A precast beam construction power management method, system, device and medium

Abstract

The application relates to a precast beam construction power management method, system, equipment and medium, belonging to the technical field of construction supervision, which comprises the following steps: first, a full-factor digital twin of a precast beam field is built, and multi-dimensional data such as site space, electrical topology, real-time load and personnel positioning are integrated; the electrical loop load, personnel position and behavior information are collected through on-site sensors and are synchronously mapped to the twin model. The power load conflict risks are predicted in combination with the construction plan, and the dynamic safety risks of personnel contacting live equipment are calculated based on the electronic fence, personnel qualifications and equipment working conditions. When the risks exceed the threshold, digital twin simulation deduction is carried out, and collaborative control measures such as loop power regulation, switching operation, regional load scheduling, personnel sound and light alarm driving and the like are executed, the power safety of construction personnel is considered, and the construction efficiency and power resource utilization rate of the precast beam field are improved.

Description

Technical Field

[0001] This application relates to the technical field of construction supervision, and in particular to a method, system, equipment and medium for power management in precast beam construction. Background Technology

[0002] As a core component in transportation engineering construction, precast beams are produced primarily in precast beam yards, involving multiple processes such as steel reinforcement processing, formwork erection, concrete pouring, steam curing, and prestressing tensioning. Each process relies on a large number of high-power electrical devices, and the power usage scenario is characterized by scattered equipment distribution, large load fluctuations, overlapping work areas, and frequent personnel movement. Temporary power safety is a core aspect of construction safety in precast beam yards, permeating the entire production process. It not only directly affects the safety of construction personnel and the integrity of equipment but also impacts construction progress and the quality of precast beams.

[0003] Currently, the power management for precast beam construction largely adopts a technical solution combining manual control and basic protection. Firstly, a basic power management system is established, clearly defining responsibilities and assigning professional electricians to handle the installation, commissioning, and daily maintenance of temporary power systems. Regarding equipment and line management, cables are laid according to specifications, using methods such as direct burial, overhead installation, or conduit protection to avoid mechanical damage. Standardized distribution boxes and switch boxes are installed, implementing the "one machine, one switch, one leakage protector, one circuit breaker" requirement. Regular manual inspections are conducted on cable insulation, leakage protector effectiveness, and equipment grounding. For load management, transformers and power lines are rationally configured based on preliminary load calculations. During construction, electricians manually read meters to monitor the power load in each area, manually disconnecting the power supply to the relevant circuits when overloads or short circuits are detected. For personnel safety management, safety training, on-site warning signs, and physical barriers are used to remind workers to stay away from live equipment. The qualifications of electricians and other special operations personnel are verified, and unqualified personnel are prohibited from operating electrical equipment.

[0004] Although it can meet basic electricity safety requirements, it relies too much on manual operation, which has drawbacks such as untimely data collection, inability to predict load conflicts in advance, and lagging monitoring of personnel safety risks. It is difficult to achieve simultaneous improvement in electricity safety and production efficiency, and it is easy to cause safety accidents or construction delays due to human negligence. Summary of the Invention

[0005] In order to maximize the production efficiency and resource utilization of precast beam yards while ensuring the safety of construction workers, this application provides a method, system, equipment and medium for power management in precast beam construction.

[0006] Firstly, this application provides a method for managing electricity consumption during precast beam construction, employing the following technical solution: A method for managing electricity consumption during precast beam construction includes: Construct a full-element digital twin of the precast beam yard, which integrates the site spatial model, equipment electrical topology, real-time load data, and personnel positioning data; The system collects multi-source data from the field in real time through a pre-deployed sensor network and synchronously maps the multi-source data to the digital twin. The multi-source data includes load data streams of each electrical circuit and personnel data streams. The personnel data streams include personnel location data streams and personnel behavior data streams. Based on the load data stream and the preset construction plan, the risk of load conflicts in future periods is predicted in a rolling manner. Based on the personnel data stream, combined with the preset electronic fence, personnel qualification database and equipment status, the dynamic comprehensive risk index between personnel and electrical equipment is calculated in real time. When the predicted load conflict risk or the dynamic comprehensive risk index exceeds the corresponding preset threshold, a multi-strategy simulation is performed in the digital twin to generate and execute collaborative scheduling instructions. The collaborative scheduling instructions include power regulation or circuit breaker control of the target electrical circuit, load adjustment of the associated area, and audible and visual alarms to drive away specific personnel.

[0007] By adopting the above technical solutions and a real-time multi-source data synchronization mapping mechanism, every change in the physical field can be instantly fed back to the digital twin. Based on the rolling prediction of load data streams and construction plans, the risk of load conflicts in future periods can be identified in advance, avoiding equipment downtime or construction delays caused by imbalances in power resource allocation, and ensuring the stability of the beam yard's production rhythm. Furthermore, the dynamic comprehensive risk index calculated by combining personnel data streams, electronic fences, personnel qualification databases, and equipment status enables intelligent monitoring of the safe distance between personnel and energized equipment, fundamentally avoiding safety hazards such as personnel misoperation and unauthorized entry into dangerous areas. When a risk threshold is triggered, the automatic generation and execution of multi-strategy simulation and collaborative scheduling instructions further transforms passive response into proactive prevention. Through power regulation, load adjustment, and personnel removal, safety and production risks are quickly mitigated, maximizing the production efficiency and resource utilization of the precast beam yard while ensuring the safety of construction personnel.

[0008] Optionally, the step of acquiring multi-source data in real time through a pre-deployed sensor network specifically includes: By deploying smart IoT circuit breakers at various levels of distribution boxes and key electrical equipment, current, voltage, and power data of electrical circuits are collected to form a load data stream; By deploying ultra-wideband positioning base stations throughout the construction site and using positioning tags worn by personnel, three-dimensional coordinate data of personnel is collected to form a personnel location data stream; By deploying visual surveillance equipment in key risk areas, video stream data is collected, and personnel behavior characteristics are extracted through image recognition to form a personnel behavior data stream. By deploying regional edge gateways in each construction zone, the load data stream, personnel location data stream, and personnel behavior data stream of this zone are aggregated and preprocessed.

[0009] Optionally, the step of rolling forecasting of load conflict risks for future periods based on the load data stream and the preset construction plan specifically includes: Obtain a pre-defined detailed construction plan, which includes the expected start time, duration, required equipment, and rated power of each process. Based on the construction plan, a reference load time-series curve for each piece of equipment under ideal operating conditions is generated; The system collects the actual load data stream within the previous time window in real time and calculates the deviation between the actual load and the benchmark load for the corresponding time period. The deviation value is used to correct the baseline load time series curve for subsequent periods to obtain the corrected predicted load curve; The corrected predicted load curve is compared with the safe capacity threshold of the power distribution system to identify the risk time window in which the total load or the load of key branches will exceed the threshold within a set future period. When a risk time window is identified, a list of high-priority devices that will be running in parallel within the corresponding future time period is extracted as the core element of load conflict risk.

[0010] By adopting the above technical solution, the time nodes, equipment configurations, and power requirements of each process in the construction plan are transformed into a benchmark load time-series curve, providing a clear and ideal reference for the allocation of power resources. This allows managers to predict the power load trend at different construction stages in advance. The calculation of the deviation between the real-time collected actual load data stream and the benchmark load, along with the dynamic correction of subsequent prediction curves, eliminates the impact of uncertainties such as differences in personnel operation and changes in equipment operating conditions during construction. This ensures that the prediction results continuously match the actual operating conditions on site, improving the accuracy and reliability of load prediction. Comparing the corrected predicted load curve with the distribution system's safe capacity threshold can identify potential overload risk windows for total load or key branches in future periods, exposing potential power conflicts in advance and allowing sufficient time for dispatch adjustments. When a risk window is identified, the extraction of a list of high-priority equipment provides clear core control targets for load dispatch, making dispatch decisions more targeted and avoiding interference with the construction schedule due to blind adjustments.

[0011] Optionally, the step of calculating the dynamic comprehensive risk index between personnel and live equipment in real time specifically includes: Based on the real-time location of personnel and the pre-set electronic fence of powered equipment, spatial proximity is calculated as a distance risk factor; By analyzing visual surveillance data streams, we can identify the degree to which personnel's actions and intentions comply with safety regulations, and use this as a behavioral risk factor. Access the personnel qualification database to determine whether the current personnel are qualified to operate or approach the target equipment, as a qualification risk factor; Obtain the real-time insulation status, operating mode, and voltage level of the target equipment as a risk factor for equipment status. Based on a preset weighted fusion model, the distance risk factor, behavioral risk factor, qualification risk factor and equipment status risk factor are comprehensively calculated to output the dynamic comprehensive risk index.

[0012] By adopting the above technical solutions, and using electronic fences to calculate the spatial proximity of personnel to live equipment, the potential risks of personnel entering dangerous areas can be detected immediately, moving the safety control point forward and avoiding electric shock accidents caused by excessive close proximity. Identifying the compliance of personnel's actions with safety regulations further refines risk assessment at the behavioral level. It not only monitors personnel location but also reveals whether their behavior exhibits dangerous tendencies such as unauthorized operation or accidental contact, shifting safety management from passive post-incident accountability to proactive pre-incident intervention. The retrieval and verification of personnel qualification databases ensures from the source that only personnel with the corresponding operating qualifications can approach or operate specific equipment, eliminating safety hazards caused by unlicensed operation. Furthermore, the inclusion of real-time insulation status, operating mode, and voltage level of equipment provides the equipment's own safety status as an important reference for risk assessment, avoiding safety risks caused by equipment failure, insulation failure, etc., making risk assessment more closely aligned with actual on-site conditions.

[0013] Optionally, the step of generating the cooperative scheduling instruction specifically includes: When the trigger condition is that the dynamic comprehensive risk index exceeds the first preset threshold, a first type of instruction set with personnel safety as the core is generated first. The first type of instruction set includes emergency power cut-off and access lock-up of risk source equipment, and sending targeted expulsion alarms to risk personnel. When the triggering condition is the aforementioned load conflict risk, a second type of instruction set is generated with the core objective of ensuring production continuity. The second type of instruction set includes scheduling non-critical equipment for delay, rotational shutdown, or power reduction based on the dynamic priority of the process. When the dynamic comprehensive risk index and the load conflict risk exist simultaneously or are related to each other, a fusion instruction set is generated. Under the premise of ensuring personnel safety, the fusion instruction set creates power supply conditions for safety intervention measures by actively adjusting the load of the associated area, or indirectly disperses the gathering of people in the risk area by adjusting the regional power supply strategy.

[0014] By adopting the above technical solutions, when the dynamic comprehensive risk index triggers the threshold, the first type of instruction set, centered on personnel safety, can be quickly activated. This involves cutting off the source of danger by urgently powering down the risk-generating equipment and locking access permissions, while simultaneously sending targeted alarms to at-risk personnel, rapidly placing them in a safe environment. This resolves personal safety crises in the shortest possible time and prevents serious accidents. The second type of instruction set, targeting load conflict risks, focuses on ensuring production. It flexibly schedules non-critical equipment based on dynamic process priorities, optimizing power resource allocation through delays, rotating shutdowns, or power reductions. This solves overload problems while avoiding unnecessary shutdowns, ensuring the smooth progress of critical construction processes and effectively balancing power safety and production efficiency. When both risks exist simultaneously or are interrelated, the fusion instruction set demonstrates stronger synergy and flexibility. Under the premise of ensuring personnel safety, it proactively adjusts the load in related areas to create conditions for safety intervention, or indirectly disperses personnel gathering in risk areas through power supply strategy adjustments. This achieves a deep integration of safety control and production scheduling, eliminating safety hazards while minimizing the impact on production.

[0015] Optionally, the steps following the execution of the cooperative scheduling instruction further include: After the coordinated scheduling instruction is executed, continuously monitor whether the load conflict risk is eliminated and whether the dynamic comprehensive risk index drops below the safety threshold; Once the risk has been confirmed to be eliminated, the safety-after-production resumption protocol is activated. Simulations were conducted in the digital twin to verify whether restoring power would lead to new load conflict risks. If the simulation verification passes, a power restoration command for the device will be generated and issued. If the simulation verification fails, a step-by-step recovery instruction is generated and issued. The step-by-step recovery instruction includes: first, scheduling other non-critical loads to temporarily relinquish the required capacity; then, restoring power to critical equipment; and finally, orderly restoring the scheduled non-critical loads.

[0016] By adopting the above technical solutions and continuously monitoring load conflict risks and dynamic comprehensive risk indices, the timing of risk mitigation can be determined, avoiding safety hazards caused by premature recovery and preventing excessive delays from impacting production schedules. The safety-based production recovery protocol, activated after risk confirmation, establishes a standardized and regulated process for production restart, ensuring a smooth recovery process. Simulation verification of power restoration in a digital twin allows for the prediction of potential new load conflict risks in advance, eliminating potential problems at their inception and avoiding secondary risks that may occur during actual recovery. If the simulation verification passes, a power restoration command is issued directly, quickly restoring production order; if the verification fails, a step-by-step recovery command is used, first scheduling non-critical loads to relinquish capacity, then restoring power to critical equipment, and finally orderly restoring non-critical loads. This tiered recovery strategy ensures priority power supply to critical processes and avoids new conflicts through dynamic load balancing, minimizing losses caused by production interruptions while ensuring safety.

[0017] Optionally, the electricity management method further includes: Continuously record all risk events, scheduling instructions, and execution results to create a traceable production log; Based on production log data, identify high-frequency risk areas, typical risk scenarios, and the actual effectiveness of different scheduling strategies; The analysis results are used to optimize the model parameters, risk factor weighting coefficients, and scheduling strategy library of the digital twin.

[0018] Secondly, this application provides a power management system for precast beam construction, which adopts the following technical solution: A power management system for precast beam construction includes: The digital twin construction and synchronization module is used to construct a full-element digital twin of the precast beam yard and synchronously map the real-time collected multi-source data to the digital twin. The digital twin integrates the site space model, equipment electrical topology, real-time load data and personnel positioning data. The multi-source data acquisition module collects multi-source data in real time through a pre-deployed sensor network. The multi-source data includes load data streams of each electrical circuit and personnel data streams. The personnel data streams include personnel location data streams and personnel behavior data streams. The load conflict risk prediction module is used to predict the load conflict risk in future periods based on the load data stream and the preset construction plan. The dynamic safety risk assessment module is used to calculate the dynamic comprehensive risk index between personnel and live equipment in real time based on the personnel data stream, combined with the preset electronic fence, personnel qualification database and equipment status. The collaborative scheduling decision and execution module is used to perform multi-strategy simulation and deduction in the digital twin when the predicted load conflict risk or the dynamic comprehensive risk index exceeds the corresponding preset threshold, generate and execute collaborative scheduling instructions, including power regulation or circuit breaker control of the target electrical circuit, load adjustment of the associated area, and audible and visual alarms to drive away specific personnel.

[0019] Thirdly, this application provides a computer device that adopts the following technical solution: A computer device includes a memory, a processor, and a computer program stored in the memory, the processor executing the computer program to implement the power management method for precast beam construction as described in the first aspect.

[0020] Fourthly, this application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium storing a computer program that can be loaded by a processor and executed as described in the first aspect, for the management of electrical power consumption during precast beam construction. Attached Figure Description

[0021] Figure 1 This is a first flowchart of an embodiment of the method of this application; Figure 2 This is a second flowchart of an embodiment of the method of this application; Figure 3 This is a third flowchart of an embodiment of the method of this application; Figure 4 This is the fourth flowchart of an embodiment of the method of this application; Figure 5 This is the fifth flowchart of an embodiment of the method of this application; Figure 6 This is the sixth flowchart of an embodiment of the method of this application. Detailed Implementation

[0022] To make the purpose, technical solution, and advantages of this application clearer, the following description is provided in conjunction with the appendix. Figures 1-6 The present application will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the application.

[0023] The first embodiment of this application discloses a method for managing electricity consumption during precast beam construction. (Refer to...) Figure 1 The electricity management method includes S110-S150: S110 constructs a full-element digital twin of the precast beam yard, which integrates the site spatial model, equipment electrical topology, real-time load data, and personnel positioning data. S120 collects multi-source data from the field in real time through a pre-deployed sensor network and synchronously maps the multi-source data to a digital twin. The multi-source data includes load data streams of each electrical circuit and personnel data streams. The personnel data streams include personnel location data streams and personnel behavior data streams. S130, based on load data streams and preset construction plans, continuously predicts the risk of load conflicts in future periods; S140, based on personnel data flow, combined with preset electronic fences, personnel qualification database and equipment status, calculates the dynamic comprehensive risk index between personnel and live equipment in real time; S150 When the predicted load conflict risk or dynamic comprehensive risk index exceeds the corresponding preset threshold, multi-strategy simulation is performed in the digital twin to generate and execute collaborative scheduling instructions. The collaborative scheduling instructions include power regulation or circuit breaker control of the target electrical circuit, load adjustment of the associated area, and audible and visual alarms to drive away specific personnel.

[0024] Specifically, for step S110, based on the construction drawings and site survey data of the precast beam yard, a 1:1 site space model is constructed using Revit software. This model includes geometric information such as beam yard zones (rebar processing area, formwork assembly area, tensioning area, curing area, etc.), roads, distribution box locations, installation locations of key electrical equipment, and the surrounding environment, ensuring a complete match between the model and the actual site. For the electrical topology, the wiring relationships of the beam yard's various levels of power distribution systems (main distribution box - sub-distribution boxes - switch boxes - electrical equipment) are analyzed. An electrical topology diagram is drawn using AutoCAD Electrical, clarifying the connection methods, cable specifications, and protection device parameters for each circuit. This topology information is then imported into a digital twin platform and linked to the site space model. Real-time load data is synchronously accessed through the interfaces of subsequently deployed smart IoT devices, while personnel positioning data adopts UWB positioning technology. Positioning base stations are deployed in key areas of the beam yard, and positioning tags are provided to on-site workers. The tags send location signals in real time, which are received by the base stations and transmitted to the digital twin platform. Ultimately, the four elements of site space, electrical topology, load data, and personnel positioning are integrated into a visual and interactive full-element digital twin.

[0025] In S120, the steps of acquiring multi-source data from the field in real time through a pre-deployed sensor network specifically include S121-S124 (not shown in the figure): S121 collects current, voltage, and power data of electrical circuits through smart IoT circuit breakers deployed in distribution boxes at all levels and key electrical equipment, forming a load data stream; S122 collects three-dimensional coordinate data of personnel by deploying ultra-wideband positioning base stations throughout the construction site and positioning tags worn by personnel, forming a personnel location data stream; S123 collects video stream data through visual monitoring equipment deployed in key risk areas, and extracts personnel behavior characteristics through image recognition to form a personnel behavior data stream; S124 aggregates and preprocesses load data streams, personnel location data streams, and personnel behavior data streams in its region through regional edge gateways deployed in each construction zone.

[0026] Specifically, for step S120, select smart IoT circuit breakers suitable for the precast beam yard industrial environment, prioritizing models with IP65 protection rating and support for RS485 / 5G dual-mode communication. Configure them appropriately based on the distribution box level and equipment power: deploy three-phase smart circuit breakers at the main distribution box and sub-distribution boxes to monitor parameters such as three-phase current, voltage, active power, reactive power, and power factor; deploy single-phase or three-phase circuit breakers at switch boxes and individual key electrical equipment (such as tensioning machines, welding machines, and curing boilers) to focus on monitoring current, voltage, and power data. After wiring is completed, perform debugging and calibration to ensure data acquisition accuracy (current error ≤ ±1%, voltage error ≤ ±0.5%). Set the acquisition frequency to once every 10 seconds. The acquired electrical parameters are converted into digital signals by the circuit breaker's built-in acquisition module and uploaded in real-time to the edge computing gateway via the communication module. After preliminary processing, a standardized load data stream is formed. Data preprocessing algorithms are deployed in the edge computing gateway. First, outliers are removed from the collected raw data. The 3σ criterion is used to identify and filter out abnormal data caused by sensor malfunctions or loose wiring (such as current sudden changes to 0 or current far exceeding the equipment's rated current). For missing data, linear interpolation is used to supplement it, ensuring the continuity of the data stream. Simultaneously, the preprocessed load data is categorized and labeled according to distribution box level (e.g., main distribution, sub-distribution, switch box), equipment type (e.g., processing equipment, tensioning equipment, curing equipment), and work area. Timestamps are added to form a structured load data stream, for example, labeled "2024-05-20 14:30:20, rebar processing area distribution box, welding machine 1, current 38A, voltage 380V, active power approximately 14.4kW," facilitating subsequent mapping with the digital twin and load analysis.

[0027] High-definition network cameras (resolution of 1080P or higher, supporting infrared night vision) are deployed in key risk areas of the beam yard (such as within 3 meters of the distribution box, high-voltage equipment area, tensioning operation area, and temporary power supply operation point). The cameras are powered by POE and fixedly installed at a high place (no less than 4 meters high) to ensure that there are no blind spots in the monitoring range. The acquisition frequency is set to 25 frames / second, and the video stream is transmitted to the video server through the network cable. A model based on the YOLOv8 object detection algorithm is deployed on the video server to identify personnel targets and clothing such as safety helmets / insulating gloves in the video stream, and personnel IDs are assigned in conjunction with the DeepSORT multi-object tracking algorithm. For action features, the detected personnel regions are cropped and input into a spatiotemporal behavior recognition model (such as CNN+LSTM) for intent inference. The recognition results are converted into structured data (such as "Person ID: 001, Location: Around the main distribution box, Behavior: Reaching out to touch the distribution box door, Time: 2024-05-20 15:10:30"), forming a preliminary personnel behavior data stream. At the same time, the recognition results are initially screened to remove misidentified data (such as birds, debris, etc.).

[0028] First, the load data stream and the personnel behavior data stream are synchronized and aligned using timestamps to ensure accurate matching of load data and personnel behavior data at the same moment. The personnel data stream is further integrated into a personnel location data stream and a personnel behavior data stream. The personnel location data stream comes from the UWB positioning system and includes information such as personnel ID, real-time coordinates, and movement speed. The personnel behavior data stream comes from visual recognition results and includes information such as behavior type, behavior duration, and violation level. Subsequently, the integrated load data stream and personnel data stream are synchronously mapped to a digital twin through an API interface to achieve real-time linkage between data and physical entities. For example, in the twin model, clicking on a certain electrical device can simultaneously view its real-time load data and the location and behavior status of surrounding personnel; when a person enters the electronic fence of an electrical device, the corresponding personnel icon in the digital twin will issue a red warning, and the real-time load status of that area will be displayed in conjunction with it.

[0029] Reference Figure 2 S130, the steps for rolling forecasting of load conflict risks in future periods based on load data streams and preset construction plans specifically include S210-S260: S210, obtain the preset detailed construction plan, which includes the expected start time, duration, required equipment and its rated power for each process; S220 generates the baseline load time-series curves for each piece of equipment under ideal operating conditions based on the construction plan; S230: Real-time acquisition of actual load data stream within the previous time window at the current moment, and calculation of the deviation between the actual load and the benchmark load for the corresponding time period; S240, the deviation value is used to correct the reference load time series curve for subsequent periods to obtain the corrected predicted load curve; S250 compares the corrected predicted load curve with the safe capacity threshold of the power distribution system to identify the risk time window in which the total load or key branch load exceeds the threshold within a set future period. S260, when a risk time window is identified, extracts a list of high-priority devices that will run in parallel within the corresponding future time period, as the core element of load conflict risk.

[0030] Specifically, for step S130, a precast beam yard construction plan management system is established. This system is connected to the beam yard production management platform. The construction plan is prepared by the production management personnel based on the beam yard production tasks, process connection requirements, and equipment resources. After preparation, the plan is uploaded to the system and synchronized to the power management platform. The construction plan must include detailed process information, specifically the estimated start time (accurate to the hour), duration, and a list of required electrical equipment for each process (such as rebar processing, formwork installation, concrete pouring, tensioning, and curing). It should also indicate the priority of each process (e.g., tensioning is high priority, curing is medium priority, and cleaning is low priority). For example, a construction plan might specify: "2024-05-21 08:00-12:00, rebar processing area, rebar cutting machine (rated power 7.5kW) and rebar bending machine (rated power 5.5kW) operating, process priority: medium; 13:00-17:00, tensioning area, 2 tensioning machines (each rated power 30kW) operating, process priority: high." This ensures the precision and operability of the construction plan, providing a basis for the subsequent generation of the baseline load curve.

[0031] First, extract the rated power and operating time of each electrical device in the construction plan. Then, considering the device's operating characteristics (e.g., welding machines operate intermittently, tensioning machines operate continuously), determine the load fluctuation pattern within the operating time. For continuously operating equipment (e.g., maintenance boilers), assume stable load within the operating time and generate a stable load time-series curve based on the rated power. For intermittently operating equipment (e.g., welding machines, rebar cutters), based on historical operating data of similar equipment, analyze its working cycle (e.g., 10 minutes of work, 5 minutes of rest) and load changes within each cycle. Combined with the duration in the construction plan, generate a fluctuating load time-series curve. For example, a welding machine with a rated power of 14kW, a working cycle of 15 minutes (10 minutes of work, 5 minutes of rest), and an operating time from 8:00 to 12:00, would have a baseline load time-series curve of 14kW from 8:00 to 8:10, 0kW from 8:10 to 8:15, and this cycle continues until 12:00. At the same time, the baseline load time-series curves of all equipment are superimposed on the time axis to generate the baseline load time-series curve of the entire beam yard. The horizontal axis of the curve is time (accurate to the minute), and the vertical axis is the load power (kW). The curve is then stored in the database of the power management platform.

[0032] The time window length is set to 15 minutes (adjustable according to actual conditions) based on the operating characteristics of the beam yard equipment. The power management platform collects the actual load data stream within the 15 minutes preceding the current moment in real time, extracts the actual load data every 10 seconds within this time period, and calculates the average actual load within this time period. The benchmark load time series curve for the corresponding time period is retrieved from the database, and the average benchmark load within the same time period is extracted. The root mean square error (RMSE) is used to calculate the deviation between the two. The deviation calculation formula is as follows: , where n is the amount of data collected within the time window, and i is the load data collected in the i-th time. For example, if the current time is 10:15, the time window is 10:00-10:15, the average actual load during this period is 28kW, the average baseline load is 30kW, and the calculated deviation is 2.1kW. The larger the deviation, the greater the deviation between the actual load and the ideal state.

[0033] Based on the deviation value, the deviation type and magnitude are determined. Deviation types include positive deviation (actual load exceeds the baseline load) and negative deviation (actual load is less than the baseline load). A deviation threshold (e.g., 5%) is set. When the deviation value exceeds the threshold, the correction process is initiated. During the correction process, based on the deviation value and combined with historical deviation data of equipment operation, a linear regression method is used to fit the deviation change trend, predict the deviation situation in subsequent periods, and then adjust the baseline load time series curve for subsequent periods. For example, if the current deviation value is 2.1kW, corresponding to a deviation rate of 7% for a baseline load of 30kW (positive deviation, exceeding the 5% threshold), analysis of historical data reveals that under the same operating conditions, the deviation value will gradually decrease to 0.5kW in the following hour. Therefore, the baseline load time series curve for the following hour is adjusted upward, with the adjustment range gradually decreasing from 2.1kW to 0.5kW, ensuring that the corrected predicted load curve better reflects the actual operating conditions. After correction, the new predicted load curve is updated to the database, replacing the original baseline load time series curve, for subsequent risk identification.

[0034] The safe capacity threshold of the power distribution system is determined by comprehensively considering the rated capacity of the system, the line carrying capacity, and the setting parameters of the protection devices. For example, the rated capacity of the main distribution box in the beam yard is 250kW. Considering line losses and protection margins, the safe capacity threshold is set to 200kW. Each key branch (such as the branch in the tensioning area and the branch in the rebar processing area) has a corresponding threshold set according to the line specifications (e.g., the threshold for the branch in the tensioning area is 100kW). The corrected predicted load curve is compared with the safe capacity threshold every 5 minutes. When the total load or the load of a key branch in a certain period of the predicted load curve (e.g., 13:30-14:00) exceeds the corresponding threshold, that period is marked as a risk time window, and the risk type (total load exceeding the standard, branch load exceeding the standard), the magnitude of the exceedance (e.g., total load 210kW, exceedance 10kW), and the duration of the exceedance are recorded. At the same time, the risk time window information is synchronized to the digital twin and the power management platform to issue an initial warning.

[0035] When a risk time window is identified, the power management platform extracts a list of equipment operating in parallel within that window. Combining this with the process priorities in the construction plan, high-priority equipment (such as tensioning machines and concrete pumps, which affect critical production processes) is selected as the core element for load conflict risk. Subsequently, a rolling forecasting model is used, updating the predicted load data every 15 minutes starting from the risk time window. This, combined with the operating plans of high-priority equipment and historical load fluctuation patterns, predicts load changes within the next hour to determine if multiple high-priority devices will operate simultaneously, leading to continuous overload. For example, if the risk time window is 13:30-14:00, during this period, there are two high-priority tensioning machines (30kW each) operating in parallel. It is predicted that a new concrete pump (40kW) will be added between 14:00-14:15, bringing the total load to 100kW, exceeding the 80kW threshold for the tensioning zone branch. Therefore, a load conflict risk is predicted for this period, and the forecast results are pushed to management personnel in real time.

[0036] Reference Figure 3 S140, the steps for real-time calculation of the dynamic comprehensive risk index between personnel and live equipment specifically include S310-S350: S310 calculates spatial proximity as a distance risk factor based on the real-time location of personnel and the preset electronic fence of powered equipment. S320 analyzes visual surveillance data streams to identify the degree to which personnel's actions and intentions comply with safety regulations, serving as a behavioral risk factor. S330 retrieves the personnel qualification database to determine whether the current personnel are qualified to operate or approach the target equipment, serving as a qualification risk factor. S340 acquires the real-time insulation status, operating mode and voltage level of the target equipment as a risk factor for equipment status. S350, based on a preset weighted fusion model, comprehensively calculates distance risk factors, behavioral risk factors, qualification risk factors, and equipment status risk factors, and outputs a dynamic comprehensive risk index.

[0037] Specifically, for step S140, electronic fences are set up for all electrical equipment (distribution boxes, high-voltage cables, and electrical equipment) in the digital twin. The electronic fences adopt GIS-based polygonal fences, and the fence range is set according to the voltage level of the equipment (e.g., the fence radius of high-voltage equipment is 3 meters, and the fence radius of low-voltage distribution boxes is 1.5 meters). The fence coordinates are precisely matched with the site spatial model.

[0038] The real-time coordinates of personnel are obtained through a UWB positioning system, and the spatial proximity between personnel and the boundary of the electronic fence with live equipment is calculated using the Euclidean distance formula. The distance calculation formula is as follows: Where (x1, y1) are the real-time coordinates of the personnel, and (x2, y2) are the coordinates of the nearest point on the electronic fence boundary. Spatial proximity is quantified as a distance risk factor, using a 0-1 rating scale: when a person is outside the fence and ≥2 meters from the boundary, the risk factor is 0 (no risk); when the distance is between 1 and 2 meters from the boundary, the risk factor is 0.3 (low risk); when the distance is between 0.5 and 1 meter from the boundary, the risk factor is 0.7 (medium risk); and when the distance is <0.5 meters from the boundary or inside the fence, the risk factor is 1.0 (high risk).

[0039] Based on the visual recognition data from S140, a deep learning model (CNN+LSTM) was used to further analyze the personnel's action intentions. This model, trained with samples from the beam yard's safety operation specifications, can identify common violations (such as touching live equipment without insulated gloves, smoking in live areas, and improper wiring) and safe behaviors (such as wearing protective equipment, staying away from live equipment, and operating according to regulations). During the analysis, the personnel's action sequences (such as reaching out-touching-stopping) were extracted and compared with the action standards in the safety specifications to calculate the compliance rate. The higher the compliance rate, the lower the behavioral risk factor. The behavioral risk factor also used a 0-1 rating system, with the specific standards as follows: fully compliant with safety specifications (such as wearing insulated gloves and staying away from fences), risk factor 0; minor violation (such as not wearing a safety helmet but not approaching live equipment), risk factor 0.2; moderate violation (such as approaching the fence but not touching the equipment), risk factor 0.6; serious violation (such as touching live equipment and improper wiring), risk factor 1.0. In addition, based on the duration of the behavior, if the violation lasts for more than 30 seconds, the risk factor will be increased by 0.1 (up to a maximum of 1.0) to ensure the accuracy of the behavior risk assessment.

[0040] A personnel qualification database is pre-built, storing basic information, qualification certificates (such as electrician's licenses and special equipment operation certificates), training records, and assessment results for all personnel working in the beam yard. Scanned copies of the original qualification certificates must be uploaded, and training records and assessment results are updated in real time. When personnel enter or approach an area where equipment is operated, the power management platform identifies the personnel ID through the UWB positioning system and automatically retrieves the corresponding information from the qualification database to determine whether the personnel are qualified to operate or approach the target equipment: if the personnel possess the corresponding qualification (such as an electrician's license) and have passed the assessment, and the training is valid, the qualification risk factor is 0; if the personnel possess the qualification but have failed the assessment or the training has expired, the qualification risk factor is 0.5; if the personnel do not possess the corresponding qualification, the qualification risk factor is 1.0.

[0041] Real-time data from the target equipment is collected via smart IoT circuit breakers and equipment status sensors, including parameters such as insulation status, operating mode, and voltage level. Insulation status is collected via insulation resistance sensors every 30 minutes. An insulation resistance value ≥ 1 MΩ is considered good insulation; 0.5 MΩ ≤ insulation resistance < 1 MΩ is considered fair insulation; and < 0.5 MΩ is considered poor insulation. Operating mode is obtained through the equipment control module and is categorized into normal operation, standby, and fault states. Voltage level is determined based on the equipment's rated parameters and real-time voltage data, and is categorized as high voltage (≥ 1000V) and low voltage (< 1000V). These parameters are quantified as equipment status risk factors using a 0-1 rating system: good insulation, normal operation, low voltage equipment – ​​risk factor 0; fair insulation, standby, low voltage equipment – ​​risk factor 0.3; poor insulation, normal operation, low voltage equipment – ​​risk factor 0.7; poor insulation, fault, high voltage equipment – ​​risk factor 1.0.

[0042] The Analytic Hierarchy Process (AHP) was used to determine the weight coefficients of each risk factor. Considering the priority of electrical safety in the precast beam yard, with personnel safety taking precedence over equipment safety, the weights for the distance risk factor and behavior risk factor were set to 0.4, 0.3, 0.15, and 0.15 respectively. These weight coefficients were confirmed to be valid after a consistency test (CR < 0.1). The weighted fusion model used a linear weighted summation formula: Dynamic Comprehensive Risk Index = Distance Risk Factor × 0.4 + Behavior Risk Factor × 0.3 + Qualification Risk Factor × 0.15 + Equipment Status Risk Factor × 0.15. The calculation result ranges from 0 to 1; a higher index indicates a higher risk level. For example, in a certain scenario, the distance risk factor is 0.7, the behavior risk factor is 0.6, the qualification risk factor is 0, and the equipment status risk factor is 0.3. The calculated dynamic comprehensive risk index is 0.7×0.4+0.6×0.3+0×0.15+0.3×0.15=0.28+0.18+0+0.045=0.505, which belongs to the medium risk level. The calculation results are synchronized to the digital twin and the power management platform in real time.

[0043] Reference Figure 4 S150, the steps for generating cooperative scheduling instructions specifically include S410-S430: S410 When the triggering condition is that the dynamic comprehensive risk index exceeds the first preset threshold, the first type of instruction set with personnel safety as the core is generated first. The first type of instruction set includes emergency power cut-off and access lock-up of risk source equipment, and sending targeted eviction alarm to risk personnel. S420, when the triggering condition is load conflict risk, generates a second type of instruction set with the core of ensuring production continuity. The second type of instruction set includes scheduling non-critical equipment for delay, rotational shutdown or power reduction based on the dynamic priority of the process. S430 When the dynamic comprehensive risk index and the load conflict risk exist simultaneously or are related to each other, a fusion instruction set is generated. Under the premise of ensuring personnel safety, the fusion instruction set creates power supply conditions for safety intervention measures by actively adjusting the load of the related area, or indirectly disperses the gathering of people in the risk area by adjusting the regional power supply strategy.

[0044] Specifically, for step S150, when the dynamic comprehensive risk index exceeds the first preset threshold (set to 0.7, adjustable according to actual site conditions), a first-class instruction set centered on personnel safety is generated. The generation and issuance of the instruction set are achieved through the linkage between the power management platform and on-site equipment. First, the platform automatically identifies risk source equipment (such as high-voltage distribution boxes near personnel, or poorly insulated electrical equipment) and sends an emergency power-off command to the corresponding smart IoT circuit breaker. Upon receiving the command, the circuit breaker immediately cuts off the power supply to the corresponding circuit and locks the equipment's operating permissions, prohibiting any personnel from manually closing the circuit until the risk is eliminated. Second, a directional evacuation alarm is sent through on-site audible and visual alarms (deployed around the risk area). The alarm sound uses a customized voice (such as "Attention! This is a high-risk area, please evacuate immediately!"), and directional reminder information is sent to at-risk personnel through the built-in speaker of the personnel positioning tag, ensuring that personnel can notice and evacuate in a timely manner. In addition, the platform records the execution status of the instructions (such as power-off time and alarm activation time) in real time and pushes it to the mobile phone of the management personnel for real-time monitoring.

[0045] When the trigger condition is a load conflict risk, a second type of instruction set is generated with the core objective of ensuring production continuity. This is achieved by using a load scheduling algorithm based on the dynamic priority of work processes to schedule non-critical equipment. First, the power management platform extracts all operating equipment within the risk time window and classifies it into high-priority (e.g., tensioning, concrete pouring), medium-priority (e.g., steel bar processing), and low-priority (e.g., cleaning, lighting) equipment according to the work process priority in the construction plan. The scheduling strategy is determined based on the extent of load exceeding the limit: If the load exceeds the limit by a small amount (<10%), low-priority equipment is scheduled for delayed operation, postponing its start-up time to the off-peak period (e.g., cleaning equipment originally scheduled to start at 13:30 is postponed to 14:30); if the load exceeds the limit by a moderate amount (10%-20%), medium-priority equipment is scheduled for rotating shutdowns, for example, two welding machines in the rebar processing area are scheduled to operate in an alternating mode (each machine runs for 30 minutes and rests for 30 minutes) to reduce the load on the same machine; if the load exceeds the limit by a large amount (>20%), medium and low-priority equipment is scheduled for reduced power, adjusting the operating power of the equipment through smart IoT circuit breakers (e.g., reducing the power of the welding machine from 14kW to 10kW) to ensure that the total load or branch load is reduced below the safe threshold, while prioritizing the normal operation of high-priority equipment to minimize the impact on production.

[0046] When the dynamic comprehensive risk index and load conflict risk coexist or are interrelated, a fusion instruction set is generated. The core principle is to coordinate load scheduling and safety intervention measures while ensuring personnel safety. If the risk correlation manifests as personnel approaching high-load operating equipment (e.g., personnel near a tensioning machine causing overload), the fusion instruction set first schedules temporary shutdowns of non-critical loads in the area (e.g., surrounding lighting equipment, small cleaning equipment) to free up power supply capacity, ensuring the normal operation of the risk source equipment (tensioning machine) and creating sufficient time for safety intervention measures (e.g., evacuating personnel). After personnel are evacuated to a safe area and the dynamic comprehensive risk index drops below the threshold, non-critical loads are gradually restored. If the risk correlation manifests as overload causing a decrease in equipment insulation performance, thereby increasing personnel safety risks, the fusion instruction set first schedules power reduction or shutdown of non-critical high-load equipment to reduce the total load and improve equipment operating status. Simultaneously, a personnel evacuation alarm is activated to prohibit personnel from approaching the equipment. After the equipment status returns to normal and the load drops to a safe range, normal equipment operation and personnel work are restored.

[0047] Reference Figure 5 S150, the steps following the execution of the coordinated scheduling instruction also include S510-S550: S510 continuously monitors whether the risk of load conflict is eliminated and whether the dynamic comprehensive risk index drops below the safety threshold after the coordinated scheduling instruction is executed. S520: Once the risk has been confirmed to be eliminated, the safety-after-production recovery protocol is initiated. S530, in a digital twin, simulates and verifies whether restoring power supply will lead to new load conflict risks; S540: If the simulation verification passes, a power restoration command for the device will be generated and issued. If the simulation verification fails, the S550 generates and issues a step-by-step recovery instruction. The step-by-step recovery instruction includes: first, scheduling other non-critical loads to temporarily relinquish the required capacity; then, restoring power to critical equipment; and finally, orderly restoring the scheduled non-critical loads.

[0048] Specifically, after the coordinated dispatch command is executed, the power management platform activates a real-time monitoring mechanism to continuously collect load data streams and risk index data. The monitoring frequency is increased to once every 5 seconds, focusing on whether the load conflict risk has been eliminated (whether the total load and branch load have dropped below the safety threshold) and whether the dynamic comprehensive risk index has dropped below the safety threshold (≤0.7). During the monitoring process, if the load conflict risk is eliminated but the dynamic comprehensive risk index still does not meet the standard, the first type of command set continues to be executed to strengthen personnel safety intervention; if the dynamic comprehensive risk index meets the standard but the load conflict risk still exists, the second type of command set continues to be executed to optimize load dispatch; if neither meets the standard, the integrated command set is adjusted to increase the intervention intensity until at least one risk is eliminated, ensuring that the risk is under control. At the same time, the monitoring data is fed back to the digital twin in real time, and managers can intuitively view the risk change trend and the effect of command execution through the twin.

[0049] Once the risk is confirmed to have been eliminated (load conflict risk eliminated and dynamic comprehensive risk index reduced to below the safety threshold), the safety-recovery production resumption protocol is activated. This protocol, pre-set in the power management platform, includes the recovery process, safety inspection items, and division of responsibilities. First, management personnel, through digital twins and on-site inspections, confirm that the risk source has been eliminated (e.g., personnel have been evacuated, equipment is in normal condition, and load is stable), fill out a risk elimination confirmation form, and upload it to the platform. Upon receiving the confirmation form, the platform automatically activates the recovery protocol. The first step is to notify on-site personnel to prepare for production resumption (e.g., checking equipment status and tidying up the work area); the second step is to conduct a comprehensive inspection of the scheduled equipment to confirm that the equipment is fault-free, wiring is normal, and insulation is good; the third step is to verify the construction plan to confirm that resuming production will not lead to new risks, ensuring the safety and standardization of the recovery process.

[0050] The current load data, equipment operating status, and construction plan are synchronized to the twin simulation module to build a simulation environment consistent with the actual scenario. Then, the system simulates the equipment operation after power restoration, following the equipment startup order in the restoration protocol (high-priority equipment starts first, followed by medium and low-priority equipment in sequence), simulating the load change trend over the next 30 minutes, with a focus on simulating the load superposition when multiple devices start simultaneously. During the simulation, the simulated load is compared in real time with the power distribution system's safe capacity threshold. If the simulated load does not exceed the threshold and no new load conflict scenarios occur, the simulation verification is considered successful. If load exceeding the limit or branch load conflicts occur during the simulation, the simulation verification is considered unsuccessful, and risk points during the simulation process (such as the simultaneous startup of two medium-priority devices causing load exceeding the limit) are recorded.

[0051] If the simulation verification passes, the power management platform automatically generates a power restoration command for the equipment. The command is issued sequentially according to equipment priority, with high-priority equipment (such as tensioning machines and concrete pumps) receiving the command first. The command specifies the equipment start-up time, operating power, and safety precautions. Upon receiving the command, the smart IoT circuit breaker starts the equipment according to preset parameters, restoring normal operation. Simultaneously, the platform monitors the load data after equipment startup in real time to ensure the load remains stable within a safe range. If abnormal load fluctuations occur, the power restoration is immediately paused, and the simulation verification is repeated. After power restoration is complete, the platform records the restoration time, equipment operating status, and other information, synchronizing it to the production log to ensure the entire restoration process is traceable.

[0052] If the simulation verification fails, a step-by-step recovery command is generated, which is implemented in three steps: First, the power management platform schedules other non-critical loads (such as cleaning equipment and auxiliary lighting equipment) to temporarily shut down, cutting off their power supply through smart IoT circuit breakers to free up the required power supply capacity and ensure that the load does not exceed the limit when the critical equipment is restored to power. For example, if the simulation finds that restoring two tensioning machines (total power 60kW) will cause the branch load to exceed the limit by 10kW, the three cleaning machines (total power 12kW) on that branch are scheduled to temporarily shut down to free up sufficient capacity. Second, a power restoration command is issued to high-priority critical equipment to start the equipment and monitor the load data in real time to ensure that the load is stable below the safety threshold. Third, after the critical equipment is running normally and the load is stable, the scheduled non-critical loads are restored in an orderly manner according to the order of "medium priority first, then low priority". The load data is monitored once after each piece of equipment is restored to avoid the accumulation of loads and the resulting new risks, until all equipment is fully restored to normal operation, completing the step-by-step recovery process.

[0053] Reference Figure 6 Furthermore, the electricity management methods also include S610-S630: S610 continuously records all risk events, scheduling instructions, and execution results, forming a traceable production log; S620, based on production log data, identifies high-frequency risk areas, typical risk scenarios, and the actual effectiveness of different scheduling strategies; S630 utilizes the analysis results to optimize the model parameters, risk factor weight coefficients, and scheduling strategy library of the digital twin.

[0054] Specifically, a production log management system is established. This system interfaces with the power management platform and digital twin, automatically collecting various types of data and recording them in a structured manner. The production log records include: risk events (event occurrence time, location, risk type, risk level, triggering factors), dispatch instructions (instruction type, issuance time, executing equipment, instruction content), execution results (instruction execution time, execution effect, reason for non-execution), and related data (load data, risk index, personnel location, equipment status), etc. The recording format uses standardized templates, such as "2024-05-20 15:10:30, tensioning area, personnel safety risk (high risk), triggering factor: personnel entering the high-voltage equipment electronic fence, dispatch instruction: emergency power outage + directional evacuation, execution time: 15:10:35, execution effect: equipment power outage, personnel evacuation, risk index decreased from 0.8 to 0.2." Simultaneously, the production log supports keyword retrieval by time, risk type, equipment name, etc., ensuring log traceability. Log data is retained for no less than one year, meeting safety production supervision requirements.

[0055] Statistical analysis was performed on risk event data in the production logs. K-means clustering was used to cluster the locations of risk events, identifying areas with high frequency of occurrence (such as the area around the main distribution box and the tensioning zone) as high-frequency risk areas. The distribution of risk types in each area was also statistically analyzed (e.g., the proportion of personnel safety risks and the proportion of load conflict risks). The triggering factors, development processes, and handling results of risk events were categorized and analyzed to extract typical risk scenarios (e.g., personnel illegally approaching high-voltage equipment, multiple high-priority devices operating simultaneously leading to overload, and safety risks caused by poor equipment insulation). The characteristics and patterns of each typical scenario were summarized. The execution effects of different scheduling strategies were quantitatively evaluated, calculating the risk elimination rate (number of risks eliminated / number of instructions issued) and the degree of production impact (production delay time caused by scheduling) for each strategy.

[0056] For digital twin model parameters, the site spatial model and electrical topology are updated based on actual environmental changes in high-frequency risk areas (such as equipment location adjustments and the addition of new electrical equipment). Combined with historical load data and risk event data, the load forecasting model parameters of the digital twin are optimized (such as time window length and interpolation algorithm coefficients) to improve the accuracy of load forecasting. For risk factor weights, the weight coefficients of each risk factor are adjusted based on the analysis results of typical risk scenarios. For example, if a high-frequency risk scenario involves safety risks caused by personnel violations, the weight of the behavioral risk factor is increased to 0.4, the distance risk factor is adjusted to 0.3, the qualification risk factor is adjusted to 0.15, and the equipment status risk factor is adjusted to 0.15, ensuring that risk assessments better reflect actual risk conditions. For the scheduling strategy library, existing strategies are optimized based on the effectiveness analysis results of different scheduling strategies (such as adjusting the rotation interval of non-critical equipment and the power reduction magnitude).

[0057] Based on the above method embodiments, the second embodiment of this application discloses a power management system for precast beam construction. The power management system for precast beam construction of this embodiment can implement any of the above-described methods for managing power consumption during precast beam construction, and the specific working processes of each module in the power management system can be referred to the corresponding processes in the above method embodiments.

[0058] For ease of understanding, an example is as follows: A power management system for precast beam construction includes: The digital twin construction and synchronization module is used to construct a full-element digital twin of the precast beam yard and synchronously map the real-time collected multi-source data to the digital twin. The digital twin integrates the site spatial model, equipment electrical topology, real-time load data and personnel positioning data. The multi-source data acquisition module collects multi-source data in real time through a pre-deployed sensor network. The multi-source data includes load data streams of each electrical circuit and personnel data streams. The personnel data streams include personnel location data streams and personnel behavior data streams. The load conflict risk prediction module is used to predict the load conflict risk in future periods based on load data streams and preset construction plans. The dynamic safety risk assessment module is used to calculate the dynamic comprehensive risk index between personnel and live equipment in real time based on personnel data flow, combined with preset electronic fences, personnel qualification databases and equipment status. The collaborative scheduling decision and execution module is used to perform multi-strategy simulation and deduction in the digital twin when the predicted load conflict risk or dynamic comprehensive risk index exceeds the corresponding preset threshold, generate and execute collaborative scheduling instructions, including power regulation or circuit breaker control of the target electrical circuit, load adjustment of the associated area, and audible and visual alarms to drive away specific personnel.

[0059] The third embodiment of this application provides a computer device, which may include a memory, a processor, and a computer program stored in the memory. The processor executes the computer program to implement a method for managing electricity consumption during precast beam construction.

[0060] The memory can communicate with the processor via a communication bus, which can be an address bus, a data bus, a control bus, etc.

[0061] Additionally, the memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device.

[0062] Furthermore, the processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0063] The fourth embodiment of this application provides a computer-readable storage medium storing a computer program that can be loaded by a processor and executed as a method for managing electricity consumption during precast beam construction.

[0064] The computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device; the program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0065] The above are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Any feature disclosed in this specification (including the abstract and drawings) may be replaced by other equivalent or similar features unless specifically stated otherwise. That is, unless specifically stated otherwise, each feature is only one example of a series of equivalent or similar features.

Claims

1. A method for managing electricity consumption during precast beam construction, characterized in that, include: Construct a full-element digital twin of the precast beam yard, which integrates the site spatial model, equipment electrical topology, real-time load data, and personnel positioning data; The system collects multi-source data from the field in real time through a pre-deployed sensor network and synchronously maps the multi-source data to the digital twin. The multi-source data includes load data streams of each electrical circuit and personnel data streams. The personnel data streams include personnel location data streams and personnel behavior data streams. Based on the load data stream and the preset construction plan, the risk of load conflicts in future periods is predicted in a rolling manner. Based on the personnel data stream, combined with the preset electronic fence, personnel qualification database and equipment status, the dynamic comprehensive risk index between personnel and electrical equipment is calculated in real time. When the predicted load conflict risk or the dynamic comprehensive risk index exceeds the corresponding preset threshold, a multi-strategy simulation is performed in the digital twin to generate and execute collaborative scheduling instructions. The collaborative scheduling instructions include power regulation or circuit breaker control of the target electrical circuit, load adjustment of the associated area, and audible and visual alarms to drive away specific personnel.

2. The method for managing electricity consumption during precast beam construction according to claim 1, characterized in that, The steps of acquiring multi-source data in real time through a pre-deployed sensor network specifically include: By deploying smart IoT circuit breakers at various levels of distribution boxes and key electrical equipment, current, voltage, and power data of electrical circuits are collected to form a load data stream; By deploying ultra-wideband positioning base stations throughout the construction site and using positioning tags worn by personnel, three-dimensional coordinate data of personnel is collected to form a personnel location data stream; By deploying visual surveillance equipment in key risk areas, video stream data is collected, and personnel behavior characteristics are extracted through image recognition to form a personnel behavior data stream. By deploying regional edge gateways in each construction zone, the load data stream, personnel location data stream, and personnel behavior data stream of this zone are aggregated and preprocessed.

3. The method for managing electricity consumption during precast beam construction according to claim 1, characterized in that, Based on the load data stream and the preset construction plan, the specific steps for rolling prediction of load conflict risks in future periods include: Obtain a pre-defined detailed construction plan, which includes the expected start time, duration, required equipment, and rated power of each process. Based on the construction plan, a reference load time-series curve for each piece of equipment under ideal operating conditions is generated; The system collects the actual load data stream within the previous time window in real time and calculates the deviation between the actual load and the benchmark load for the corresponding time period. The deviation value is used to correct the baseline load time series curve for subsequent periods to obtain the corrected predicted load curve; The corrected predicted load curve is compared with the safe capacity threshold of the power distribution system to identify the risk time window in which the total load or the load of key branches will exceed the threshold within a set future period. When a risk time window is identified, a list of high-priority devices that will be running in parallel within the corresponding future time period is extracted as the core element of load conflict risk.

4. The method for managing electricity consumption during precast beam construction according to claim 1, characterized in that, The steps for calculating the dynamic comprehensive risk index between personnel and live equipment in real time specifically include: Based on the real-time location of personnel and the pre-set electronic fence of powered equipment, spatial proximity is calculated as a distance risk factor; By analyzing visual surveillance data streams, we can identify the degree to which personnel's actions and intentions comply with safety regulations, and use this as a behavioral risk factor. Access the personnel qualification database to determine whether the current personnel are qualified to operate or approach the target equipment, as a qualification risk factor; Obtain the real-time insulation status, operating mode, and voltage level of the target equipment as a risk factor for equipment status. Based on a preset weighted fusion model, the distance risk factor, behavioral risk factor, qualification risk factor and equipment status risk factor are comprehensively calculated to output the dynamic comprehensive risk index.

5. The method for managing electricity consumption during precast beam construction according to claim 1, characterized in that, The steps for generating the cooperative scheduling instructions specifically include: When the trigger condition is that the dynamic comprehensive risk index exceeds the first preset threshold, a first type of instruction set with personnel safety as the core is generated first. The first type of instruction set includes emergency power cut-off and access lock-up of risk source equipment, and sending targeted expulsion alarms to risk personnel. When the triggering condition is the aforementioned load conflict risk, a second type of instruction set is generated with the core objective of ensuring production continuity. The second type of instruction set includes scheduling non-critical equipment for delay, rotational shutdown, or power reduction based on the dynamic priority of the process. When the dynamic comprehensive risk index and the load conflict risk exist simultaneously or are related to each other, a fusion instruction set is generated. Under the premise of ensuring personnel safety, the fusion instruction set creates power supply conditions for safety intervention measures by actively adjusting the load of the associated area, or indirectly disperses the gathering of people in the risk area by adjusting the regional power supply strategy.

6. The method for managing electricity consumption during precast beam construction according to claim 5, characterized in that, The steps following the execution of the cooperative scheduling instruction also include: After the coordinated scheduling instruction is executed, continuously monitor whether the load conflict risk is eliminated and whether the dynamic comprehensive risk index drops below the safety threshold; Once the risk has been confirmed to be eliminated, the safety-after-production resumption protocol is activated. Simulations were conducted in the digital twin to verify whether restoring power would lead to new load conflict risks. If the simulation verification passes, a power restoration command for the device will be generated and issued. If the simulation verification fails, a step-by-step recovery instruction is generated and issued. The step-by-step recovery instruction includes: first, scheduling other non-critical loads to temporarily relinquish the required capacity; then, restoring power to critical equipment; and finally, orderly restoring the scheduled non-critical loads.

7. The method for managing electricity consumption during precast beam construction according to claim 1, characterized in that, The electricity management method also includes: Continuously record all risk events, scheduling instructions, and execution results to create a traceable production log; Based on production log data, identify high-frequency risk areas, typical risk scenarios, and the actual effectiveness of different scheduling strategies; The analysis results are used to optimize the model parameters, risk factor weighting coefficients, and scheduling strategy library of the digital twin.

8. A power management system for precast beam construction, characterized in that, The method for managing electricity consumption during precast beam construction as described in any one of claims 1 to 7 includes: The digital twin construction and synchronization module is used to construct a full-element digital twin of the precast beam yard and synchronously map the real-time collected multi-source data to the digital twin. The digital twin integrates the site space model, equipment electrical topology, real-time load data and personnel positioning data. The multi-source data acquisition module collects multi-source data in real time through a pre-deployed sensor network. The multi-source data includes load data streams of each electrical circuit and personnel data streams. The personnel data streams include personnel location data streams and personnel behavior data streams. The load conflict risk prediction module is used to predict the load conflict risk in future periods based on the load data stream and the preset construction plan. The dynamic safety risk assessment module is used to calculate the dynamic comprehensive risk index between personnel and live equipment in real time based on the personnel data stream, combined with the preset electronic fence, personnel qualification database and equipment status. The collaborative scheduling decision and execution module is used to perform multi-strategy simulation and deduction in the digital twin when the predicted load conflict risk or the dynamic comprehensive risk index exceeds the corresponding preset threshold, generate and execute collaborative scheduling instructions, including power regulation or circuit breaker control of the target electrical circuit, load adjustment of the associated area, and audible and visual alarms to drive away specific personnel.

9. A computer device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the power management method for precast beam construction as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer program is stored and can be loaded by a processor and executed as described in any one of claims 1 to 7 for the management of electricity for precast beam construction.

Images