Real-time monitoring and low-carbon optimization control system for carbon emission of building electrical construction

Abstract

This invention discloses a real-time monitoring and low-carbon optimization control system for carbon emissions during building electrical construction. This invention relates to the field of industrial program control system technology and is the first to achieve industrial-grade closed-loop control in a building electrical construction scenario. Unlike existing technologies that rely solely on data statistics or passive monitoring, this invention achieves a complete control closed loop of "data acquisition-analysis-optimization-regulation-feedback," which can proactively optimize construction carbon emissions and effectively reduce the total carbon emissions during the electrical construction phase. The invention features a central optimization control layer that communicates with the edge computing layer and incorporates a built-in low-carbon optimization control model for electrical construction. It receives data uploaded from each edge computing layer for global scheduling, calculates and generates a global optimization scheduling scheme, which includes equipment operating parameters, process sequence, and temporary power supply allocation strategies.

Description

Technical Field

[0001] This invention relates to the field of industrial program control system technology, specifically to a real-time monitoring and low-carbon optimization control system for carbon emissions from building electrical construction. Background Technology

[0002] Carbon emission control in the construction sector is increasingly becoming a focus of industry attention. However, current carbon emission control in the construction sector mostly focuses on the building operation phase. For the construction phase, especially the high-emission, high-dynamic, and frequently temporary equipment access sub-scenarios of building electrical construction, there is a serious lack of suitable industrial control systems. This makes it impossible to achieve real-time energy consumption monitoring and closed-loop optimization of control parameters, resulting in a huge blind spot in carbon emission control during the construction phase.

[0003] Existing general carbon emission data integration technologies, such as the multi-source heterogeneous carbon emission data integration system disclosed in Chinese patent document CN202510394307.2, can only realize the collection, screening, integration of multi-source emission data at the enterprise level and carbon emission prediction at the planning level. Its core technology is still offline data processing, lacking the closed-loop control capability at the industrial level. It cannot adjust control parameters in real time for dynamic construction processes. More importantly, this technology has not been architecturally designed for the complex scenarios of building electrical construction, and cannot adapt to the real-time access and industrial control needs of temporary equipment and dynamic working conditions during construction. It can only realize post-event data statistics and cannot achieve proactive low-carbon optimization of construction carbon emissions.

[0004] Furthermore, some existing building energy regulation technologies only target fixed operating equipment in completed buildings (such as central air conditioning and fixed lighting), with the control objective being operational energy consumption optimization that balances personnel comfort. They completely fail to cover the building construction phase. Due to the extensive use of temporary power and mobile construction equipment during the construction phase, the dynamic changes in work processes are extremely rapid, and fixed centralized control architectures cannot be reused in construction scenarios. At the same time, existing site monitoring technologies often only focus on the operating status of the equipment itself and fault alarms, which are passive data monitoring tools. They also lack the global scheduling optimization and actuator closed-loop control capabilities of the industrial control field, and cannot meet the proactive low-carbon management needs of the building electrical construction phase. Summary of the Invention

[0005] To address the technical problems in existing technologies, such as the lack of carbon emission control during the building construction phase, the absence of a closed-loop industrial control system adapted to the dynamic scenarios of electrical construction, and the inability to achieve real-time low-carbon optimization during the construction process, this invention provides a real-time monitoring and low-carbon optimization control system for carbon emissions during building electrical construction. This invention fills the technical gap in industrial control-type carbon emission management in the building electrical construction scenario, and provides a feasible real-time monitoring and low-carbon optimization control solution for the building electrical construction phase.

[0006] The system of the present invention adopts a four-layer distributed industrial control architecture, including a distributed sensing layer, an edge computing layer, a central optimization control layer, and an execution and regulation layer; Distributed sensing layer: Deployed in various electrical construction areas of the construction site, including dynamically connected temporary power monitoring terminals, construction equipment status sensors and carbon emission collection terminals, to achieve second-level collection of power consumption data, equipment operating data and real-time carbon emission data; It also supports plug-and-play access for temporary construction equipment; Edge computing layer: Local nodes deployed in various electrical construction areas are used to perform real-time preprocessing of local data collected by the distributed sensing layer, and to execute abnormal data removal and smoothing filtering algorithms; Simultaneously, in response to sudden local carbon emission anomalies or network communication anomalies, implement rapid preliminary edge local control and network outage self-governance based on state adaptation; Central Optimization Control Layer: Communicates with the edge computing layer, has a built-in dedicated low-carbon optimization control model for electrical construction, performs global scheduling by receiving data uploaded from each edge computing layer, and calculates and generates a global optimization scheduling scheme; The optimized scheduling scheme includes equipment operating parameters, process operation sequence, and temporary power supply allocation strategy; The execution control layer is connected to the central optimization control layer and the edge computing layer. It is used to receive and parse the global optimization scheduling scheme or the preliminary control instructions of the edge nodes, make low-level dynamic adjustments to the operating power, operation sequence and temporary power supply allocation parameters of the construction equipment in each construction area, and feed back the adjusted status data to the distributed perception layer to form a complete control closed loop.

[0007] Furthermore, during real-time preprocessing, the edge computing layer employs a time-step-based weighted filtering algorithm to process the original acquired power, resulting in a smoothed operating power.

[0008] Furthermore, the electrical construction low-carbon optimization control model of the central optimization control layer is solved with the objective function of minimizing real-time carbon emissions, and satisfies multiple linear constraints including current process progress constraints, equipment rated load constraints, and temporary power supply capacity constraints at the construction site.

[0009] Compared with the prior art, the beneficial effects of the present invention by adopting the above technical solution are as follows: (1) Innovation of closed-loop control system: For the first time, industrial-grade closed-loop control has been realized in the construction of building electrical systems. Unlike the simple data statistics or passive monitoring of existing technologies, it realizes a complete control closed loop of "data acquisition-analysis-optimization-regulation-feedback". It can actively optimize construction carbon emissions and effectively reduce the total carbon emissions in the electrical construction phase.

[0010] (2) Breakthrough in scenario adaptability: In view of the temporary and dynamic characteristics of electrical construction, a distributed control architecture was designed to support the dynamic plug-and-play access of temporary construction equipment. The equipment access response time reaches the second level, which is far superior to the minute-level access capability of the existing fixed architecture. This solves the problem that the existing technology cannot adapt to the discrete and dynamic working conditions of construction scenarios.

[0011] (3) Improved precision: The exclusive low-carbon optimization model for electrical construction is customized for the characteristics of construction procedures, taking into account the constraints of carbon emission optimization, construction progress, and safe electricity use. It achieves maximum optimization of carbon emissions without affecting any construction progress.

[0012] (4) Real-time response and fault tolerance during network outage: The system realizes high-frequency data acquisition and control response. Combined with the edge adaptive and autonomous collaborative control mechanism of mining, the system can not only capture and curb high carbon emission anomalies in the construction process in real time, but also maintain the continuity and stability of local industrial-grade control in the harsh weak communication environment on site. The control response time is extremely short, which fully meets the needs of rapid changes in construction procedures on the construction site. Attached Figure Description

[0013] Figure 1 This is a system architecture block diagram of the present invention; Figure 2 This is a schematic diagram of the workflow of the present invention. Detailed Implementation

[0014] To make the objectives, technical solutions, and beneficial effects of this invention clearer, the following description is provided in conjunction with the appendix. Figure 1-2 The present invention has been described in great detail with specific embodiments. It should be understood that the specific embodiments provided herein are merely for explaining the technical details of the present invention in detail, so that those skilled in the art can fully implement the present invention, and are not intended to be an absolute limitation on the scope of the present invention.

[0015] Example 1: This embodiment details the application of a real-time carbon emission monitoring and low-carbon optimization control system for building electrical construction in the electrical construction phase of a large commercial building. The electrical construction phase of this commercial building involves a large and highly distributed temporary power demand. According to the process division, the construction site of this commercial building is clearly divided into three core construction areas in physical space: the pipeline installation area in the north, the temporary power supply area in the south, and the equipment commissioning area in the west. The system of this invention is fully deployed in this project and operates in a closed loop strictly in accordance with a four-layer distributed industrial control architecture.

[0016] I. Physical Deployment and Network Registration Mechanism of Distributed Sensing Layer The system implemented targeted underlying sensor network deployments in the three areas mentioned above. In the pipeline installation area of ​​the North Zone, 10 highly sensitive temporary power consumption monitoring terminals were deployed. These monitoring terminals are directly connected to the electrical circuit through current transformers and voltage probes, and are responsible for collecting the power consumption and operating characteristics of high-power power tools (such as large electric threading machines, metal cutting machines, heavy-duty electric hammers, etc.) used during pipeline installation. In the temporary power supply area of ​​the South Zone, 5 power supply monitoring terminals were deployed. These terminals are installed in the main distribution cabinet and box-type substation, and collect key power supply data of the temporary transformers on site in real time, including total apparent power, three-phase imbalance, and harmonic content. In the equipment commissioning area of ​​the West Zone, 8 equipment status sensors were deployed, and the Modbus-RTU industrial fieldbus was used to cyclically read the internal status register data of large electrical commissioning equipment (such as chiller unit commissioning test instrument and fire-fighting pump control cabinet).

[0017] All terminals have a strictly set standard data acquisition cycle of 1 second. It is particularly worth emphasizing that, in order to adapt to the temporary and mobile characteristics of construction equipment, the perception layer has an embedded dynamic plug-and-play identification mechanism. When construction workers connect a new temporary construction device with radio frequency identification (RFID) or specific electrical characteristics to the intelligent temporary distribution box at the construction site, the perception terminal adds a message to the upper-level network broadcast device within 500ms of detecting the load current, automatically completes the device identity resolution and network registration. Subsequently, the perception layer converts the various physical quantities acquired into real-time carbon emission conversion data through the built-in algorithm and publishes it to the upper layer using the MQTT protocol.

[0018] II. Data Preprocessing and Rapid Anomaly Intervention in the Edge Computing Layer The system has configured edge computing nodes equipped with industrial-grade ARM processors for the North, South and West zones respectively, and installed these nodes in the secondary control boxes of each zone. The edge computing nodes not only serve as data aggregation gateways, but more importantly, they are responsible for the preprocessing of local high-frequency data and initial intervention in emergency situations.

[0019] During operation, edge computing nodes verify and clean the received data reported every second, removing distorted data frames caused by electromagnetic interference such as electrical sparks. Taking the pipeline installation area in the North District as an example, during a peak construction period on a certain day, the edge node monitoring program detected a sudden surge in the instantaneous power of a newly connected high-power cutting machine. According to the local carbon emission conversion matrix calculation, this sudden condition caused the local carbon emission rate to exceed the preset reasonable threshold within 3 seconds. At this time, in order to avoid control lag caused by network link queuing in the central optimization scheduling, the edge computing node immediately initiated local rapid preliminary regulation. The edge node directly issued a "current peak pruning" command to the execution mechanism that manages the circuit. By utilizing the fast switching characteristics of the intelligent solid-state relay, the upper limit of the power output of the circuit where the cutting machine is located was softly limited, thereby curbing unnecessary carbon emission accumulation at the incipient stage of the anomaly. Subsequently, the node timestamped this action and reported it to the central node, waiting for further takeover by the global optimization command.

[0020] III. Global Scheduling and Mathematical Model Solution of the Central Optimization Control Layer In the project's main control room, a central server node (i.e., the central node) equipped with a high-performance computing environment was deployed. This central node, as the brain of the system, subscribes to and receives aggregated preprocessed data uploaded from three regional edge computing nodes in real time, and inputs this massive amount of data into a low-carbon optimization control model specifically built for electrical construction. This control model can calculate and search for the globally optimal carbon emission scheduling path in the complex collaborative network of equipment on the construction site.

[0021] The low-carbon optimization control model for central electrical construction uses real-time carbon emission minimization as the objective function, and its mathematical expression is set as follows: ; The physical and mathematical meanings of the relevant variables in the above formula are explained as follows: This represents the total carbon emissions from construction within the system planning time window. Representing the The carbon emission factor of each construction equipment (this factor is not a static constant, but a dynamic coefficient calibrated by the central server based on the equipment's nameplate efficiency, current age, and the properties of the power supply it is connected to). Representing the The operating power allocated to each construction piece of equipment within the scheduling cycle; Representing the The operating time of each construction equipment (i.e., the length of time it is continuously powered on and put into production). The mathematical summation symbol represents the summation of all elements participating in the global scheduling during the current time period. The calculation is performed by accumulating the data from each construction device.

[0022] To ensure that the global scheduling scheme can be seamlessly integrated into the actual engineering site without violating the project schedule and safety regulations, the control model is subject to three types of extremely strict linear constraints: The first type is schedule constraints: ,in, This represents the total time threshold set in advance for the current process within the system. The existence of this schedule constraint ensures that when the system attempts to achieve low-carbon goals by suspending or delaying high-carbon emission equipment, no matter the combination, the cumulative working time of any process will not exceed the time limit, thus perfectly guaranteeing the construction progress.

[0023] The second type is load constraint: ,in, Representing the The rated maximum load (nameplate power or safety limit power) of each construction equipment is constrained by this constraint. This ensures that no matter how the low-carbon optimization algorithm iterates and allocates power, it will not forcibly increase the power of a single piece of equipment, causing it to operate beyond its limit. This absolutely guarantees the electromechanical safety of the construction equipment.

[0024] The third category is power supply constraints: ,in, This represents the maximum temporary power supply capacity that a temporary transformer or backup generator at the construction site can provide. Since construction sites are often limited by the capacity of the distribution network, this power supply constraint ensures that the total power demand of the entire construction site at any given time is strictly locked within a safe threshold, effectively preventing large-scale power outages caused by overload.

[0025] The central node iteratively optimizes the objective function and its constraints using embedded high-level heuristic algorithms (such as an improved particle swarm optimization algorithm). During a specific time period, the central layer generates the following highly targeted global optimization scheduling scheme: For the North Zone, the central model detected that if three high-energy-consuming large metal cutting machines were started simultaneously at the current moment, it would not only cause the local power supply peak to exceed the warning line, but also cause the carbon emission efficiency of the overall equipment in the area to plummet. Therefore, the scheduling plan reorganized the timing of these three cutting machines to stagger the peak. Specifically, the instruction was to maintain the normal operation of machine No. 1, while the start-up time of machines No. 2 and No. 3 was staggered by 15 minutes. This fine adjustment significantly reduced the peak power of the North Zone by 12%, and the overall carbon emission of this process also decreased by 16%.

[0026] For the coordination across the South and West zones, the central node analyzes the historical and real-time load curves reported by the power supply monitoring terminal in the South zone to determine that the South zone currently has sufficient redundant capacity. To address the power capacity bottleneck faced by the ongoing commissioning of heavy-duty chiller units in the West zone, the central node generates a dynamic power supply and distribution dispatch command. This command requires the bus tie intelligent circuit breaker between the South and West zones to close, flexibly transferring 10% of the idle power supply capacity in the South zone to the West zone. This measure not only removes the power blockade of the equipment commissioning in the West zone and improves the commissioning and operation efficiency, but also significantly improves the overall load balancing rate of the temporary transformer, thereby reducing the energy waste caused by no-load and light-load conditions.

[0027] For the western region itself, within the tolerance range allowed by the debugging accuracy, the central optimization model sent frequency reduction messages to several frequency converter debugging devices in the western region, reducing their overall operating power by 8%. Thanks to the additional power supply allocated from the southern region and the active reduction of power within the western region, the debugging progress of the equipment was not hindered in any way, and deep carbon reduction was achieved.

[0028] IV. Action feedback and control loop closure of the execution and regulation layer After the central node completes the model solution within milliseconds, it immediately sends the packaged control parameters (including the target value of equipment operating power, process start and stop flags, control words of the bus tie cabinet, etc.) to the terminal modules (such as programmable logic controllers PLCs, frequency converters and intelligent miniature circuit breakers) deployed in various regions through the TCP / IP industrial-grade network. After receiving the control frame, the execution terminal drives the physical relay to engage or adjusts the duty cycle of the power electronic switch to quickly complete the control of the operating parameters of the physical equipment on site.

[0029] Most importantly, after the execution and control layer completes a series of electrical actions, it does not stop working. The execution device immediately sends the action completion confirmation information, along with the actual steady-state status data of the device after the response, back to the distributed sensing layer and simultaneously reports to the edge nodes and the central node. In this way, the physical and information loops of "data acquisition-analysis-optimization-control issuance-status feedback" are seamlessly closed. Based on this, the central system performs rolling optimization and scheme iteration every 5 seconds, giving the entire low-carbon control system a strong dynamic self-correction capability.

[0030] The continuous tracking and verification of the project over two months in Example 1 showed that after the deployment of this system, the total carbon emissions of the overall electrical construction phase of the project were reduced by 18%, while the peak power supply of the main power supply at the construction site was reduced by 12%. Most importantly, many key progress nodes of the project were achieved on time or ahead of schedule, which fully verified the excellent effectiveness and reliability of the four-layer control system.

[0031] Example 2: This embodiment is based on the four-layer core architecture of Embodiment 1. It focuses on the harsh communication environment (such as weak signal areas such as underground garage pipeline construction and deep foundation pit lightning protection grounding construction) and the severe data disturbance caused by high-frequency start-stop of equipment in building electrical construction. It explores and details the adaptive collaborative control method based on state smoothing and edge network outage autonomous mechanism. The introduction of this method completely solves the fatal defects of the traditional architecture, such as control system paralysis and control failure caused by network communication interruption and data spikes. The application scenario of this embodiment is the electrical supporting construction of deep underground integrated pipe gallery of a large municipal infrastructure hub.

[0032] In the confined space of deep underground utility tunnels, which are encased in thick reinforced concrete, the electrical construction process exhibits two significant characteristics. First, the data collected by the sensing layer suffers from severe high-frequency disturbances: for example, welding machines that frequently perform spot welding operations and submersible sewage pumps that frequently start and stop due to water level changes. The electrical characteristics of these devices determine that their power consumption will experience instantaneous spikes of several times or even more than ten times their rated power. Second, the network communication links are extremely fragile: when edge computing nodes transmit data to the central ground node via wireless relays, severe network packet loss often occurs due to physical obstructions and electromagnetic interference from equipment, resulting in intermittent communication links.

[0033] To eliminate the problem of severe data jitter caused by high-frequency start-stop of equipment, the edge computing layer of this invention introduces a first-order low-pass time series smoothing filter mechanism when preprocessing the data collected by the distributed sensing layer. When the edge node receives the raw power sequence pushed by the sensing layer at high frequency, it does not directly use the data to trigger an alarm or upload, but instead performs dynamic weighting processing on it.

[0034] The smoothing filtering algorithm specifically adopts the following mathematical iterative formula: ; In the above formula, the meanings of the relevant independent variables and constants are explained in a very rigorous manner: Representing the The smoothed operating power of a construction device at the k-th time step calculated by the system (this value is used to filter transient pulses and reflect the true long-term thermal effect and carbon emission accumulation rate). The representative perception layer is in the first The data collected at the [number]th time step The original acquisition power of each construction device; The filter coefficient is represented by the edge node expert system. In this underground utility tunnel embodiment, the expert system dynamically calibrates the high-frequency pulsed equipment (such as welding machines) to an appropriate value between 0.25 and 0.35 to give the current raw data a lower weight, thereby effectively suppressing the penetration of burrs. This represents the current time step number, indicating that this is a time-progressive process of a discrete control system; Representing the The construction equipment in the previous calculation period (i.e., the 1st ... The smoothed operating power history value retained after subtracting 1 time step.

[0035] By applying this smoothing filtering algorithm, the edge computing nodes are like adding an electrical damper to the data stream, successfully removing data spikes caused by high-frequency start-stop. This preprocessing avoids the central layer optimization model receiving false load surge signals and misjudging that the entire utility tunnel is facing overload risk, thereby preventing the system from frequently triggering large-scale shutdowns and erroneous control oscillations.

[0036] Meanwhile, in response to the vulnerability of network links caused by signal obstruction in underground spaces, the system of this invention sets up a highly sensitive real-time network status assessment engine at the handshake layer between edge nodes and central nodes. Edge nodes continuously monitor the receipt of heartbeat detection packets and statistically calculate the degree of degradation of network communication quality.

[0037] The evaluation indicators for network outage self-governance are defined by the following monitoring formula: ; The meanings of the corresponding control parameters in the above formula are explained below: Representing the edge node at the th The network packet loss rate is calculated at each time step (this is a dimensionless value between 0 and 1). This represents the number of data packets detected by the edge node that were not acknowledged and returned by the central node within the set detection time sliding window, and which were determined to be lost. This represents the total number of heartbeat packets sent from the edge node to the central node within the same sliding window.

[0038] A packet loss rate threshold is pre-written within the system's underlying program. (In this harsh environment embodiment, to balance the stability and sensitivity of control, it is set to 15%). At some point during construction, the proximity of a large tunnel boring machine caused strong spatial electromagnetic shielding, and the edge node of the underground utility tunnel detected the current packet loss rate. It rapidly climbed and exceeded the packet loss rate threshold. At this point, the system determines that the communication network in the area is in a state of disconnection or extreme unreliability. The system kernel triggers a hardware-level interrupt command, and the edge node immediately and proactively cuts off the high-frequency data synchronization behavior to the central node (to prevent further channel congestion) and seamlessly switches to the "edge disconnection autonomous mode".

[0039] In the edge-disconnected autonomous mode, due to the temporary loss of the powerful global computing resources and global perspective of the central optimization control layer, the edge node calls the miniaturized local scheduling model solidified in its memory to take over the underlying control of all temporary equipment in the utility tunnel section. Its primary purpose is to suppress the disorderly expansion of local carbon emissions by relying on its own computing power, under the premise of ensuring the absolute safety of local construction and uninterrupted construction.

[0040] Edge nodes employ a simplified local prediction objective function for autonomous regulation and optimization, the formula of which is: ; The specific meanings of the relevant variables in the above formula are explained in detail below: This represents the local estimated carbon emissions calculated after entering the autonomous mode without internet access (only covering the energy consumption estimate within the area under the jurisdiction of this edge node). Representing the local first Carbon emission factor of each construction equipment (here, the subscript j is used differently than usual, specifically referring to equipment units that have been transferred to the jurisdiction of a local independent control domain after losing central supervision). Representing the The construction equipment was in the first The operating power obtained after smoothing and filtering each time step; Representing the The remaining operating time of each construction device (this value no longer depends on the cloud, but is derived by the edge node based on the initial input parameters of the local human-machine interface (HMI) or the fitting decay law of local historical operating time). Represents all local first-order autonomous jurisdictions within the marginal autonomous jurisdiction. Summation is performed on each controlled device.

[0041] Meanwhile, the autonomous process of edge-based microgrid disconnection is not an unlimited search for optimization; it must be strictly constrained by the rigid physical bottleneck of the local microgrid island. The key local constraints it must follow are expressed as follows: ; in, This represents the maximum safe output capacity of the local temporary power distribution box for the underground utility tunnel.

[0042] During the aforementioned extreme period of network outage and self-governance, multiple welders simultaneously ignited arcs within the utility tunnel, coupled with the operation of multiple high-power fans, causing the total power demand calculated using a smoothing algorithm to rapidly approach [the required level]. Upon reaching the critical value, the edge node's control program immediately uses the locally persistent device priority list (e.g., welding, as a critical path process, has the highest priority, while temporary exhaust fans and some backup lighting have a lower priority) to forcibly reduce or cut off the operating power of several secondary exhaust fans and some auxiliary water pumps via the underlying relay control cables. This ensures limited... Capacity resources can be tilted and stably supplied to core welding equipment, successfully avoiding the overall tripping of the main distribution box due to overload, realizing the self-rescue of the local power grid and low-carbon micro-circulation in extreme environments.

[0043] When the source of strong electromagnetic interference moves away, the low-frequency, minute detection messages in the edge node's backend detect that the packet loss rate remains stable at a set threshold for two consecutive minutes. When the edge node's adaptive engine determines that the external network link has been fully restored, the node converts the accumulated device status during the autonomous period into historical logs and energy consumption lists, and pushes them in batches to the central optimization control layer in a compressed package format with breakpoint resume. After the data synchronization handshake is successful, the edge node releases its local autonomous state, and the control of the construction site is smoothly transferred back to the global scheduling module of the central layer, and the cross-regional linkage optimization strategy in Example 1 continues to be executed. Example 2 provides a detailed analysis of how the system maintains more than 99% online control continuity in harsh site conditions with extremely unstable network environments through the organic combination of smoothing algorithms and network outage autonomous logic, demonstrating unparalleled robustness.

[0044] Example 3: When the scheduling scheme described in the above embodiments is implemented at the execution control layer, if the electrical physical characteristics of the power equipment cannot be monitored and controlled at high resolution, any low-carbon control strategy will be ineffective. This embodiment focuses on the calculation mechanism of the underlying electrical operating parameters and the refinement process of the compensation action of the execution control layer. There are a large number of loads with significant inductive or capacitive properties at the construction site of building electrical systems (such as AC asynchronous motor spindle systems and large power frequency transformers). These devices will exchange a large amount of non-functional power between the power grid and the devices during operation. If the central system only uses the apparent total power as the optimization parameter input, it will lead to serious distortion of carbon emission conversion and cause the control strategy to fall into deviation.

[0045] In order to separate this part of reactive power that does not generate useful work but occupies power supply capacity and generates ohmic heat loss on the line, the sensing layer has a built-in high-precision Fast Fourier Transform (FFT) analyzer, whose core task is to accurately extract active power.

[0046] The formula for calculating actual active power is: ; The electrical control variables in the above formula are explained in detail below: Representing the The real-time active power of each device (this is the energy form that truly drives the mechanical load to do work, and it is also the only benchmark data for the central low-carbon optimization model to map carbon emission factors). Representing the Real-time operating effective voltage of each device; Representing the Real-time operating effective current of each device; Representing the The power factor of a device (i.e., the cosine of the phase difference between the voltage waveform and the current waveform, used to quantify the active conversion efficiency of electrical energy into mechanical energy).

[0047] When the central system determines that a certain heavy construction equipment with high impedance in Implementation Example 1 or Implementation Example 2 (defined as the first...) When a device (such as a heavy-duty submersible pump array used for continuous dewatering of deep foundation pits) is operating in a state of extremely low efficiency and high energy consumption and issues a carbon reduction command to the execution control layer, the execution layer will not take a drastic power-off restart measure. Instead, after receiving an industrial message containing the target operating parameters, the execution control layer issues the command to the designated intelligent variable frequency drive (VFD) and dynamic var compensator (SVG) by the internal bus controller.

[0048] The variable frequency drive utilizes space vector pulse-width modulation (SVPWM) technology to smoothly and continuously reduce the power supply frequency of the submersible pump array stator windings (e.g., from 50 Hz to 35 Hz at a slope). In this control process, the sensing layer continuously feeds back to the high-speed PID loop within the execution control layer. and More importantly, frequency reduction often causes changes in the internal magnetic field state of the motor, leading to... A sharp drop in the power factor, if left unchecked, will lead to a significant increase in reactive current on the line. The heat generated by line losses due to Joule's law will offset the carbon emission savings achieved through frequency reduction. Therefore, while the frequency converter adjusts the frequency, the reactive power compensation controller monitors the power factor in real time. The drop in power factor is quickly measured, and the internal thyristor components are used to connect the corresponding capacity of the capacitor bank network to the distribution bus. By injecting leading current into the grid to offset the lagging current of the motor, the execution layer forcibly stabilizes the overall power factor of the sub-bus firmly above the excellent range of 0.95.

[0049] Through this meticulous electromechanical coordination and control at the micro level, the system not only precisely reduces the active power reflecting mechanical energy consumption at the source, but also... (It achieves a substantial reduction in direct carbon emissions) and completely eliminates redundant circulating reactive current in the line (eliminating indirect power line losses and hidden carbon emissions). This embodiment deeply reveals how the system, when facing complex electromechanical loads at construction sites, ensures that the top-level abstract algorithm scheduling instructions can be executed in the physical world in the most energy-efficient and precise way through high-order analysis of electrical parameters and closed-loop joint action of control devices. It fully demonstrates the rigor and excellent performance of the industrial-grade carbon emission optimization control system in field applications.

[0050] The above detailed embodiments are merely for illustrating the specific technical solutions, implementation paths, and working mechanisms of the present invention under different extreme environments, and are not intended to limit the scope of protection or creative ideas of the present invention. Any equivalent substitutions, reorganizations of logical functional modules, and non-substantial parameter changes that can be easily made by those skilled in the art based on common knowledge in the field, after a deep understanding of the core distributed architecture and adaptive collaborative control technology disclosed in the present invention, should undoubtedly be covered and fall within the scope of protection claimed by the present invention.

Claims

1. A real-time monitoring and low-carbon optimization control system for carbon emissions from building electrical construction, characterized by: The system adopts a four-layer distributed industrial control architecture, including a distributed sensing layer, an edge computing layer, a central optimization control layer, and an execution and regulation layer. The distributed sensing layer is deployed in various electrical construction areas of the construction site, including dynamically accessed temporary power consumption monitoring terminals, construction equipment status sensors, and carbon emission collection terminals. It is used to realize the second-level collection of power consumption data, equipment operating data, and real-time carbon emission data, and supports plug-and-play access of temporary construction equipment. The edge computing layer is deployed at local nodes in each electrical construction area to perform real-time preprocessing of the local data collected by the distributed sensing layer and to execute abnormal data removal and smoothing filtering algorithms. Simultaneously, in response to sudden local carbon emission anomalies, rapid initial edge-based local control is implemented based on state adaptation; It continuously monitors the heartbeat detection packet receipt between the edge computing layer and the central optimization control layer to obtain the network packet loss rate. When the network packet loss rate exceeds the preset packet loss rate threshold, it triggers the network disconnection autonomous mode for network communication anomalies, and the edge computing layer takes over the underlying control of the devices in the local partition to perform local autonomous regulation and optimization. The central optimization control layer is communicatively connected to the edge computing layer and has a built-in low-carbon optimization control model for electrical construction. It performs global scheduling by receiving data uploaded from each edge computing layer and calculates and generates a global optimization scheduling scheme. The optimized scheduling scheme includes equipment operating parameters, process operation sequence, and temporary power supply allocation strategy; The execution control layer is communicatively connected to the central optimization control layer and the edge computing layer. It is used to receive and parse the global optimization scheduling scheme and the preliminary control instructions of the edge nodes, perform low-level dynamic adjustments to the operating power, operation sequence and temporary power supply allocation parameters of construction equipment in each construction area, and feed back the adjusted status data to the distributed perception layer to form a complete control closed loop.

2. The real-time monitoring and low-carbon optimization control system for carbon emissions from building electrical construction as described in claim 1, characterized in that: The distributed sensing layer has an embedded dynamic plug-and-play identification mechanism. When temporary construction equipment is connected to the intelligent temporary power distribution box at the construction site, the sensing terminal adds a message to the upper-level network broadcasting device after detecting a change in load current, and automatically completes the device identity resolution and network registration. The distributed sensing layer converts the acquired physical quantities into real-time carbon emission conversion data using a built-in algorithm and publishes it to the edge computing layer via the MQTT protocol.

3. The real-time monitoring and low-carbon optimization control system for carbon emissions from building electrical construction according to claim 1, characterized in that: When performing real-time preprocessing, the edge computing layer uses a time-step-based weighted filtering algorithm to process the original collected power and obtain the smoothed operating power. The mathematical iterative formula for the weighted filtering algorithm is: ； in, Representing the The construction equipment was in the first The smoothed operating power is calculated at each time step; The representative perception layer is in the first The data collected at the [number]th time step The original acquisition power of each construction device; Represents the filter coefficients; Represents the current time step number; Representing the The construction equipment was in the first Subtract one time step from the smoothed historical operating power value.

4. The real-time monitoring and low-carbon optimization control system for carbon emissions from building electrical construction according to claim 1, characterized in that: The electrical construction low-carbon optimization control model of the central optimization control layer is solved by optimizing the total carbon emissions of construction as the objective function. The mathematical expression of the objective function is: ； in, This represents the total carbon emissions from construction within the system planning time window. Representing the The carbon emission factor of a construction equipment is a dynamic coefficient comprehensively calibrated based on the equipment's nameplate efficiency, cumulative operating time, and the attributes of the power supply it is connected to. Representing the The operating power allocated to each construction piece of equipment within the scheduling cycle; Representing the Operating time of each construction device; This represents the cumulative calculation of all construction equipment participating in the global scheduling during the current time period.

5. The real-time monitoring and low-carbon optimization control system for carbon emissions from building electrical construction according to claim 4, characterized in that: The low-carbon optimization control model for electrical construction satisfies the following three types of linear constraints during the solution process: Schedule constraint: The total operating time of all construction equipment shall not exceed the total project duration threshold pre-entered in the system for the current process; Load constraint: The operating power allocated to each construction device shall not exceed the rated load of that construction device; Power supply constraints: The total operating power allocated to all construction equipment shall not exceed the rated capacity of the temporary power supply on the construction site.

6. The real-time monitoring and low-carbon optimization control system for carbon emissions from building electrical construction according to claim 1, characterized in that: The offline autonomy of the edge computing layer includes: The edge computing layer continuously monitors the heartbeat probe packet receipt status between itself and the central optimization control layer, and calculates the network packet loss rate. The network packet loss rate is the ratio of the number of data packets lost within a set detection time sliding window to the total number of heartbeat data packets sent. When the network packet loss rate exceeds a preset packet loss rate threshold, the edge computing layer cuts off high-frequency data synchronization to the central optimization and control layer and switches to the edge network disconnection autonomous mode. In the aforementioned autonomous edge disconnection mode, the edge computing layer calls a miniaturized local scheduling model embedded in memory to take over the underlying control of all temporary devices within the local partition. It performs autonomous regulation and optimization with the goal of local carbon emission estimation, while simultaneously meeting the rated power supply capacity constraints of the local temporary power distribution box. When the network packet loss rate remains stable below the packet loss rate threshold for a preset recovery time, the edge computing layer will push the accumulated device state transition history logs and energy consumption list during the autonomous period to the central optimization control layer in batches through the breakpoint resume method. After completing the data synchronization, the local autonomous state will be terminated and control will be handed over back to the central optimization control layer.

7. The real-time monitoring and low-carbon optimization control system for carbon emissions from building electrical construction according to claim 1, characterized in that: The distributed sensing layer has a built-in fast Fourier transform analyzer for accurately extracting the active power of construction equipment. The formula for calculating the active power is: ； in, Representing the Real-time active power of each device; Representing the Real-time operating effective voltage of each device; Representing the Real-time operating effective current of each device; Representing the The power factor of a device is the cosine of the phase difference between the voltage waveform and the current waveform.

8. The real-time monitoring and low-carbon optimization control system for carbon emissions from building electrical construction according to claim 7, characterized in that: The execution control layer includes an intelligent variable frequency drive and a dynamic reactive power compensation controller. The intelligent frequency converter driver uses space voltage vector control technology to continuously adjust the power supply frequency of the stator winding of the construction equipment, thereby achieving stable adjustment of the equipment's operating power. The dynamic reactive power compensation controller is used to monitor the power factor changes of the construction equipment in real time during the frequency adjustment process of the intelligent frequency converter, and inject a capacitor bank of corresponding capacity into the distribution bus through the internal thyristor component to inject leading current into the power grid to offset the lagging current of the motor, thereby stabilizing the overall power factor of the distribution bus above the preset target value.

9. The real-time monitoring and low-carbon optimization control system for carbon emissions from building electrical construction according to claim 1, characterized in that: After completing the electrical action, the execution control layer sends the action completion confirmation information, along with the actual state data of the steady-state operation of the equipment after the response, back to the distributed sensing layer, and simultaneously reports it to the edge computing layer and the central optimization control layer. The central optimization control layer performs rolling optimization and scheme iteration at fixed time intervals, and dynamically corrects the global optimization scheduling scheme based on the status data fed back by the execution control layer.

Images