[0016] The main technical advantages of this invention are reflected in the following aspects: This solution addresses the industry pain points of high uncertainty in traditional distribution network line loss calculation data, lack of dynamic and static line loss correction mechanisms, weak adaptability of fixed calculation models, and insufficient accuracy in electricity theft detection. It establishes a dynamic line loss correction system based on energy storage unit calibration and a static line loss correction system based on voltage sensing, achieving unified processing of the entire power grid data through a service subsystem. This solution adopts a hierarchical, closed-loop, and interconnected line loss correction logic, breaking through the technical limitations of traditional fixed calculation formulas. It effectively avoids calculation errors caused by power grid operating condition fluctuations, equipment aging, environmental interference, and data deviations, significantly improving the accuracy and operational stability of segmented line loss calculations. It can accurately identify abnormal line loss deviations, providing reliable data support for accurate judgment of electricity theft behavior in distribution networks, and is adaptable to the complex and ever-changing actual operating conditions of distribution networks. Attached Figure Description

[0017] Figure 1 : System architecture diagram of this invention. Detailed Implementation

[0018] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings, so that the technical solution of the present invention can be more easily understood and mastered.

[0019] This invention is a transformer substation electricity theft analysis system based on segmented line loss calculation. It is applied to the monitoring of line loss status and identification of electricity theft behavior in various segments of low-voltage distribution lines in urban and rural power distribution networks. The entire system consists of a service subsystem, several reusable energy storage units deployed at different transmission nodes in the power distribution network, and several voltage sensors configured on the corresponding segments of low-voltage distribution lines. The entire system relies on a unified data acquisition, calculation, and scheduling logic to complete the entire process of acquiring basic data of segmented lines, calculating layered line losses, comparing line loss deviations, iteratively correcting the calculation formula, and analyzing the correlation of abnormal line losses.

[0020] Reference Figure 1As shown, the service subsystem is an integrated data acquisition, processing, and scheduling carrier deployed in the back-end management and control area of ​​the distribution network. This carrier is built on an industrial server cluster and is responsible for coordinating the reception, aggregation, processing, and distribution of various operating data of low-voltage distribution lines in various segments throughout the network, as well as the issuance of scheduling instructions for various front-end equipment. The service subsystem is equipped with a complete line loss calculation model. The line loss calculation model is divided into two independent modules, dynamic line loss calculation module and static line loss calculation module, according to the inherent characteristics of distribution network line losses. The two modules respectively carry out special calculations and formula corrections for the variable loss and fixed loss generated during the operation of the segmented low-voltage distribution lines. Reusable energy storage units are not a commonly known equipment name in the distribution network field. Reusable energy storage units are integrated energy storage and controllable transmission devices deployed in batches at each transmission node of the distribution network. These devices have two stable and switchable operating modes. The first operating mode is the local power supply mode, which supplies power to various electrical devices at this node. The second operating mode is the cross-node transmission mode, which transmits power to other transmission nodes across the current node. Each reusable energy storage unit is pre-set with unique power supply and power consumption characteristics during the factory deployment stage. Power supply characteristics refer to the power output characteristics, current output characteristics, and voltage output characteristics exhibited by the energy storage unit during the continuous output of power. Power consumption characteristics refer to the power acceptance characteristics, current acceptance characteristics, and voltage acceptance characteristics exhibited by the energy storage unit during the continuous reception of external power. These two types of characteristic parameters serve as the benchmark for theoretical line loss calculations and participate in the system's calculation process over a long period of time. The voltage sensors used in this invention are specifically configured for segmented low-voltage power distribution lines. Each voltage sensor is installed at the beginning and end of the segmented low-voltage power distribution line. During operation, the device continuously senses the grid voltage status at the corresponding installation location and generates raw voltage data, providing a basic data source for calculating static line losses. After the entire system is put into normal operation, the service subsystem will continuously collect segmented current operation data and overall power grid operating condition data of each segment of low-voltage distribution lines in the entire network according to a fixed sampling period. All the raw data collected will be uniformly integrated into the line loss calculation model as the basic data source for the whole process calculation. When the system starts the dynamic line loss calibration process according to the preset operation strategy, the service subsystem will generate the corresponding scheduling instructions and send them to the reusable energy storage unit at the designated location, controlling the energy storage unit to switch from the local node power supply mode to the cross-node transmission mode. The reusable energy storage unit in the cross-node transmission mode will strictly execute the segmented power supply transmission task preset by the system. The service subsystem will simultaneously retrieve the power supply characteristics of the transmitting end reusable energy storage unit and the power consumption characteristics of the receiving end reusable energy storage unit pre-stored in the system. Combined with the segmented current data that has been collected in real time, the three types of information will be sent to the dynamic line loss calculation module to carry out the calculation work.

[0021] The dynamic line loss calculation module internally incorporates theoretical dynamic line loss calculation logic and actual dynamic line loss calculation logic. The formula for calculating theoretical dynamic line loss is as follows: in, This represents the theoretical dynamic line loss corresponding to the current segment of the low-voltage distribution line. This value is a baseline variable loss value calculated based on standardized transmission operating conditions. This represents the real-time output power of the reusable energy storage unit at the transmitting end. This parameter is directly extracted from the preset power supply characteristics of the reusable energy storage unit at the transmitting end. This represents the real-time received power of the reusable energy storage unit at the receiving end. This parameter is directly extracted from the preset power consumption characteristics of the reusable energy storage unit at the receiving end. The function represents the real-time acquisition of segmented low-voltage distribution line current operation data by the service subsystem. This is a function for calculating the dynamic losses of AC lines in a power distribution network. It is built upon the relationship between power and current within the AC transmission system and can accurately calculate the theoretical variable losses under standardized operating conditions based on three types of input parameters. The formula for calculating the actual dynamic line loss is as follows: The specific meanings of each parameter in the formula are as follows: This represents the actual dynamic line loss corresponding to the current segment of the low-voltage distribution line. This value is the variable loss value generated under the actual operating conditions of the line. This represents the average line voltage throughout the entire operation of a segmented low-voltage distribution line. The function represents the inherent line impedance of the segmented low-voltage distribution line itself. This is a pre-installed function for calculating the actual dynamic loss within the line loss calculation model. This function calculates the current actual variable loss of the line based on the line's basic electrical parameters in real time. After obtaining the theoretical dynamic line loss and the actual dynamic line loss, the dynamic line loss calculation module further calculates the difference between the two values ​​to obtain the dynamic line loss deviation. The formula for calculating the dynamic line loss deviation is: Formula parameters This parameter represents the dynamic line loss deviation corresponding to the segmented low-voltage distribution line. It is obtained by calculating the difference between the theoretical dynamic line loss and the actual dynamic line loss and then taking the absolute value. It can intuitively reflect the calculation error in the current dynamic line loss calculation formula. After generating the dynamic line loss deviation, the dynamic line loss calculation module will adjust the overall parameters and correct the logic of the dynamic line loss calculation formula built into the module according to the deviation value, and complete the iterative optimization of the dynamic loss calculation logic.

[0022] During the synchronous operation of the dynamic line loss calibration process, voltage sensors configured on each segment of the low-voltage distribution line continuously collect real-time voltage data at the beginning and end of the corresponding line. All raw voltage data collected by the sensors is transmitted to the service subsystem, which then forwards it to the static line loss calculation module. The static line loss calculation module uses the received real-time voltage data to calculate the actual static line loss of the segmented low-voltage distribution line. The system has a pre-stored theoretical static line loss benchmark value for each segmented low-voltage distribution line. The static line loss calculation module compares the calculated actual static line loss with the system's preset theoretical static line loss item by item. Based on the differences obtained after the comparison, it performs parameter correction and logic optimization on the static line loss calculation formula built into the module. The theoretical static line loss calculation formula is as follows: in, This represents the theoretical static line loss corresponding to the current segment of the low-voltage distribution line. This value is the initial baseline value for the fixed loss of the line. This represents the rated reference voltage corresponding to the transformers at both ends of the segmented line. The function represents the initial operating state parameters of segmented low-voltage power distribution lines and their associated transformer equipment. The system uses a pre-defined fixed loss baseline calculation function to calculate the fixed loss value under standard operating conditions based on the equipment's rated parameters. The actual static line loss calculation formula is as follows: in, This represents the actual static line loss corresponding to the current segment of the low-voltage distribution line. This value is the fixed loss generated under the current actual operating conditions of the line and transformer. This represents the real-time voltage data collected at the starting point of a segmented low-voltage distribution line. The function represents the real-time voltage data collected at the termination end of a segmented low-voltage distribution line. This module, a static line loss calculation module, incorporates actual fixed loss calculation functions. It calculates the fixed losses generated by the line itself and the transformer equipment based on the real-time voltage status at both ends of the line. Dividing the line loss calculation model into dynamic and static modules allows for precise differentiation between two different types of line losses. Calculations and corrections are performed separately for variable and fixed losses. Standardized transmission operations using reusable energy storage units establish a stable and reliable theoretical benchmark for dynamic line losses. Real-time voltage data collected by voltage sensors accurately reflects the real-time changes in fixed line losses. The two correction processes work together to continuously optimize the overall computational capability of the line loss calculation model from different loss dimensions. Simultaneously, the service subsystem's unified operation mode for full data acquisition and command scheduling effectively reduces uncertainties in the original data input and output stages, ensuring the stability and continuity of the data source during system operation. This lays a solid foundation for subsequent line loss calculation and electricity theft investigation.

[0023] The reusable energy storage unit is equipped with an adaptive update module for characteristic parameters. This module is an embedded computing and processing unit integrated on the control motherboard inside the reusable energy storage unit. This module is not a known functional module in the field of distribution network energy storage equipment. It is a dedicated functional structure added by this invention to achieve dynamic parameter calibration. The adaptive update module maintains a real-time communication connection with the service subsystem throughout the process. The core function of the module is to periodically receive real-time operating parameters of the distribution network issued by the service subsystem, and to iteratively update the power supply characteristic parameters and power consumption characteristic parameters pre-stored locally inside the reusable energy storage unit based on the externally accessed real-time operating parameters, so as to achieve dynamic self-calibration of the power interaction characteristic parameters of the transmission node. The complete operation steps of this module are executed sequentially according to time. In the first step, when the reusable energy storage unit is in any of the following operating states—normal standby, local power supply, or cross-node power transmission—the feature parameter adaptive update module will proactively initiate a communication request at a fixed time period preset by the system to establish a stable communication link with the service subsystem. In the second step, after receiving the communication request, the service subsystem summarizes the current overall network operating status and the real-time operating parameters of the distribution network around the corresponding transmission node, and pushes the integrated operating data to the feature parameter adaptive update module. In the third step, the feature parameter adaptive update module retrieves the historical power supply feature parameters and historical power consumption feature parameters currently stored locally by the energy storage unit, and performs parameter iterative calculations in combination with the newly received real-time operating parameters. In the fourth step, after all parameter iterative calculations are completed, the module will solidify and store the new power supply feature parameters and power consumption feature parameters in the local non-volatile storage area, directly replacing the original historical parameters in the subsequent full-process line loss benchmark calculation.

[0024] The feature parameter adaptive update module embeds a dedicated parameter iteration calculation formula, the specific expression of which is: The specific meanings of each parameter in the formula are as follows: This represents the node feature parameters after this iteration update. These parameters encompass all sub-parameters corresponding to power supply and power consumption characteristics. This represents the historical feature parameters stored locally in the module before the current update action was executed. This represents the system's preset parameter update adjustment coefficient. This coefficient is configured in advance based on the drastic changes in the overall operating conditions of the distribution network, and can control the magnitude of parameter iteration. This represents the change in real-time operating parameters of the distribution network issued by the service subsystem. This parameter directly reflects the magnitude of change in the current operating status of the power grid surrounding the transmission node compared to the previous update cycle. The feature parameter adaptive update module, based on this calculation logic, iteratively calculates each of the power, voltage, and current parameters included in the power supply characteristics and the power, voltage, and current parameters included in the power consumption characteristics. This ensures that all feature parameters of the energy storage unit related to power interaction change synchronously with the real-time operating conditions of the power grid. Traditional energy storage devices have their operating characteristic parameters fixed at the factory, and these parameters do not change during long-term operation, making them unsuitable for adapting to the evolving operating conditions of the distribution network over time. The feature parameter adaptive update module added in this invention, through periodic parameter iterative updates, ensures that the baseline characteristics of the reusable energy storage unit always match the actual operating state of the transmission node. This completely avoids the problem of deviation between the theoretical line loss baseline and the actual operating conditions of the line caused by fixed parameters, thereby ensuring the validity and authenticity of the baseline data in the dynamic line loss calibration process.

[0025] The reusable energy storage unit is equipped with an impedance adaptive matching module during the segmented power transmission task. The impedance adaptive matching module is a control unit integrated inside the power output circuit of the reusable energy storage unit. This module is not a standard structure of conventional power transmission equipment in the distribution network. It is a dedicated functional structure specially designed for the standardization of power transmission conditions in this invention. The core function of the impedance adaptive matching module is to collect the line impedance parameters of the current segmented low-voltage distribution line in real time, and autonomously adjust the power transmission output impedance of the reusable energy storage unit itself based on the collected impedance data. This ensures that the power transmission circuit of the transmitting energy storage unit, the intermediate segmented low-voltage distribution line, and the receiving circuit of the receiving energy storage unit form a standardized impedance matching state, ensuring that the cross-node power transmission operation is always in ideal operating conditions. The complete working steps of this module are executed sequentially according to the power transmission operation process. First, after the reusable energy storage unit receives the cross-node power transmission dispatch instruction issued by the service subsystem and officially starts the segmented power transmission task, the impedance adaptive matching module immediately activates its internal impedance detection circuit and starts the real-time line impedance detection function. Second, the impedance detection circuit inside the module continuously collects the real-time line impedance parameters of the segmented low-voltage distribution lines along the complete transmission loop, while simultaneously reading the fixed input impedance parameters of the reusable energy storage unit at the receiving end. Third, the module performs joint calculations on the collected line impedance parameters and the input impedance parameters of the energy storage unit at the receiving end to obtain the optimal output impedance target value that the reusable energy storage unit needs to achieve under the current power transmission conditions. Fourth, the module drives the internally integrated impedance adjustment device to gradually adjust its own power transmission output impedance until the output impedance value is consistent with the calculated target value. Fifth, throughout the entire cycle of the cross-node power transmission operation, the module continuously and cyclically executes the entire set of actions of line impedance acquisition, joint calculation, and impedance adjustment, dynamically adapting to the real-time fluctuations in the impedance of the segmented low-voltage distribution lines.

[0026] The impedance adaptive matching module embeds a line impedance matching calculation formula, the specific expression of which is: in, This represents the final power output impedance formed after the reusable energy storage unit has been adjusted. This represents the impedance parameters of the segmented low-voltage distribution line acquired in real time by the impedance adaptive matching module. This represents the factory-set fixed input impedance parameters of the reusable energy storage unit at the receiving end. represents the line impedance matching correction coefficient, which is pre-configured in advance in combination with the transmission characteristics of the AC low-voltage distribution lines in the distribution network. The impedance adjustment device installed inside the module adopts a continuous stepless adjustment structure, which can achieve smooth changes in the output impedance. During the adjustment process, there will be no instantaneous mutations in voltage and current, ensuring the stability of the operation state of the power transmission loop. The impedance of the sectionalized low-voltage distribution lines will fluctuate in real time due to the combined influence of various external factors such as environmental temperature, line aging, and temporary load access. After the impedance mismatch problem occurs inside the power transmission loop, a large amount of additional useless loss will be generated. Such additional losses will directly interfere with the line loss detection and calculation results. Through the operation mode of dynamic impedance matching, the impedance adaptive matching module can maintain the power transmission loop in the standard operating condition for a long time, effectively eliminating the interference of additional losses caused by the impedance mismatch problem, and enabling the calculation and correction of dynamic line loss to be carried out only for the inherent losses of the sectionalized low-voltage distribution lines themselves, greatly improving the purity and calculation accuracy of the dynamic line loss calibration results.

[0027] The voltage sensor is equipped with a clock drift self-calibration unit inside. The clock drift self-calibration unit is a timing control unit integrated on the internal control circuit of each voltage sensor. This unit does not belong to the standard configuration of conventional voltage acquisition devices. It is a dedicated functional unit specially added by the present invention to achieve synchronous acquisition of voltage data across the network. The clock drift self-calibration unit remains continuously bound to the global clock signal output by the service subsystem in real time. The core operating function of the unit is to compare the deviation between the local acquisition clock of the voltage sensor and the global clock issued by the service subsystem in real time, and complete the compensation and correction action of the local acquisition timing based on the calculated clock deviation, ultimately achieving synchronous acquisition of voltage sensor data for all sectionalized lines across the network. The complete working steps of this unit are executed in sequence according to the device operation timing. In the first step, after all voltage sensors are powered on and start up and enter the normal operating state, the internal clock drift self-calibration unit will continuously receive the global clock pulse signal broadcast by the service subsystem to establish a global clock synchronization link. In the second step, the timing operation circuit inside the unit compares the local clock counting result of the sensor with the global clock counting result every fixed time interval to accurately calculate the clock drift deviation between the two types of clocks. In the third step, the unit generates a corresponding timing compensation instruction based on the solved clock drift deviation and adjusts the acquisition trigger moment of the local voltage data acquisition circuit through the instruction. In the fourth step, during the whole process of long-term operation of the voltage sensor, the unit will continuously and cyclically execute the whole set of actions of clock comparison, deviation calculation, and timing compensation to continuously suppress the timing offset problem caused by the natural drift of the hardware clock.

[0028] The clock drift self-calibration unit embeds clock deviation calculation and timing compensation operation formulas, and the two sets of formulas are as follows: The parameters in the first set of formulas This represents the clock drift deviation between the local clock and the global clock of the voltage sensor. This represents the timing value output by the local hardware clock of the voltage sensor. This represents the global clock timing value that is uniformly generated and broadcast by the service subsystem. The second set of formula parameters... This represents the actual data acquisition trigger time used by the voltage sensor after timing compensation adjustment. The acquisition circuit inside the voltage sensor will strictly start the voltage data acquisition action according to this time. The global clock generated by the service subsystem serves as the unified timing reference for all sensing devices in the network. It is broadcast synchronously to all voltage sensors to ensure that all sensing devices receive completely consistent timing reference signals. During long-term power-on operation, the clock circuits inside electronic hardware devices will inevitably experience natural drift. The clock drift amplitude of voltage sensors at different locations varies significantly, which will directly cause misalignment in the timing of voltage data acquisition at both ends of segmented low-voltage distribution lines. The timing misalignment problem will further lead to significant deviations in the static line loss calculation results. The clock drift self-calibration unit, through continuous clock self-calibration and timing compensation, enables all voltage sensors in the network to maintain synchronized acquisition timing over a long period of time, ensuring the consistency of the original voltage data on which the static line loss calculation is based in the time dimension, and solidifying the foundation of static line loss correction work from the data acquisition level.

[0029] The service subsystem is equipped with a data preprocessing module and a timing misalignment correction algorithm. The data preprocessing module is a dedicated computing unit within the service subsystem responsible for integrating, cleaning, and standardizing raw data. It is a core intermediate processing link in the data transmission chain of the entire system. The timing misalignment correction algorithm is a dedicated computing logic fully embedded within the data preprocessing module. It is mainly used to solve the problem of inconsistent timing of multi-source heterogeneous data acquisition in the distribution network. The data preprocessing module and the timing misalignment correction algorithm work together to complete the preprocessing of segmented low-voltage distribution line current data, voltage data, and power grid operating condition data. The complete workflow of this part of the structure is executed sequentially according to the data flow order. The first step is that the service subsystem uniformly summarizes the raw current operation data, raw voltage data, and raw power grid condition data uploaded by various acquisition devices, energy storage units, and sensors across the entire network. All summarized data is then transmitted in batches to the data preprocessing module. The second step is that the data preprocessing module first distinguishes the acquisition clock and sampling period corresponding to different data sources, and calculates the timing misalignment between different types of data in each group. The third step is that the module calls the internally embedded timing misalignment correction algorithm to perform time axis translation and alignment processing on the raw data with timing deviations. At the same time, it performs dimension unification operation and invalid information removal operation on the raw data packets. The fourth step is that after all data processing actions are completed, the module generates standardized power supply characteristic information with unified format and complete timing coupling, and stably transmits this type of standardized data to the line loss calculation module to participate in the subsequent line loss calculation work of the entire process.

[0030] The specific expression for the timing misalignment correction algorithm is as follows: in, After the timing alignment process is completed, at the system standard time... Standardized data generated below This represents the raw, unprocessed data uploaded by various devices. This represents the time-series misalignment calculated between two different data sources. The data preprocessing module shifts the time axis of the original data based on this misalignment, ultimately achieving time-series alignment of the multi-source data. When performing invalid field removal, the module intelligently identifies and removes redundant identifier fields and blank fields unrelated to line loss calculation within the original data packet. When performing dimension unification, it converts the original data with different precisions and units output by different acquisition devices into a unified data format and precision specified by the system. Different types of data acquisition devices within the distribution network have different sampling periods and communication transmission delays. These hardware differences directly cause time-series chaos and inconsistent data dimensions in multi-source acquisition data. Directly using unprocessed raw data for line loss calculation introduces a large number of additional calculation errors. The data preprocessing module, combined with the time-series misalignment correction algorithm, can complete data purification and time-series alignment from the data source, comprehensively eliminating various interference factors at the data level and effectively improving the reliability and stability of the overall calculation results of the line loss calculation model.

[0031] The dynamic line loss calculation module incorporates a deviation source tracing and correction mechanism. This mechanism is a complete logic system for operating condition factor correlation analysis and parameter correction within the dynamic line loss calculation module. This logic system is not a conventional correction method in the field of distribution network line loss calculation. It is a proprietary operating logic specifically designed by this invention to achieve accurate correction. The core operating logic of the deviation source tracing and correction mechanism is to establish a one-to-one correlation between the dynamic line loss deviation obtained by the dynamic line loss calculation module and various power grid operating condition factors such as load fluctuations of segmented low-voltage distribution lines, transmission distance, and power transmission capacity. It quantifies the influence weight of different operating condition factors on line loss deviation, and then matches the corresponding parameter correction coefficients according to the weight ratio of various operating condition factors, ultimately completing the targeted correction of the dynamic line loss calculation formula. The complete execution steps of the deviation source tracing and correction mechanism are executed sequentially according to the operational logic. First, after the dynamic line loss calculation module obtains the dynamic line loss deviation, it immediately retrieves all operating condition factor data corresponding to the current segment of the low-voltage distribution line. These operating condition factors specifically include parameters that directly affect the line's variable losses, such as the line's real-time load status, transmission distance, and real-time transmission power. Second, the weight calculation unit within the deviation source tracing and correction mechanism calculates the sub-item line loss deviation generated by each operating condition factor acting alone, while simultaneously determining the contribution weight of each operating condition factor to the overall dynamic line loss deviation. Third, the module sorts the parameters according to their weight values ​​from high to low, accurately identifying the main causes of the current dynamic line loss deviation. Fourth, based on the identified main causes, the module matches the corresponding exclusive correction coefficient from the system's preset coefficient library. Fifth, the module uses the matched correction coefficient to adjust the core operational parameters within the dynamic line loss calculation formula item by item, ultimately completing the targeted correction of the dynamic line loss calculation formula.

[0032] The deviation source correction mechanism internally includes the calculation formulas for the weight of operating condition factors and the calculation formulas for the correction of line loss parameters, which are as follows:

[0033] First set of formula parameters Representing the The influence weights corresponding to the power grid operating condition factors, Representing the The sub-item dynamic line loss deviation generated under the individual action of each operating condition factor. This represents the total number of operating condition factors currently involved in the weighting calculation. The second set of formula parameters. This represents the calculation parameters within the dynamic line loss calculation formula after the correction is completed. This represents the original calculation parameters within the formula before the correction action is executed. Representing the Each operating condition factor has its own specific correction and adjustment coefficient. The operating condition factors selected for the entire mechanism are all physical quantities that can directly change the magnitude of variable losses during the operation of segmented low-voltage distribution lines in the distribution network. The weighted calculation results can intuitively distinguish the degree of influence of different factors on line loss deviation. Traditional line loss correction methods mostly adopt a unified overall correction operation mode, which cannot distinguish the specific causes of line loss deviation, often resulting in over-correction or under-correction. The deviation source tracing correction mechanism can accurately locate the true source of dynamic line loss deviation and carry out targeted parameter adjustments, ensuring that the correction process of the dynamic line loss model fully conforms to the actual loss patterns of segmented low-voltage distribution lines, effectively improving the scientific nature and practical effectiveness of the dynamic line loss calculation formula correction operation.

[0034] The static line loss calculation module is bound to an environmental parameter compensation mechanism. This mechanism is a parameter compensation logic within the static line loss calculation module that integrates external environmental parameters with equipment aging status parameters. This logic is an optimized operating logic specific to this invention. The core function of the environmental parameter compensation mechanism is to continuously access the ambient temperature and humidity parameters around the segmented low-voltage distribution lines, as well as the aging time sequence parameters of the lines and transformer supporting equipment. It couples these two types of external parameters into the original theoretical static line loss calculation formula, thereby achieving dynamic compensation and real-time updating of the theoretical static line loss benchmark value. The complete operation of this mechanism follows the static line loss calculation process. First, before each static line loss comparison, the static line loss calculation module actively reads the ambient temperature and humidity data collected by external environmental sensors, and simultaneously retrieves the long-term aging time-series data of the line and transformer equipment recorded in the system's backend. Second, the environmental parameter compensation mechanism converts the collected environmental parameters and equipment aging parameters into corresponding quantitative compensation coefficients. Third, the module substitutes the calculated compensation coefficients into the original theoretical static line loss calculation formula to perform compensation calculations on the original theoretical static line loss benchmark value, generating a new theoretical static line loss value. Fourth, the module uses the compensated and updated theoretical static line loss value to perform item-by-item comparison with the calculated actual static line loss. Fifth, the module combines the comparison results of the two types of values ​​to complete the parameter correction and logic optimization of the static line loss calculation formula.

[0035] The specific expression for the compensation calculation formula corresponding to the environmental parameter compensation mechanism is as follows: The specific meanings of each parameter in the formula are as follows: This represents a completely new theoretical static line loss generated after compensation for environmental parameters and equipment aging parameters. This represents the original, preset theoretical static line loss baseline value of the system before the compensation operation was performed. This represents the compensation coefficient corresponding to ambient temperature and humidity. This represents the comprehensive real-time environmental temperature and humidity parameters around the railway line. This represents the compensation coefficient corresponding to the aging state of the equipment. This represents a comprehensive set of parameters related to the aging process of power lines and transformers. External environmental factors, such as temperature and humidity, directly affect the resistivity of low-voltage power distribution line conductors. Changes in resistivity lead to corresponding changes in the line's fixed losses. Furthermore, the aging process caused by long-term equipment operation gradually alters transformer core losses and line insulation losses. Both of these parameters continuously influence the actual value of static line losses. Traditional technical solutions maintain a fixed theoretical benchmark value for static line losses once set, completely neglecting changes in losses due to environmental variations and equipment aging. This results in a continuous deviation between the theoretical benchmark value and the actual fixed losses of the line. The environmental parameter compensation mechanism dynamically updates the theoretical static line loss benchmark value, ensuring that the benchmark value always closely matches the actual loss state during long-term line operation. This effectively reduces the inherent gap between theoretical and actual static line losses, comprehensively improving the calculation accuracy of the entire static line loss correction process.

[0036] The line loss calculation model incorporates a dynamic-static line loss bidirectional linkage calibration unit. This unit is the core logical unit within the model, enabling data exchange and collaborative optimization between the dynamic and static line loss modules. This unit is not a known functional structure in existing distribution network line loss calculation technologies. It completely breaks the original state of independent operation between the two types of line loss calculation modules. On the one hand, it sets the fully corrected static line loss parameters as the baseline threshold used in the dynamic line loss calculation process. On the other hand, it collects the average deviation generated during the long-term operation of dynamic line loss according to a fixed statistical period, and uses this average to iteratively update the benchmark value of static line loss. Ultimately, it forms a complete calibration system in which the dynamic and static line loss parameters interact and iterate in a closed loop. The complete two-way linkage workflow of this unit is executed sequentially according to the data interaction order. In the first step, the dynamic and static line loss two-way linkage calibration unit continuously reads the set of static line loss parameters output by the static line loss calculation module after completing all correction work. This set of parameters is uniformly set as the baseline threshold used by the dynamic line loss calculation module when it performs calculations. In subsequent calculations, the dynamic line loss calculation module will use this baseline threshold as the basis for fixed line loss to complete the calculation of variable loss and fixed loss. In the second step, the unit summarizes all dynamic line loss deviation data generated in multiple consecutive time periods according to the system's preset statistical period and calculates the average value of dynamic line loss deviation over a long period. In the third step, the unit uses the calculated average value of dynamic line loss deviation as a correction amount and inputs it back into the static line loss calculation module to iteratively update the theoretical benchmark value corresponding to the static line loss. In the fourth step, under the scheduling and control of the two-way linkage calibration unit, the dynamic line loss calculation module and the static line loss calculation module continuously and cyclically execute the entire process of data interaction, parameter iteration, and formula correction to achieve overall collaborative optimization of the line loss calculation model.

[0037] The dynamic and static line loss bidirectional linkage calibration unit internally includes baseline threshold calculation formulas and static benchmark iteration formulas, as follows: First set of formula parameters This represents the baseline threshold used uniformly in the dynamic line loss calculation process. This represents the average actual static line loss of segmented low-voltage distribution lines over a long period. This value objectively reflects the true level of fixed line losses. (Second set of formula parameters) This represents a completely new theoretical static line loss after the completion of iterative updates. This represents the original theoretical static line loss used by the system before the execution of this iteration. This represents the average value of dynamic line loss deviation over a long period. The baseline threshold used in the dynamic line loss calculation process primarily serves to delineate the boundary between fixed and variable line losses. The average deviation generated by long-term dynamic line loss operation can indirectly reflect the implicit deviation of the static line loss benchmark value. In traditional technical solutions, the calculation and correction of dynamic and static line losses are completely independent, and the optimization results of the two modules cannot be shared. The bidirectional linkage calibration unit for dynamic and static line losses constructs a closed-loop collaborative optimization system, allowing the optimization results of the static line loss module to effectively support the calculation of the dynamic line loss module. In turn, the long-term operating data of the dynamic line loss module optimizes the benchmark value of the static line loss module. The two technical methods work together to form a synergistic gain effect, comprehensively improving the overall operating condition adaptability and comprehensive calculation accuracy of the line loss calculation model.

[0038] The service subsystem incorporates a multi-path cross-calibration mechanism. This mechanism is a polling and cross-verification logic specifically designed for multiple reusable energy storage units within the same topological region. This scheduling strategy is an operating mechanism exclusively designed for this invention. Based on the segmented topology of the distribution network, the multi-path cross-calibration mechanism uniformly divides reusable energy storage units that are spatially adjacent and interconnected by low-voltage distribution lines into the same calibration group. It controls the combination of energy storage units within the same calibration group to take turns switching between different transmission paths and different transmission power levels to perform cross-node transmission calibration tasks. It conducts cross-verification of all dynamic line loss deviations obtained under multi-path and multi-operating conditions, accurately eliminating random errors generated during a single transmission operation. The complete working steps of this mechanism are executed sequentially according to the scheduling logic. First, the service subsystem, based on the geographical topology of the distribution network and the line segmentation rules, divides reusable energy storage units that are spatially adjacent and interconnected by transmission links into the same independent calibration group. Second, the multi-path cross-calibration mechanism generates an ordered polling scheduling sequence, controlling different combinations of energy storage units within the calibration group according to the sequence, switching to different transmission paths and different transmission power levels. Third, after each group of transmission paths and power levels completes a complete cross-node transmission calibration task, the system synchronously records the dynamic line loss deviation data generated under the corresponding operating conditions. Fourth, the mechanism cross-compares and comprehensively analyzes all line loss deviation data generated by multiple transmission paths and multiple operating conditions within the same calibration group, identifying and eliminating random error data that deviates from the overall data distribution pattern. Fifth, the system uses the effective deviation data after random error elimination to perform the final correction of the dynamic line loss calculation formula.

[0039] The multi-path cross-calibration mechanism internally includes a formula for calculating the mean of multi-path deviations and a formula for determining effective deviations, which are as follows:

[0040] First set of formula parameters This represents the average value of the dynamic line loss deviation across multiple transmission paths. This represents the total number of transmission paths participating in this cross-calibration work. Representing the The dynamic line loss deviation corresponding to each transmission path. Parameters within the second set of formulas. This represents the final effective dynamic line loss deviation after removing random errors. This represents the system's preset random error threshold. Only when the difference between the deviation value corresponding to a single path and the overall mean is less than this threshold will the data be considered valid and participate in subsequent calculations. The polling scheduling sequence generated by the system ensures that each energy storage unit combination within the calibration group can participate in the calibration work equally. Different transmission power levels can simulate various load conditions during the daily operation of the distribution network. Single cross-node transmission calibration operations are easily affected by uncertain factors such as instantaneous fluctuations in grid conditions and accidental external interference, resulting in random errors. The deviation data obtained under a single operating condition cannot objectively reflect the inherent problems in the line loss calculation model. The multi-path cross-calibration mechanism can effectively filter out error data caused by random interference through repeated verification actions across multiple operating conditions and paths, retaining valid deviation data that can truly reflect model defects. This allows the correction of the dynamic line loss calculation formula to be based on stable and reliable data, comprehensively improving the stability of the dynamic line loss correction results and the anti-interference capability of the entire system.

[0041] The data preprocessing module incorporates an anomaly type identification and differentiated repair mechanism. This mechanism is a complete logical system within the data preprocessing module that distinguishes different causes of data anomalies and executes corresponding processing strategies. This logical system does not fall within the scope of conventional data processing technology for power distribution networks. The anomaly type identification and differentiated repair mechanism can accurately identify two typical anomaly states in the original collected data. The first type is data anomalies caused by faults in various hardware devices of the power grid, and the second type is data distortion caused by electricity theft. The mechanism performs differentiated processing operations for the two types of abnormal data with different causes, ensuring the normal operation of different functional links of the system. The complete execution steps of this mechanism are performed sequentially according to the data processing order. First, after receiving the raw data collected from the entire network, the data preprocessing module immediately initiates its internal anomaly type identification and differentiated repair mechanism. Second, the mechanism calls upon the system's preset voltage fluctuation threshold, current mutation threshold, and operating condition anomaly threshold to judge each piece of raw data against exceeding limits. Third, the mechanism combines the fluctuation pattern, duration, and overall distribution characteristics of the data anomaly to accurately distinguish the cause of the data anomaly, determining whether the current anomaly belongs to equipment failure or electricity theft distortion. Fourth, for abnormal data caused by equipment failure, the mechanism performs data filling and data masking operations, using normal and valid data from adjacent time positions to complete the missing data, while masking the faulty abnormal data to prevent it from participating in the line loss calculation process. Fifth, for data distortion caused by electricity theft, the mechanism performs anomaly feature marking operations, completely preserving the original distorted data and attaching a unique feature label, then synchronously transmitting the marked data to the line loss calculation link and the electricity theft analysis link.

[0042] The anomaly type identification and differentiated repair mechanism includes anomaly detection formulas and fault data completion formulas, which are as follows: The first set of formulas is for data anomaly detection, and the parameters within the formulas are... This represents the raw data currently being analyzed. This represents the normal baseline data under the corresponding operating conditions. The first set of formulas represents the anomaly detection threshold for the corresponding data type. This threshold is further subdivided into several sub-thresholds, such as voltage fluctuation threshold, current surge threshold, and abnormal operating condition threshold. When the difference between the original data and the normal baseline data exceeds the corresponding threshold, the mechanism determines that the data set is abnormal. The second set of formulas is the equipment fault data completion formula, with parameters within the formula... This represents the replacement data after the supplementation is completed. This represents the normal, valid data from the time series preceding the current abnormal data. This represents the normal, valid data at the next time-series node following the current abnormal data. Abnormal data generated by equipment failure typically exhibits short-term abrupt changes and irregular fluctuations, while data distortion caused by electricity theft typically exhibits long-term, continuous deviations from the normal value range. The mechanism relies on these morphological characteristics to distinguish between different anomaly types. Traditional data processing methods uniformly remove all abnormal data exceeding a threshold. This approach simultaneously eliminates both equipment failure anomaly data and electricity theft distortion data. On one hand, this results in a missing data source for line loss calculation, leading to incorrect model corrections. On the other hand, it directly loses the abnormal features corresponding to electricity theft, making subsequent electricity theft identification impossible. The anomaly type identification and differentiated repair mechanism, through anomaly classification and differentiated processing strategies, effectively shields the line loss calculation model from interference by equipment failure data, ensuring the accuracy of line loss model correction, while also fully retaining the abnormal data features corresponding to electricity theft, guaranteeing the effectiveness and accuracy of electricity theft identification.

[0043] Of course, the above are just typical examples of the present invention. In addition, the present invention may have many other specific embodiments. All technical solutions formed by equivalent substitution or equivalent transformation fall within the scope of protection claimed by the present invention.

Claims

1. A transformer substation electricity theft analysis system based on segmented line loss calculation, characterized in that, include: Service subsystem, several reusable energy storage units deployed at different transmission nodes of the distribution network, and several voltage sensors configured on segmented low-voltage distribution lines; The service subsystem is internally configured with a line loss calculation model and an electricity theft analysis model. The service subsystem collects current operation data and grid condition data of segmented low-voltage distribution lines to generate power supply characteristic information. The line loss calculation model calculates the corresponding segmented line loss based on the power supply characteristic information. The electricity theft analysis model inputs the corresponding segmented line loss into the power supply topology network to output the corresponding electricity theft information. The line loss calculation model is internally divided into a dynamic line loss calculation module and a static line loss calculation module. The reusable energy storage unit is configured with a local node power supply mode and an inter-node transmission mode, and the power supply characteristics of each reusable energy storage unit are... The power consumption characteristics are preset known parameters; when the reusable energy storage unit operates in cross-node transmission mode, it performs segmented power supply transmission tasks. The service subsystem, based on the preset known power supply characteristics, the power consumption characteristics of the reusable energy storage unit at the receiving end, and the collected segmented current data, generates the theoretical dynamic line loss and actual dynamic line loss of the segmented low-voltage distribution line through the dynamic line loss calculation module; the dynamic line loss calculation module compares the theoretical dynamic line loss with the actual dynamic line loss to generate line loss deviation, and the dynamic line loss calculation module corrects the dynamic line loss calculation parameters based on the line loss deviation; the voltage sensor is used to collect the voltage data of the corresponding segmented low-voltage distribution line in real time; The service subsystem inputs real-time voltage data into the static line loss calculation module, which calculates the actual static line loss of the segmented low-voltage distribution line based on the real-time voltage data. The static line loss calculation module compares the actual static line loss with the preset theoretical static line loss, and corrects the static line loss calculation parameters based on the comparison results.

2. The transformer substation electricity theft analysis system based on segmented line loss calculation according to claim 1, characterized in that, The reusable energy storage unit has a built-in feature parameter adaptive update module, which is communicatively connected to the service subsystem. The adaptive update module for feature parameters is used to periodically receive real-time operating parameters of the distribution network, and iteratively update its own pre-stored power supply feature parameters and power consumption feature parameters based on the real-time operating parameters to complete the dynamic self-calibration of node feature parameters.

3. The transformer substation electricity theft analysis system based on segmented line loss calculation according to claim 2, characterized in that, When the reusable energy storage unit performs cross-node segmented power transmission tasks, it has a built-in impedance adaptive matching module. The impedance adaptive matching module is used to collect the line impedance parameters of the current segmented low-voltage distribution line in real time, and adaptively adjust its own power transmission output impedance based on the line impedance parameters, so that the power transmission conditions of the transmitting end and the receiving end energy storage unit are in a standardized matching state.

4. The transformer substation electricity theft analysis system based on segmented line loss calculation according to claim 1, characterized in that, Each of the voltage sensors has a built-in clock drift self-calibration unit, which is bound to the global clock signal of the service subsystem; The clock drift self-calibration unit is used to compare the deviation between the local acquisition clock and the global clock in real time, and correct the acquisition timing based on the clock deviation compensation to realize the synchronous acquisition of voltage data of the entire line segment.

5. The transformer substation electricity theft analysis system based on segmented line loss calculation according to claim 1, characterized in that, The data preprocessing module of the service subsystem has a built-in timing misalignment correction algorithm. The data preprocessing module is used to calculate the timing deviation of current data, voltage data, and operating condition data in different acquisition cycles, and to complete the misalignment compensation and alignment of multi-source heterogeneous data based on the timing deviation, and output standardized power supply characteristic information with fully coupled timing.

6. The transformer substation electricity theft analysis system based on segmented line loss calculation according to claim 1, characterized in that, The dynamic line loss calculation module has a built-in deviation source correction mechanism. The deviation source correction mechanism is used to associate the load fluctuation, transmission distance and power transmission power operating condition factors corresponding to the line loss deviation. Based on the weight ratio of different operating condition factors, the corresponding dynamic parameter correction coefficient is matched in a targeted manner to achieve targeted correction of line loss deviation.

7. The transformer substation electricity theft analysis system based on segmented line loss calculation according to claim 1, characterized in that, The static line loss calculation module is bound to an environmental parameter compensation mechanism; the static line loss calculation module is used to access the ambient temperature and humidity of the line and the equipment aging time parameters, and couple the environmental parameters and equipment aging parameters to the theoretical static line loss calculation formula.

8. The transformer substation electricity theft analysis system based on segmented line loss calculation according to claim 6 or 7, characterized in that, The line loss calculation model incorporates a bidirectional linkage calibration unit for dynamic and static line loss. This bidirectional linkage calibration unit uses the corrected static line loss parameters as the baseline threshold for dynamic line loss calculation, while simultaneously using the long-term mean deviation of dynamic line loss to iteratively correct the static line loss benchmark value.

9. The transformer substation electricity theft analysis system based on segmented line loss calculation according to claim 3, characterized in that, The service subsystem has a built-in multi-path cross-calibration mechanism. The multi-path cross-calibration mechanism is used to control the energy storage units in the same topology calibration group to switch different transmission paths and different transmission powers to complete multiple rounds of calibration, cross-verify the dynamic line loss deviation under multiple paths, and eliminate the random error of a single transmission calibration.

10. The transformer substation electricity theft analysis system based on segmented line loss calculation according to claim 5, characterized in that, The data preprocessing module has a built-in anomaly type identification and differential repair mechanism. This mechanism is used to distinguish between data anomalies caused by equipment failure and data distortion caused by electricity theft. Differentiated processing methods, such as data missing masking and anomaly feature marking, are used for the two types of abnormal data.

Images