Engineering project intelligent distribution box for high-precision carbon emission data acquisition

Abstract

The invention relates to the technical field of intelligent power distribution and green construction management, and discloses a high-precision carbon emission data acquisition engineering project intelligent power distribution box, which comprises an engineering project construction power distribution box, a communication network and an intelligent power distribution cloud platform, and is characterized in that the power distribution box integrates a data acquisition module, an edge calculation module and a safety linkage control module; and the cloud platform is used for collecting electric parameters, identifying the type of access equipment by using a built-in portrait model, supporting off-line calculation and power-off control, deploying a dynamic carbon emission calculation engine, determining a comprehensive dynamic carbon emission factor based on a real-time power supply structure of the regional power grid, and accurately calculating the real-time carbon emission of a loop in combination with periodic energy consumption. According to the invention, through an end-cloud collaborative architecture, refined monitoring of a construction machinery equipment level, carbon accounting based on dynamic factors and active intervention on illegal idling are realized, and the problems of low precision and lack of real-time management and control means of traditional static accounting are solved.

Description

Technical Field

[0001] This invention relates to the field of intelligent power distribution and green construction management technology, specifically to an intelligent power distribution box for engineering projects that collects high-precision carbon emission data. Background Technology

[0002] As a key area of ​​energy consumption and carbon emissions, the construction industry is gradually undergoing a green transformation in its construction process. Construction sites involve a large number of high-energy-consuming mechanical equipment such as tower cranes, construction hoists, and welding machines. Electricity consumption is one of the main sources of carbon emissions. Therefore, building a system that can monitor, calculate, and manage the carbon emissions from construction electricity consumption is of great practical significance for implementing green construction evaluation standards and achieving carbon neutrality and management throughout the entire life cycle of construction projects.

[0003] Currently, carbon emission monitoring at construction sites relies on traditional electricity metering instruments or energy consumption monitoring systems. These existing technologies adopt a centralized data acquisition mode, uploading the total electricity consumption or branch electricity consumption at the construction site to the backend server, and then performing static multiplication calculations based on the annual average carbon emission factor published by the country or region to obtain the total carbon emissions for a certain period. Although some systems have realized remote meter reading and basic data statistics functions, they have accumulated certain technical expertise in terms of data real-time performance, granularity, and the accuracy of accounting models.

[0004] However, existing monitoring technologies still have shortcomings when facing the demands of refined and dynamic carbon management. First, existing accounting methods mostly use static annual average carbon emission factors, ignoring the impact of real-time fluctuations in the proportion of renewable energy connected to the regional power grid on the carbon intensity of electricity. This results in accounting results that cannot accurately reflect the environmental cost differences brought about by various power supply structures at the time of electricity consumption. Second, traditional metering terminals lack edge computing and intelligent identification capabilities, and can only record energy consumption values ​​without automatically identifying specific load equipment types. They are unable to distinguish between effective operating energy consumption and illegal idling energy consumption, and are prone to data loss in the unstable network environment of construction sites. Therefore, this invention proposes a high-precision carbon emission data acquisition intelligent distribution box for engineering projects to address the shortcomings of existing technologies. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an intelligent power distribution box for engineering projects that collects high-precision carbon emission data. This solves the problems of low accuracy, lack of equipment-level fine-grained monitoring capabilities, and lagging data acquisition and control in the current carbon emission accounting field, which are caused by the use of static factors. It also addresses the issues of lagging data acquisition and control that prevent the implementation of energy conservation and emission reduction strategies.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solution: an intelligent power distribution box for engineering projects with high-precision carbon emission data acquisition, comprising an engineering project construction power distribution box arranged at the construction site, an intelligent power distribution cloud platform deployed on the server, and a communication network connecting the two.

[0007] The construction power distribution box of the project serves as a core node on the edge side, integrating a power input unit, a switch protection unit, a metering unit, a data acquisition unit, an edge computing unit, a safety linkage control module, and a communication module. It is capable of distributing power and acquiring high-frequency data for electrical equipment on the construction site.

[0008] The edge computing unit has a built-in construction load equipment profiling model, which can perform localized identification and processing of the collected electrical parameter data, determine the equipment type of the access circuit, and upload the processed structured data.

[0009] The intelligent power distribution cloud platform receives data through the data access module and executes the core accounting logic using the dynamic carbon emission calculation engine. The core accounting logic breaks through the limitations of the traditional fixed carbon emission factor and can calculate the real-time carbon emission of each outgoing circuit based on the cumulative energy consumption during the data upload period and the comprehensive dynamic carbon emission factor determined by the real-time power structure of the regional power grid.

[0010] The system also receives control commands from the edge or cloud via the safety linkage control module, driving the switch protection unit to perform physical actions, forming a closed loop from monitoring to control.

[0011] Preferably, to address the issue of decreased data acquisition accuracy caused by strong electromagnetic interference at the construction site, the construction distribution box for the project incorporates electromagnetic compatibility design in its physical spatial structure, dividing it into a high-voltage isolation area and a low-voltage shielding area. The high-voltage and low-voltage units are physically separated. The data acquisition unit employs a high-precision Hall current sensor based on the closed-loop magnetic balance principle on the load side of the branch circuit breaker. The high-precision Hall current sensor is directly connected to the line, ensuring electrical isolation while achieving high-fidelity acquisition of non-sinusoidal large currents and minute leakage currents, providing high-quality basic data for subsequent edge computing feature extraction.

[0012] Preferably, to achieve automatic identification and accurate profiling of construction machinery and equipment types, the edge computing unit executes an edge intelligent algorithm based on load feature analysis. The edge intelligent algorithm monitors current waveform changes in real time, captures feature window data when load state switching is detected, and extracts a multi-dimensional feature vector containing the starting impact current multiple, starting duration, steady-state power factor, and total harmonic distortion rate of the current waveform. By calculating the similarity distance between the current feature vector and the benchmark feature template in the pre-set model library, when the matching degree meets the confidence requirement, the loop is automatically labeled with the equipment type (such as tower crane, hoist, etc.), thereby achieving equipment-level energy consumption and carbon emission tracking without manual intervention.

[0013] Preferably, to address the data loss issue caused by unstable network environments at construction sites, the system incorporates an offline computing and breakpoint resume mechanism. When the communication network is interrupted, the edge computing unit automatically switches modes, uses locally cached carbon emission factors to perform a preliminary estimate of real-time data, and stores it in a non-volatile circular buffer. After the network is restored, missing data segments are located by comparing timestamps and then retransmitted in packages. After receiving the retransmitted data, the cloud platform corrects and updates the offline estimate based on the actual grid carbon emission factors at historical moments, ensuring the continuity and accuracy of carbon emission accounting data.

[0014] Preferably, to achieve real-time intervention in high-energy-consumption or illegal electricity use, the safety linkage control module is decoupled from the high-voltage circuit through opto-isolation technology. The control level output by the edge computing unit drives the opto-isolated relay, thereby connecting the external circuit of the shunt trip coil or electric operating mechanism equipped with the circuit breaker, and actuating the mechanical tripping mechanism. This method realizes the physical control of the high-voltage on / off state by the low-voltage control signal, ensuring that the power supply can be cut off in milliseconds when safety policies are triggered or carbon budgets are exceeded.

[0015] Preferably, the dynamic carbon emission calculation engine adopts a spatiotemporal-based dynamic factor generation mechanism, which solves the problem that traditional fixed factors cannot reflect the fluctuations in carbon intensity on the power production side. The spatiotemporal-based dynamic factor generation mechanism first uses GPS geographic coordinates to map the project to the specific regional power grid, and then obtains the current power structure data of the regional power grid in real time through the data interface (i.e., the real-time grid-connected power of different power generation types such as thermal power, hydropower, and wind power). The system calculates the real-time average carbon emission intensity on the physical side based on the carbon emission coefficients of various power sources throughout their entire life cycle, and then integrates it with the policy-based annual benchmark factor according to weights to generate a comprehensive carbon emission factor that changes dynamically over time, so that the calculation results truly reflect the cleanliness of the power grid at the time of electricity consumption.

[0016] Preferably, in the specific carbon emission calculation process, the system employs a calculation logic of time-series alignment and integral accumulation. The system retrieves the power data records and the comprehensive dynamic carbon emission factor sequence within the time window to be calculated, and strictly aligns the two sets of data with different sampling frequencies on the time axis through zero-order hold interpolation. Subsequently, the Discrete Riemann sum algorithm is used to multiply and accumulate the average active power, duration, and corresponding dynamic carbon emission factor within each tiny time slice, thereby obtaining the cumulative total carbon emissions within that time window.

[0017] Preferably, to improve the energy efficiency of construction power supply, the intelligent power distribution cloud platform integrates an energy efficiency diagnostic module, which can identify the illegal idling status of equipment. By jointly analyzing the switch closing status, real-time power, no-load threshold, and duration, the system can determine whether the equipment is in an inefficient energy consumption state. Combined with the project's carbon budget management mechanism, when the cumulative carbon emissions exceed the standard or the illegal idling is not addressed for a long time, the cloud platform can generate an active intervention command, which, after edge-side verification, automatically triggers a power outage operation to enforce green construction management requirements.

[0018] Preferably, to visually represent the spatial distribution of carbon emissions, the system incorporates a visualization management module and 3D building information modeling (BIM) technology. By establishing a mapping relationship between power distribution circuits and construction machinery elements in the 3D BIM model, the system converts the calculated real-time carbon emission intensity into visual signals, driving changes in the surface material color of the corresponding machinery model in the digital twin scene. This visualization method enables managers to intuitively locate high-carbon-emission hotspots in the 3D scene, assisting in on-site scheduling and management.

[0019] Preferably, this invention also constructs an open data ecosystem, achieving standardized data output through an application programming interface (API) gateway. The system can not only encapsulate electricity and carbon emission data in JSON format and report it to the government regulatory platform according to green construction monitoring standards, but also seamlessly integrate with enterprise resource planning systems, providing accurate carbon emission inventory data for construction cost accounting and promoting the in-depth mining and application of data value.

[0020] This invention provides an intelligent power distribution box for engineering projects that collects high-precision carbon emission data. It has the following beneficial effects:

[0021] 1. This invention integrates an edge computing unit and a built-in construction load equipment profile model into the construction distribution box of an engineering project. It uses feature vectors such as starting inrush current and harmonic distortion rate to perform localized identification of circuit loads. This allows the system to automatically distinguish between different construction machinery such as tower cranes and construction hoists without manual input. This solves the problem that traditional electricity meters can only measure total energy consumption but cannot distinguish load types, and improves the granularity and real-time performance of carbon emission data collection.

[0022] 2. This invention utilizes the dynamic carbon emission calculation engine of the intelligent power distribution cloud platform, abandoning the traditional static annual average factor calculation method. Instead, it combines real-time power structure data of the regional power grid to generate a comprehensive dynamic carbon emission factor. Based on the real-time changes in the proportion of different energy sources such as thermal power, hydropower, and wind power in the power grid, it can calculate the actual carbon emissions generated by electricity consumption behavior at each moment, truly reflecting the environmental cost of electricity consumption and providing data support for peak-shifting electricity consumption and green construction.

[0023] 3. This invention establishes an end-to-cloud collaborative proactive control system through the cooperation of a safety linkage control module and an energy efficiency diagnosis module. The system not only has the ability to perform offline calculations and resume data transmission in a network outage environment to ensure data integrity, but also automatically triggers the shunt trip to perform a power-off operation when it detects that the equipment is running idling illegally or the total carbon emissions of the project exceed the budget threshold. This transforms carbon emission management from passive post-event statistics to proactive pre-event control and in-event intervention, thus achieving energy conservation and emission reduction goals. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the overall system architecture and electrical principle of the distribution box of the present invention;

[0025] Figure 2 This is a schematic diagram of the edge-side processing and device identification process of the present invention;

[0026] Figure 3 This is a schematic diagram illustrating the carbon emission calculation based on dynamic factors according to the present invention. Detailed Implementation

[0027] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] See attached document Figure 1 This invention provides a smart power distribution box for engineering projects that collects high-precision carbon emission data, including: a construction power distribution box, a communication network, and a smart power distribution cloud platform.

[0029] The construction distribution box is set up at the construction site. It is responsible for power distribution, circuit protection, data acquisition, and edge computing for the electrical equipment at the construction site.

[0030] The communication network connects the construction power distribution box with the intelligent power distribution cloud platform, and is used to realize the bidirectional transmission of data and control commands.

[0031] The intelligent power distribution cloud platform is deployed on the server side. It is responsible for data storage, carbon emission core algorithm conversion, energy efficiency diagnosis, and generation of control strategies.

[0032] The construction distribution box integrates a power input unit, a switch protection unit, a metering unit, a data acquisition unit, an edge computing unit, a safety linkage control module, and a communication module.

[0033] The power input unit receives electrical energy from the external power grid. The switch protection unit is electrically connected to the power input unit and includes a circuit breaker and a leakage current protection device. The switch protection unit is used to perform circuit on / off control and overload / short circuit protection. The metering unit adopts a branch-independent metering architecture, with an independent metering channel set up for each outgoing circuit in the construction distribution box. The data acquisition unit is connected to the metering unit and is equipped with a high-precision Hall current sensor and a three-phase voltage acquisition probe. The data acquisition unit is configured to synchronously acquire electrical parameters of each outgoing circuit at a preset sampling frequency. The electrical parameter data output by the data acquisition unit includes instantaneous voltage, instantaneous current, frequency, active power, reactive power, power factor, and harmonic content. The edge computing unit is communicatively connected to the data acquisition unit. The unit has a built-in microprocessor and memory. The edge computing unit is configured to receive electrical parameter data output by the data acquisition unit, and perform noise reduction and timestamp processing on the electrical parameter data. The edge computing unit internally stores a construction load equipment profile model. The edge computing unit uses the construction load equipment profile model to compare the characteristics of the acquired current waveform and identify the type and operating status of the equipment connected to the circuit. The safety linkage control module is electrically connected to the intelligent circuit breaker or magnetic latching relay in the switch protection unit. The safety linkage control module is configured to receive local commands from the edge computing unit or remote commands from the intelligent power distribution cloud platform, and drive the switch protection unit to perform trip power cut or power limiting operations. The communication module is connected to the edge computing unit and sends the processed data to the intelligent power distribution cloud platform through the communication network.

[0034] The intelligent power distribution cloud platform includes a data access module, a dynamic carbon emission calculation engine, an energy efficiency diagnosis module, a load anomaly identification module, and a visualization management module.

[0035] The data access module receives data packets uploaded by the communication module and stores the parsed data in the time-series database. The dynamic carbon emission calculation engine is the core processing unit of the smart power distribution cloud platform. Based on the dynamic carbon emission factor, the dynamic carbon emission calculation engine performs real-time carbon emission calculation on the uploaded power data. The dynamic carbon emission calculation engine first calculates the cumulative energy consumption of the monitoring circuit within a specific time period. The calculation formula is as follows:

[0036]

[0037] Among them, E k The t represents the cumulative power consumption of the k-th loop during the calculation period; N represents the total number of sampling points during the calculation period; i P represents the i-th sampling time; k (t i ) indicates that the k-th loop at sampling time t i The instantaneous active power sample value; Δt represents the time interval between two adjacent samples; This represents the sampled values ​​P of instantaneous active power from 1 to N. k (t i Summation is performed.

[0038] The dynamic carbon emission calculation engine then determines the dynamic carbon emission factor for the current moment, using the following logic:

[0039] λ(t i )=α·F nat +β·F grid (t i ,R);

[0040] Wherein, λ(t) i ) represents the sampling time t i Comprehensive dynamic carbon emission factor; F nat This represents the preset annual benchmark carbon emission factor for the State Grid Corporation of China; F grid (t i R) represents the real-time average carbon emission intensity of the regional power grid R on the physical side at time t, reflecting the current power structure ratio of the regional power grid; α and β are the baseline factor weight coefficient and dynamic factor weight coefficient, respectively, and satisfy α+β=1.

[0041] The dynamic carbon emission calculation engine ultimately calculates the real-time carbon emissions of this loop, using the following formula:

[0042]

[0043] Among them, C k This represents the total carbon emissions of the k-th loop within the calculation period.

[0044] The energy efficiency diagnostic module analyzes the power factor and load rate of each circuit and generates energy efficiency optimization suggestions; the load anomaly identification module identifies overload, phase loss, short circuit and harmonic exceedance anomalies based on current waveform characteristics and harmonic content data; the visualization management module generates a graphical interface to display real-time energy consumption curves, carbon emission trends and equipment operating status, and provides an API interface for data interaction with external systems.

[0045] See attached document Figure 1The construction distribution box adopts a modular cabinet structure design. The interior of the construction distribution box is strictly divided into a strong current isolation area and a weak current shielding area in terms of physical space, so as to minimize the impact of electromagnetic interference on microelectronic devices.

[0046] The high-voltage isolation area houses the power input unit and switch protection unit. The low-voltage shielded area houses the data acquisition unit, edge computing unit, communication module, and DC24V regulated power supply module for powering the low-voltage modules.

[0047] The power input unit includes a three-phase five-wire incoming terminal block, a main disconnect switch, and a T1-level power surge protector. The three-phase five-wire incoming terminal block is connected to the L1 phase line, L2 phase line, L3 phase line, N neutral line, and PE ground line of the external power grid, respectively.

[0048] The switch protection unit adopts a tree-like hierarchical cascade structure. The switch protection unit includes a main circuit breaker located on the main bus incoming side and multiple branch circuit breakers located on the branch outgoing side. Each branch circuit breaker independently controls a specific outgoing circuit, which is directly connected to specific construction machinery loads such as tower cranes, construction elevators, welding machines, or concrete mixers.

[0049] The metering unit and data acquisition unit adopt an embedded integrated design in physical implementation, and each outgoing circuit is configured independently. For each branch circuit breaker in the switch protection unit, the data acquisition unit has a set of current acquisition components and a set of voltage acquisition circuits physically installed at the load side outgoing terminal of the branch circuit breaker.

[0050] The current acquisition component uses a high-precision Hall current sensor based on the closed-loop magnetic balance principle. This high-precision Hall current sensor is directly connected to the L1, L2, and L3 copper busbars or cables on the output side of the branch circuit breaker. Internally, the high-precision Hall current sensor integrates a Hall sensing element, a magnetic core, and a negative feedback compensation circuit. It is configured to linearly convert the large current signal flowing through the primary side into a small current signal on the secondary side. The linearity of the high-precision Hall current sensor is better than 0.1%, and its bandwidth covers the DC to 10kHz range, ensuring accurate capture of high-frequency harmonic components during the frequency conversion startup of construction equipment.

[0051] The voltage acquisition circuit is directly connected in parallel to the output terminals of the branch circuit breaker via a voltage divider network composed of high-precision metal film resistors. The voltage acquisition circuit acquires the voltages of L1 relative to the neutral line, L2 relative to the neutral line, and L3 relative to the neutral line, respectively.

[0052] The data acquisition unit also includes a multi-channel synchronous sampling analog-to-digital converter (ADC) and a floating-point digital signal processor (DSP).

[0053] The multi-channel synchronous sampling analog-to-digital converter is configured with 24-bit resolution to ensure that the quantization error meets the requirements of the 0.2-level metrological standard throughout the entire measurement range. The multi-channel synchronous sampling analog-to-digital converter simultaneously receives analog signals from the high-precision Hall current sensor and analog signals from the voltage acquisition circuit, and performs synchronous conversion to eliminate phase errors.

[0054] The floating-point digital signal processor reads the conversion results from the multi-channel synchronous sampling analog-to-digital converter via a high-speed serial peripheral interface (SPI). Internally, the floating-point digital signal processor runs a Fast Fourier Transform (FFT) algorithm to calculate in real time the fundamental amplitude of voltage and current, as well as the amplitude and phase of the 2nd to 50th harmonic components.

[0055] The edge computing unit employs an industrial-grade embedded microprocessor based on the ARM Cortex-A (register and operating mode) series architecture or equivalent performance. The edge computing unit establishes a bidirectional communication connection with the data acquisition unit via an RS-485 industrial bus or an onboard CAN (serial communication protocol) bus, periodically reading the processed electrical parameter data.

[0056] The edge computing unit's motherboard integrates a large-capacity industrial-grade non-volatile memory (eMMC). This industrial-grade non-volatile memory is used to locally store historical power data, cumulative carbon emission data, and device status logs with UTC (Coordinated Universal Time) timestamps during communication network outages.

[0057] The safety linkage control module includes multiple opto-isolated relay output interfaces and a power drive circuit. The output of the safety linkage control module is electrically connected to the shunt circuit breaker accessories in the switch protection unit via control cables. The shunt circuit breaker accessories specifically include shunt trip coils or electric operating mechanisms.

[0058] When the edge computing unit determines that a power-off operation is required based on its internal logic, it sends a control level to the safety linkage control module. The safety linkage control module then drives the corresponding opto-isolation relay to close, thereby connecting the external control circuit of the shunt trip coil, which in turn drives the mechanical tripping mechanism inside the branch circuit breaker to physically cut off the power supply to the corresponding outgoing circuit.

[0059] The communication module connects to the edge computing unit via a UART asynchronous serial interface or a 100Mbps Ethernet interface. The communication module integrates an industrial-grade, full-network compatible wireless communication module and a high-gain RF antenna to establish a wireless data transmission link with the intelligent power distribution cloud platform.

[0060] See attached document Figure 2The data cleaning and alignment process performed by the edge computing unit aims to eliminate random noise interference from the complex electromagnetic environment at the construction site on the original sampling data, and to solve the time asynchrony problem during multi-channel parallel acquisition, providing a standardized data foundation for subsequent carbon emission accounting and equipment profiling.

[0061] The edge computing unit first establishes a first-in-first-out (FIFO) data queue based on a circular buffer. The edge computing unit receives the original discrete sampling sequence from the data acquisition unit through a high-speed bus. The original discrete sampling sequence contains the instantaneous voltage value, instantaneous current value, and instantaneous power value corresponding to each sampling moment.

[0062] To address the high-frequency electromagnetic pulse interference generated by the frequent start-stop of high-power frequency converters at construction sites, the edge computing unit applies a sliding window weighted mean filtering algorithm to denoise the original discrete sampling sequence. Instead of directly using single sample values ​​as the basis for calculation, the edge computing unit performs weighted smoothing on multiple consecutive sampling points based on a preset time window. The specific calculation logic of the sliding window weighted mean filtering algorithm is as follows:

[0063]

[0064] Among them, V filter (t) represents the filtered electrical parameter value at time t; W represents the length of the sliding window, which is an integer between 5 and 20 based on the sampling frequency of the data acquisition unit; j represents the index of the sampling point shifted forward from the current time; V raw (tj) represents the original sampled value acquired at time tj; ω j This represents the weight coefficient of the index of the j-th sampling point, where ω is the weight coefficient. j The value of decreases exponentially as j increases, ensuring that the sampled data at the current moment dominates the filtering result.

[0065] After the noise reduction process is completed, the edge computing unit performs a zero-point drift correction operation. Since the Hall current sensor may have a small zero-point bias voltage when there is no current, the edge computing unit has a preset zero-point cutoff threshold. When the filtered current value is less than the zero-point cutoff threshold, the edge computing unit forces the current value and power value at that moment to be set to zero, thereby eliminating the cumulative error of sensor temperature drift in standby power consumption calculation.

[0066] Subsequently, the edge computing unit performs time alignment and timestamp marking on the multiple data streams. Since the multiple analog-to-digital converters in the data acquisition unit may have microsecond-level conversion time differences, the edge computing unit uses its internal high-precision real-time clock (RTC) as a unified time reference. The edge computing unit maintains clock synchronization with the smart power distribution cloud platform through the Network Time Protocol (NTP).

[0067] The edge computing unit uses a linear interpolation algorithm to map the asynchronously sampled data from each loop onto a unified whole-second or whole-millisecond time axis. For any standard timestamp T sync If no actual sampling occurs at that moment, the edge computing unit uses the two nearest actual sampling points (t) before and after that moment. prev V prev ) and (t next V next Estimate the value at that moment:

[0068]

[0069] Among them, V sync (T sync ) indicates that at the standard timestamp T sync Aligned data values ​​at; t prev and t next These represent the actual sampling times; V prev and V next These represent the actual sampled values.

[0070] After cleaning and alignment, the edge computing unit encapsulates the data of each loop into a structured data packet containing a unified UTC timestamp, device loop identifier ID, standard voltage value, standard current value, and active power value, and stores the structured data packet in the shared memory area of ​​the edge computing unit for use by the device profiling and recognition module.

[0071] See attached document Figure 2 The edge computing unit executes a construction load equipment profiling and identification method. This method aims to automatically parse the physical equipment type of the access circuit based on the principle of non-intrusive load monitoring (NILM) and high-frequency sampled electrical parameter data, thereby providing equipment-level semantic information for refined carbon emission accounting.

[0072] The edge computing unit first performs transient event detection. The edge computing unit monitors the root mean square current sequence output by the data acquisition unit in real time. When the edge computing unit detects that the current change value of two adjacent sampling cycles exceeds the preset event trigger threshold, the edge computing unit determines that a load state switching event has occurred in the current circuit, and extracts data segments of 2 to 5 seconds before and after the event trigger time as feature window data to be analyzed.

[0073] The edge computing unit then performs feature extraction on the feature window data to construct a multi-dimensional load feature vector. The specific feature dimensions extracted by the edge computing unit include:

[0074] Starting inrush current multiple: the ratio of the maximum instantaneous peak current to the average steady-state operating current within the characteristic window;

[0075] Startup process duration: the length of time required for the current to rise from the trigger threshold to its peak value and then fall back to 95% of its steady-state value.

[0076] Steady-state power factor: the average power factor of the equipment after it has entered a stable operating phase;

[0077] Total Harmonic Distortion (THD) of Current Waveform: Focus on extracting the amplitude proportion of the 3rd, 5th and 7th odd harmonic components;

[0078] Current fluctuation variance: reflects the degree of dispersion of current amplitude during equipment operation;

[0079] The load feature vector constructed by the edge computing unit is represented as follows:

[0080]

[0081] Among them, K surge Indicates the starting inrush current multiple; T startup Indicates the duration of the startup process; PF steady Represents steady-state power factor; THD i The set representing the distortion rates of the 3rd, 5th, and 7th harmonics; This represents the variance of current fluctuation.

[0082] The non-volatile memory inside the edge computing unit contains a pre-built library of construction load equipment profile models, which stores the baseline feature templates of five typical construction equipment: tower cranes, construction hoists, concrete mixers, AC welding machines, and on-site lighting equipment.

[0083] The edge computing unit uses a weighted Euclidean distance algorithm to calculate the currently extracted load feature vector F. load The similarity distance between the edge computing unit and each baseline feature template in the construction load equipment profiling model library is used. For different types of construction equipment, the edge computing unit adopts differentiated discrimination logic.

[0084] For AC welding machines: the edge computing unit focuses on detecting the variance of current fluctuations. and steady-state power factor PF steady If the current is detected to exhibit high-frequency, violent pulse fluctuations (i.e. If the power factor is extremely high and the steady-state power factor is below 0.6, the edge computing unit identifies the circuit device as an AC welding machine.

[0085] For tower cranes and construction hoists: the edge computing unit focuses on detecting the set of harmonic distortion (THD) of the 3rd, 5th, and 7th harmonics. iand startup process duration T startup If the startup process is detected to last for a long time (usually more than 2 seconds) and is accompanied by specific frequency harmonic components (generated by the frequency converter driver), the edge computing unit will identify the loop device as a tower crane or construction hoist.

[0086] For on-site lighting equipment: If the starting inrush current multiple K is detected... surge Smaller and current fluctuation variance When operating at near-zero power (i.e., constant power), the edge computing unit identifies the loop device as a lighting load.

[0087] When the calculated minimum similarity distance is less than the preset confidence threshold, the edge computing unit writes the successfully matched device type label into the configuration register of the loop and uploads the device type label to the smart power distribution cloud platform in subsequent data packets. If the minimum similarity distance is greater than the confidence threshold, the edge computing unit marks the device as an unknown load and triggers the original waveform data upload mechanism for the smart power distribution cloud platform to perform manual auxiliary labeling and model iteration updates.

[0088] See attached document Figure 2 The edge computing unit performs offline computing and breakpoint resume mechanism. The offline computing and breakpoint resume mechanism is designed to ensure that the construction power distribution box can still maintain continuous recording of energy consumption data and localized estimation of carbon emissions in the event of unstable or interrupted communication network, and ensure the integrity of data after network recovery.

[0089] The edge computing unit periodically sends heartbeat detection data packets to the smart power distribution cloud platform through the communication module. When the edge computing unit does not receive a response confirmation signal from the smart power distribution cloud platform within three consecutive heartbeat cycles, the edge computing unit determines that it is currently in a network interruption state and automatically switches to offline working mode.

[0090] Before entering offline working mode, the edge computing unit maintains a dynamic carbon emission factor cache table in its local memory. The dynamic carbon emission factor cache table is updated periodically by the smart power distribution cloud platform through downlink commands during normal network periods. The dynamic carbon emission factor cache table stores the preset carbon emission factor values ​​at the hourly granularity corresponding to the current date.

[0091] During offline operation, the edge computing unit continues to perform data acquisition and processing tasks. Based on the current time determined by its internal real-time clock, the edge computing unit indexes the corresponding local carbon emission factor λ from the dynamic carbon emission factor cache table. local (t), edge computing units utilize local carbon emission factor λ local (t) Estimate the carbon emissions in real time based on the electricity data of the current collection period. The estimation formula is as follows:

[0092] C offline (t)=P meas (t)·Δt·λ local (t);

[0093] Among them, C offline (t) represents the carbon emissions at time t calculated offline; P meas (t) represents the measured active power at time t; Δt represents the sampling interval; λ local (t) represents the local carbon emission factor at time t read from the cache table.

[0094] The edge computing unit packages the calculated power data, estimated carbon emission data, equipment status code, and UTC timestamp into fixed-length binary records. The edge computing unit sequentially writes the binary records into the onboard industrial-grade non-volatile memory's circular data buffer. The circular data buffer is equipped with a power-loss protection mechanism to prevent data loss due to power failure of the distribution box.

[0095] When the communication module detects that the communication network has been restored, the edge computing unit initiates the breakpoint resume process. The edge computing unit first sends a breakpoint query request to the intelligent power distribution cloud platform. The intelligent power distribution cloud platform then returns the timestamp T of the last record successfully received in the database. last_ack .

[0096] Edge computing units are based on timestamp T last_ack The edge computing unit locates the starting address of the unuploaded data within the circular data buffer. It then packages the unuploaded data records into multiple batch data packets according to the Maximum Transmission Unit (MTU) limit.

[0097] The edge computing unit sends batch data packets sequentially according to the time sequence. After sending each batch data packet, the edge computing unit waits for confirmation from the smart power distribution cloud platform. If the sending fails, the edge computing unit executes an exponential backoff retransmission strategy.

[0098] After receiving the data packets uploaded offline, the smart power distribution cloud platform uses the more accurate historical actual carbon emission factors stored on the smart power distribution cloud platform (if there is a deviation from the local cache factor) to correct and update the offline estimated carbon emission data, thereby ensuring the accuracy of the final archived data. After receiving the confirmation receipt of all data packets, the edge computing unit clears the corresponding data in the circular data buffer, releases the storage space, and completes the breakpoint resume operation.

[0099] See attached document Figure 3The dynamic carbon emission calculation engine in the intelligent power distribution cloud platform executes a dynamic carbon emission factor construction strategy. This strategy aims to solve the technical problem in traditional carbon accounting methods that fail to reflect real-time changes in the power grid's energy structure due to the use of annual fixed average factors, which leads to accounting deviations.

[0100] The dynamic carbon emission calculation engine first determines the basic geospatial and temporal parameters required for the calculation. The engine then reads the GPS geographic coordinates information entered during the registration of the construction distribution box and maps the construction project to the specific regional power grid supply range based on the geographic coordinates. The regional power grid supply range specifically refers to the provincial power grid or the inter-provincial regional power grid.

[0101] The dynamic carbon emission calculation engine establishes a communication connection with the external power dispatch and trading center database or third-party renewable energy big data platform through an encrypted data exchange interface (API). The dynamic carbon emission calculation engine obtains the real-time power structure data of the regional power grid at the current time t at a preset time refresh frequency (e.g., every 15 minutes or every 1 hour).

[0102] Real-time power structure data specifically includes real-time grid-connected power values ​​for different power generation types within the regional power grid. These power generation types include, but are not limited to, coal-fired power generation, gas-fired power generation, large-scale hydropower generation, onshore and offshore wind power generation, centralized and distributed photovoltaic power generation, and nuclear power generation.

[0103] The dynamic carbon emission calculation engine calculates the real-time average carbon emission intensity of the power grid in this region at time t based on real-time power structure data. The calculation logic is as follows:

[0104]

[0105] Among them, F grid (t,R) represents the real-time average carbon emission intensity of the regional power grid R on the physical side at time t (unit: kilogram CO2 equivalent, kgCO2 / kWh); K represents the total number of power generation types; m represents the number of power generation types; P gen (m,t) represents the real-time grid-connected power generation of the m-th power generation type at time t (unit: megawatts, MW); γ m This represents the life-cycle carbon emission coefficient for the m-th type of power generation.

[0106] The dynamic carbon emission calculation engine has a pre-built database of life-cycle carbon emission coefficients. This database stores carbon emission coefficient values ​​per unit of electricity for various power sources, derived from IPCC (Intergovernmental Panel on Climate Change) standards or Life Cycle Assessment (LCA) databases (for example, the coefficient for coal-fired power generation is significantly higher than that for wind power generation).

[0107] To balance policy compliance and physical accuracy in carbon emission accounting, the dynamic carbon emission calculation engine further introduces a weighted synthesis mechanism. The dynamic carbon emission calculation engine will calculate the real-time average carbon emission intensity F on the physical side. grid (t,R) is weighted and fused with the annual baseline carbon emission factor released by the national competent authority to generate the final comprehensive dynamic carbon emission factor λ(t) used for accounting:

[0108] λ(t)=ω policy ·F baseline +ω phys ·F grid (t,R);

[0109] Among them, F baseline This refers to the average grid baseline emission factor for the previous year published by the State Grid or regional grid, which is usually used as the legal basis for policy-based carbon accounting; ω policy ω represents the weighting coefficient of the policy benchmark factor. phys This represents the weighting coefficient of the physical real-time factor.

[0110] The intelligent power distribution cloud platform allows administrators to configure weighting coefficients according to different accounting scenarios. In the scenario of compliant carbon reporting, the intelligent power distribution cloud platform will use ω... policy Set to 1, and set ω phys Set to 0; in scenarios involving refined low-carbon construction management and green energy consumption analysis, the intelligent power distribution cloud platform increases ω. phys The value of this value allows the comprehensive dynamic carbon emission factor λ(t) to sensitively reflect the reduction in grid carbon intensity during periods of high wind power generation at night or high photovoltaic power generation at midday, thereby guiding construction units to arrange high-energy-consuming operations during low-carbon periods.

[0111] See attached document Figure 3 The dynamic carbon emission calculation engine in the intelligent power distribution cloud platform executes real-time carbon emission calculation logic. The core of this logic lies in aligning and integrating high-frequency sampled energy consumption data with time-varying dynamic carbon emission factors over time, thereby achieving calculus-integral level calculation of carbon emissions. The dynamic carbon emission calculation engine first retrieves the k-th outgoing circuit from a specific construction distribution box within the time window [T] from the time-series database. start ,T end All uploaded data records within [the specified range] are generated by the edge computing unit, and each record contains a unified world coordinated timestamp t. j And the average active power value P within the data upload period ΔT corresponding to that timestamp. avg (k,t j ).

[0112] At the same time, the dynamic carbon emission calculation engine retrieves the same time window [T] from the dynamic carbon emission factor database. start ,T end The integrated dynamic carbon emission factor sequence within the [database] is calculated using a zero-order hold-below interpolation operation, since the update frequency of power dispatch data (e.g., every 15 minutes) is typically lower than the upload frequency of edge data (e.g., every 1 minute).

[0113] Specifically, for any upload timestamp t j The dynamic carbon emission calculation engine retrieves the comprehensive dynamic carbon emission factor sequence and finds the effective time interval [T]. factor_m ,T factor_m+1 ), making T factor_m ≤t j <T factor_m+1 The dynamic carbon emission calculation engine will calculate the factor value λ(T) corresponding to the effective time interval. factor_m Assign the value to time t j Instantaneous emission factor λ sync (t j This allows for strict alignment of the power data sequence and the emission factor sequence on the time axis.

[0114] After data alignment is completed, the dynamic carbon emission calculation engine uses the Discrete Riemann sum algorithm to calculate the total carbon emissions of the loop within the time window to be calculated. The calculation process is described as follows:

[0115]

[0116] Among them, C total (k) represents the cumulative carbon emissions of the k-th outgoing circuit within the time window to be calculated, in kilograms of carbon dioxide equivalent; N represents the total number of data upload records within the time window to be calculated; P avg (k,t j ) indicates that the k-th loop is at time t j The average active power, in watts; ΔT represents the data upload cycle of the edge computing unit, in hours; 1000 represents the conversion of watts to kilowatts; λ sync (t j ) represents time t j The corresponding aligned integrated dynamic carbon emission factor is expressed in kilograms of carbon dioxide equivalent per kilowatt-hour.

[0117] The dynamic carbon emission calculation engine not only calculates the carbon emissions of a single loop, but also performs multi-dimensional aggregate calculations. Based on the device type tags uploaded by the edge computing unit, the engine aggregates the carbon emissions of different loops belonging to the same type (e.g., all tower cranes). total(k) is accumulated to generate a report on the carbon emissions of individual equipment.

[0118] The dynamic carbon emission calculation engine will calculate the C emissions of all outgoing circuits in the construction distribution box. total (k) is accumulated to generate a node carbon emission report for a single distribution box. The dynamic carbon emission calculation engine writes the calculated carbon emission data into the distributed ledger database. Each carbon emission record in the distributed ledger database contains the original power value on which the calculation is based, the carbon emission factor value used, and the timestamp of the calculation, supporting subsequent carbon audits and data traceability.

[0119] See attached document Figure 3 The dynamic carbon emission calculation engine in the intelligent power distribution cloud platform performs construction carbon intensity analysis. The construction carbon intensity analysis aims to build a mapping relationship between carbon emission values ​​and the actual quantity of the project, thereby deriving carbon emission performance indicators per unit of work.

[0120] The dynamic carbon emission calculation engine first determines the target equipment loop to be analyzed and its corresponding time window. The dynamic carbon emission calculation engine then calls the cumulative carbon emission C of the target equipment loop within the specified time window, obtained from the aforementioned real-time carbon emission calculation logic. total (k).

[0121] In order to obtain effective workload data on the denominator side, the intelligent power distribution cloud platform establishes a data interoperability link with external smart construction site management systems or equipment-specific IoT platforms through encrypted interfaces.

[0122] For different types of construction machinery, the dynamic carbon emission calculation engine requests the effective workload W. job (k) have different physical dimension definitions:

[0123] For tower cranes and construction hoists, the effective workload W job (k) is defined as lifting ton-meters, which is the product of the lifting mass (tons) and the vertical displacement (meters);

[0124] For concrete mixers or mortar mixers, the effective working capacity W job (k) is defined as the volume (cubic meters) of concrete produced by mixing;

[0125] For AC welding machines, the effective workload W job (k) is defined as the effective welding time (hours) or the amount of welding material consumed.

[0126] When workload data cannot be directly obtained through external APIs, the dynamic carbon emission calculation engine executes a physical inversion algorithm based on the principle of energy conservation. The dynamic carbon emission calculation engine utilizes the electrical energy data uploaded by the target equipment loop and the equipment's inherent energy conversion efficiency parameter η. mech(Mechanical efficiency) is used to inversely estimate the effective workload. Taking a construction hoist as an example, the dynamic carbon emission calculation engine converts the electrical energy consumed during the ascent process into an increase in gravitational potential energy, thereby estimating the effective workload W. est .

[0127] The dynamic carbon emission calculation engine calculates the carbon emission intensity index of the construction equipment based on the acquired cumulative carbon emissions and effective workload. The calculation formula is as follows:

[0128]

[0129] Among them, I CI (k) represents the carbon emission intensity of the k-th device loop; C total (k) represents the cumulative carbon emissions during the statistical period; W job (k) represents the measured or estimated value of the effective workload within the same statistical period.

[0130] The dynamic carbon emission calculation engine further calculates the carbon emission intensity index I. CI (k) and the industry-standard carbon intensity benchmark value I pre-installed in the cloud database baseline The dynamic carbon emission calculation engine compares the actual intensity with the benchmark value, using the formula D. dev :

[0131]

[0132] When the deviation D dev When the value is positive and exceeds the preset tolerance range (e.g., +10%), the dynamic carbon emission calculation engine marks the device as a low-carbon-efficiency operating unit. The dynamic carbon emission calculation engine generates a diagnostic report that includes the device ID, the period of exceedance, the actual intensity value, and suggested optimization measures. The diagnostic report is then sent to the visualization management module for highlighting and display, while providing the energy efficiency diagnosis module with a basis for control decisions.

[0133] See attached document Figure 3 The energy efficiency diagnosis module and load anomaly identification module in the intelligent power distribution cloud platform work together to perform energy efficiency diagnosis tasks. The core of the energy efficiency diagnosis task is to extract non-productive energy waste behaviors and potential electrical safety hazards from massive time-series electrical parameter data, and trigger physical-level control actions accordingly.

[0134] The energy efficiency diagnostic module first executes the logic for identifying illegal idling of construction machinery. The energy efficiency diagnostic module continuously reads the loop switch status bit S uploaded by the edge computing unit. status and real-time active power P real (t), the energy efficiency diagnostic module has preset no-load power thresholds P for different equipment types. idle_thand the idling duration threshold T idle_limit .

[0135] When the energy efficiency diagnostic module detects that a certain loop meets the following combined conditions, the energy efficiency diagnostic module determines that the target equipment loop is in an illegal idling state:

[0136] Switch status bit S status Displayed as closed (logic 1); Real-time active power P real (t) remains below the no-load power threshold P idle_th The duration of the state ΔT duration The idling duration threshold T was exceeded. idle_limit .

[0137] For violations of idling status, the energy efficiency diagnosis module generates an abnormal event record containing the device ID, start time of idling, and cumulative idling energy consumption, and pushes the abnormal event record to the visualization management module.

[0138] The load anomaly identification module primarily performs electrical fault early warning based on three-phase unbalance and current harmonic analysis. The module calculates the three-phase current unbalance U of the current circuit. I The calculation formula is as follows:

[0139]

[0140] Among them, I a ,I b ,I c These are the effective values ​​of the three-phase currents; I avg It is the arithmetic mean of the three-phase currents.

[0141] When the calculated three-phase current imbalance U I When the preset safety limit is exceeded, or when the total harmonic distortion rate exceeds the national standard limit, the load anomaly identification module generates an electrical fault warning signal. Based on the above diagnostic results, the intelligent power distribution cloud platform executes a hierarchical closed-loop linkage control strategy:

[0142] The first level of control is carbon budget early warning control, with the intelligent power distribution cloud platform monitoring the total cumulative carbon emissions (C) of the entire construction project in real time. project_total When the cumulative total carbon emissions C project_total Achieve the project's phased carbon emission budget B budget When the power distribution reaches 80%, the intelligent power distribution cloud platform automatically sends a yellow warning message to project managers via SMS gateway or APP push service.

[0143] The second level of control is remote active intervention control. The intelligent power distribution cloud platform triggers the active intervention mechanism when one of the following two situations occurs:

[0144] Scenario 1: The energy efficiency diagnostic module detects that a certain circuit is in an illegal idling state for a long time, and no manual intervention is detected within a certain period of time after a reminder is sent to the management personnel (i.e. the equipment is still not shut down);

[0145] Scenario 2: Receiving a demand-side power curtailment order from the grid or the project's cumulative carbon emissions C project_total Budget B has been exhausted budget .

[0146] After the active intervention mechanism is triggered, the intelligent power distribution cloud platform generates an encrypted tripping control command. The platform then sends this command to the corresponding construction distribution box via the communication network. The edge computing unit within the construction distribution box receives and decrypts the command. The edge computing unit then verifies the command's validity and the target circuit ID.

[0147] After the verification is passed, the edge computing unit outputs a high-level drive signal to the safety linkage control module. The relay contacts inside the safety linkage control module close, connecting the coil power supply of the shunt trip unit equipped with the target circuit breaker. After the shunt trip unit coil is energized, it generates electromagnetic attraction, which pushes the mechanical locking mechanism of the circuit breaker to trip, thereby physically cutting off the power supply to the circuit and forcibly stopping the equipment from idling or exceeding emission standards, realizing a closed loop from digital diagnosis to physical control.

[0148] See attached document Figure 1 As the human-computer interaction front end of the intelligent power distribution cloud platform, the visualization management module is built on graphics rendering technology. The visualization management module establishes a full-duplex communication channel with the backend server through a long connection protocol to achieve millisecond-level real-time data refresh.

[0149] The visualization management module is configured to generate and display the carbon and electricity dual-flow Sankey diagram at the engineering site. The visualization management module extracts the total incoming power data of the power input unit and the branch power data of each metering unit from the time-series database. The visualization management module maps the total incoming power data to the source node width of the Sankey diagram and maps each branch power data to the branch node width of the Sankey diagram. Based on the real-time carbon emissions calculated by the dynamic carbon emission calculation engine, the visualization management module dynamically adjusts the color saturation of each branch node. When the carbon emission intensity of a certain circuit exceeds the preset threshold, the visualization management module renders the branch node as a red highlight.

[0150] The visualization management module also includes a 3D building information model fusion display submodule. This submodule maintains a mapping table that records the correspondence between the ID of each outgoing circuit of the construction distribution box and the globally unique identifier of the corresponding construction machinery primitive object in the 3D building information model. Based on the real-time carbon emission intensity index of each circuit, the 3D building information model fusion display submodule drives the 3D building information model engine to change the surface material color of the corresponding construction machinery primitive object, thereby intuitively presenting the physical location distribution of high-carbon emission equipment in the 3D digital twin scene.

[0151] The visualization management module provides multi-dimensional statistical report generation functions. It can generate three-dimensional correlation reports of energy consumption, carbon emissions, and costs, categorized by hour, shift, and equipment type. The visualization management module multiplies the cumulative carbon emissions by the preset carbon tax unit price or carbon trading price to calculate the virtual carbon cost, and then overlays the virtual carbon cost with the actual electricity cost to generate a comprehensive energy cost curve.

[0152] In terms of data interaction, the visual management module integrates a standardized application programming interface (API) gateway. The API gateway supports a RESTful architecture and provides data services to external systems via encrypted protocols.

[0153] Application Programming Interface (API) gateways specifically provide the following two types of data interaction services:

[0154] The first category is the government regulatory platform docking service. In accordance with the "Green Construction Monitoring Data Exchange Standard" stipulated by the housing and construction department, the application interface gateway encapsulates the power data collected by the construction distribution box and the carbon emission data calculated by the smart power distribution cloud platform into a JSON format data packet. The application interface gateway pushes the JSON format data packet to the government green construction supervision cloud platform on a regular basis through the HTTP POST request method to meet the compliance reporting requirements.

[0155] The second category is enterprise internal system integration services. The application programming interface gateway grants data reading permissions to the enterprise resource planning (ERP) system of construction companies. The ERP system obtains the carbon emission inventory for a specified project and time period by calling the interface provided by the application programming interface gateway. The ERP system then uses the carbon emission inventory to calculate construction costs and prepare the company's environmental, social and governance (ESG) reports.

[0156] In addition, the visualization management module also supports exporting encrypted raw data files in CSV or Excel format. The raw data files contain raw sampling data and calculation process data with digital signatures, which can be used for third-party audit traceability in the event of carbon emission data disputes.

Claims

1. A smart power distribution box for engineering projects with high-precision carbon emission data acquisition, characterized in that, This includes construction distribution boxes, communication networks, and intelligent power distribution cloud platforms for engineering projects; The construction distribution box of the project is used to distribute power, collect data and perform edge computing on the electrical equipment at the construction site. The intelligent power distribution cloud platform is deployed on the server side and is used for data storage, carbon emission accounting and generation control strategies. The communication network is used to connect the construction distribution box of the project and the intelligent power distribution cloud platform to achieve two-way communication. The construction power distribution box of the project integrates a power input unit, a switch protection unit, a metering unit, a data acquisition unit, an edge computing unit, a safety linkage control module, and a communication module. The switch protection unit is electrically connected to the power input unit and is used to perform circuit on / off control. The metering unit sets up independent metering channels for each outgoing circuit of the switch protection unit. The data acquisition unit is used to collect electrical parameter data for each outgoing circuit of the switch protection unit. The edge computing unit is connected to the data acquisition unit and is used to identify the electrical parameter data and determine the type of equipment connected to the outgoing circuit using a built-in construction load equipment profiling model. The edge computing unit then sends the processed data and equipment type tags to the intelligent power distribution cloud platform through the communication module. The safety linkage control module is electrically connected to the switch protection unit. The safety linkage control module is used to receive control commands from the edge computing unit and drive the switch protection unit to perform a trip power-off operation. The intelligent power distribution cloud platform includes a data access module and a dynamic carbon emission calculation engine. The dynamic carbon emission calculation engine is used to calculate the real-time carbon emissions of the outgoing circuit based on the cumulative energy consumption during the data upload cycle and the comprehensive dynamic carbon emission factor determined based on the real-time power structure of the regional power grid.

2. The intelligent power distribution box for high-precision carbon emission data acquisition in engineering projects according to claim 1, characterized in that, The construction distribution box of the project is physically divided into a high-voltage isolation area and a low-voltage shielding area. The power input unit and the switch protection unit are arranged in the high-voltage isolation area, and the data acquisition unit, the edge computing unit and the communication module are arranged in the low-voltage shielding area. The data acquisition unit is physically equipped with a current acquisition component and a voltage acquisition circuit at the load side output terminal of the branch circuit breaker of the switch protection unit. The current acquisition component adopts a high-precision Hall current sensor based on the closed-loop magnetic balance principle. The high-precision Hall current sensor is directly connected to the line on the output side of the branch circuit breaker.

3. The intelligent power distribution box for high-precision carbon emission data acquisition in engineering projects according to claim 1, characterized in that, The edge computing unit is used to identify the electrical parameter data using a built-in construction load equipment profiling model, specifically including: Real-time monitoring of current changes in the electrical parameter data; when a load state switching event is detected, data from a feature window is captured. A feature extraction operation is performed on the feature window data to construct a load feature vector. The dimensions of the load feature vector include at least the starting impact current multiple, the starting process duration, the steady-state power factor, and the total harmonic distortion rate of the current waveform. Call the pre-set construction load equipment portrait model library and calculate the similarity distance between the currently extracted load feature vector and the baseline feature template in the construction load equipment portrait model library; When the minimum similarity distance is less than the preset confidence threshold, the successfully matched device type label is written into the loop configuration. The edge computing unit distinguishes AC welding machines, tower cranes, construction hoists and lighting loads based on the current fluctuation variance, steady-state power factor and harmonic component characteristics in the load feature vector.

4. The intelligent power distribution box for high-precision carbon emission data acquisition in engineering projects according to claim 1, characterized in that, The edge computing unit is also used to perform offline computing and breakpoint resume mechanism; When the communication network is determined to be interrupted, the edge computing unit switches to offline working mode. The edge computing unit uses the local carbon emission factor in the local dynamic carbon emission factor cache table to estimate the real-time carbon emission of the power data in the current collection period, and writes the estimation result and timestamp into the circular data buffer of the local non-volatile memory. Once the communication network is determined to be restored, the edge computing unit sends a breakpoint query request to the smart power distribution cloud platform, locates the unuploaded data records based on the returned last confirmation timestamp, packages and sends them to the smart power distribution cloud platform, and the smart power distribution cloud platform uses historical actual carbon emission factors to correct and update the offline estimated carbon emission data.

5. The intelligent power distribution box for high-precision carbon emission data acquisition in engineering projects according to claim 1, characterized in that, The safety linkage control module includes an output interface for an opto-isolated relay. The branch circuit breaker in the switch protection unit is equipped with a shunt trip coil or an electric operating mechanism. The edge computing unit sends a control level to the safety linkage control module to drive the opto-isolation relay to close, thereby connecting the external control circuit of the shunt trip coil, which in turn drives the mechanical tripping mechanism of the branch circuit breaker to operate and physically cuts off the power supply to the corresponding outgoing circuit.

6. The intelligent power distribution box for high-precision carbon emission data acquisition in engineering projects according to claim 1, characterized in that, The dynamic carbon emission calculation engine, in conjunction with the real-time power structure of the regional power grid, determines the comprehensive dynamic carbon emission factor, specifically including: Read the GPS geographic coordinates information of the construction distribution box when it was registered for the project, and map the construction project to the specific regional power grid supply range; The real-time power structure data of the regional power grid at the current moment is obtained through the data exchange interface. The real-time power structure data includes the real-time grid-connected power values ​​of different power generation types. Based on the real-time power structure data and the carbon emission coefficients of various power sources throughout their entire life cycle, the real-time average carbon emission intensity of the regional power grid on the physical side is calculated. The real-time average carbon emission intensity on the physical side is then weighted and fused with the annual benchmark carbon emission factor issued by the national competent authority according to a preset weight to generate the comprehensive dynamic carbon emission factor.

7. The intelligent power distribution box for high-precision carbon emission data acquisition in engineering projects according to claim 6, characterized in that, The dynamic carbon emission calculation engine calculates the real-time carbon emissions of the outgoing circuit, specifically including: Retrieve data records uploaded by the edge computing unit within the time window to be calculated from the time series database; Retrieve the comprehensive dynamic carbon emission factor sequence within the time window to be calculated from the dynamic carbon emission factor database; The timestamps of the data records are matched with the effective time intervals of the comprehensive dynamic carbon emission factor sequence to align the power data sequence and the carbon emission factor sequence on the time axis. The discrete Riemann sum algorithm is used to multiply and sum the aligned average active power, data upload cycle and the comprehensive dynamic carbon emission factor to obtain the cumulative carbon emissions of the outgoing circuit within the time window to be calculated.

8. The intelligent power distribution box for high-precision carbon emission data acquisition in engineering projects according to claim 1, characterized in that, The intelligent power distribution cloud platform also includes an energy efficiency diagnostic module; The energy efficiency diagnosis module is used to perform illegal idling identification. When it is detected that the switch status of the outgoing circuit is closed, and the real-time active power is continuously less than the preset no-load power threshold and exceeds the idling duration threshold, the outgoing circuit is determined to be in an illegal idling state. The intelligent power distribution cloud platform is used to monitor the total cumulative carbon emissions of the project in real time. When the total cumulative carbon emissions of the project reach the preset carbon budget threshold or when long-term illegal idling is detected without human intervention, an active intervention mechanism is triggered to generate a trip control command and send it to the edge computing unit. After verification by the edge computing unit, the power is cut off through the safety linkage control module.

9. The intelligent power distribution box for high-precision carbon emission data acquisition in engineering projects according to claim 1, characterized in that, The intelligent power distribution cloud platform also includes a visualization management module, which contains a 3D building information model fusion display sub-module. The 3D building information model fusion display submodule maintains a mapping table that records the correspondence between the outgoing circuit IDs of the construction distribution boxes of the project and the construction machinery primitive objects in the 3D building information model; The visualization management module is used to drive the 3D building information model engine to change the surface material color of the corresponding construction machinery primitive objects according to the real-time carbon emission intensity index of each outgoing circuit, and to present the physical location distribution of high carbon emission equipment in the 3D digital twin scene.

10. The intelligent power distribution box for high-precision carbon emission data acquisition in engineering projects according to claim 9, characterized in that, The visual management module integrates an application programming interface (API) gateway; The application programming interface gateway is used to encapsulate the collected power data and calculated carbon emission data into JSON format data packets in accordance with the green construction monitoring data exchange standard, and push them to the government's green construction supervision cloud platform on a regular basis. The application programming interface gateway is also used to open a data interface to the enterprise resource planning system, allowing the enterprise resource planning system to obtain the carbon emission inventory of a specified project for construction cost accounting.

Images