Intelligent digital electric box electricity utilization monitoring management system based on Internet of Things

Abstract

The invention relates to the technical field of intelligent monitoring of power distribution systems, and discloses an intelligent digital electric box power utilization monitoring management system based on the Internet of Things. The system comprises a data acquisition module, a data processing and analysis module, an intelligent control module, an Internet of Things communication module and a user interaction and management module. The data acquisition module acquires electrical and environmental parameters in real time; the data processing and analyzing module constructs an electricity utilization fingerprint database of the equipment, calculates a real-time threshold value through a dynamic threshold value engine, and realizes equipment-level anomaly recognition in combination with an anomaly detection engine; the intelligent control module executes a control action according to a detection result; the Internet of Things communication module realizes two-way data transmission; and the user interaction and management module provides a visual operation interface. According to the invention, through a dynamic threshold mechanism, equipment fingerprint identification and multi-level early warning control, equipment-level hidden faults are accurately identified, and the power utilization safety and the intelligent management level are improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent monitoring technology for power distribution systems, and in particular to a smart digital electrical box power monitoring and management system based on the Internet of Things. Background Technology

[0002] Traditional low-voltage power distribution systems have long relied on mechanical protection devices and basic electrical parameter monitoring for power consumption monitoring, resulting in systemic flaws in their technical architecture. At the hardware level, existing solutions mostly employ branch-level total current monitoring combined with fixed-threshold circuit breakers, capable only of responding to overloads and short circuits, lacking the ability to identify equipment-level anomalies, such as current waveform distortion caused by partial short circuits in motor windings. At the data analysis level, while mainstream smart distribution boxes have data uploading capabilities, their static threshold mechanisms cannot adapt to dynamic scenarios such as changes in environmental temperature and humidity, and the coordinated operation of multiple devices, leading to frequent false disconnections in high-temperature environments or missed alarms when multiple devices are started simultaneously. Regarding functional scalability, existing systems generally lack the ability to build equipment fingerprint databases, unable to identify equipment types through current harmonic characteristics or establish equipment health assessment models based on historical data, making predictive maintenance difficult.

[0003] Furthermore, existing interactive systems are mostly limited to displaying fault status, lacking both dynamic threshold configuration interfaces and waveform comparison functions required for anomaly tracing, making it difficult for users to quickly locate the source of the fault when faced with warnings. More importantly, the functional modules in existing technologies are isolated: environmental sensor data is not used in threshold calculations, communication modules only transmit data without implementing security authentication, and there is a lack of linkage mechanisms between the control module and the fire protection system, forming a technical break in the closed-loop management of safety hazards. These deficiencies collectively limit the effectiveness of existing systems in key scenarios such as electrical fire prevention and energy efficiency optimization. Summary of the Invention

[0004] The technical problem to be solved by this invention is that the existing power distribution system has safety hazards. To address this, we propose a smart digital power box power monitoring and management system based on the Internet of Things.

[0005] To achieve the above objectives, this application adopts the following technical solution: an IoT-based smart digital electrical box power monitoring and management system, comprising:

[0006] Data acquisition module: used to collect electrical and environmental parameters of branch circuits and key nodes of low-voltage power distribution system in real time. Electrical parameters include current, voltage, and power, and environmental parameters include temperature and humidity.

[0007] Data processing and analysis module: Deployed in the industrial-grade MCU built into the smart electrical box; used to extract features from the collected data to build the device's "electricity fingerprint database", and calculate real-time thresholds through a dynamic threshold engine, combined with an anomaly detection engine to achieve device-level anomaly identification;

[0008] Intelligent control module: used to execute control actions based on anomaly detection results;

[0009] IoT communication module: used to enable bidirectional data transmission between modules and with external terminals and cloud platforms;

[0010] User interaction and management module: This includes a display screen on the surface of the smart electrical box, which provides a visual interface for users to configure, monitor and control the system.

[0011] Preferably, the data acquisition module includes:

[0012] Electrical parameter sensors are used to collect instantaneous current, current waveform, line voltage, and phase voltage, and to calculate active power and reactive power based on current and voltage.

[0013] Environmental and condition sensors are used to collect contact point temperature, leakage current, and high-frequency pulse signals in the current waveform;

[0014] The data preprocessing unit is used to filter, periodically self-calibrate and perform analog-to-digital conversion on the raw data, and temporarily store high-frequency sampled data.

[0015] Preferably, the "electricity consumption fingerprint database" in the data processing and analysis module is constructed for:

[0016] Extract the steady-state characteristics, transient characteristics and periodic characteristics of the equipment. The steady-state characteristics include the effective value of current, power value and power factor during normal operation. The transient characteristics include the peak current, start-up time and harmonic content during startup. The periodic characteristics include the start-up and shutdown cycle and power fluctuation range.

[0017] Based on the branch device type manually entered by the user, the device “electricity fingerprint” is automatically generated through high-frequency sampling for 7-14 days. When the device is replaced or added, the fingerprint database is updated by relearning through the sudden change characteristics of the current waveform.

[0018] Preferably, the dynamic threshold engine in the data processing and analysis module is used to calculate real-time thresholds based on the following factors:

[0019] Environmental Correction Factors Based on the current ambient temperature T and humidity H, through a function calculate, It is a monotonically increasing function; the higher the temperature, the greater the humidity. The larger;

[0020] Behavioral baseline Based on historical habits and normal ranges, the electricity load of the same branch circuit exhibits regularity at different times. By constructing a behavioral baseline using historical data, we can reflect the normal range of electricity consumption. ,in This represents the average current for the same period in history, and the data is sourced from locally stored historical current data. This represents the standard deviation of current for the same period in history. This is the confidence level coefficient, set according to the stability of the power consumption scenario; higher stability results in a higher confidence level. The smaller the value, the larger the value; when the current is detected multiple times consecutively... If the device is otherwise functioning normally, and a user adds a new device, the baseline will be automatically updated. ,in Weights for the new data, To ensure the baseline adapts slowly to user habits, It is a reference for the normal current range at a certain moment, and more than 95% of normal electricity consumption behavior should fall within this range;

[0021] Equipment synergy factor , Due to the cumulative effect of multiple devices operating simultaneously, the total current will temporarily increase when multiple devices on a branch start up at the same time. Therefore, the threshold for the device coordination factor needs to be relaxed to avoid misjudgment. ,in For the i-th device within the branch, the data source is the "electricity consumption fingerprint database". This is an indicator function; it is 1 when the device is running and 0 when it is stopped, determined by current characteristics. For equipment The weight is calculated as the ratio of the rated power of the equipment to the total rated power of the branch; the greater the power, the higher the weight. This reflects the cumulative effect of active devices on a branch line, such as multiple high-power devices operating simultaneously. Increase the threshold accordingly to accommodate temporary high loads; as equipment is gradually shut down... The threshold decreases, returning to normal.

[0022] Time decay factor This is used to measure the impact of time periods on electricity consumption characteristics. Electricity load characteristics differ at different times, and the sensitivity of the threshold needs to be adjusted using a time decay factor. For piecewise functions: Peak electricity consumption periods: Larger values ​​result in a more relaxed threshold; off-peak electricity consumption periods: The value is small, and the threshold is strict; the time decay factor is matched with the user's daily routine.

[0023] Final dynamic threshold Based on the above factors, real-time device-level and tributary-level thresholds are generated: .

[0024] Preferably, the anomaly detection engine in the data processing and analysis module employs a three-level criterion linkage detection:

[0025] Primary criterion: Real-time current of the device The current is compared with the device-level dynamic threshold to determine whether the current is abnormal. The device-level dynamic threshold is based on the steady-state current in the device fingerprint. and environmental correction factors Through formula Calculation, where Safety factor for equipment type; for example, a certain heating equipment. This is its normal operating current. If the measured current consistently exceeds this value... If so, it is determined to be an abnormal current.

[0026] Secondary criterion: Whether the harmonic distortion rate (THD) of the detection equipment exceeds [the specified threshold]. and power fluctuations Does it exceed ,in These are normal harmonic values ​​from the fingerprint database. To allow for deviation, The steady-state power in the fingerprint database, To allow for fluctuation ratios; for example, in the fingerprints of motor-type devices. Higher than normal, if the measured THD level is significantly higher than normal. This may be due to a short circuit in the motor windings; power fluctuations in heating equipment are usually small. Exceeding This may be due to poor contact of the heating element;

[0027] Level 3 criterion: Detect whether abnormal current and leakage characteristics occur simultaneously; leakage characteristics are branch leakage current. Exceed ,in This is the normal leakage current baseline. Allowable deviation.

[0028] Preferably, the data processing and analysis module also includes an early warning level classification unit, used for:

[0029] Level 1 Warning: When a single parameter deviates slightly from the threshold and the temperature is normal, only an APP push notification will be triggered;

[0030] Level 2 alert: When multiple parameters are abnormal, the app will issue a notification and a local audible and visual alarm, and the system will continue to monitor the situation.

[0031] Level 3 warning: When a high-risk anomaly occurs, the circuit of the socket where the faulty device is located will be automatically cut off and an emergency alarm will be pushed.

[0032] Preferably, the intelligent control module includes:

[0033] Actuators, including smart circuit breakers installed in the main circuit of branch circuits and smart relays in high-risk equipment sockets;

[0034] The control logic unit is used to maintain power supply during a level 1 warning, continuously track abnormal trends during a level 2 warning, and trigger an intelligent relay to cut off the power supply to the faulty equipment during a level 3 warning.

[0035] Preferably, the IoT communication module includes:

[0036] The local communication unit uses an industrial bus to connect the sensors inside the electrical box with the edge computing unit to transmit raw collected data.

[0037] The remote communication unit uses 4G, WiFi, Ethernet, RS485, etc. to transmit analysis results and control commands to the cloud and user terminals;

[0038] The security encryption unit uses a symmetric encryption algorithm to encrypt transmitted data and uses a unique identifier to authenticate the device.

[0039] Preferably, the user interaction and management module includes:

[0040] The local interaction unit displays real-time status and warning information, and provides manual control buttons via a touch screen built into the electrical box.

[0041] The remote interaction unit provides functions such as device status visualization, historical data query, threshold configuration, device management, and remote control via computer, tablet, mobile phone, or web platform.

[0042] Preferred cloud platforms include:

[0043] The data storage unit is used to store 30 days of high-frequency data and 36 months of historical data in the cloud. The historical data includes raw collected data, anomaly records, and device health index.

[0044] The deep analysis unit is used to calculate the equipment health index H using the formula. Implementation, in which For the number of abnormal occurrences, For current exceeding the limit, the weighting coefficients a and b are: a reflects the impact of abnormal frequency, and b reflects the impact of abnormal severity. The lower the H value, the higher the risk of equipment aging or failure.

[0045] The model optimization unit updates the "electricity fingerprint" recognition algorithm quarterly through big data training.

[0046] The technical effects and advantages of this invention are as follows:

[0047] In this invention, data acquisition, processing and analysis, intelligent control, communication and user interaction are integrated. The integrated data acquisition module obtains electrical and environmental parameters in real time, and the data processing and analysis module constructs a device power fingerprint database and calculates dynamic thresholds. Combined with multi-level anomaly detection and intelligent control modules, accurate fault identification and response are achieved. It has the advantages of adapting to environmental changes and multi-device collaborative scenarios through a dynamic threshold mechanism, constructing a device power fingerprint database to achieve accurate anomaly identification, and integrating multi-level early warning and intelligent control to improve safety. Attached Figure Description

[0048] The disclosure of this invention is illustrated with reference to the accompanying drawings. It should be understood that the drawings are for illustrative purposes only and are not intended to limit the scope of protection of this invention. In the drawings, the same reference numerals are used to refer to the same parts:

[0049] Figure 1 This is the structural topology diagram of the present invention;

[0050] Figure 2 This is a block diagram of the module structure of the present invention. Detailed Implementation

[0051] It is readily understood that, based on the technical solution of this invention, those skilled in the art can propose various interchangeable structural methods and implementations without altering the essential spirit of the invention. Therefore, the following detailed embodiments and accompanying drawings are merely illustrative examples of the technical solution of this invention and should not be considered as the entirety of the invention or as limitations or restrictions on the technical solution of this invention.

[0052] Reference Figures 1-2 As shown, this invention provides a technical solution: an IoT-based smart digital electrical box power monitoring and management system, including a data acquisition module, a data processing and analysis module, an intelligent control module, an IoT communication module, and a user interaction and management module. The data acquisition module collects electrical and environmental parameters of branch circuits and key nodes of the low-voltage power distribution system in real time. The data processing and analysis module is deployed in an industrial-grade MCU built into the smart electrical box, extracting features from the collected data to construct a power consumption fingerprint database for the device, and calculating real-time thresholds through a dynamic threshold engine, combined with an anomaly detection engine to achieve device-level anomaly identification. The intelligent control module executes control actions based on the anomaly detection results. The IoT communication module enables bidirectional data transmission between modules and with external terminals and cloud platforms. The user interaction and management module includes a display screen on the surface of the smart electrical box, providing a visual interface for users to configure, monitor, and control the system.

[0053] The smart electrical box is equipped with a display screen, which allows users to control the on / off status of each branch circuit via touch buttons on the screen, and also allows users to view the electrical data of each branch circuit in real time.

[0054] This application further proposes a data acquisition module including an electrical parameter sensor for acquiring instantaneous current, current waveform, line voltage, and phase voltage, and calculating active power and reactive power based on current and voltage; an environmental and state sensor for acquiring contact point temperature, leakage current, and high-frequency pulse signals in the current waveform; and a data preprocessing unit for filtering, periodically self-calibrating, and performing analog-to-digital conversion on the raw data, and temporarily storing high-frequency sampled data.

[0055] The electrical parameter sensors are measurement devices deployed at critical nodes of the circuit. Specifically, they employ high-precision Hall effect sensors combined with differential voltage probes. These sensors acquire instantaneous current and voltage waveform data by directly contacting the conductor between the live and neutral wires, and calculate power parameters based on Ohm's law. The environmental and status sensors are composite sensors integrating temperature detection and leakage current monitoring. They utilize thermocouple arrays combined with leakage current transformers. Installed inside the electrical box near connectors and branch live wires, they monitor contact point temperature rise and leakage current signals in real time, while simultaneously capturing high-frequency pulses in the current waveform to identify arc faults. The data preprocessing unit, a circuit module for noise reduction and format conversion of the raw signal, employs digital filters combined with a self-calibration circuit. It eliminates instantaneous interference through moving average filtering, periodically triggers zero-point calibration to compensate for sensor drift, and converts analog signals into digital signals before storing them in a buffer to ensure the stability and integrity of subsequent analysis data.

[0056] Specifically, electrical parameter sensors are distributed across the branch circuit live wires and key nodes of the main branch. By synchronously acquiring instantaneous current and voltage waveforms, they can accurately calculate the active and reactive power of each branch. Environmental and status sensors are installed inside the electrical box at the joints and on the surface of the branch live wires. By monitoring the temperature changes and leakage current intensity at the contact points, combined with high-frequency pulse signal analysis, they can identify abnormal heating or arc discharge caused by poor contact or insulation aging. The data preprocessing unit processes the raw signals output by the sensors in real time. For example, it uses a Butterworth low-pass filter to eliminate high-frequency noise interference and uses an analog-to-digital converter to quantize the analog signal into a digital signal. At the same time, it periodically triggers a self-calibration program, such as performing zero-point calibration every 30 minutes, to eliminate drift errors caused by long-term sensor operation. The preprocessed data is temporarily stored in a high-speed cache to provide high-precision input for subsequent feature extraction.

[0057] This application further proposes the construction of an electricity fingerprint database in the data processing and analysis module, including the extraction of steady-state characteristics, transient characteristics, and periodic characteristics of the equipment. The steady-state characteristics include the effective value of current, power value, and power factor during normal operation. The transient characteristics include the peak current, start-up time, and harmonic content during startup. The periodic characteristics include the start-up and shutdown cycle and power fluctuation range. Based on the branch equipment type manually entered by the user, the electricity fingerprint of the equipment is automatically generated through high-frequency sampling for 7-14 days. When the equipment is replaced or added, the fingerprint database is updated by relearning triggered by the sudden change characteristics of the current waveform.

[0058] Among them, steady-state characteristics are parameters such as current, power, and power factor of the equipment under stable operating conditions, which are collected and calculated by current transformers and power analysis modules, and are used to characterize the baseline power consumption behavior of the equipment during normal operation; transient characteristics are dynamic parameters such as current peak value and harmonic content generated during equipment start-up or shutdown, which are extracted by high-speed sampling circuits and fast Fourier transform algorithms, and are used to identify abnormal fluctuations during equipment start-up and shutdown; periodic characteristics are the recurring start-up and shutdown intervals and power change patterns during equipment operation, which are mined from historical data by time series analysis algorithms, and are used to determine whether the equipment deviates from the predetermined operating mode; high-frequency sampling automatically generates power consumption fingerprints by continuously collecting equipment operation data, such as recording current waveforms at a sampling rate of more than 1000 times per second, and combining them with machine learning algorithms to generate a unique power consumption feature template for the equipment; current waveform mutation feature triggers relearning, which means that when a mutation signal that does not match the fingerprint database is detected in the branch current waveform, such as when the waveform distortion rate exceeds the preset threshold due to the addition of new equipment, the fingerprint database update process is automatically initiated.

[0059] Specifically, the construction of the electricity fingerprint database achieves device-level electricity behavior modeling through multi-dimensional feature extraction and adaptive learning mechanisms. Steady-state features are used to establish baseline parameters for normal device operation; for example, the effective current value of an air conditioner during operation is maintained between 5A and 6A. Transient features capture the current surge characteristics at the moment of device startup; for example, the peak current of a motor can reach more than three times the rated value when starting. If an abnormal increase in the peak value is detected, it may indicate a winding short circuit. Periodic features analyze the operating patterns of the device; for example, if a device starts every 2 hours and runs continuously for 15 minutes, an abnormally shortened cycle may indicate an overload risk. Through high-frequency sampling of 7-14 days, the system can completely record the electricity consumption patterns of the device under different operating conditions. For example, data is collected during the daily peak electricity consumption period to cover the full load state of the device. When the user manually enters the device type, the system automatically associates it with preset feature extraction rules; for example, lighting devices do not need to analyze the start-stop cycle, while motor devices need to focus on monitoring harmonic content. When a device is replaced or added, the sudden change in current waveform features triggers the fingerprint database update; for example, adding a microwave oven causes high-frequency pulse groups in the current waveform, and the system automatically starts the feature extraction process and generates the electricity fingerprint of the new device.

[0060] The dynamic threshold engine in the data processing and analysis module is used to calculate real-time thresholds based on the following factors:

[0061] Environmental Correction Factors Based on the current ambient temperature T and humidity H, through a function calculate, It is a monotonically increasing function; the higher the temperature, the greater the humidity. The larger;

[0062] Behavioral baseline Based on historical habits and normal ranges, the electricity load of the same branch circuit exhibits regularity at different times. By constructing a behavioral baseline using historical data, we can reflect the normal range of electricity consumption. ,in This represents the average current for the same period in history, and the data is sourced from locally stored historical current data. This represents the standard deviation of current for the same period in history. This is the confidence level coefficient, set according to the stability of the power consumption scenario; higher stability results in a higher confidence level. The smaller the value, the larger the value; when the current is detected multiple times consecutively... If the device is otherwise functioning normally, and a user adds a new device, the baseline will be automatically updated. ,in Weights for the new data, To ensure the baseline adapts slowly to user habits, It is a reference for the normal current range at a certain moment, and more than 95% of normal electricity consumption behavior should fall within this range;

[0063] Equipment synergy factor , Due to the cumulative effect of multiple devices operating simultaneously, the total current will temporarily increase when multiple devices on a branch start up at the same time. Therefore, the threshold for the device coordination factor needs to be relaxed to avoid misjudgment. ,in For the i-th device within the branch, the data source is the "electricity consumption fingerprint database". This is an indicator function; it is 1 when the device is running and 0 when it is stopped, determined by current characteristics. For equipment The weight is calculated as the ratio of the rated power of the equipment to the total rated power of the branch; the greater the power, the higher the weight. This reflects the cumulative effect of active devices on a branch line, such as multiple high-power devices operating simultaneously. Increase the threshold accordingly to accommodate temporary high loads; as equipment is gradually shut down... The threshold decreases, returning to normal.

[0064] Time decay factor This is used to measure the impact of time periods on electricity consumption characteristics. Electricity load characteristics differ at different times, and the sensitivity of the threshold needs to be adjusted using a time decay factor. For piecewise functions: peak electricity consumption periods (e.g., daytime): Larger values ​​allow for a more relaxed threshold; off-peak electricity consumption periods (e.g., nighttime): The value is small, and the threshold is strict; the time decay factor is matched with the user's daily routine.

[0065] Final dynamic threshold Based on the above factors, real-time device-level and tributary-level thresholds are generated: .

[0066] Among them, the environmental correction factor refers to the coefficient that dynamically adjusts the threshold based on environmental parameters. It is implemented by mapping the threshold to a linear or nonlinear function to solve the problem of increased current fluctuations caused by decreased heat dissipation capacity of equipment in high temperature and high humidity environments. The behavioral baseline reflects the historical data model of users' electricity consumption habits. It is implemented by calculating the mean and standard deviation of the current historical data stored locally and combining it with the confidence coefficient to generate a dynamic range. It is implemented by solving the problem that fixed thresholds cannot adapt to the different electricity consumption patterns of users. The equipment coordination factor is the threshold compensation coefficient for multiple devices running simultaneously. It is implemented by calculating the weight of the rated power data of the devices in the electricity fingerprint database and combining it with the current waveform characteristics to judge the operating status of the devices. It is implemented by solving the problem of false judgment of instantaneous current exceeding the limit caused by the superposition of multiple devices. The time decay factor is a parameter that adjusts the threshold sensitivity based on the time period. It is implemented by using a clock module to obtain the current time and mapping it to a decay coefficient through a preset piecewise function. It is implemented by solving the problem of false alarms caused by the difference in electricity load characteristics at different time periods.

[0067] The anomaly detection engine in the data processing and analysis module employs a three-level criterion-based linkage detection approach:

[0068] Primary criterion: Real-time current of the device The current is compared with the device-level dynamic threshold to determine whether the current is abnormal. The device-level dynamic threshold is based on the steady-state current in the device fingerprint. and environmental correction factors Through formula Calculation, where Safety factor for equipment type; for example, a certain heating equipment. This is its normal operating current. If the measured current consistently exceeds this value... If so, it is determined to be an abnormal current.

[0069] Secondary criterion: Whether the harmonic distortion rate (THD) of the detection equipment exceeds [the specified threshold]. and power fluctuations Does it exceed ,in These are normal harmonic values ​​from the fingerprint database. To allow for deviation, The steady-state power in the fingerprint database, To allow for fluctuation ratios; for example, in the fingerprints of motor-type devices. Higher than normal, if the measured THD level is significantly higher than normal. This may be due to a short circuit in the motor windings; power fluctuations in heating equipment are usually small. Exceeding This may be due to poor contact of the heating element;

[0070] Level 3 criterion: Detect whether abnormal current and leakage characteristics occur simultaneously; leakage characteristics are branch leakage current. Exceed ,in This is the normal leakage current baseline. Allowable deviation.

[0071] The data processing and analysis module also includes an early warning level classification unit, used for:

[0072] Level 1 Warning: When a single parameter deviates slightly from the threshold and the temperature is normal, only an APP push notification will be triggered;

[0073] Level 2 alert: When multiple parameters are abnormal, the app will issue a notification and a local audible and visual alarm, and the system will continue to monitor the situation.

[0074] Level 3 warning: When a high-risk anomaly occurs, the circuit of the socket where the faulty device is located will be automatically cut off and an emergency alarm will be pushed.

[0075] Specifically, the early warning level classification unit divides early warnings into three levels by monitoring the number and type of abnormal parameters in real time. When a single parameter is detected to deviate slightly and the temperature is within the normal range, the system only pushes information through the mobile application to avoid excessive interference with the user. When multiple parameters are abnormal at the same time, the system activates the local audible and visual alarm device while pushing a reminder, and determines whether to escalate the response by continuously monitoring the abnormal trend. When a high-risk abnormal combination is detected, the system automatically cuts off the power supply to the socket circuit where the faulty device is located and sends an emergency alarm message to the user.

[0076] The intelligent control module includes:

[0077] Actuators, including smart circuit breakers installed in the main circuit of branch circuits and smart relays in high-risk equipment sockets;

[0078] The control logic unit is used to maintain power supply during a level 1 warning, continuously track abnormal trends during a level 2 warning, and trigger an intelligent relay to cut off the power supply to the faulty equipment during a level 3 warning.

[0079] The IoT communication module includes:

[0080] The local communication unit uses an industrial bus to connect the sensors inside the electrical box with the edge computing unit to transmit raw collected data.

[0081] The remote communication unit uses 4G, WiFi, Ethernet, RS485, etc. to transmit analysis results and control commands to the cloud and user terminals;

[0082] The security encryption unit uses a symmetric encryption algorithm to encrypt transmitted data and uses a unique identifier to authenticate the device.

[0083] Specifically, when the system detects a Level 1 warning, the control logic unit maintains power supply and only pushes reminder information to the user through the communication module. At this time, the smart circuit breaker and relay remain closed. When a Level 2 warning is triggered, the control logic unit continuously monitors the changing trend of abnormal parameters. If the abnormality continues to worsen, it is upgraded to a Level 3 warning, and a local audible and visual alarm is activated. When a Level 3 warning is triggered, the control logic unit prioritizes sending instructions to the smart relay to cut off the power supply circuit of the socket where the high-risk device is located. If leakage or short circuit occurs in the main circuit, the smart circuit breaker is triggered simultaneously to cut off the power supply to the main circuit. For example, when leakage and excessive current are detected in a device connected to a socket, the control logic unit only disconnects the relay of that socket, while the main circuit breaker remains closed, and other devices can still operate normally.

[0084] The user interaction and management module includes:

[0085] The local interaction unit displays real-time status and warning information, and provides manual control buttons via a touch screen built into the electrical box.

[0086] The remote interaction unit provides functions such as device status visualization, historical data query, threshold configuration, device management, and remote control via computer, tablet, mobile phone, or web platform.

[0087] The cloud platform includes:

[0088] The data storage unit is used to store 30 days of high-frequency data and 36 months of historical data in the cloud. The historical data includes raw collected data, anomaly records, and device health index.

[0089] The deep analysis unit is used to calculate the equipment health index H using the formula. Implementation, in which For the number of abnormal occurrences, For current exceeding the limit, the weighting coefficients a and b are: a reflects the impact of abnormal frequency, and b reflects the impact of abnormal severity. The lower the H value, the higher the risk of equipment aging or failure.

[0090] The model optimization unit updates the "electricity fingerprint" recognition algorithm quarterly through big data training.

[0091] The technical scope of this invention is not limited to the content described above. Those skilled in the art can make various modifications and variations to the above embodiments without departing from the technical concept of this invention, and all such modifications and variations should fall within the protection scope of this invention.

Claims

1. A smart digital electrical box power monitoring and management system based on the Internet of Things, characterized in that, include: Data acquisition module: used to collect electrical and environmental parameters of branch circuits and key nodes of low-voltage power distribution system in real time. The electrical parameters include current, voltage, and power, and the environmental parameters include temperature and humidity. Data processing and analysis module: Deployed in the industrial-grade MCU built into the smart electrical box; used to extract features from the collected data to build the device's "electricity fingerprint database", and calculate real-time thresholds through a dynamic threshold engine, combined with an anomaly detection engine to achieve device-level anomaly identification; Intelligent control module: used to execute control actions based on anomaly detection results; IoT communication module: used to enable bidirectional data transmission between modules and with external terminals and cloud platforms; User interaction and management module: This includes a display screen on the surface of the smart electrical box, which provides a visual interface for users to configure, monitor and control the system.

2. The IoT-based smart digital electrical box power monitoring and management system according to claim 1, characterized in that, The data acquisition module includes: Electrical parameter sensors are used to collect instantaneous current, current waveform, line voltage, and phase voltage, and to calculate active power and reactive power based on current and voltage. Environmental and condition sensors are used to collect contact point temperature, leakage current, and high-frequency pulse signals in the current waveform; The data preprocessing unit is used to filter, periodically self-calibrate and perform analog-to-digital conversion on the raw data, and temporarily store high-frequency sampled data.

3. The IoT-based smart digital electrical box power monitoring and management system according to claim 1, characterized in that: The "electricity consumption fingerprint database" in the data processing and analysis module is constructed for: Extract the steady-state characteristics, transient characteristics and periodic characteristics of the equipment. The steady-state characteristics include the effective value of current, power value and power factor during normal operation. The transient characteristics include the peak current, start-up time and harmonic content during startup. The periodic characteristics include the start-up and shutdown cycle and power fluctuation range. Based on the branch device type manually entered by the user, the device's "electricity fingerprint" is automatically generated through high-frequency sampling over 7-14 days. When the device is replaced or added, the fingerprint database is updated by relearning based on the sudden change characteristics of the current waveform.

4. The IoT-based smart digital electrical box power monitoring and management system according to claim 1, characterized in that: The dynamic threshold engine in the data processing and analysis module is used to calculate real-time thresholds based on the following factors: Environmental Correction Factors Based on the current ambient temperature T and humidity H, through a function calculate, It is a monotonically increasing function; the higher the temperature, the greater the humidity. The larger; Behavioral baseline Based on historical habits and normal ranges, the electricity load of the same branch circuit exhibits regularity at different times. By constructing a behavioral baseline using historical data, we can reflect the normal range of electricity consumption. ,in This represents the average current for the same period in history, and the data is sourced from locally stored historical current data. This represents the standard deviation of current for the same period in history. This is the confidence level coefficient, set according to the stability of the power consumption scenario; higher stability results in a higher confidence level. The smaller the value, the larger the value; when the current is detected multiple times consecutively... If the device is otherwise functioning normally, and a user adds a new device, the baseline will be automatically updated. ,in Weights for the new data, To ensure the baseline adapts slowly to user habits, It is a reference for the normal current range at a certain moment; Equipment synergy factor , Due to the cumulative effect of multiple devices operating simultaneously, the total current will temporarily increase when multiple devices on a branch start up at the same time. Therefore, the threshold for the device coordination factor needs to be relaxed to avoid misjudgment. ,in For the i-th device within the branch, the data source is the "electricity consumption fingerprint database". This is an indicator function; it is 1 when the device is running and 0 when it is stopped, determined by current characteristics. For equipment The weight is calculated as the ratio of the rated power of the equipment to the total rated power of the branch; the greater the power, the higher the weight. Reflects the cumulative effect of active equipment on branch lines; Time decay factor This is used to measure the impact of time periods on electricity consumption characteristics. Electricity load characteristics differ at different times, and the sensitivity of the threshold needs to be adjusted using a time decay factor. For piecewise functions: Peak electricity consumption periods: Larger values ​​result in a more relaxed threshold; off-peak electricity consumption periods: The value is small, and the threshold is strict; the time decay factor is matched with the user's daily routine. Final dynamic threshold Based on the above factors, real-time device-level and tributary-level thresholds are generated: .

5. The IoT-based smart digital electrical box power monitoring and management system according to claim 1, characterized in that: The anomaly detection engine in the data processing and analysis module employs a three-level criterion-based linkage detection approach: Primary criterion: Real-time current of the device The current is compared with a device-level dynamic threshold to determine whether it is abnormal. The device-level dynamic threshold is based on the steady-state current in the device fingerprint. and environmental correction factors Through formula Calculation, where Safety factor for equipment type; Secondary criterion: Whether the harmonic distortion rate (THD) of the detection equipment exceeds [the specified threshold]. and power fluctuations Does it exceed ,in These are normal harmonic values ​​from the fingerprint database. To allow for deviation, The steady-state power in the fingerprint database, To allow for a certain percentage of fluctuation; Level 3 criterion: Detect whether abnormal current and leakage current characteristics occur simultaneously, wherein the leakage current characteristic is branch leakage current. Exceed ,in This is the normal leakage current baseline. Allowable deviation.

6. The IoT-based smart digital electrical box power monitoring and management system according to claim 5, characterized in that, The data processing and analysis module also includes an early warning level classification unit, used for: Level 1 Warning: When a single parameter deviates slightly from the threshold and the temperature is normal, only an APP push notification will be triggered; Level 2 alert: When multiple parameters are abnormal, the app will issue a notification and a local audible and visual alarm, and the system will continue to monitor the situation. Level 3 warning: When a high-risk anomaly occurs, the circuit of the socket where the faulty device is located will be automatically cut off and an emergency alarm will be pushed.

7. The IoT-based smart digital electrical box power monitoring and management system according to claim 1, characterized in that: The intelligent control module includes: Actuators, including smart circuit breakers installed in branch main circuits and smart relays in high-risk equipment sockets; The control logic unit is used to maintain power supply during a level 1 warning, continuously track abnormal trends during a level 2 warning, and trigger an intelligent relay to cut off the power supply to the faulty equipment during a level 3 warning.

8. The IoT-based smart digital electrical box power monitoring and management system according to claim 7, characterized in that, The IoT communication module includes: The local communication unit uses an industrial bus to connect the sensors inside the electrical box with the edge computing unit to transmit raw collected data. The remote communication unit uses 4G, WiFi, Ethernet, and RS485 to transmit analysis results and control commands to the cloud and user terminals. The security encryption unit uses a symmetric encryption algorithm to encrypt transmitted data and uses a unique identifier to authenticate the device.

9. The IoT-based smart digital electrical box power monitoring and management system according to claim 8, characterized in that, The user interaction and management module includes: The local interaction unit displays real-time status and warning information, and provides manual control buttons via a touch screen built into the electrical box. The remote interaction unit provides functions such as device status visualization, historical data query, threshold configuration, device management, and remote control via computer, tablet, mobile phone, or web platform.

10. The IoT-based smart digital electrical box power monitoring and management system according to claim 9, characterized in that: The cloud platform includes: The data storage unit is used to store 30 days of high-frequency data and 36 months of historical data from the cloud. The historical data includes raw collected data, anomaly records, and device health index. The deep analysis unit is used to calculate the equipment health index H using the formula. Implementation, in which For the number of abnormal occurrences, For current exceeding the limit, the weighting coefficients a and b are: a reflects the impact of abnormal frequency, and b reflects the impact of abnormal severity. The lower the H value, the higher the risk of equipment aging or failure. The model optimization unit updates the "electricity fingerprint" recognition algorithm quarterly through big data training.

Images