Intelligent fire-fighting platform electrical safety and energy-saving linkage method and system
By constructing a hierarchical hybrid monitoring network and an AI prediction engine, combined with drone diagnostics, the problem of synergy between electrical safety and energy saving in the smart fire protection platform was solved, achieving low power consumption, early fault identification, and quantified carbon emission reduction, thereby improving the system's response speed and management efficiency.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-17
- Publication Date
- 2026-04-03
AI Technical Summary
Existing smart fire protection platforms suffer from deep coordination issues in electrical safety and energy conservation. Safety monitoring and energy management are disconnected, and the continuous high-power monitoring method of high-precision sensors increases the system deployment and operation costs. Response speed depends on personnel confirmation and cannot achieve early warning of hidden faults.
A hierarchical hybrid monitoring network is constructed, combining an AI prediction and early warning engine with a basic equipment alarm engine. Drones are used for refined diagnosis and dynamic linkage response to form a closed loop of synergistic optimization for safety and energy conservation. Equipment health prediction models are used to identify hidden faults in advance, and the benefits of safety investment are quantified through carbon emission reduction.
It enables on-demand sensor wake-up, reduces energy consumption and costs, identifies hidden faults in advance, improves response speed and accuracy, quantifies the carbon emission reduction benefits of safety investment, and forms a complete technical closed loop from prediction to diagnosis to benefit quantification.
Smart Images
Figure CN121789367A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of smart fire protection and energy management technology, and in particular to a method and system for linking electrical safety and energy saving in a smart fire protection platform. Background Technology
[0002] Electrical fires, due to their concealment, suddenness, and severity of consequences, have long been among the leading causes of various fires. Fires caused by electrical reasons account for a high proportion of the total number of fires, especially in residential areas, industrial enterprises, and infrastructure. Aging electrical wiring, overload, leakage, and equipment failure are the main causes. Existing smart fire protection platforms, through the deployment of sensor networks, can remotely monitor parameters such as current, voltage, temperature, and residual current of electrical circuits in real time, and display the data and trigger threshold alarms through the platform, significantly improving regulatory efficiency. However, current mainstream smart fire protection platforms still have significant challenges in the deep integration of electrical safety and energy conservation. First, most existing technologies treat safety monitoring and energy management as two independent systems. The safety system focuses on exceeding limits and alarms, and the abundant electrical data it collects is only used to determine "whether there is danger" rather than to analyze "whether energy efficiency is optimized." Second, in pursuit of comprehensive monitoring, existing solutions typically deploy and activate sensors with the same performance at all monitoring points, 24 / 7. This "continuously high-power" monitoring method results in significant energy consumption for the vast sensor network, especially for advanced sensors with high precision and high-frequency sampling (such as full-waveform fault arc detectors and continuous thermal imagers). This contradicts the initial goal of energy conservation and emission reduction and increases the deployment and long-term operating costs of the system. Furthermore, the response logic of current platforms often remains at the level of "alarm to notification to manual verification." After an alarm occurs, it still heavily relies on personnel to arrive on-site for confirmation and initial handling, and the response speed is limited by the arrival time of personnel. Therefore, this paper proposes a method and system for linking electrical safety and energy conservation in a smart fire protection platform. Summary of the Invention
[0003] The purpose of this invention is to address the problem that traditional early warning systems cannot directly quantify "safety early warning" into "energy saving and carbon emission reduction benefits," leading to a disconnect in management. This invention proposes a method and system for linking electrical safety and energy saving in a smart fire protection platform. This solution achieves early diagnosis through the H(t) model, providing an accurate energy saving calculation basis for the CSI integral model. The combination of the two forms a complete technical closed loop from prediction and diagnosis to benefit quantification and incentive feedback.
[0004] To achieve the above objectives, the present invention adopts the following technical solution: A method for linking electrical safety and energy conservation in a smart fire protection platform includes the following steps: S1. Construct a hierarchical hybrid monitoring network: including a basic layer monitoring unit that continuously monitors key electrical circuits 24 / 7, and an enhanced layer monitoring unit that is deployed at the end point, can be remotely woken up, and is in low power mode by default; S2. Establish a dual-core triggering mechanism: including an AI prediction and early warning engine that performs trend and pattern analysis based on the data from the basic layer monitoring unit, and a basic equipment alarm engine that performs real-time monitoring based on fixed thresholds. S3. Execute dynamic linkage response strategy: When the AI prediction and early warning engine outputs an early warning signal, the platform automatically wakes up the associated enhancement layer monitoring unit to perform refined diagnosis and generate preventive maintenance instructions. When the alarm engine of the basic equipment outputs an alarm signal, the platform automatically dispatches a drone carrying a detection payload to the alarm area, and determines the fire level and location based on the image data transmitted back by the drone. S4. Form a closed loop for synergistic optimization of safety and energy conservation: By analyzing energy consumption data related to early warning and response events, identify energy-saving opportunities and quantify carbon emission reductions; S5. Establish a health baseline and conduct in-depth diagnosis: Based on the AI prediction and early warning engine's learning of the historical operating data of specific electrical equipment, establish a health behavior baseline for the equipment; When real-time operating data shows subtle deviations from the baseline, the enhancement layer monitoring unit for the device is preemptively activated for in-depth diagnostics before generating an early warning signal.
[0005] As a preferred embodiment, the base layer monitoring unit in S1 includes at least a sensor for monitoring current, voltage, and residual current; The enhancement layer monitoring unit includes at least one or more of a high-precision temperature sensor, a fault arc detector, or a thermal imaging module.
[0006] As a preferred embodiment, the early warning signal generation of the AI prediction and early warning engine in S2 is based on the analysis of the trend of harmonic content change of electrical circuits, the trend of three-phase imbalance deterioration, the deviation of load curve from historical baseline, or the mode of insulation performance degradation.
[0007] As a preferred embodiment, scheduling the drone in S3 includes: the platform sending a flight mission instruction containing the target coordinates to a preset intelligent drone hangar; The drone hangar controls the drones to take off autonomously, cruise, and transmit real-time data streams containing visible light and infrared thermal imaging.
[0008] As a preferred embodiment, the fire level determination is achieved by analyzing the images transmitted back by the UAV using an image recognition algorithm to identify the smoke color, flame shape, and heat spread range, and thereby classifying the fire into levels such as early overheating, initial open flame, or intense burning.
[0009] As a preferred embodiment, the quantification of carbon emission reduction in S4 specifically involves: the system measuring and recording the equipment idle and standby energy consumption avoided due to the execution of the preventive maintenance instructions, or the normal operating energy efficiency restored due to the elimination of hidden dangers, and converting it into carbon emission reduction equivalents according to the preset carbon emission factor.
[0010] An electrical safety and energy-saving linkage system for a smart fire protection platform, used to implement a method for linking electrical safety and energy saving in a smart fire protection platform, comprising: Hybrid monitoring layer: Composed of base layer monitoring units and enhancement layer monitoring units; Mobile reconnaissance layer for unmanned aerial vehicles (UAVs): including at least one intelligent UAV system with automatic take-off and landing and mission execution capabilities, and a hangar; Edge computing gateway: used to aggregate data from the hybrid monitoring layer; Cloud-edge collaborative intelligent platform: Communicates between the edge computing gateway and the UAV mobile reconnaissance layer; The cloud-edge collaborative intelligent platform includes: a data middleware platform, a dual-core analysis engine, an intelligent rule engine, a visualization and command module, and an energy efficiency and carbon management module; The data platform is used for unified access and processing of multi-source data; The dual-core analysis engine integrates an AI prediction and early warning module and a real-time alarm monitoring module; The intelligent rule engine is used to automatically execute a preset linkage response strategy based on the output of the dual-core analysis engine. The visualization and command module is used to integrate digital twin scenes with the real-time view of drones; The energy efficiency and carbon management module is used to analyze the energy consumption associated with safety incidents and calculate carbon emission reductions.
[0011] As a preferred embodiment, the intelligent rule engine incorporates multiple virtual intelligent agent agents; The intelligent agent agent includes at least one of the following: a security guardian agent aimed at maximizing security, an energy efficiency steward agent aimed at ensuring production continuity, and a reconnaissance expert agent responsible for assessing fire conditions. The intelligent rule engine generates comprehensive decisions through game theory and consensus algorithms among the multiple intelligent agents.
[0012] As a preferred embodiment, the system further includes a blockchain evidence storage module, which is used to generate hash values for the early warning records generated by the AI prediction and early warning engine, the handling records executed by the UAV mobile reconnaissance layer, and the carbon emission reductions calculated by the energy efficiency and carbon management module, and store them on the blockchain.
[0013] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. The hybrid monitoring architecture proposed in this invention, which combines a normally-on base layer and an on-demand wake-up enhancement layer, overturns the traditional mode of continuous high-power monitoring across the entire domain. The high-precision enhancement layer sensor is only activated when AI predicts a risk or the base layer triggers an alarm, and remains in a dormant state with extremely low power consumption at other times. This not only directly and significantly reduces the energy consumption and cost of the entire sensing network during long-term operation and extends the lifespan of battery-powered sensors, but also reflects the green and sustainable design concept of the smart fire protection system itself.
[0014] 2. This invention introduces an AI-based health prediction model based on equipment behavior baselines, enabling the identification of early, latent fault modes (such as slow degradation of insulation performance and slight increases in contact resistance) from massive electrical operation data that traditional threshold alarms cannot detect. This significantly advances the timeline for hazard detection, providing a valuable window for preventative maintenance and truly achieving "prevention before the event," thus reducing the probability of electrical fires at the source.
[0015] 3. This invention addresses the pain points of "invisibility, inaccessibility, and unclear detection" in complex scenarios such as large-scale facilities, high-rise buildings, and underground spaces by integrating drones as automatically dispatchable aerial intelligent nodes into the coordinated response chain. Upon receiving an alarm, the drone automatically takes off, utilizing its onboard dual-optical payload to quickly provide a visualized, thermally imaged global perspective, and determines the intelligence level using a comprehensive fire threat scoring model (such as Formula 2 in the specification). This enables the command center to obtain accurate and intuitive on-site information immediately, preventing personnel from blindly entering high-risk environments and providing crucial support for scientific decision-making and precise rescue. Attached Figure Description
[0016] Figure 1 This is a flowchart of a smart fire protection platform electrical safety and energy-saving linkage method proposed in this invention. Detailed Implementation
[0017] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.
[0018] Example, refer to Figure 1 An intelligent fire protection platform electrical safety and energy-saving linkage system includes: Hybrid monitoring layer: Composed of base layer monitoring units and enhancement layer monitoring units; Mobile reconnaissance layer for unmanned aerial vehicles (UAVs): including at least one intelligent UAV system with automatic take-off and landing and mission execution capabilities, and a hangar; Edge computing gateway: Used to aggregate data from the hybrid monitoring layer; Cloud-edge collaborative intelligent platform: communication connection between edge computing gateway and UAV mobile reconnaissance layer; The cloud-edge collaborative intelligent platform includes: a data middleware platform, a dual-core analysis engine, an intelligent rule engine, a visualization and command module, and an energy efficiency and carbon management module; A data platform is used for unified access and processing of multi-source data; Dual-core analysis engine, integrating AI prediction and early warning module and real-time alarm monitoring module; The intelligent rule engine is used to automatically execute preset linkage response strategies based on the output of the dual-core analysis engine; The visualization and command module is used to integrate digital twin scenes with the real-time view of drones; The energy efficiency and carbon management module is used to analyze the energy consumption associated with safety incidents and calculate carbon emission reductions. The blockchain evidence storage module is used to generate hash values for the early warning records generated by the AI prediction and early warning engine, the handling records executed by the drone mobile reconnaissance layer, and the carbon emission reductions calculated by the energy efficiency and carbon management module, and store them on the blockchain.
[0019] The intelligent rules engine contains multiple virtual intelligent agent agents; Intelligent agent agents include at least one of the following: a security guardian agent aimed at maximizing safety, an energy efficiency steward agent aimed at ensuring production continuity, and a reconnaissance expert agent responsible for assessing fire conditions. The intelligent rule engine generates comprehensive decisions through game theory and consensus algorithms among multiple intelligent agents.
[0020] A method for linking electrical safety and energy conservation in a smart fire protection platform includes the following steps: Construct a layered hybrid monitoring network: including a basic layer monitoring unit that continuously monitors key electrical circuits 24 / 7, and an enhanced layer monitoring unit that is deployed at the end points, can be remotely woken up, and is in low-power mode by default; The base layer monitoring unit includes at least sensors for monitoring current, voltage, and residual current; The enhancement layer monitoring unit includes at least one or more of the following: a high-precision temperature sensor, a fault arc detector, or a thermal imaging module.
[0021] Establish a dual-core triggering mechanism: including an AI prediction and early warning engine that performs trend and pattern analysis based on data from basic layer monitoring units, and a basic equipment alarm engine that performs real-time monitoring based on fixed thresholds; The AI prediction and early warning engine generates early warning signals based on the analysis of the trends in harmonic content changes in electrical circuits, the deterioration trend of three-phase imbalance, the deviation of load curves from historical baselines, or the degradation mode of insulation performance.
[0022] Execute dynamic linkage response strategy: When the AI prediction and early warning engine outputs an early warning signal, the platform automatically wakes up the associated enhancement layer monitoring unit to perform fine-grained diagnosis and generate preventive maintenance instructions; When real-time operating data shows subtle deviations from the baseline, the enhancement layer monitoring unit for the device is preemptively activated to perform in-depth diagnosis H(t) before generating an early warning signal. To achieve AI-driven predictive early warning and build a quantitative model for equipment health, the system continuously collects multi-dimensional time-series data on key electrical equipment (such as transformers and high-voltage motors) under normal operating conditions, including but not limited to three-phase current. , , ,Voltage and vibration spectrum Furthermore, feature extraction is performed using an autoencoder to form a baseline feature vector for the device in a healthy state. When real-time monitoring data flows in, the AI engine calculates the overall health index of the device at the current time t. ; : in: The equipment is in The health index at any given time ranges from 100 to 100. The closer the value is to 1, the healthier the animal is. : Real-time feature vectors extracted from monitoring data at all times; Real-time feature vectors relative to health benchmarks Mahalanobis distance; : The real-time spectrum is constantly acquired and converted by a vibration sensor or a high-frequency current sensor; The baseline spectrum under the health condition of the equipment; : No. Real-time values of key electrical characteristic parameters, such as total harmonic distortion, power factor, and unbalance. : The baseline value of the corresponding characteristic parameter in a healthy state; are the weight coefficients of each item, and satisfy ; The weights are obtained by training with historical fault data to maximize the recognition rate of early faults.
[0023] Using the above formula: The system sets a health warning threshold H_threshold. When < H_threshold, the AI prediction warning engine determines that the device has a risk of early hidden faults and generates a warning signal; At the same time, the rule engine will silently send a wake-up instruction to the high-precision partial discharge detector or vibration analyzer (enhanced layer monitoring unit) associated with the device and in the sleep state, and start in-depth diagnosis, so as to locate and confirm potential hazards before the traditional threshold alarm is triggered, and achieve the effect of predictive warning.
[0024] By introducing an AI health prediction model based on the device behavior baseline, early and hidden fault modes (such as slow degradation of insulation performance and slight increase in contact resistance) that cannot be detected by traditional threshold alarms can be identified from a large amount of electrical operation data. This significantly advances the time point of potential hazard discovery, provides a valuable disposal window for preventive maintenance, truly achieves "preventing problems before they occur", and reduces the probability of electrical fires at the source.
[0025] When the basic equipment alarm engine outputs an alarm signal, the platform automatically dispatches a drone carrying a detection payload to the alarm area, and determines the fire level and locates the position based on the image data transmitted back by the drone; Dispatching the drone includes: The platform sends a flight task instruction containing the target coordinates to a preset intelligent drone hangar; The drone hangar controls the drone to take off autonomously, cruise, and transmit a real-time data stream containing visible light and infrared thermal imaging.
[0026] Among them, when the drone arrives at the alarm airspace after being dispatched, the infrared thermal imager and visible light camera carried by it will transmit image data synchronously.
[0027] The platform automates the on-site risk through the following fire threat comprehensive scoring model. The fire threat comprehensive scoring model is based on the smoke characteristics score identified by a convolutional neural network through visible light images, the highest temperature value identified in the infrared thermal image, and the total area of the pixel region exceeding the warning temperature threshold in the thermal image to achieve the fire threat scoring.
[0028] The fire level determination is to analyze the image data transmitted back by the drone through an image recognition algorithm, identify the smoke color, flame shape, and heat diffusion range, and accordingly divide into levels such as early overheating, initial fire, or intense combustion; Specifically, the fire situation levels are divided into observation level, early warning level, and emergency level. When the fire is at the observation level and marked as low-risk overheating, notify the inspection personnel to verify. When the fire is at the warning level, activate the on-site audible and visual alarms and send a detailed report to the fire safety officer; When the fire triggers the emergency level, the system automatically shuts off non-fire-fighting power, activates emergency evacuation broadcasts, and pushes the drone video stream and coordinates to the fire and rescue command center with one click.
[0029] Forming a closed loop for synergistic optimization of safety and energy conservation: By analyzing energy consumption data related to early warning and response events, energy-saving opportunities are identified and carbon emission reductions are quantified; The quantification of carbon emission reduction is specifically as follows: the system measures and records the idle and standby energy consumption of equipment avoided due to the execution of preventive maintenance instructions, or the normal operating energy efficiency restored due to the elimination of hidden dangers, and converts it into carbon emission reduction equivalent according to the preset carbon emission factor.
[0030] Establish a health baseline and conduct in-depth diagnosis: Based on the AI prediction and early warning engine's learning of the historical operating data of specific electrical equipment, establish a health behavior baseline for the equipment; To make the value of safety investments explicit, this invention designs a carbon safety score (CSI) generation mechanism. After each effective early warning and response loop is completed, the system automatically calculates the comprehensive benefits created by this event and generates a corresponding numerical score (CSI). This refers to the direct energy savings (unit: kWh) resulting from this event. The calculation method is as follows: in, and These are the average operating power of the i-th device before and after repair; This is the runtime after the repair. It is the standby power of the j-th unnecessary circuit that is shut down by the system policy. It is the cumulative shutdown time; Grid carbon emission factor (unit: tCO2e / kWh); This is an estimated value of the potential direct property losses avoided through this early warning (unit: 10,000 yuan). Carbon conversion loss coefficient (unit: tCO2e / ten thousand yuan); The CSI points, along with event numbers, key data hash values, timestamps, and other data, are submitted to the blockchain notarization module to generate an immutable certificate of rights. These points can be used for internal security performance evaluations, redeeming maintenance services, or as reliable data for third-party organizations to assess a company's green security performance and provide green financial incentives (such as insurance discounts). By employing a carbon security points generation mechanism to deeply analyze the energy consumption data behind each early warning and response event, the energy saved by eliminating hidden dangers and the potential losses avoided are calculated and scientifically converted into carbon emission reduction equivalents. This makes the enterprise's safety investment measurable, reportable, and verifiable, improving the level of internal management refinement. Early diagnosis is achieved through the H(t) model, providing an accurate energy saving calculation basis for the CSI points model. The combination of the two forms a complete technical closed loop from prediction and diagnosis to benefit quantification and incentive feedback.
[0031] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.
Claims
1. A method for linking electrical safety and energy saving in a smart fire protection platform, characterized in that, Includes the following steps: S1. Construct a hierarchical hybrid monitoring network: including a basic layer monitoring unit that continuously monitors key electrical circuits 24 / 7, and an enhanced layer monitoring unit that is deployed at the end point, can be remotely woken up, and is in low power mode by default; S2. Establish a dual-core triggering mechanism: including an AI prediction and early warning engine that performs trend and pattern analysis based on the data from the basic layer monitoring unit, and a basic equipment alarm engine that performs real-time monitoring based on fixed thresholds. S3. Execute dynamic linkage response strategy: When the AI prediction and early warning engine outputs an early warning signal, the platform automatically wakes up the associated enhancement layer monitoring unit to perform refined diagnosis and generate preventive maintenance instructions. When the alarm engine of the basic equipment outputs an alarm signal, the platform automatically dispatches a drone carrying a detection payload to the alarm area, and determines the fire level and location based on the image data transmitted back by the drone. S4. Form a closed loop for synergistic optimization of safety and energy conservation: By analyzing energy consumption data related to early warning and response events, identify energy-saving opportunities and quantify carbon emission reductions.
2. The method for linking electrical safety and energy saving in a smart fire protection platform according to claim 1, characterized in that, The base layer monitoring unit described in S1 includes at least sensors for monitoring current, voltage, and residual current; The enhancement layer monitoring unit includes at least one or more of a high-precision temperature sensor, a fault arc detector, or a thermal imaging module.
3. The method for linking electrical safety and energy saving in a smart fire protection platform according to claim 1, characterized in that, The early warning signal generation of the AI prediction and early warning engine in S2 is based on the analysis of the trend of harmonic content change, the trend of three-phase imbalance deterioration, the deviation of load curve from historical baseline, or the mode of insulation performance degradation of electrical circuits.
4. The method for linking electrical safety and energy saving in a smart fire protection platform according to claim 1, characterized in that, The scheduling of drones in S3 includes: the platform sending flight mission instructions containing target coordinates to a preset intelligent drone hangar; The drone hangar controls the drones to take off autonomously, cruise, and transmit real-time data streams containing visible light and infrared thermal imaging.
5. The method for linking electrical safety and energy saving in a smart fire protection platform according to claim 1, characterized in that, The fire level determination is achieved by analyzing images transmitted back by drones using image recognition algorithms to identify smoke color, flame shape, and heat spread range, and then classifying them into levels such as early overheating, initial open flame, or intense burning.
6. The method for linking electrical safety and energy saving in a smart fire protection platform according to claim 1, characterized in that, The quantification of carbon emission reduction in S4 specifically involves the system measuring and recording the equipment's idle and standby energy consumption avoided due to the execution of the preventive maintenance instructions, or the normal operating energy efficiency restored due to the elimination of hidden dangers, and converting it into carbon emission reduction equivalents according to the preset carbon emission factor.
7. A method for linking electrical safety and energy saving in a smart fire protection platform according to any one of claims 1-6, characterized in that, The method further includes S5: Establish a health baseline and conduct in-depth diagnosis: Based on the AI prediction and early warning engine's learning of the historical operating data of specific electrical equipment, establish a health behavior baseline for the equipment; When real-time operating data shows subtle deviations from the baseline, the enhancement layer monitoring unit for the device is preemptively activated for in-depth diagnostics before generating an early warning signal.
8. A smart fire protection platform electrical safety and energy-saving linkage system, used to implement the smart fire protection platform electrical safety and energy-saving linkage method according to any one of claims 1-7, characterized in that, include: Hybrid monitoring layer: Composed of base layer monitoring units and enhancement layer monitoring units; Mobile reconnaissance layer for unmanned aerial vehicles (UAVs): including at least one intelligent UAV system with automatic take-off and landing and mission execution capabilities, and a hangar; Edge computing gateway: used to aggregate data from the hybrid monitoring layer; Cloud-edge collaborative intelligent platform: Communicates between the edge computing gateway and the UAV mobile reconnaissance layer; The cloud-edge collaborative intelligent platform includes: a data middleware platform, a dual-core analysis engine, an intelligent rule engine, a visualization and command module, and an energy efficiency and carbon management module; The data platform is used for unified access and processing of multi-source data; The dual-core analysis engine integrates an AI prediction and early warning module and a real-time alarm monitoring module; The intelligent rule engine is used to automatically execute a preset linkage response strategy based on the output of the dual-core analysis engine. The visualization and command module is used to integrate digital twin scenes with the real-time view of drones; The energy efficiency and carbon management module is used to analyze the energy consumption associated with safety incidents and calculate carbon emission reductions.
9. The intelligent fire protection platform electrical safety and energy-saving linkage system according to claim 8, characterized in that, The intelligent rule engine contains multiple virtual intelligent agent agents; The intelligent agent agent includes at least one of the following: a security guardian agent aimed at maximizing security, an energy efficiency steward agent aimed at ensuring production continuity, and a reconnaissance expert agent responsible for assessing fire conditions. The intelligent rule engine generates comprehensive decisions through game theory and consensus algorithms among the multiple intelligent agents.
10. The intelligent fire protection platform electrical safety and energy-saving linkage system according to claim 9, characterized in that, The system also includes a blockchain evidence storage module, which is used to generate hash values for the early warning records generated by the AI prediction and early warning engine, the handling records executed by the UAV mobile reconnaissance layer, and the carbon emission reductions calculated by the energy efficiency and carbon management module, and store them on the blockchain.