Microspace robot and fire early warning system thereof based on multi-dimensional temperature model

Abstract

The invention relates to a micro-space robot and a fire early warning system thereof based on a multi-dimensional temperature model, and belongs to the technical field of edge intelligence and Internet of Things. The robot has a miniaturized structure, integrates an infrared temperature sensor, an ambient air temperature sensor and a communication module, and has edge computing capability. A plurality of robots communicate through a wireless gateway, the gateway uploads data to a cloud server through 4G / WiFi, and multi-point and multi-dimensional temperature data collection and intelligent judgment are achieved. The system is based on various early warning models such as temperature threshold judgment, continuous temperature rise trend analysis, short-time rapid rise identification, individual deviation comparison and group abnormity linkage, the early warning capability in high-fire risk places such as express delivery courier stations is remarkably improved, fire hazards are found in advance, and losses are effectively reduced.

Description

Technical Field

[0001] This invention belongs to the field of Internet of Things sensing terminals and fire safety early warning technology, specifically involving a micro-space intelligent robot with microstructure and multi-dimensional temperature intelligent judgment capabilities, and its fire early warning system based on edge algorithms and multi-model fusion. Background Technology

[0002] In urban last-mile logistics locations such as express delivery stations, warehouses, and enclosed stacking areas, traditional fire prevention methods like smoke detectors, cameras, and manual patrols are insufficient to respond promptly to the initial stages of a fire due to their limited space, high-density stacking of goods, unattended personnel, and numerous blind spots in monitoring. This poses serious safety hazards. Especially with lithium-ion batteries, flammable foam packaging, and miscellaneous storage conditions, the risk of overheating, if not identified early, can quickly escalate into a fire, causing significant economic losses and a crisis of trust in the platform. Therefore, developing an intelligent micro-robot system that can be embedded in high-density stacking scenarios, possesses early detection capabilities, and supports remote early warning and local edge detection has significant practical value and widespread application potential. Summary of the Invention

[0003] This invention aims to provide a micro-space robot and its fire early warning system based on a multi-dimensional temperature model, in order to solve the problems of traditional fire early warning systems such as difficulty in deployment in micro-space environments (such as stacking gaps and enclosed corners), serious perception blind spots, and slow response. It proposes a micro-intelligent robot system that supports local edge intelligent judgment and is adaptable to various micro-space deployment methods, and constructs a highly sensitive fire early warning mechanism based on a multi-dimensional temperature model to achieve earlier and more accurate identification of fire hazards.

[0004] The technical solution of this invention includes the following key components and functions:

[0005] 1. Structural Design of Microrobots

[0006] The following components are integrated inside the main shell of the micro-space intelligent robot: Infrared temperature sensor to collect the surface temperature of objects within the field of view; An ambient temperature sensor collects the temperature of the surrounding air. Communication modules, including but not limited to BLE communication modules (such as nRF52832), for low-power data interaction; MCU edge computing chips (such as the STM32F0 series) are used for local algorithm computation; The power supply module uses a button cell battery, which can be replaced for operation; The housing supports magnetic, cable tie, or embedded structures.

[0007] 2. Temperature Acquisition and Judgment Models (Five Types)

[0008] The edge computing module has the following pre-defined multi-model judgment mechanism:

[0009] Model 1: Fixed threshold judgment

[0010] Setting parameters: Infrared temperature threshold T1, e.g., T1 = 65℃ Air temperature threshold T2, e.g., T2 = 55℃ A Level 1 warning is triggered when the current collected value satisfies Tir ≥ T1 or Tair ≥ T2.

[0011] Model 2: Identification of Continuous Temperature Rise Trend

[0012] Setting parameters: Determine if window W1 = 3 minutes; Cumulative temperature rise ΔT ≥ 8℃; Temperature values ​​are sampled every 30 seconds. If the infrared or air temperature increases monotonically within W1 time and the cumulative increase exceeds ΔT, an alarm is triggered.

[0013] Model 3: Short-term rapid rise identification

[0014] Setting parameters: Initial temperature T0; The boost factor K = 1.2; Time window W2 = 30 seconds; If the current temperature Tcurr ≥ K × T0 within W2, it is identified as a short-term temperature rise anomaly, triggering an alarm.

[0015] Model 4: Comparison of Individual Bias

[0016] Setting parameters: Historical average temperature Tavg (7-day sliding window average); Deviation ratio P1 = 1.15; If the current temperature Tcurr ≥ P1 × Tavg, it is identified as an independent anomaly of the robot.

[0017] Model 5: Group Interaction Judgment

[0018] Define N robots as the same area; If more than 30% of the nodes meet any of the abnormal models within the same 10-minute period, a regional-level group warning will be triggered. The cloud server receives and broadcasts regional abnormal signals.

[0019] 3. Data communication mechanism

[0020] Each robot collects and performs edge detection once at regular intervals, including but not limited to every 30 seconds; Gateway data and alarm events are uploaded to the cloud server using the MQTT protocol; The cloud and user terminals perform real-time data synchronization and alarm display, and generate deployment image reports.

[0021] 4. Binding and Identification Mechanism

[0022] Each robot is assigned a unique device number upon leaving the factory; It supports binding deployment information via QR code, including: number, deployment point photo, and remarks; The user terminal interface displays: temperature curves, alarm records, deployment point information, fault status, etc. Attached Figure Description

[0023] Figure 1 is a schematic diagram of the stacking deployment, showing a typical deployment scenario where the intelligent robot for micro-space described in this invention is installed in the gap between the stacks of goods and the back of the shelves in an express delivery station.

[0024] Figure 2 shows the layout of the sealed gaps in the electrical box, demonstrating how the robot is magnetically attached to the side of the distribution box to closely monitor abnormal electrical heat sources.

[0025] Figure 3 shows a monitoring diagram of a sudden temperature rise in a cold chain container, demonstrating how a robot deployed at the bottom of the container identifies the abrupt temperature increase from 5°C to 18°C.

[0026] Figure 4 shows a multi-point anomaly group early warning diagram, which illustrates the distributed deployment of multiple robots and the synchronization of early warning information to the Bluetooth gateway when local temperature anomalies occur, triggering a group collaborative judgment mechanism. Detailed Implementation

[0027] To make the objectives, technical solutions, and beneficial effects of this invention clearer and more explicit, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. This invention is not limited to the specific embodiments described below; any equivalent substitutions or improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

[0028] Example 1: Deployment of temperature sensors for stacking goods at express delivery stations

[0029] In a city's express delivery station, two microspace intelligent robots, as described in this invention, are deployed in the gap between the back of the shelves in the goods stacking area and the center of the stack. Each robot is magnetically fixed to the metal shelf structure, and its embedded infrared temperature sensor and air temperature sensor collect the surface temperature of the target cargo box and the local ambient temperature, respectively.

[0030] In this embodiment, the system sets the infrared temperature alarm threshold to 60℃, and the edge computing module executes the temperature judgment model every 30 seconds. When a robot detects that its infrared temperature is ≥ 60℃, the fixed threshold early warning model is automatically triggered, and the early warning information is uploaded to the cloud server in real time via BLE communication with the nearest gateway.

[0031] After receiving the alarm data, the server automatically pushes a mobile notification to the duty management terminal and retrieves the robot's binding information, including the shelf number, deployment location photo, and installation notes, to assist duty personnel in quickly locating the abnormal area and conducting on-site investigation. Actual handling confirmed that the issue was a pre-heating state caused by abnormal heating of electrical components in the goods, with an alarm response time of no more than one minute.

[0032] Example 2: Slow-heat identification of high-level pallets in warehouses

[0033] In an automated storage and retrieval system, several intelligent robots are deployed in the top pallet area, with an average spacing of no more than 2 meters. This area suffers from poor ventilation, making it inaccessible to conventional smoke detectors, and poses a safety hazard due to the potential for heat-generating equipment to enter the warehouse.

[0034] Each robot is secured to the steel beam at the edge of the shelf with plastic straps. The parameters of the continuous temperature rise trend model are set as follows: cumulative temperature rise threshold: 8℃, monitoring window time: 10 minutes.

[0035] During system operation, robot MWSR-A092 detected that its ambient air temperature slowly increased from 26.2℃ to 34.5℃, and the process met the condition of monotonically increasing. Because the temperature rise did not exceed the fixed threshold, the system did not immediately issue a high-temperature alarm, but instead identified the behavior as an isolated slow-heating trend through the edge model.

[0036] Upon receiving an isolated anomaly, the cloud server, in conjunction with the platform, generated a "heat island warning" and prompted manual review. On-site inspection revealed that the pallet box was sealed and poorly ventilated, containing an unpowered electric heater assembly, causing the equipment casing to continue heating. This enabled proactive risk management.

[0037] Example 3: Monitoring of Sudden Temperature Rise in Cold Chain Containers

[0038] In the cold chain transportation of vaccine-related pharmaceutical products, the robot described in this invention is fixedly installed at the bottom of the inner wall of an insulated container, with infrared sensors facing the outer surface of the goods, for continuous monitoring of the stability of the cold chain environment.

[0039] Before deployment, the device recorded an initial temperature of 5°C, and the threshold parameters for the rapid temperature rise warning model were set as follows: Short window: 3 minutes Temperature rise factor: K = 2.5 (i.e., temperature rise ≥ 2.5 times the initial temperature)

[0040] During transportation, when the container reached noon, the robot detected that the temperature had risen rapidly to 18°C, far exceeding the set threshold. The device immediately triggered a "short-term rapid rise" alarm, broadcasting the alarm frame to the gateway via BLE and uploading it to the logistics platform via the 4G module.

[0041] Upon receiving the alert, the platform automatically sent a phone call to the driver and highlighted the vehicle's status in red on the user's visual interface. Manual verification ultimately revealed that a power outage in the cold chain equipment had prematurely exposed the insulation failure, significantly preventing the vaccine batch from spiraling out of control.

[0042] Example 4: Deployment of enclosed hot zone on the back of the electrical box

[0043] In the power distribution room of a logistics center, the gap between the power control box and the rear wall is less than 8cm, making it impossible to place traditional smoke detectors. The micro-robot described in this invention is inserted into the gap using a long-arm tool and then magnetically attached to the metal casing on the back of the power box.

[0044] The robot collects infrared temperature and air temperature data every 60 seconds. The system is configured with an individual historical deviation model, referencing the device's average sliding temperature over the past 72 hours (T_avg = 36.8℃), and setting the deviation ratio P = 1.15. If the real-time temperature T_curr ≥ 42.3℃, an early warning is triggered.

[0045] In actual operation, the temperature of the device was ≥ 43.1℃ in three consecutive samplings. The edge detection module actively executed the deviation anomaly reporting process and marked it as "continuous overheating". After inspection by engineers, it was found that a component in the electrical cabinet had been operating under high load for a long time, and the wiring port was overheating. If not dealt with, it may cause breakdown or fire.

[0046] Example 5: Early warning of simultaneous abnormal groups at multiple points

[0047] In Zone A of a large express delivery distribution center, a total of 20 micro-space intelligent robots are deployed, distributed across the same cargo flow structure. The distance between the robots is controlled within 3 meters, and the gateway is responsible for the centralized uploading of area broadcast data.

[0048] The system activated the group linkage model, setting the group linkage alarm threshold ratio to 30%. During a night shift operation, seven robots triggered the infrared temperature deviation model within 10 minutes, with each temperature value exceeding 1.3 times its average historical temperature.

[0049] With the abnormal nodes accounting for 35%, the system immediately generated a regional-level high-temperature warning and automatically displayed the affected robot numbers and deployment photos on the terminal. Staff investigated the abnormal area based on the location of the bound shelves and discovered a large batch of suspected lithium battery packages densely piled up in a draftless aisle, posing an extremely high risk of heat buildup. These packages have been relocated and disposed of.

Claims

1. A micro-space robot and its fire early warning system based on a multi-dimensional temperature model, characterized in that, include: At least one set of micro-robots equipped with infrared temperature sensors and ambient air temperature sensors are used to simultaneously collect and monitor the surface temperature of a target object and the temperature of its surrounding microenvironment; the robots also include an edge computing module that periodically executes temperature judgment algorithms; the robots are equipped with a wireless communication module that connects to a cloud server via 4G or WiFi communication; the cloud server connects to a user terminal, which displays the collected data, early warning information, and deployment photos.

2. The system according to claim 1, wherein the edge computing module sets a fixed temperature threshold, and triggers an alarm if the infrared or air temperature is ≥ the threshold.

3. The system according to claim 1, wherein the edge computing module is used to determine the continuous temperature rise trend, and if the temperature continues to rise and exceeds the cumulative threshold within a set time period, an early warning is triggered.

4. The system according to claim 1, wherein an early warning is triggered when the temperature rise exceeds a preset multiple of the initial temperature within a set short-term window.

5. The system according to claim 1, wherein the system determines whether the current robot temperature is higher than a set percentage of its own historical average; if it exceeds this percentage, an early warning is triggered.

6. The system according to claim 1, wherein the system determines whether the current temperature of a certain robot is higher than a set proportion of the average temperature of multiple robots in the same area, and triggers an early warning if it exceeds this proportion.

7. The system according to claim 1, wherein if multiple robots simultaneously exhibit abnormal temperatures exceeding a set threshold ratio within a set time period, a regional group early warning is triggered.

8. The system according to claim 1, wherein robot deployment information supports the display of robot number, shelf number, deployment photo and name by the terminal.

Images