A remote intelligent meter reading method and a remote intelligent meter reading system

Abstract

The application discloses a remote intelligent meter reading method and a remote intelligent meter reading system, and particularly relates to the field of intelligent meter reading, and aims to solve the problem that in the existing remote intelligent meter reading process, only address fields and frame check are relied on for response confirmation, and it is difficult to identify abnormal responses, device aging and coupling loop drift in complex power line environments; by extracting a synchronization convergence fingerprint vector in a carrier synchronization establishment process and a current transient feature vector during response frame transmission, a dynamic corresponding relationship between communication synchronization behavior and electrical transient behavior is constructed, a topology maintenance comparison is performed on a current response pulse, a current transient feature vector prediction trajectory is deduced based on historical meter reading records, a homologous determination result is generated by combining a trajectory deviation amount and a transient topology maintenance degree, a dynamic coupling stability sequence is constructed, and stable uploading and abnormal identification of remote intelligent meter reading results are realized, so that the automation level of meter reading management and the user service experience are improved.

Description

Technical Field

[0001] This invention relates to the field of smart meter reading technology, and more specifically, to a remote smart meter reading method and a remote smart meter reading system. Background Technology

[0002] With the continuous development of the power Internet of Things and smart distribution networks, remote smart meter reading based on power line carrier communication has been widely used in residential communities, industrial parks, commercial complexes and old distribution substations. The concentrator sends meter reading commands to various metering instruments through the power line carrier network, and the metering instruments return the corresponding carrier response frames, realizing the remote automatic collection of user electricity consumption data, water consumption data or other metering data.

[0003] In actual operation, power lines simultaneously serve the dual functions of power supply and communication. The lines are subject to complex factors such as load switching, motor start-up and shutdown, harmonic disturbances, line impedance fluctuations, and local contact aging. The coupling loop status, analog front-end characteristics, and carrier transmission response processes of different metering instruments also differ, which can easily lead to phenomena such as synchronization establishment fluctuations, current transient drift, and local pulse instability during the carrier response process.

[0004] Most existing remote smart meter reading methods rely solely on address fields, frame verification results, or communication success rates to determine the validity of current response frames. They lack the ability to analyze the dynamic correspondence between carrier synchronization establishment and electrical transient response behavior. When abnormal node spoofing, carrier response drift, coupling loop anomalies, or equipment aging occur, abnormal response frames may still be misclassified as normal meter reading results and uploaded to the concentrator, affecting the authenticity and stability of meter reading data. Furthermore, existing solutions typically lack the ability to continuously track dynamic fluctuations during historical meter readings, failing to identify the coupled stable state between communication synchronization behavior and current transient behavior during long-term meter operation. This makes them ill-suited to the reliability, authenticity, and continuous stability requirements of remote smart meter reading in complex power line carrier environments. Therefore, providing a remote smart meter reading method that addresses these issues, enabling remote, accurate, and efficient data acquisition from metering instruments, reducing the cost of manual on-site meter reading, and improving the automation level of energy management and user service experience, is a pressing problem that needs to be solved. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide a remote intelligent meter reading method and a remote intelligent meter reading system to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A remote smart meter reading method includes the following steps: S1. During the period when the terminal receives the preamble of the carrier response frame of the metering instrument, the transient frequency offset drift during the phase-locked loop synchronization establishment process is captured, and combined with the symbol synchronization pulse obtained based on the carrier signal demodulation, a synchronization convergence fingerprint vector is constructed. S2. Collect the current response pulse sequence during the transmission of the corresponding response frame in the power line carrier coupling loop, extract the leading edge slope and pulse width of the current response pulse, and construct the current transient feature vector. S3. Perform a topology preservation comparison of adjacent current response pulses on the transient current feature vector to obtain the transient topology preservation degree; S4. Using the synchronous convergence fingerprint vector as the driving quantity, read the correspondence between the driving quantity and the current transient feature vector in the historical meter reading record, estimate the predicted trajectory of the current transient feature vector, and calculate the trajectory deviation with the measured current transient feature vector. Combine the transient topology preservation degree to generate the same source judgment result. When the judgment result is the same source, accept the current response frame; otherwise, start the address re-reading. S5. After the current response frame is accepted, based on the trajectory deviation and transient topology preservation in the historical successful meter reading records of the metering instrument, the local fluctuation discrete quantity is statistically analyzed to construct a dynamic coupling stability sequence. S6. When the dynamic coupling stability sequence shows a continuous convergence state, the metering data obtained from parsing the current response frame is encapsulated as meter reading results and uploaded to the concentrator via power line carrier.

[0008] As a further aspect of the present invention, in step S1, constructing the synchronously converged fingerprint vector specifically includes: During the period when the terminal receives the carrier response frame preamble, quadrature demodulation is performed on the carrier signal to obtain the baseband in-phase component and extract the symbol synchronization pulse; The maximum frequency deviation during the frequency traction phase is extracted from the error voltage sequence output by the phase-locked loop phase detector, and the number of symbol cycles that it takes for the maximum frequency deviation to fall back to the error voltage range corresponding to the stable locked state of the phase-locked loop is counted as the transient frequency deviation drift. The time offset between the demodulated symbol synchronization pulse and the ideal timing reference corresponding to the local recovery clock is measured symbol by symbol to obtain the timing jitter of each symbol, and the overshoot peak detection is performed on the carrier phase transient response corresponding to the transition edge of each symbol to obtain the phase recovery overshoot. The transient frequency offset drift, timing jitter of each symbol, and phase recovery overshoot are arranged in the order of phase-locked loop convergence time to form a synchronous convergence fingerprint vector.

[0009] As a further aspect of the present invention, in step S2, constructing the current transient feature vector specifically includes: During the entire duration of the response frame, the power line carrier coupling loop is sampled through a current transformer to obtain a current response pulse time sequence synchronized with the symbol period of the response frame; Within each symbol period, the continuous rising interval of the current response pulse is identified. The time taken for the current to rise from the steady-state baseline to the pulse peak position is taken as the rising time of the leading edge. The rate of change of the current amplitude relative to time within the continuous rising interval is calculated as the leading edge slope. The time taken for the measured current to deviate from the steady-state baseline and recover to the neighborhood of the steady-state baseline is taken as the pulse width of the corresponding pulse; The leading edge slope and pulse width corresponding to each symbol period are arranged sequentially to construct the transient current feature vector.

[0010] As a further aspect of the present invention, obtaining the transient topology preservation degree in S3 specifically includes: The transient current feature vector is retrieved from the previous successful meter reading record as the current feature reference vector. In the current transient current feature vector, the leading edge slope ratio sequence of adjacent current response pulses corresponding to different symbol periods is calculated, and the difference is compared with the leading edge slope ratio sequence of the corresponding adjacent pulses in the current feature reference vector. Calculate the pulse width ratio sequence of adjacent pulses in the current transient feature vector and compare it with the pulse width ratio sequence of corresponding adjacent pulses in the current feature reference vector. The comparison results of the difference between the leading edge slope ratio sequence of adjacent pulses and the comparison results of the pulse width ratio sequence are converted into transient topology preservation degree through normalization.

[0011] As a further aspect of the present invention, in step S4, generating the homology determination result specifically includes: Based on the correspondence between the synchronous convergence fingerprint vector and the current transient feature vector in the successful meter reading records of the metering instrument, a nonlinear mapping model from the synchronous convergence fingerprint vector to the current transient feature vector is established. Input the current synchronous converged fingerprint vector into the nonlinear mapping model to obtain the corresponding current transient feature vector prediction trajectory; The leading edge slope and pulse width of each pulse in the predicted trajectory are compared with the leading edge slope and pulse width of the corresponding pulse position in the current measured current transient characteristic vector, and the resulting difference sequence is combined to calculate the trajectory deviation. The distribution range of trajectory deviation corresponding to each successful meter reading of the statistical metering instrument is used to determine the allowable range of trajectory deviation. If the current trajectory deviation falls within the allowable trajectory deviation range, and the current transient topology retention is consistent with the direction of change of the transient topology retention corresponding to the previous successful meter reading records, it is determined that the current response frame is consistent with the historical successful meter reading records, and the current response frame is accepted; otherwise, address-based re-reading is initiated.

[0012] As a further aspect of the present invention, in step S5, constructing a dynamic coupling stability sequence by statistically analyzing local fluctuation discrete quantities specifically includes: After receiving the current response frame, the trajectory deviation and transient topology preservation degree generated at each successful meter reading are extracted in chronological order to form a trajectory deviation sequence and a transient topology preservation degree sequence; The local dispersion of the sampled values ​​within a set sampling window in the trajectory deviation sequence is calculated as the local fluctuation dispersion of the trajectory deviation. The local fluctuation discreteness is calculated in the same way for the transient topology preservation sequence, and the above indices are arranged in time order to form a dynamic coupling stability sequence.

[0013] As a further aspect of the present invention, in step S6, when the dynamic coupling stability sequence exhibits a continuous convergence state, encapsulating the metering data obtained from parsing the current response frame into a meter reading result and uploading it to the concentrator via power line carrier specifically includes: Extract the local fluctuation discrete sequence corresponding to the trajectory deviation and the local fluctuation discrete sequence corresponding to the transient topology preservation from the dynamic coupling stability sequence; Within the historical sampling window, the local fluctuation discreteness of the trajectory deviation and the local fluctuation discreteness of the transient topology preservation degree corresponding to adjacent historical sampling windows are compared respectively. When the local fluctuation discreteness of both items decreases continuously, it is determined that the dynamic coupling fluctuation is in a continuous convergence state. During continuous convergence, the metering data obtained from parsing the current response frame is encapsulated into a meter reading result message and uploaded to the concentrator via power line carrier.

[0014] On the other hand, the present invention provides a remote intelligent meter reading system, comprising: The data acquisition module is used to perform timed automatic reading of various metering instruments in the target area, synchronize the freeze time reference of each energy meter, extract physical layer features when receiving carrier response frames, and perform integrity verification on the acquired data to form a traceable usage acquisition record. The anomaly identification and supplementary data collection module is used to automatically mark the energy meter as an abnormal node and start the supplementary data collection mechanism with address re-reading when the same source determination fails or the dynamic coupling stability sequence becomes unstable during the continuous meter reading cycle of the same meter. The coupling evaluation module is used to perform trend analysis on the trajectory deviation sequence and transient topology retention sequence of each electricity meter's previous readings, monitor carrier communication quality and equipment online status, and generate early warning prompts. The data analysis module is used to perform time-of-use statistics, load trend identification, and abnormal energy consumption behavior detection on the accepted metering data according to preset time periods, and push the analysis results to the user terminal. The strategy correction module is used to accumulate the historical results of the same source determination and the dynamic coupling stability sequence of each metering instrument, and adaptively correct the acceptance threshold and retry strategy parameters of the same source determination based on the communication success rate and abnormal node distribution in the operation log.

[0015] The technical effects and advantages of the remote intelligent meter reading method and system of the present invention are as follows: This invention establishes a dynamic correspondence between communication synchronization behavior and electrical transient behavior by simultaneously extracting the synchronization convergence fingerprint vector during the phase-locked loop (PLL) synchronization establishment phase and the current transient feature vector in the power line carrier coupling loop during the metering instrument carrier response process. This enables joint determination of the homology of metering instrument response frames. Compared to traditional meter reading methods that rely solely on address fields or frame verification results, this effectively reduces the risk of erroneous acceptance of abnormal response frames in complex power line environments. By performing topology preservation comparison on adjacent current response pulses and combining the synchronization convergence fingerprint vector to infer the current transient feature vector and predict the trajectory, dynamic analysis of the consistency between communication synchronization behavior and electrical response behavior can be achieved, improving the ability to identify spoofed responses, coupling anomalies, and equipment aging conditions during remote smart meter reading. Furthermore, a dynamic coupling stability sequence is constructed based on historical successful meter reading records to continuously track the local fluctuations and discrete states of trajectory deviation and transient topology preservation. Metering data is only uploaded when the dynamic coupling stability sequence shows continuous convergence, which helps improve the long-term stability and data reliability of remote smart meter reading results. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of a remote intelligent meter reading method according to the present invention; Figure 2 This is a schematic diagram of the structure of a remote intelligent meter reading system according to the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0018] Example 1 Figure 1 The present invention provides a remote smart meter reading method, which includes the following steps: S1. During the period when the terminal receives the preamble of the carrier response frame of the metering instrument, the transient frequency offset drift during the phase-locked loop synchronization establishment process is captured, and combined with the symbol synchronization pulse obtained based on the carrier signal demodulation, a synchronization convergence fingerprint vector is constructed. S2. Collect the current response pulse sequence during the transmission of the corresponding response frame in the power line carrier coupling loop, extract the leading edge slope and pulse width of the current response pulse, and construct the current transient feature vector. S3. Perform a topology preservation comparison of adjacent current response pulses on the transient current feature vector to obtain the transient topology preservation degree; S4. Using the synchronous convergence fingerprint vector as the driving quantity, read the correspondence between the driving quantity and the current transient feature vector in the historical meter reading record, estimate the predicted trajectory of the current transient feature vector, and calculate the trajectory deviation with the measured current transient feature vector. Combine the transient topology preservation degree to generate the same source judgment result. When the judgment result is the same source, accept the current response frame; otherwise, start the address re-reading. S5. After the current response frame is accepted, based on the trajectory deviation and transient topology preservation in the historical successful meter reading records of the metering instrument, the local fluctuation discrete quantity is statistically analyzed to construct a dynamic coupling stability sequence. S6. When the dynamic coupling stability sequence shows a continuous convergence state, the metering data obtained from parsing the current response frame is encapsulated as meter reading results and uploaded to the concentrator via power line carrier.

[0019] In step S1, a synchronously converged fingerprint vector is constructed.

[0020] In the carrier receiving front-end of the concentrator or acquisition terminal, the power line carrier signal input through the coupling circuit is first processed by a bandpass filter to remove the power frequency component and high-frequency random noise before being input to the local oscillator mixer unit. The carrier center frequency is 132kHz, and the passband range of the bandpass filter is set to 125kHz to 140kHz to ensure that the effective frequency band of the carrier during the preamble is completely preserved. Subsequently, the carrier is synchronously demodulated using two reference signals orthogonal to the local oscillator signal to obtain the baseband in-phase component and quadrature component. The in-phase component is used for subsequent phase-locked loop synchronization analysis, and the quadrature component is used to assist in judging the stability of carrier phase recovery. After demodulation, a symbol synchronization pulse is established using the fixed symbol flipping rhythm in the preamble. The synchronization establishment start point is the continuous detection of eight stable preamble transitions. The symbol period length is determined according to the standard symbol width in the current carrier communication protocol. In this embodiment, the symbol period is set to 416 microseconds. To avoid false triggering due to transient surges in the line, a consistency check is performed on the synchronization pulse intervals within three consecutive symbol periods during the establishment of the symbol synchronization pulse. When the deviation between adjacent pulse intervals is less than 5% of the corresponding symbol period length, the current synchronization pulse sequence is determined to meet the stable synchronization condition, and the corresponding pulse edge time is recorded as the time reference point for subsequent phase-locked loop convergence analysis. Throughout the process, the in-phase component and the synchronization pulse are buffered and recorded using a unified clock reference. The buffer length covers the entire preamble duration. In this embodiment, the preamble length is set to 32 symbol periods to ensure that all dynamic changes during the phase-locked loop establishment process are completely preserved.

[0021] During the preamble synchronization establishment process, the error voltage change curve at the output of the phase-locked loop (PLL) phase detector is read in real time, and the error voltage sequence is time-aligned according to the symbol period. Since the PLL undergoes a frequency pulling process during the initial carrier access phase, the error voltage exhibits a significant unidirectional offset and fallback process. Therefore, this embodiment uses the frequency offset state corresponding to the maximum absolute value of the error voltage as the maximum frequency offset state. In implementation, the error voltage sequence is converted to a local PLL reference to obtain the transient frequency offset drift at the corresponding time. The frequency offset is measured in Hz, and the sampling frequency is set to 16 samples per symbol period. Subsequently, the error voltage fallback process is continuously monitored. When the error voltage stably falls within the ±0.03V range for four consecutive symbol periods, the PLL is considered to have entered a stable locked state. The ±0.03V error voltage range is determined based on long-term meter reading test results in the field. This range corresponds to a stable locked interval with frequency offset fluctuations less than 18Hz, effectively avoiding misjudgments caused by transient disturbances in the line. Then, the number of symbol cycles from the occurrence of the maximum frequency deviation state to the entry into the stable locked state is counted. The statistical results are usually between 6 and 18 symbol cycles. Due to differences in the degree of crystal aging, coupling impedance and analog front end, different metering instruments will have different dynamic characteristics in their frequency deviation fall-off process.

[0022] Using the theoretical symbol edge of the locally recovered clock output as the ideal timing reference, the actual detected symbol synchronization pulse edge times are read one by one, and the time offset between the two is calculated to form a symbol-by-symbol timing jitter sequence. In this embodiment, the timing jitter is recorded using a nanosecond-level measurement method, with a sampling accuracy set to 50 nanoseconds. When drift in the same direction occurs within multiple consecutive symbol periods, the corresponding drift process is completely preserved without averaging, to ensure that the dynamic change trajectory during the phase-locked loop convergence process is not weakened. Simultaneously, the phase error change curve inside the phase-locked loop is continuously read at each symbol transition edge. When an instantaneous peak value exceeding the steady-state phase value is detected during the phase error recovery process, the overshoot amplitude corresponding to this peak value is recorded as the phase recovery overshoot. In this embodiment, the phase overshoot is recorded in angular form, and the overshoot detection window covers two symbol periods after the symbol transition. If multiple local peak values ​​appear within the window range, the peak value with the largest absolute value is selected as the phase recovery overshoot corresponding to the current symbol. After completing the periodic analysis of all symbols, the transient frequency offset drift, timing jitter of each symbol, and phase recovery overshoot are arranged in chronological order to form a synchronous convergent fingerprint vector.

[0023] In S2, a transient current feature vector is constructed.

[0024] A wideband current transformer is connected in series in the carrier coupling circuit corresponding to the metering instrument to continuously sample the high-frequency transient current in the line during the carrier response period. In this embodiment, the bandwidth of the current transformer is set to 10kHz to 500kHz, and the sampling frequency is set to 1MHz to ensure that the transient current pulses formed during carrier modulation are completely preserved. During the sampling process, the symbol timing boundary in the carrier communication protocol is used as the basis for time segmentation, and the current data collected during the entire response frame is divided into periods, with each symbol period corresponding to an independent current analysis segment. Since there are both power frequency components and random load disturbances in the power line, a high-pass filter is added at the sampling input, with the high-pass cutoff frequency set to 8kHz to filter out low-frequency load fluctuation components and retain only the high-frequency current pulse response formed during carrier transmission. For all current analysis segments in the same response frame, continuous buffering is performed according to the order of symbol appearance, and the symbol position number corresponding to each analysis segment is recorded synchronously. In actual operation, due to differences in power amplification links, coupling impedances and line connection states, the transient current pulses generated during carrier transmission of different metering instruments will have stable differences in rise speed and duration. Therefore, continuous sampling and recording of the current pulses during the entire duration of the response frame can accurately reflect the transient transmission status of the current metering instrument.

[0025] The average current value of consecutive sampling points within the initial region of each current analysis segment is statistically analyzed and used as the steady-state baseline for the current symbol period. The statistical interval of the steady-state baseline is the first 20% of the sampling range of the current symbol period. The current current analysis segment is then scanned along the time axis. When the current values ​​of four consecutive sampling points exceed the steady-state baseline by 5%, the current moment is considered to have entered a continuous rising interval. The 5% deviation ratio is determined based on long-term test results in residential power distribution environments, effectively suppressing random noise-induced false triggering even under high load fluctuations. After identifying the continuous rising interval, the current change process is continuously tracked. When the current reaches the maximum sampling value in the current symbol period, the time elapsed from the start of the continuous rising interval to the peak position is recorded as the rising edge duration. Subsequently, the current change between adjacent sampling points within the continuous rising interval is statistically analyzed segment by segment, and the rate of change is calculated based on the sampling time interval to obtain the rising edge slope of the corresponding pulse. For abnormal rate of change caused by local spikes, if it exceeds three times the overall average rate of change of the current continuous rising interval, the corresponding sampling segment is removed to avoid transient interference causing abnormal amplification of the rising edge slope. Since the differences in power amplifier build-up speed and output impedance of different metering instruments directly affect the pulse leading edge establishment process, the rising time and slope of the leading edge can stably characterize the transient emission characteristics of different metering instruments.

[0026] After identifying that the current pulse has entered a sustained rising range, the entire pulse duration is tracked along the time axis, and the position where the current first continuously deviates from the steady-state baseline is recorded as the pulse start point. In this embodiment, the deviation start time is confirmed by having three consecutive sampling points all above the steady-state baseline by 5%, avoiding misjudgment caused by random spikes in the line. Subsequently, the current fall-off process is continuously monitored. When the current re-enters the range of 3% above and below the steady-state baseline and remains stable for three consecutive sampling points, the current pulse is considered to have recovered to the steady-state baseline neighborhood. The 3% baseline neighborhood range is determined based on the results of on-site meter reading tests under different load conditions, and can stably reflect the pulse end state in both residential transformer substations and industrial power distribution environments. Then, the complete time length experienced from the pulse start point to the recovery to the steady-state baseline neighborhood is counted and used as the pulse width corresponding to the current pulse. After completing the analysis of all symbol periods, the leading edge slope and pulse width of the corresponding symbol period are written into the feature sequence buffer in the order of symbol appearance, and each feature item is kept to correspond to the symbol position number, so that the transient current response relationship corresponding to different symbol periods is completely preserved.

[0027] In S3, the transient topology preservation degree is obtained.

[0028] After the current metering instrument completes a normal response and passes the communication integrity verification, the current transient feature vector saved during the most recent successful meter reading is read from the historical successful meter reading records as the current feature reference vector. In this embodiment, only meter reading data that has passed the same source determination and has a stable continuous communication status is retained in the historical successful meter reading records to avoid abnormal response frames being written into the reference data, which would cause subsequent comparison distortion. After reading the current feature reference vector, all leading edge slope values ​​in the current current transient feature vector are traversed in the order of symbol appearance, and a set of ratio calculation units is formed by the leading edge slopes corresponding to two adjacent symbol periods. For example, the leading edge slope corresponding to the second symbol period is divided by the leading edge slope corresponding to the first symbol period to form the first set of leading edge slope ratios, the leading edge slope corresponding to the third symbol period is divided by the leading edge slope corresponding to the second symbol period to form the second set of leading edge slope ratios, and so on to form a complete leading edge slope ratio sequence. To avoid extreme anomalies in individual pulses due to transient line interference, in this embodiment, when a certain leading-edge slope value exceeds four times the average leading-edge slope of all current response frames, the corresponding slope value is marked as an abnormal slope point, and the average of the two adjacent valid slope values ​​is used to replace it. Subsequently, in the same manner, a corresponding reference leading-edge slope ratio sequence is constructed from the current characteristic reference vector. After constructing the two ratio sequences, the ratio differences at corresponding positions are compared sequentially, and the difference at each position is recorded. In this embodiment, the comparison results are recorded using absolute difference. When the ratio difference at a certain position is higher than 0.35 for three consecutive symbol periods, it is determined that the current metering instrument has abnormal transient fluctuations within the corresponding symbol interval, and the corresponding difference position is marked as an abnormal pulse segment. Since the dynamic establishment behavior of different metering instruments in power amplification links, line impedances, and coupling loops has stable characteristics, the relationship between the leading-edge slope changes of adjacent pulses remains stable under long-term normal operation. By constructing the leading-edge slope ratio sequence and performing position-by-position difference comparison, it is possible to accurately identify whether the current transient pulse structure has experienced abnormal drift.

[0029] After completing the leading-edge slope ratio sequence analysis, all pulse width values ​​in the current transient feature vector are read in sign order, and a width ratio sequence is constructed using the proportional relationship between adjacent pulse widths. In this embodiment, the width ratio at the corresponding position is formed by dividing the width of the next pulse by the width of the previous pulse, thereby preserving the relative topological relationship during the pulse width change process throughout the entire response frame. Then, in the same way as the leading-edge slope ratio sequence, a corresponding reference width ratio sequence is constructed from the historical current feature reference vector, and difference comparison is performed position by position. For positions where the width difference deviates significantly from the historical stable state, this embodiment uses a continuous position association confirmation method for verification. That is, when two adjacent ratio positions simultaneously show a width difference exceeding 0.4, it is determined that the corresponding pulse structure has undergone local topological drift, thereby avoiding misjudgment caused by random line noise affecting a single pulse. After statistically analyzing the differences in the leading-edge slope ratio and pulse width ratio, normalization is performed on all difference results. In this embodiment, the normalization process uses the maximum stable difference range in historical successful meter reading records as the scaling benchmark. The normalization range for the leading-edge slope difference is set to 0 to 1.2, and the normalization range for the pulse width difference is set to 0 to 1.5. After normalization, a unified aggregation based on preset weights is performed on the normalized difference results corresponding to each location. The weights can be set based on experience according to historical meter reading records, but consistency must be ensured. Transient topology preservation is generated in reverse according to the degree of difference. When all normalized difference results remain at a low level overall, the corresponding transient topology preservation remains high; when the local difference increases continuously, the corresponding transient topology preservation decreases synchronously.

[0030] In step S4, a homology determination result is generated.

[0031] After the metering instrument completes multiple normal meter readings consecutively, the synchronous convergence fingerprint vector and the corresponding transient current feature vector at that time are extracted from the historical successful meter reading records, and corresponding pairing is performed according to the same meter reading cycle. In this embodiment, only historical meter reading records with stable continuous communication and passing the same source determination are selected as training samples, and meter reading data with line transient disturbances, failed re-reading, or abnormal response are removed to avoid abnormal samples participating in model training and causing the mapping relationship to drift. Then, a nonlinear mapping model from the synchronous convergence fingerprint vector to the transient current feature vector is established. In this embodiment, a three-layer nonlinear mapping structure is adopted, where the input layer corresponds to the transient frequency offset drift, symbol timing jitter, and phase recovery overshoot in the synchronous convergence fingerprint vector, the intermediate mapping layer adopts a continuous fully connected structure, the number of hidden nodes is set to 64, and the output layer corresponds to the leading edge slope and pulse width sequence in the transient current feature vector. During the training process, the historical synchronous convergence fingerprint vectors are input into the model in chronological order, and the corresponding historical transient current feature vectors are used as the target output. The mapping layer connection parameters are updated through error backpropagation. In this embodiment, the training iterations are set to 120 rounds. When the overall prediction error decreases by less than 0.5% in 10 consecutive iterations, parameter updates are stopped and the current model structure is solidified. After training is completed, the synchronous convergence fingerprint vector extracted during the current meter reading process is input into the nonlinear mapping model. Based on the historically established correspondence between synchronous dynamics and current dynamics, the model outputs the corresponding current transient feature vector prediction trajectory. Each pulse position in the prediction trajectory corresponds to the predicted leading edge slope and the predicted pulse width, and maintains a correspondence with the sign position in the current measured current transient feature vector.

[0032] After the nonlinear mapping model outputs the predicted trajectory of the current transient characteristic vector, all predicted pulses in the predicted trajectory are traversed according to their sign positions. The predicted leading-edge slope and predicted pulse width at the corresponding positions are read, and the actual leading-edge slope and actual pulse width at the corresponding positions in the current measured current transient characteristic vector are also read. Subsequently, a step-by-step difference calculation is performed on the predicted and measured values. In this embodiment, the absolute difference method is used to record the corresponding deviation results, and a leading-edge slope difference sequence and a pulse width difference sequence are formed according to their sign positions. For abnormal spike deviations at a single pulse position, when the corresponding difference exceeds five times the average of all differences, the position is marked as a sudden interference point, and the average of the two adjacent effective differences is used to replace it, avoiding distortion of the overall trajectory deviation caused by transient line glitches. Then, the leading-edge slope difference sequence and the pulse width difference sequence are merged according to their sign positions, and the overall difference distribution is statistically analyzed to form the trajectory deviation corresponding to the current meter reading process. In this embodiment, the trajectory deviation is recorded in the form of the overall root mean square, so that local drift at multiple consecutive pulse positions can be uniformly reflected. Subsequently, all trajectory deviations in the historical successful meter reading records are read, and their distribution range under long-term stable operation is statistically analyzed. In this embodiment, the trajectory deviations corresponding to the most recent 50 successful meter reading records are used to construct a stable distribution range, and the range of 15% above and below the average of the historical trajectory deviations is taken as the allowable range of trajectory deviations.

[0033] After calculating the current trajectory deviation, it is determined whether the current trajectory deviation is within the historical trajectory deviation allowable range. If the current trajectory deviation is within the allowable range, the transient topology retention rate corresponding to the current meter reading process is read, and continuous direction analysis is performed on the transient topology retention rate change process in the historical successful meter reading records. In this embodiment, the transient topology retention rate in the 10 most recent consecutive successful meter reading records is read in chronological order, and the change direction between adjacent meter reading cycles is counted. When the change direction of the current transient topology retention rate is consistent with the historical stable change direction, it is determined that the current pulse structure change process is consistent with the historical normal communication state. If the current trajectory deviation falls within the allowable range but the change direction of the transient topology retention rate reverses, it is determined that although the overall trajectory of the current response process is close to the historical stable state, the local pulse structure has experienced abnormal drift and does not meet the same-source acceptance condition. When the trajectory deviation and the change direction of the transient topology retention rate simultaneously meet the stability condition, it is determined that the current response frame maintains the same source relationship with the historical successful meter reading records, and the current metering data is allowed to enter the subsequent meter reading result processing flow. If any condition is not met, the address-based recopying process is initiated. In this embodiment, the recopying instruction carries the unique communication address of the current metering instrument and the corresponding freeze time identifier, so that the metering instrument returns to the complete response frame and re-executes the synchronous convergence analysis and current transient trajectory determination until a valid response result that meets the same source relationship condition is obtained.

[0034] In S5, the discrete values ​​of local fluctuations are statistically analyzed to construct a dynamic coupling stability sequence.

[0035] After the current response frame passes the same-source determination and is accepted, the corresponding trajectory deviation and transient topology retention are read from the current meter reading record and written to the historical meter reading record buffer of the corresponding meter in the order of the current meter reading time. In this embodiment, each meter establishes an independent historical operation record area to store the changes in trajectory deviation and transient topology retention of the meter in continuous meter reading cycles for a long time. To avoid abnormal meter reading results affecting subsequent stability analysis, only meter reading results that have completed same-source acceptance and passed data verification are allowed to enter the historical record area. Abnormal meter reading records that fail to be reread, have communication interruptions, or whose trajectory deviation exceeds the allowable range are not written. After the current meter reading result is written, the trajectory deviation of the corresponding meter in the most recent consecutive successful meter reading cycles is read in the order of meter reading time and arranged in chronological order to form a trajectory deviation sequence. In this embodiment, the length of the historical sequence is set to the most recent 60 successful meter reading cycles. When the number of records is less than 60, the number of existing records is used for subsequent analysis. Subsequently, the transient topology retention in the corresponding historical meter reading cycle is read in the same way, and a transient topology retention sequence is formed. Since trajectory deviation reflects the overall trajectory consistency between synchronous convergence behavior and transient current behavior, while transient topology retention reflects the stable state of the pulse local structure, both reflect the communication stability and transient response stability of the current metering instrument during long-term meter reading. For continuously stable metering instruments, the trajectory deviation sequence and transient topology retention sequence usually change slowly. When line contacts age, coupling impedance changes, or power amplifier links malfunction, both sequences will exhibit synchronous fluctuations and diffusion. Therefore, by storing these two dynamic indicators over a long period and arranging them in chronological order, the dynamic coupling change process of the metering instrument under long-term operation can be completely recorded.

[0036] The trajectory deviation values ​​corresponding to six consecutive successful meter readings within the current sampling window are read sequentially over time, and the average value of all trajectory deviation values ​​within the window is calculated first. In this embodiment, the window average value is obtained by directly summing all trajectory deviation values ​​within the window and dividing by the number of samples. Subsequently, each trajectory deviation value within the window is read sequentially, and the deviation magnitude between each trajectory deviation value and the window average value is calculated. The deviation magnitude is recorded as the absolute value of the corresponding trajectory deviation value minus the window average value. After completing the calculation of all deviation magnitudes, all deviation magnitudes within the window are accumulated, and the accumulated result is divided by the number of effective samples within the current window to obtain the local dispersion corresponding to the current sampling window. Since this local dispersion directly reflects the fluctuation range of the trajectory deviation values ​​around the average state within the current sampling window, it is used as the local fluctuation dispersion value corresponding to the trajectory deviation value. When the trajectory deviation values ​​within the window remain close for a long time, the overall deviation magnitude is small, and the local fluctuation dispersion value decreases accordingly; when the trajectory deviation values ​​begin to spread significantly during continuous meter reading, the corresponding deviation magnitude increases accordingly, and the local fluctuation dispersion value increases accordingly. To avoid individual abnormal sampling values ​​from having an excessive impact on the overall dispersion, in this embodiment, when the deviation between a certain trajectory deviation and the window average exceeds three times the average deviation of all deviations in the current window, the corresponding sampling value is marked as an abnormal fluctuation point, and the abnormal sampling value is replaced by the median value of the other valid sampling values ​​in the window before the local dispersion statistics are re-executed.

[0037] In step S6, when the dynamic coupling stability sequence exhibits a continuous convergence state, the metering data obtained from parsing the current response frame is encapsulated as meter reading results and uploaded to the concentrator via power line carrier.

[0038] After the current metering instrument completes a successful meter reading and generates a dynamic coupling stability sequence, the local fluctuation discrete quantities corresponding to all trajectory deviations and transient topology preservation in the dynamic coupling stability sequence are read in chronological order. In this embodiment, six consecutive successful meter reading records constitute a historical sampling window, which is then moved sequentially along the time axis in steps of one meter reading cycle to form a continuous historical sampling window sequence. Subsequently, the local fluctuation discrete quantities of trajectory deviations corresponding to two adjacent historical sampling windows are read sequentially, and a window-by-window size comparison is performed. When the local fluctuation discrete quantity of trajectory deviations corresponding to the later historical sampling window is continuously smaller than that corresponding to the earlier historical sampling window, it is determined that the fluctuation diffusion of trajectory deviations in the continuous meter reading cycle is weakening. Then, the same method is used to perform continuous comparisons between adjacent historical sampling windows on the local fluctuation discrete quantities corresponding to transient topology preservation. When the local fluctuation discrete quantity of transient topology preservation corresponding to the later historical sampling window is continuously lower than that corresponding to the earlier historical sampling window, it is determined that the local fluctuation degree of the current pulse topology is synchronously weakening. In this embodiment, when the continuous decrease in the state lasts for at least three adjacent historical sampling windows, the corresponding local fluctuation is considered to have entered a stable convergence phase. The determination length of three consecutive historical sampling windows is determined based on long-term meter reading test results of residential distribution transformer areas, which can effectively avoid misjudgment caused by a single accidental fluctuation even under high-frequency load fluctuation environments. For historical sampling windows where the local fluctuation dispersion suddenly increases in the opposite direction during continuous comparison, when the corresponding increase exceeds 40% of the average local fluctuation dispersion of the previous historical sampling window, the historical sampling window is marked as an abnormal disturbance window, and the current continuous convergence determination process is interrupted. The continuous convergence state detection is restarted from the next historical sampling window, thereby avoiding the incorrect acceptance of abnormal response frames due to sudden disturbances in the local line.

[0039] In the continuous convergence state, the metering data parsed from the current response frame is encapsulated into a meter reading result message and uploaded to the concentrator via power line carrier. Specifically, after determining that the current dynamic coupling fluctuation is in a continuous convergence state, the corresponding metering data content in the current response frame is read. In this embodiment, the metering data includes the current cumulative energy value, freeze time data, time-of-use electricity price metering data, and the current metering instrument operating status identifier. Subsequently, according to the carrier communication message format corresponding to the concentrator, the current metering data is encapsulated, and the current metering instrument communication address, the current meter reading timestamp, and the corresponding freeze time identifier are written into the message header. In this embodiment, the timestamp is uniformly generated using the concentrator's local clock, with a time precision set to the second level. The freeze time identifier is consistent with the freeze time when the current meter reading task is issued, thereby ensuring that the data between different metering instruments has a unified metering time reference. After message encapsulation, the integrity of the current meter reading result message is checked. In this embodiment, a 16-bit cyclic redundancy check code is used to check the entire message content. When the check result is correct, the current meter reading result message is written to the carrier transmission buffer and uploaded to the corresponding concentrator via the power line carrier coupling loop. If two consecutive carrier transmission failures are detected during the upload process, the current dynamic coupling stability sequence state is reread. The current meter reading result message is only allowed to be resent if the continuous convergence state remains unchanged. If the continuous convergence state has failed, the current result upload is stopped immediately, and the address-based re-reading process is restarted to avoid continuing to upload untrusted meter reading data when the dynamic coupling state has become unstable.

[0040] Example 2 The difference between Embodiment 2 and Embodiment 1 is that this embodiment introduces a remote intelligent meter reading system.

[0041] Figure 2 A schematic diagram of a remote intelligent meter reading system according to the present invention is provided. The remote intelligent meter reading system includes: The data acquisition module is used to perform timed automatic reading of various metering instruments in the target area, synchronize the freeze time reference of each energy meter, extract physical layer features when receiving carrier response frames, and perform integrity verification on the acquired data to form a traceable usage acquisition record. The anomaly identification and supplementary data collection module is used to automatically mark the energy meter as an abnormal node and start the supplementary data collection mechanism with address re-reading when the same source determination fails or the dynamic coupling stability sequence becomes unstable during the continuous meter reading cycle of the same meter. The coupling evaluation module is used to perform trend analysis on the trajectory deviation sequence and transient topology retention sequence of each electricity meter's previous readings, monitor carrier communication quality and equipment online status, and generate early warning prompts. The data analysis module is used to perform time-of-use statistics, load trend identification, and abnormal energy consumption behavior detection on the accepted metering data according to preset time periods, and push the analysis results to the user terminal. The strategy correction module is used to accumulate the historical results of the same source determination and the dynamic coupling stability sequence of each metering instrument, and adaptively correct the acceptance threshold and retry strategy parameters of the same source determination based on the communication success rate and abnormal node distribution in the operation log.

[0042] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0043] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0044] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0045] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0046] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0047] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0048] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0049] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0050] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for remote intelligent meter reading, characterized in that Includes the following steps: S1. During the period when the terminal receives the preamble of the carrier response frame of the metering instrument, the transient frequency offset drift during the phase-locked loop synchronization establishment process is captured, and combined with the symbol synchronization pulse obtained based on the carrier signal demodulation, a synchronization convergence fingerprint vector is constructed. S2. Collect the current response pulse sequence during the transmission of the corresponding response frame in the power line carrier coupling loop, extract the leading edge slope and pulse width of the current response pulse, and construct the current transient feature vector. S3. Perform a topology-preserving comparison of adjacent current response pulses on the transient current feature vector to obtain the transient topology-preserving degree, specifically including: The transient current feature vector is retrieved from the previous successful meter reading record as the current feature reference vector. In the current transient current feature vector, the leading edge slope ratio sequence of adjacent current response pulses corresponding to different symbol periods is calculated, and the difference is compared with the leading edge slope ratio sequence of the corresponding adjacent pulses in the current feature reference vector. Calculate the pulse width ratio sequence of adjacent pulses in the current transient feature vector and compare it with the pulse width ratio sequence of corresponding adjacent pulses in the current feature reference vector. The comparison results of the difference in the sequence of leading edge slope ratio of adjacent pulses and the comparison results of the difference in the sequence of pulse width ratio are converted into transient topology preservation through normalization. S4. Using the synchronous convergence fingerprint vector as the driving quantity, read the correspondence between the driving quantity and the current transient feature vector in the historical meter reading records, estimate the predicted trajectory of the current transient feature vector, and calculate the trajectory deviation from the measured current transient feature vector. Combined with the transient topology preservation degree, generate a homogeneity determination result. When the determination result is homogeneous, accept the current response frame; otherwise, initiate a re-reading with address. Specifically, this includes: Based on the correspondence between the synchronous convergence fingerprint vector and the current transient feature vector in the successful meter reading records of the metering instrument, a nonlinear mapping model from the synchronous convergence fingerprint vector to the current transient feature vector is established. Input the current synchronous converged fingerprint vector into the nonlinear mapping model to obtain the corresponding current transient feature vector prediction trajectory; The leading edge slope and pulse width of each pulse in the predicted trajectory are compared with the leading edge slope and pulse width of the corresponding pulse position in the current measured current transient characteristic vector, and the resulting difference sequence is combined to calculate the trajectory deviation. The distribution range of trajectory deviation corresponding to each successful meter reading of the statistical metering instrument is used to determine the allowable range of trajectory deviation. If the current trajectory deviation falls within the trajectory deviation allowable range, and the current transient topology retention is consistent with the direction of change of the transient topology retention corresponding to the previous successful meter reading records, it is determined that the current response frame is consistent with the historical successful meter reading records, and the current response frame is accepted; otherwise, a re-reading with address is initiated. S5. After the current response frame is accepted, based on the trajectory deviation and transient topology preservation in the historical successful meter reading records of the metering instrument, the local fluctuation discrete quantity is statistically analyzed to construct a dynamic coupling stability sequence. S6. When the dynamic coupling stability sequence shows a continuous convergence state, the metering data obtained from parsing the current response frame is encapsulated as meter reading results and uploaded to the concentrator via power line carrier.

2. The remote intelligent meter reading method of claim 1 wherein, In step S1, constructing the synchronously converged fingerprint vector specifically includes: During the period when the terminal receives the carrier response frame preamble, quadrature demodulation is performed on the carrier signal to obtain the baseband in-phase component and extract the symbol synchronization pulse; The maximum frequency deviation during the frequency traction phase is extracted from the error voltage sequence output by the phase-locked loop phase detector, and the number of symbol cycles that it takes for the maximum frequency deviation to fall back to the error voltage range corresponding to the stable locked state of the phase-locked loop is counted as the transient frequency deviation drift. The time offset between the demodulated symbol synchronization pulse and the ideal timing reference corresponding to the local recovery clock is measured symbol by symbol to obtain the timing jitter of each symbol, and the overshoot peak detection is performed on the carrier phase transient response corresponding to the transition edge of each symbol to obtain the phase recovery overshoot. The transient frequency offset drift, timing jitter of each symbol, and phase recovery overshoot are arranged in the order of phase-locked loop convergence time to form a synchronous convergence fingerprint vector.

3. The remote intelligent meter reading method of claim 1 wherein, In step S2, constructing the transient current feature vector specifically includes: During the entire duration of the response frame, the power line carrier coupling loop is sampled through a current transformer to obtain a current response pulse time sequence synchronized with the symbol period of the response frame. Within each symbol period, the continuous rising interval of the current response pulse is identified. The time taken for the current to rise from the steady-state baseline to the pulse peak position is taken as the rising time of the leading edge. The rate of change of the current amplitude relative to time within the continuous rising interval is calculated as the leading edge slope. The time taken for the measured current to deviate from the steady-state baseline and recover to the neighborhood of the steady-state baseline is taken as the pulse width of the corresponding pulse; The leading edge slope and pulse width corresponding to each symbol period are arranged sequentially to construct the transient current feature vector.

4. The remote intelligent meter reading method of claim 1 wherein, In S5, constructing a dynamic coupled stability sequence by statistically analyzing local fluctuation discrete quantities specifically includes: After receiving the current response frame, the trajectory deviation and transient topology preservation degree generated at each successful meter reading are extracted in chronological order to form a trajectory deviation sequence and a transient topology preservation degree sequence; The local dispersion of the sampled values ​​within a set sampling window in the trajectory deviation sequence is calculated as the local fluctuation dispersion of the trajectory deviation. The local fluctuation discrete quantities are calculated in the same way for the transient topology preservation sequence. The trajectory deviation quantity and the local fluctuation discrete quantities corresponding to the transient topology preservation are arranged in time order to form a dynamic coupling stability sequence.

5. The remote intelligent meter reading method according to claim 1, characterized in that, In step S6, when the dynamic coupling stability sequence exhibits continuous convergence, the metering data obtained from parsing the current response frame is encapsulated as meter reading results and uploaded to the concentrator via power line carrier. This specifically includes: Extract the local fluctuation discrete sequence corresponding to the trajectory deviation and the local fluctuation discrete sequence corresponding to the transient topology preservation from the dynamic coupling stability sequence; Within the historical sampling window, the local fluctuation discreteness of the trajectory deviation and the local fluctuation discreteness of the transient topology preservation degree corresponding to adjacent historical sampling windows are compared respectively. When the local fluctuation discreteness of both items decreases continuously, it is determined that the dynamic coupling fluctuation is in a continuous convergence state. During continuous convergence, the metering data obtained from parsing the current response frame is encapsulated into a meter reading result message and uploaded to the concentrator via power line carrier.

6. A remote intelligent meter reading system, used to implement the remote intelligent meter reading method according to any one of claims 1-5, characterized in that, include: The data acquisition module is used to perform timed automatic reading of various metering instruments in the target area, synchronize the freeze time reference of each energy meter, extract physical layer features when receiving carrier response frames, and perform integrity verification on the acquired data to form a traceable usage acquisition record. The anomaly identification and supplementary data collection module is used to automatically mark the energy meter as an abnormal node and start the supplementary data collection mechanism with address re-reading when the same source determination fails or the dynamic coupling stability sequence becomes unstable during the continuous meter reading cycle of the same meter. The coupling evaluation module is used to perform trend analysis on the trajectory deviation sequence and transient topology retention sequence of each electricity meter's previous readings, monitor carrier communication quality and equipment online status, and generate early warning prompts. The data analysis module is used to perform time-of-use statistics, load trend identification, and abnormal energy consumption behavior detection on the accepted metering data according to preset time periods, and push the analysis results to the user terminal. The strategy correction module is used to accumulate the historical results of the same source determination and the dynamic coupling stability sequence of each metering instrument, and adaptively correct the acceptance threshold and retry strategy parameters of the same source determination based on the communication success rate and abnormal node distribution in the operation log.

Images