Electric energy metering box anti-stealing electricity intelligent monitoring control method and system

Abstract

The present application relates to the technical field of electric energy metering protection, and discloses an electricity larceny prevention intelligent monitoring control method and system for an electric energy metering box. The system comprises an electricity larceny prevention feature recognition module, a security strategy matching module, a dynamic parameter adjustment module and an intervention control execution module. The electricity larceny prevention feature recognition module collects voltage ripple and current ripple features of the metering box in real time and outputs an electricity larceny prevention feature data set; the security strategy matching module queries a pre-set electricity larceny mode feature library based on the feature data set and outputs a target security strategy identifier and initial security strategy parameters; the dynamic parameter adjustment module corrects the initial parameters based on the feature data set and generates real-time security strategy parameters; and the intervention control execution module executes hierarchical intervention actions according to the real-time parameters. The system can comprehensively capture changes in electric energy parameters, accurately identify electricity larceny behavior, adapt to different electricity use scenarios and realize a complete electricity larceny prevention process from monitoring to intervention.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electric energy metering protection, in particular to an electric energy metering box anti-stealing electricity intelligent monitoring control method and system. BACKGROUND

[0002] In the power supply network, the electric energy metering box as the core carrier of electric energy consumption metering, its operation safety and metering accuracy are directly related to the economic benefit of the power enterprise and the stable operation of the power system. With the continuous growth of the number of power users and the increasing complexity of electricity consumption scenarios, the stealing electricity behavior presents the development trend of diversification, concealment and intelligence, and the traditional electric energy metering box protection means has been difficult to meet the actual needs of the current anti-stealing electricity work.

[0003] At present, the common electric energy metering box protection measures on the market are mainly physical lock protection, regular manual inspection and simple current and voltage abnormal alarm. The physical lock protection method can only limit the opening behavior of the metering box outside, and cannot deal with the hidden stealing electricity operation such as internal wiring tampering and shunt around the meter, and the lock itself is easy to be damaged or copied, and the protection reliability is low. The regular manual inspection mode is limited by the inspection cycle, the number of personnel and the geographical range, and it is difficult to realize real-time monitoring of a large number of scattered metering boxes, and the stealing electricity behavior often cannot be found until a long time after the occurrence, resulting in a large economic loss of the power enterprise.

[0004] The simple current and voltage abnormal alarm system can issue an alarm when detecting that the parameter exceeds the set threshold, but this kind of system usually adopts a fixed threshold judgment method, which cannot be flexibly adjusted according to the normal parameter fluctuation of different electricity consumption scenarios and different time periods. For example, during the peak period of industrial electricity consumption, the normal current and voltage fluctuation range will increase significantly, and if the system still judges according to the threshold of the regular period, a large number of false alarm information will be generated, which not only increases the work burden of the operation and maintenance personnel, but also may cause the real stealing electricity alarm to be ignored. In addition, this kind of system can only realize the alarm function, lacks the further identification, analysis and active intervention ability of the stealing electricity behavior, cannot form a complete anti-stealing electricity closed loop from monitoring to disposal, and is difficult to effectively curb the occurrence of stealing electricity behavior.

[0005] With the continuous upgrading of stealing electricity technology, new stealing electricity means such as tampering with voltage and current waveform by using power electronic equipment and interfering with the normal work of metering chip through wireless signal are emerging, and the traditional anti-stealing electricity system has obvious deficiencies in the comprehensiveness and accuracy of stealing electricity feature identification, which cannot effectively capture the subtle parameter changes caused by these new stealing electricity behaviors, resulting in a passive situation of anti-stealing electricity work, and a real-time, intelligent and self-adaptive anti-stealing electricity monitoring system is needed to solve the above problems. SUMMARY

[0006] The present application aims to provide an electricity metering box electricity larceny prevention intelligent monitoring control method and system to solve the problems in the background art.

[0007] To achieve the above-mentioned purpose, the present application provides an electricity metering box electricity larceny prevention intelligent monitoring system, which comprises an electricity larceny prevention feature identification module, a security strategy matching module, a dynamic parameter adjustment module and an intervention control execution module; the electricity larceny prevention feature identification module is used to collect the voltage ripple feature and the current ripple feature of the electricity metering box in real time and output the electricity larceny prevention feature dataset; the security strategy matching module is connected with the electricity larceny prevention feature identification module and is used to query the preset electricity larceny mode feature library according to the electricity larceny prevention feature dataset, output the target security strategy identifier and the corresponding initial security strategy parameter; the dynamic parameter adjustment module is connected with the security strategy matching module and is used to dynamically correct the initial security strategy parameter based on the electricity larceny prevention feature dataset to generate the real-time security strategy parameter; the intervention control execution module is connected with the dynamic parameter adjustment module and is used to execute the hierarchical intervention action according to the real-time security strategy parameter.

[0008] Preferably, the electricity larceny prevention feature identification module generates the electricity larceny prevention feature dataset by the following way: extracting the mutation frequency feature and the distortion amplitude feature in the voltage ripple feature; extracting the phase shift feature and the abnormal harmonic feature in the current ripple feature; combining the mutation frequency feature, the distortion amplitude feature, the phase shift feature and the abnormal harmonic feature into a multi-dimensional feature vector as the electricity larceny prevention feature dataset.

[0009] Preferably, the security strategy matching module outputs the target security strategy identifier by the following way: calculating the similarity weight value of the electricity larceny prevention feature dataset and each historical electricity larceny feature template in the electricity larceny mode feature library; selecting all historical electricity larceny feature templates with the similarity weight value exceeding the preset threshold value as the matching feature template set; determining the target security strategy identifier according to the strategy priority coefficient corresponding to the matching feature template set.

[0010] Preferably, the dynamic parameter adjustment module generates the real-time security strategy parameter by the following way: obtaining the ripple mutation proportion feature and the phase shift trend feature in the electricity larceny prevention feature dataset; calculating the voltage compensation weight according to the ripple mutation proportion feature; calculating the current compensation weight according to the phase shift trend feature; weighting and fusing the threshold value parameter in the initial security strategy parameter by using the voltage compensation weight and the current compensation weight to obtain the real-time threshold value parameter as a component of the real-time security strategy parameter.

[0011] Preferably, the system further comprises a system delay compensation module; the system delay compensation module is configured to invoke a pre-stored metering response delay duration when the anti-theft electricity feature recognition module is working; perform time domain compensation on the voltage ripple feature and the current ripple feature according to the metering response delay duration, and output the compensated ripple feature to the anti-theft electricity feature recognition module.

[0012] Preferably, the intervention control execution module executes the hierarchical intervention action by the following manner: when the real-time safety strategy parameter contains a primary intervention identifier, starting an electromagnetic locking device of a metering box protective shell; when the real-time safety strategy parameter contains a middle-level intervention identifier, triggering a breaking control circuit of a metering loop; when the real-time safety strategy parameter contains a high-level intervention identifier, activating a wireless alarm unit of a positioning tracking device.

[0013] Preferably, the safety strategy matching module is further configured to track a change rate feature of the anti-theft electricity feature data set in real time; update a strategy priority coefficient of the matching feature template set according to the change rate feature; and feed back the updated strategy priority coefficient to the dynamic parameter adjustment module.

[0014] Preferably, the intervention control execution module is further configured to continuously collect a vibration frequency feature of the metering box protective shell when executing the hierarchical intervention action; and when the vibration frequency feature exceeds a preset safety threshold, upgrading the middle-level intervention identifier to the high-level intervention identifier.

[0015] Preferably, the dynamic parameter adjustment module is further configured to acquire a current running state parameter of the electric energy metering box before generating the real-time safety strategy parameter; and when a difference degree between the current running state parameter and the initial safety strategy parameter exceeds a preset tolerance, starting a parameter difference verification process.

[0016] The parameter difference verification process comprises the following steps: calling a benchmark parameter set under a historical normal working condition; calculating a deviation coefficient of the current running state parameter and the benchmark parameter set; and when the deviation coefficient is less than a preset abnormality judgment value, replacing the initial safety strategy parameter with the benchmark parameter set as an input benchmark of the real-time safety strategy parameter.

[0017] Preferably, the application further comprises an anti-theft electricity intelligent monitoring control method for an electric energy metering box, which comprises all the modules and method processes of the anti-theft electricity intelligent monitoring control system for an electric energy metering box.

[0018] Compared with the prior art, the application has the following beneficial effects:

[0019] The anti-theft electricity feature recognition module collects the voltage ripple features and current ripple features of the electric energy metering box in real time and outputs an anti-theft electricity feature data set. Compared with the traditional method of only monitoring the current and voltage values, the anti-theft electricity feature recognition module can more comprehensively and meticulously capture the parameter change information in the electric energy metering process. The voltage ripple features and current ripple features contain the subtle fluctuation rules of the electric energy signal in the time dimension. Even if new electricity stealing methods such as tampering with the waveform and interfering with the metering chip are used, unique abnormal traces will be left on these ripple features, thereby providing more abundant and accurate basic data for subsequent electricity stealing behavior identification and helping to discover hidden electricity stealing operations that are difficult to detect by traditional monitoring methods.

[0020] The security policy matching module is connected with the anti-theft electricity feature recognition module and queries a preset electricity stealing mode feature library according to the anti-theft electricity feature data set to output a target security policy identifier and corresponding initial security policy parameters, thereby realizing rapid classification of electricity stealing behaviors and matching of preliminary response schemes. The preset electricity stealing mode feature library can integrate the feature rules corresponding to various known electricity stealing behaviors. When the anti-theft electricity feature data set collected by the system matches the features of a certain electricity stealing mode in the library, the possible electricity stealing type can be quickly located, and the initial security policy parameters for the type of electricity stealing behavior can be called, thereby avoiding the response delay caused by the need for reanalysis and judgment of different electricity stealing behaviors in traditional systems and improving the timeliness of the disposal of electricity stealing behaviors.

[0021] The dynamic parameter adjustment module dynamically corrects the initial security policy parameters based on the anti-theft electricity feature data set to generate real-time security policy parameters, thereby effectively solving the problems of poor adaptability and high false positive rate caused by the use of fixed thresholds or fixed strategies in traditional systems. In different electricity consumption scenarios, the normal voltage ripple features and current ripple features of the electric energy metering box will be different. For example, the ripple fluctuation rules of residential electricity consumption and industrial electricity consumption are different, and the electricity consumption ripple features of the same user will also change at different times. The dynamic parameter adjustment module can flexibly adjust the initial security policy parameters according to the real-time collected feature data and the actual situation of the current electricity consumption scenario, so that the security policy is always adapted to the current normal electricity consumption parameter fluctuation rules, the false alarm or missed alarm situations caused by improper parameter settings are reduced, and the adaptability of the system to complex electricity consumption environments is improved.

[0022] The intervention control execution module executes hierarchical intervention actions according to the real-time safety policy parameters, changes the limitation that the traditional system can only alarm but cannot actively dispose, and forms a complete anti-electricity-stealing process from monitoring, identification, analysis to intervention. The hierarchical intervention actions can be differentially set according to the severity, influence range and potential risk of the electricity-stealing behavior. For example, for slight and suspected electricity-stealing anomalies, light intervention actions such as data recording and abnormal marking can be performed to further observe the parameter change trend; for explicit and serious electricity-stealing behaviors, medium intervention actions such as cutting off the auxiliary power supply of the metering box and locking the data uploading function of the metering chip can be performed to prevent the electricity-stealing behavior from continuing; for the electricity-stealing behavior of maliciously damaging the metering equipment and causing large power loss, the power operation and maintenance platform can be linked to send on-site disposal instructions to start higher-level intervention measures. This hierarchical intervention method can not only avoid the influence of excessive intervention on normal power consumption, but also take effective measures according to the actual situation of the electricity-stealing behavior, maximally reduce the economic loss of the power enterprise, and maintain the stability of the power supply order. BRIEF DESCRIPTION OF DRAWINGS

[0023] Figure 1 A timing diagram of the electricity-stealing prevention intelligent monitoring system of the electric energy metering box according to the present application;

[0024] Figure 2 A flowchart for generating the electricity-stealing prevention feature data set;

[0025] Figure 3 A flowchart for generating the real-time safety policy parameters;

[0026] Figure 4 A flowchart for updating the strategy priority coefficient;

[0027] Figure 5 A flowchart for parameter difference verification. DETAILED DESCRIPTION

[0028] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0029] Please refer to Figure 1 The present application provides an electricity-stealing prevention intelligent monitoring system for an electric energy metering box, which comprises an electricity-stealing prevention feature identification module, a safety policy matching module, a dynamic parameter adjustment module and an intervention control execution module. The specific implementation is as follows:

[0030] The anti-theft electricity feature recognition module collects voltage ripple features and current ripple features of the electric energy metering box in real time and outputs an anti-theft electricity feature data set. The security policy matching module is connected with the anti-theft electricity feature recognition module, queries a preset electricity theft mode feature library according to the anti-theft electricity feature data set, and outputs a target security policy identifier and corresponding initial security policy parameters. The dynamic parameter adjustment module is connected with the security policy matching module, dynamically corrects the initial security policy parameters based on the anti-theft electricity feature data set, and generates real-time security policy parameters. The intervention control execution module is connected with the dynamic parameter adjustment module, and executes a hierarchical intervention action according to the real-time security policy parameters. Through the cooperative operation of the modules, the system realizes real-time monitoring and dynamic protection of the electric energy metering box.

[0031] Embodiment 1: refer to Figure 2 In the process of generating the anti-theft electricity feature data set, the anti-theft electricity feature recognition module performs multi-dimensional analysis of the voltage ripple features and the current ripple features. The voltage ripple features are captured in real time by a high-precision sampling circuit, and the signals are input into a digital signal processor after being converted by an analog-to-digital converter. In the voltage ripple analysis unit, the system first detects the time-domain waveform of the voltage signal and identifies the transient jump points in the waveform. By counting the number of jump points per unit time, a mutation frequency characteristic value is quantified. This characteristic value reflects the abnormal fluctuation intensity of the voltage signal. At the same time, the voltage waveform is compared point by point with a standard sinusoidal reference signal to calculate the root mean square value of the amplitude deviation of each sampling point, generating a distortion amplitude feature. This feature represents the severity of voltage waveform distortion.

[0032] The current ripple features are collected by a Rogowski coil sensor and input into the feature extraction channel after isolation amplification. The acquisition of the phase shift feature uses a double-channel correlation analysis method: the current signal and the reference voltage signal are input into a phase detection circuit, and the real-time phase angle is calculated by measuring the time difference between zero-crossing points. The system records the phase angle variation of ten consecutive power frequency cycles and takes the standard deviation as the phase shift characteristic value. The abnormal harmonic feature analysis uses the fast Fourier transform technique to perform frequency spectrum decomposition on the current signal. The system pays special attention to the frequency spectrum components other than the 3rd, 5th, and 7th integer harmonics, calculates the percentage of non-integer harmonic energy in total harmonic energy, and generates an abnormal harmonic characteristic value. The above four characteristic values are integrated into a four-dimensional feature vector by a data packaging unit, forming an anti-theft electricity feature data set. The data set uses a fixed-length data frame structure, each frame contains a timestamp, a characteristic value array, and a check code, and is transmitted to the security policy matching module through a serial communication interface.

[0033] The security policy matching module receives the anti-electricity theft feature data set, and starts a feature matching engine. The engine calls the historical electricity theft feature templates stored in the electricity theft mode feature library, each template containing a standardized feature vector and an associated policy identifier. The similarity weight value is calculated using the vector space model: the real-time feature vector is multiplied by each historical template vector, and then divided by the product of the vector module to obtain the cosine similarity coefficient. The coefficient is multiplied by the preset feature weight matrix, and the final output is a similarity weight value in the range of 0 to 1.

[0034] The system sets a dynamic similarity threshold, which is automatically adjusted according to the power grid load rate. When the similarity weight value exceeds the current threshold, the corresponding historical template is selected into the matching feature template set. After the template set is generated, the policy selection unit retrieves the policy priority coefficient associated with each template. The coefficient is maintained by the historical data analysis subsystem and dynamically updated according to the policy execution success rate and timeliness. The system adopts the maximum priority principle, selects the policy identifier corresponding to the template with the highest coefficient as the target security policy identifier, and outputs the initial security policy parameter set of the policy.

[0035] During the feature matching process, the system continuously monitors the stability of the feature vector. When the feature vector change of three consecutive sampling periods is less than a set value, the similarity threshold requirement is automatically increased to enhance the strictness of the matching. After the matching result is generated, the system encapsulates the target security policy identifier and the initial security policy parameters into a policy instruction package, and sends it to the dynamic parameter adjustment module through a parallel data bus.

[0036] The entire implementation process adopts a pipeline architecture, and the voltage and current signal acquisition, feature extraction, and policy matching are operated in parallel. The feature extraction unit adopts a double buffering mechanism to ensure data processing continuity. The policy matching engine realizes multi-threaded query, supporting simultaneous similarity calculation for 256 historical templates. CRC check and retransmission mechanism is adopted for data transmission to ensure the reliability of instruction transmission. The system completes a complete feature recognition and policy matching cycle every 200 milliseconds, realizing real-time monitoring and response.

[0037] Example 2: refer to Figure 3 During the generation of real-time security policy parameters, the dynamic parameter adjustment module performs a dynamic correction process of the initial security policy parameters. After receiving the target security policy identifier and the initial security policy parameter set from the security policy matching module, the module first extracts the key feature components from the anti-electricity theft feature data set. The acquisition of the ripple mutation proportion feature is realized through the voltage waveform analysis unit: the system sets a 200 millisecond sliding time window, and counts the number of mutation points of the voltage sampling value exceeding the ±2% rated voltage threshold within the window. The mutation point number is divided by the total sampling point number in the window to generate a ripple mutation proportion feature value in the range of 0 to 1. The feature value is updated every 50 milliseconds and recorded in a ring buffer.

[0038] The phase shift trend feature is extracted by phase sequence tracking technology. The current phase detection unit continuously outputs phase shift data, and the system retains the phase shift record of the last 20 power frequency cycles. The phase change amount of adjacent cycles is calculated by the first-order difference algorithm, and then the difference sequence is processed by five-cycle moving average to generate a characteristic value representing the phase change trend. The characteristic value is in units of degrees per second, and a positive value indicates that the phase is continuously lagging, and a negative value indicates that the phase is continuously leading.

[0039] The voltage compensation weight is calculated based on the ripple mutation proportion feature. The system presets a proportion-weight mapping table to divide the ripple mutation proportion feature value into five intervals: 0-0.2 interval maps weight coefficient 0.3, 0.2-0.4 interval maps 0.5, 0.4-0.6 interval maps 0.7, 0.6-0.8 interval maps 0.9, and 0.8-1.0 interval maps 1.0. When the feature value is at the interval boundary, the linear interpolation method is used to determine the maximum weight. The mapping relationship is fixed in the programmable read-only memory, and the parameters can be updated through the configuration interface. The generation of the current compensation weight depends on the phase shift trend feature. The system maintains a phase trend history queue with a length of 10, and processes the historical data using the exponential weighted moving average algorithm. The latest feature value is given a weight coefficient of 0.5, and the historical data weight decreases by a decay factor of 0.8. The weighted calculation result is input into the S-shaped curve conversion function, and the current compensation weight in the interval of 0 to 1 is output. The conversion function sets a saturation interval of ±15 degrees per second, and when the input value exceeds this range, the weight is automatically locked to the extreme value.

[0040] The generation of the real-time threshold parameter performs a weighted fusion operation, extracting the upper voltage threshold, the lower voltage threshold, and the current threshold from the initial safety policy parameter set. The voltage compensation weight acts on the voltage threshold parameter, multiplying the initial upper voltage threshold and lower voltage threshold by (1 + 0.5 x voltage compensation weight) and (1 - 0.5 x voltage compensation weight), respectively. The current compensation weight acts on the current threshold parameter, multiplying the initial current threshold by (1 + 0.6 x current compensation weight). The fused parameters are range-constrained by a convex combination algorithm: when the calculation result exceeds the safe operating range of the device, it is automatically truncated to the boundary value; when the parameter variation amplitude exceeds 20% of the previous value, the gradual adjustment mechanism is started, gradually transitioning to the target value within three operation periods. The final generated real-time safety policy parameter set includes the corrected upper voltage threshold, lower voltage threshold, current threshold, and intervention level parameters. The parameter set is in binary encoding format, including version number, parameter value, checksum, and other fields, and is transmitted to the intervention control execution module through a high-speed data bus. The system performs a complete parameter adjustment cycle every 100 milliseconds, implementing a dedicated operation pipeline in the FPGA chip: the feature extraction unit, weight calculation unit, and parameter fusion unit work in three parallel stages, with the output of the previous stage serving as the input of the next stage. The double-buffering mechanism is set in the data path to ensure continuous output without interruption. The parameter version management unit records the parameter serial number of each adjustment, supporting parameter rollback and operation traceability.

[0041] In terms of special working conditions, when the ripple mutation proportion characteristic value exceeds 0.9 for five consecutive times, the system automatically switches to the emergency correction mode. In this mode, the voltage compensation weight is directly set to the maximum value of 1.0, and the parameter update period is shortened to 20 milliseconds. When the grid frequency fluctuates by more than ±0.5 Hz, the phase shift trend feature calculation automatically switches to the frequency adaptive mode, using a dynamic reference frequency to recalculate the phase shift. The historical records of all corrected parameters are stored in the non-volatile memory to form a parameter adjustment log for diagnostic analysis.

[0042] Example 3: The system delay compensation module starts the compensation process when the anti-stealing electricity feature recognition module is working. The module calls the pre-stored metering response delay duration from the non-volatile memory, which is determined through laboratory calibration and field testing and includes three components: signal transmission delay, sampling and holding delay, and data processing delay. The metering response delay duration is stored in milliseconds, with a precision of microseconds. The time domain compensation of voltage ripple features and current ripple features is achieved using digital signal processing technology. The system establishes a sampling data queue, and the queue length is dynamically adjusted according to the delay duration. For each newly collected voltage and current sampling point, the compensation algorithm calculates its correct position on the time axis.

[0043] The time domain compensation process uses the following compensation formula:

[0044]

[0045] wherein: denotes the current sample index, is the compensated signal value, is the historical sample value, is the integer delay component, is the filter order, is the window function coefficient. The formula realizes signal reconstruction by weighted moving average, and the window function adopts Kaiser window design, and the main lobe width and sidelobe attenuation are configured according to the signal characteristics.

[0046] The compensated ripple feature is output to the electricity larceny prevention feature recognition module. After the voltage signal is compensated, the phase distortion degree of the waveform is reduced, and the time positioning accuracy of the mutation point is improved. After the current signal is compensated, the phase relationship of the harmonic component is corrected, and the reliability of the non-integer harmonic detection is enhanced. The compensation module performs delay calibration every 8 milliseconds, and automatically adjusts the delay parameter according to the change of the environment temperature. The temperature sensor monitors the temperature of the circuit board in real time, and the temperature-delay correction curve is stored in the lookup table.

[0047] After the intervention control execution module receives the real-time safety policy parameter, the intervention level identifier in the parameter is analyzed. When the parameter contains the primary intervention identifier, the module sends an activation instruction to the electromagnetic locking device of the meter box protection shell. The electromagnetic locking device includes a driving circuit and a mechanical lock body. The driving circuit generates a strong magnetic field after receiving a 12V DC pulse signal, and pushes the lock tongue into the locking groove. The locking state is detected by a Hall sensor, and the feedback signal is returned to the execution module. The locking device is designed as a power-off self-locking type, and will only be released when a specific unlocking instruction is received. When the real-time safety policy parameter contains the intermediate intervention identifier, the module triggers the breaking control circuit of the metering circuit. The breaking control circuit adopts a double-redundancy design, and the main circuit uses a large-capacity magnetic latching relay, and the standby circuit uses a solid-state switch. The relay coil driving voltage is 24V, and the attraction time is less than 10 milliseconds. The solid-state switch uses an IGBT device, with a rated current of 100A and a breaking time of less than 500 microseconds. The breaking instruction is sent to the main and standby circuits at the same time, and the state monitoring circuit detects the breaking execution in real time. After breaking, the circuit impedance monitoring unit verifies the open state to prevent virtual connection or sticking phenomenon.

[0048] When the real-time security policy parameter contains the advanced intervention identifier, the module activates the wireless alarm unit of the positioning tracking device. The wireless alarm unit contains a cellular communication module and a GPS positioning chip. The cellular module supports the 4G LTE standard, has a built-in SIM card, and sends alarm information in the form of a data packet in a specific format through the mobile network. The GPS chip obtains latitude and longitude coordinates with a positioning accuracy of meters. After the alarm unit is started, it first establishes a network connection and then sends alarm data containing the device number, location information, timestamp, and event type. The alarm data is protected by the AES encryption algorithm, and integrity verification is enabled during transmission.

[0049] During the execution of the hierarchical intervention action, the system maintains state monitoring. The operating current of the electromagnetic locking device is sampled in real time, and a retry mechanism is started when the current is abnormal. The breaking state of the breaking control circuit is monitored by a voltage probe, and the line voltage should drop to a safe range after breaking. The sending state of the wireless alarm unit is confirmed by the network response, and the standby frequency band is automatically switched when the sending fails. All execution actions record detailed logs, including instruction sending time, execution result, device state, and other parameters. Log data is stored in ferroelectric memory in a loop, with a storage period of not less than 30 days. The system sets an intervention action interlocking mechanism, with a 100-millisecond delay between primary intervention and intermediate intervention to prevent action conflicts. When the advanced intervention is triggered, the undo function of other intervention actions is automatically disabled. The execution module uses a multi-core processor architecture, with separate cores for instruction analysis, action execution, and state monitoring. The communication interface uses an optoelectronic isolation design to enhance anti-interference capability. The power management system provides multiple independent power supplies to ensure stable control circuit voltage during high-current breaking actions.

[0050] The execution module also has a self-checking function that detects parameters such as electromagnetic locking device resistance, breaking circuit contact resistance, and wireless signal strength daily at a fixed time. The self-checking results are compared with historical data, and a warning message is generated if the deviation exceeds the limit. The maintenance interface supports parameter configuration and action testing, and the intervention action in test mode is replaced by an indicator light display without actual mechanical operation. All configuration changes require double authentication, and operation logs are permanently saved.

[0051] Example 4: refer to Figure 4 After completing the initial feature matching, the security policy matching module starts a real-time tracking mechanism to continuously monitor the change rate feature of the anti-theft electricity feature data set. This feature is obtained by calculating the Euclidean distance change of the feature vector in consecutive sampling periods, and the system sets a sliding window with a length of 5, calculating the average change rate of the feature vector in the window every 100 milliseconds. The change rate feature value is divided into five levels, corresponding to different policy adjustment modes. When the change rate feature value is detected to be continuously rising, the module starts the dynamic update program of the policy priority coefficient.

[0052] The updating of the strategy priority coefficient is based on the correlation analysis of the change rate characteristic value and the historical matching template. The system maintains a priority coefficient mapping table to record the coefficient adjustment range corresponding to different change rate intervals. The mapping table is continuously optimized according to the field operation data, and the latest version of the mapping relationship is shown in Table 1.

[0053] Table 1: Change rate characteristics and strategy priority coefficient adjustment mapping relationship.

[0054]

[0055] The updated strategy priority coefficient is fed back to the dynamic parameter adjustment module in real time through a dedicated data channel. The feedback data includes coefficient value, update timestamp, and expiration date, and adopts differential transmission to reduce data volume. The dynamic parameter adjustment module recalculates the weight parameters according to the new priority coefficient to realize the coordinated updating of the security strategy parameters.

[0056] The intervention control execution module synchronously starts the vibration monitoring function during the execution of the hierarchical intervention action. The three-axis MEMS acceleration sensor continuously collects the vibration signals of the metering box protective shell at a sampling frequency of 2 kHz. The vibration frequency characteristic extraction adopts digital filtering technology. After the signal passes through a 0.1-500 Hz band-pass filter, frequency domain feature analysis is performed. The system calculates the main frequency component of the vibration signal every 50 milliseconds, and simultaneously counts the energy distribution in the 0-100 Hz frequency band.

[0057] The vibration safety threshold is set in a multi-level configuration manner. The primary threshold is set to an acceleration amplitude of 0.5 m / s² for general abnormal vibration; the intermediate threshold is set to 2.0 m / s² for obvious destructive vibration; and the high-level threshold is set to 5.0 m / s² for serious destructive behavior. When the vibration frequency characteristic value exceeds the currently set safety threshold, the system starts the intervention level upgrade program. The intervention identification upgrade process follows a strict logical judgment process. When the vibration characteristic value exceeds the intermediate threshold for 3 consecutive sampling periods, the system automatically upgrades the intermediate intervention identification to the high-level intervention identification. After the upgrade instruction is generated, the intermediate intervention action currently being executed is first paused, and then the execution sequence is reinitialized according to the high-level intervention process. The mechanical lock remains continuously effective during the upgrade process to avoid the occurrence of a security protection window period.

[0058] The vibration monitoring data is correlated with the intervention execution status, and the system records the time, amplitude, frequency characteristics of each vibration overrun event and the corresponding intervention action type. These data are used to optimize the vibration threshold parameters, and the vibration pattern and illegal electricity stealing behavior are established by machine learning algorithm. Long-term operation data shows that the vibration signal in a specific frequency range has a high correlation with the illegal opening of the metering box. The system uses a double verification mechanism to prevent false escalation. When the vibration characteristics overrun, it needs to meet the spatial consistency condition: the acceleration measurement values of the three axes need to maintain a certain proportional relationship to exclude random vibration interference. Before the escalation instruction is executed, the system will be delayed for 20 milliseconds for secondary verification to confirm the persistence and regularity of the vibration characteristics. The escalation operation records detailed logs, including vibration data snapshots, decision basis, execution results, etc. The vibration monitoring unit has a self-calibration function. The zero-point calibration program is automatically executed at zero o'clock every day to eliminate sensor drift errors. Sensitivity calibration is performed once every quarter to verify the measurement accuracy using a standard vibration source. Calibration data are stored in a separate memory, and calibration history can be traced and queried. When the sensor is abnormal, the system automatically switches to the backup sensor and generates a maintenance alert.

[0059] After the intervention level is upgraded, the system enters the enhanced monitoring mode, the vibration sampling frequency is increased to 5 kHz, and the data analysis window is shortened to 20 milliseconds. The wireless alarm unit starts continuous transmission mode, and the position information update frequency is increased from once every minute to once every 10 seconds. At the same time, the peripheral device linkage function is started, and a warning signal is sent to the adjacent metering box to form a regional protection network.

[0060] Example 5: see Figure 5 The dynamic parameter adjustment module performs a running state verification process before generating real-time security policy parameters. This module obtains the current running state parameters through the monitoring unit built-in the metering box, including voltage effective value, current effective value, active power, reactive power, power factor, and frequency measurement value. The voltage effective value measurement uses a true effective value conversion chip with a sampling period of 10 milliseconds and a measurement accuracy of 0.2 level. The current effective value is obtained through a Hall sensor array, and the sensor output is amplified through an isolation amplifier and then enters a 16-bit analog-to-digital converter. The power calculation unit uses a time division multiplier principle to calculate the active power and reactive power in real time. All running state parameters are updated every 20 milliseconds and transmitted to the dynamic parameter adjustment module in the form of data packets.

[0061] The parameter difference calculation adopts a multi-dimensional comparison algorithm. The system compares the current operating state parameters with the corresponding items in the initial safety policy parameters one by one. The voltage difference calculation calculates the relative deviation percentage of the current voltage effective value and the initial voltage threshold value. The current difference calculation calculates the relative deviation percentage of the current current effective value and the initial current threshold value. The power factor difference calculation calculates the deviation degree of the current power factor value and the initial power factor range. The system sets a preset tolerance parameter group, including voltage tolerance percentage, current tolerance percentage, and power factor tolerance value. When any one of the differences exceeds the corresponding preset tolerance, the parameter difference verification process is triggered.

[0062] After the parameter difference verification process is started, the system accesses the distributed database to retrieve the baseline parameter set under historical normal working conditions. The baseline parameter set includes statistical values of operating parameters under the same time period and the same load condition in the past 30 days, including voltage average value range, current average value range, power factor typical value, and its allowed fluctuation range. Data retrieval is based on multi-dimensional indexes such as time stamp, load type, and environment temperature to ensure the applicability of the baseline parameters. The baseline parameter set is stored in compressed data format and restored to complete parameter structure after decompression.

[0063] The system regards the current operating state parameter group and the baseline parameter group as two points in a multi-dimensional space and calculates their relative distance in the multi-dimensional feature space. This calculation considers the different dimensions and importance of each parameter, giving higher weight factors to key parameters such as voltage and current. The correlation between parameters is also considered in the calculation process to avoid repeated calculation of parameters that depend on each other. The deviation coefficient output value is a dimensionless value between 0 and 1, and the smaller the value, the closer to the baseline state.

[0064] The preset abnormality judgment value is determined according to power grid operation experience and historical data analysis. The judgment value sets multiple levels: below 0.1 represents a completely normal state, 0.1 to 0.3 represents a slight deviation state, 0.3 to 0.5 represents a moderate deviation state, and above 0.5 represents a significant abnormal state. When the calculated deviation coefficient is less than 0.3, the system determines that the current operating state belongs to normal working condition fluctuation, not abnormality caused by electricity stealing. At this time, the system replaces the initial safety policy parameters with the baseline parameter set as the input reference of the real-time safety policy parameters.

[0065] The parameter replacement operation adopts a smooth transition method. The system does not immediately switch the parameters completely, but gradually adjusts the parameter values from the initial values to the baseline values within 5 calculation periods. Each period adjusts 20% of the difference to avoid system oscillation caused by parameter mutation. During the parameter replacement process, the system continuously monitors the change trend of the operating state parameters. If a new deviation is found between the replaced parameters and the current operating state, the replacement will be paused and re-evaluated.

[0066] The management of the benchmark parameter set adopts a rolling update mechanism. The system automatically updates the benchmark parameter set every morning, and the normal operation data of the same period in the past 30 days is included in the statistical range, and the data of the earliest day is excluded. During the update process, a weighted average algorithm is used, and recent data is given a higher weight to ensure that the benchmark parameters can reflect the latest running state. All benchmark parameter update records are saved in the audit log, including update time, parameter value before update, parameter value after update, and other detailed information. The system also has a benchmark parameter validity verification program. Before each use of the benchmark parameter, its statistical significance is checked, and the sample size of the benchmark parameter set is required to meet the minimum statistical requirement, and the parameter distribution meets the normal distribution characteristics. If the benchmark parameter set does not meet the use conditions, the system will return to use the initial security policy parameter, and generate a warning information of insufficient benchmark data. The warning information is sent to the maintenance center, prompting the need to supplement the normal operation data of the period.

[0067] The entire parameter verification process adopts a redundant computing architecture, the main processor performs the main computing task, and the coprocessor performs the same calculation in parallel for result verification. The calculation results of the two processors are compared in real time, and when the difference exceeds the allowed range, a third-party arbitration program is started. All computing processes have detailed process records, including input parameters, intermediate results, final conclusions, etc. These records are saved in the circular buffer and can be used for post-analysis and audit.

[0068] It should be noted that, in the present document, relational terms such as first and second and the like can be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0069] While embodiments of the application have been shown and described, it is to be understood that the application is not limited to the details of the embodiments described, since numerous modifications and changes can be made to the embodiments without departing from the spirit and scope of the application as defined by the appended claims and their equivalents.

Claims

1. An electricity metering box anti-stealing electricity intelligent monitoring system, characterized in that, Comprise: The anti-electricity theft feature recognition module is used for real-time collection of voltage ripple features and current ripple features of the electric energy metering box, and outputs an anti-electricity theft feature data set; the security strategy matching module is connected with the anti-electricity theft feature recognition module, is used for querying a preset electric theft mode feature library according to the anti-electricity theft feature data set, and outputs a target security strategy identifier and corresponding initial security strategy parameters; the dynamic parameter adjustment module is connected with the security strategy matching module, is used for dynamic correction of the initial security strategy parameters based on the anti-electricity theft feature data set, and generates real-time security strategy parameters; the intervention control execution module is connected with the dynamic parameter adjustment module, and is used for executing hierarchical intervention actions according to the real-time security strategy parameters; The dynamic parameter adjustment module generates the real-time security strategy parameters by the following mode: acquiring a ripple mutation proportion feature and a phase offset trend feature in the anti-electricity theft feature data set; calculating a voltage compensation weight according to the ripple mutation proportion feature; calculating a current compensation weight according to the phase offset trend feature; weighting and fusing threshold value parameters in the initial security strategy parameters by using the voltage compensation weight and the current compensation weight to obtain real-time threshold value parameters as a component part of the real-time security strategy parameters.

2. The electricity metering box anti-stealing electricity intelligent monitoring system according to claim 1, characterized in that, The anti-electricity theft feature recognition module generates the anti-electricity theft feature data set by the following mode: extracting a mutation frequency feature and a distortion amplitude feature in the voltage ripple feature; extracting a phase offset feature and an abnormal harmonic feature in the current ripple feature; combining the mutation frequency feature, the distortion amplitude feature, the phase offset feature and the abnormal harmonic feature into a multi-dimensional feature vector as the anti-electricity theft feature data set.

3. The electricity metering box anti-stealing electricity intelligent monitoring system according to claim 1, characterized in that, The security strategy matching module outputs the target security strategy identifier by the following mode: calculating a similarity weight value of the anti-electricity theft feature data set and each historical electric theft feature template in the electric theft mode feature library; selecting all historical electric theft feature templates with a similarity weight value exceeding a preset threshold value as a matching feature template set; determining the target security strategy identifier according to a strategy priority coefficient corresponding to the matching feature template set.

4. The electricity metering box anti-stealing electricity intelligent monitoring system according to claim 1, characterized in that, Further comprise: The system delay compensation module is used for calling a pre-stored metering response delay duration when the anti-electricity theft feature recognition module works; performing time domain compensation on the voltage ripple feature and the current ripple feature according to the metering response delay duration, and outputting the compensated ripple feature to the anti-electricity theft feature recognition module.

5. The electricity metering box anti-stealing electricity intelligent monitoring system according to claim 1, characterized in that, The intervention control execution module executes the hierarchical intervention actions by the following mode: when the real-time security strategy parameters contain a primary intervention identifier, starting an electromagnetic locking device of a metering box protective shell; when the real-time security strategy parameters contain a middle-level intervention identifier, triggering a breaking control circuit of a metering loop; when the real-time security strategy parameters contain a high-level intervention identifier, activating a wireless alarm unit of a positioning tracking device.

6. The electricity metering box anti-stealing electricity intelligent monitoring system according to claim 3, characterized in that, The security policy matching module is further configured to track a change rate feature of the anti-theft electricity feature data set in real time, update a policy priority coefficient of the matching feature template set according to the change rate feature, and feed back the updated policy priority coefficient to the dynamic parameter adjustment module.

7. The electricity metering box anti-stealing electricity intelligent monitoring system according to claim 5, characterized in that, The intervention control execution module is further configured to continuously collect a vibration frequency feature of the meter protection shell when the hierarchical intervention action is executed, and upgrade the medium-level intervention identifier to the high-level intervention identifier when the vibration frequency feature exceeds a preset security threshold. 8.The electric energy metering box anti-stealing electricity intelligent monitoring system according to claim 1, characterized in that, The dynamic parameter adjustment module is further configured to acquire a current operating state parameter of the electric energy metering box before generating the real-time security policy parameter, and start a parameter difference verification procedure when a difference degree between the current operating state parameter and the initial security policy parameter exceeds a preset tolerance. The parameter difference verification procedure includes: calling a benchmark parameter set under historical normal working conditions; calculating a deviation coefficient of the current operating state parameter and the benchmark parameter set; and replacing the initial security policy parameter with the benchmark parameter set as an input benchmark of the real-time security policy parameter when the deviation coefficient is less than a preset abnormality determination value.

9. An electricity metering box anti-stealing electricity intelligent monitoring control method, characterized in that, All modules and method procedures of the electric energy metering box anti-theft electricity intelligent monitoring control system according to any one of claims 1 to 8 are included.

Images