CN-121982858-A - Intelligent fire-fighting early warning method and system for booster station

CN121982858ACN 121982858 ACN121982858 ACN 121982858ACN-121982858-A

Abstract

The application provides an intelligent fire-fighting early warning method and system for a booster station, and relates to the technical field of fire-fighting early warning, wherein the method comprises the steps of deploying a fixed sensing unit in key equipment of the booster station, establishing a new energy booster station intelligent fire-fighting network, and monitoring equipment states through a preset period; transmitting the multi-source monitoring data to a data centralized control center through a wireless sensor network based on LoRa to conduct fire risk prediction, conducting rechecking verification on the preliminary early warning signals to obtain rechecking verification results, determining risk levels according to the fire risk prediction results if fire conditions are confirmed, triggering fire control linkage control instructions, and issuing the fire control linkage control instructions to a mobile fire control robot to conduct fire control early warning. The application solves the technical problem that the fire-fighting early warning is inaccurate due to the complex electromagnetic environment in the booster station and the false alarm and omission of the traditional detector in the prior art, and improves the fire-fighting early warning accuracy through the fire risk prediction and rechecking verification of the data centralized control center.

Inventors

ZHANG TINGFA
ZHANG JIN
Xiong Tiance
WAN XIAOQIANG
JIN YUHAN
LU NAN

Assignees

江西大唐国际新能源有限公司
大唐可再生能源试验研究院有限公司

Dates

Publication Date: 20260505
Application Date: 20260114

Claims (8)

1. An intelligent fire protection early warning method for a booster station is characterized by comprising the following steps: a fixed sensing unit is deployed in key equipment of the booster station, a new energy booster station intelligent fire-fighting network is established, equipment state monitoring is carried out through a preset period, and multi-source monitoring data are obtained; transmitting the multisource monitoring data to a data centralized control center through a wireless sensing network based on LoRa to predict fire risks, and generating a preliminary early warning signal; performing rechecking verification on the preliminary early warning signal to obtain a rechecking verification result; If the rechecking verification result is a fire confirmation result, determining a risk level according to the fire risk prediction result; Based on the risk level, triggering a fire-fighting linkage control instruction, and issuing the fire-fighting linkage control instruction to the mobile fire-fighting robot for fire-fighting early warning.
2. The intelligent fire protection early warning method for the booster station of claim 1, wherein the fixed sensing unit comprises a pyrolytic particle detector, a composite electric fire detector, a distributed optical fiber temperature sensing module, a high-definition visible light and infrared thermal imaging dual-mode cradle head camera.
3. The intelligent fire protection early warning method for a booster station of claim 1, wherein the performing the rechecking verification on the preliminary early warning signal to obtain a rechecking verification result comprises: The method comprises the steps of obtaining synchronous visible light video stream and infrared thermal imaging images of an early warning area through a fixed sensing unit; performing image recognition processing based on the visible light video stream, extracting video image characteristics, wherein the video image characteristics at least comprise visual characteristics of smoke texture, flame color and form and arc flash; Extracting thermal characteristics of the highest temperature point, the temperature gradient distribution, the area and the shape of the abnormal high temperature area based on the infrared thermal imaging graph, and drawing a temperature change trend graph; Judging whether fire signs exist or not based on the video image characteristics and the temperature change trend graph; and if the fire sign exists, the rechecking verification result is that the fire condition is confirmed.
4. The intelligent fire-fighting early warning method for a booster station according to claim 3, wherein if the rechecking verification result is that no fire condition is confirmed, audible and visual warning of a corresponding area is released, meanwhile, the reason generated by the preliminary early warning signal is analyzed, an interference sample is generated, and the interference sample is used as a negative sample to iteratively optimize the fire risk prediction model.
5. The intelligent fire protection early warning method for a booster station of claim 1, wherein if the rechecking verification result is a fire confirmation result, determining a risk level according to a fire risk prediction result comprises: Acquiring a fire risk probability value of the fire risk prediction result and a confidence coefficient for representing the prediction reliability; Based on the confidence level, presetting a probability threshold interval, wherein the preset probability threshold interval comprises a preset probability first threshold and a preset probability second threshold, and the preset probability first threshold is larger than the preset probability second threshold; matching the fire risk prediction result with the preset probability threshold value, and determining a corresponding risk level: If the fire risk probability value is larger than the first threshold value of the preset probability, the risk level is a first-level risk; If the fire risk probability value is greater than the preset probability second threshold value and less than or equal to the preset probability first threshold value, the risk level is a secondary risk; And if the fire risk probability value is smaller than or equal to the second threshold value of the preset probability, the risk level is three-level risk.
6. The intelligent fire protection early warning method for a booster station of claim 5, wherein based on the risk level, triggering a fire protection coordinated control instruction, issuing to a mobile fire protection robot for fire protection early warning, comprising: if the risk level is three-level risk, generating a patrol work order and feeding the patrol work order back to the data centralized control center; if the risk level is the secondary risk, starting an audible and visual alarm device, mobilizing the mobile fire-fighting robot to go to an alarm area for standby, and feeding back real-time reconnaissance data; if the risk level is a first-level risk, a fire-fighting linkage control instruction is generated according to a fire area and is issued to a mobile fire-fighting robot for triggering type inspection and cutting off an associated power supply loop, wherein the mobile fire-fighting robot is provided with an environment sensing sensor and a fire-extinguishing module.
7. The intelligent fire protection early warning method for a booster station of claim 1, wherein if the risk level is a primary risk, generating a fire protection coordinated control instruction according to a fire area, and issuing the fire protection coordinated control instruction to a mobile fire protection robot for fire protection early warning, comprising: The data centralized control center locates a fire area based on the fire risk prediction result; Acquiring an area map of the booster station, planning a path based on the real-time position of the mobile fire-fighting robot and the fire area, and determining an optimal navigation path by taking the shortest action time and the minimum energy consumption as targets; Based on the characteristics of the fire area and the fire risk prediction result, generating a fire extinguishing scheme through the data centralized control center; the fire area, the optimal navigation path and the fire extinguishing scheme are issued to the mobile fire-fighting robot, the mobile fire-fighting robot moves to the fire area, the mounted double-light cradle head is used for identifying and locking a fire source, the self posture and the angle of the spraying device are adjusted, and when a remote confirmation instruction is received, the spraying of the fire extinguishing agent is started; in the fire extinguishing process, the mobile fire-fighting robot returns a scene image in real time, and after the fire extinguishing operation is finished, the mobile fire-fighting robot feeds back a treatment result to the data centralized control center.
8. An intelligent fire early warning system for a booster station, characterized by the steps for implementing the intelligent fire early warning method for a booster station according to any one of claims 1 to 7, comprising: the equipment state monitoring module is used for deploying a fixed sensing unit in key equipment of the booster station, establishing a new energy booster station intelligent fire-fighting network, and monitoring the equipment state through a preset period to obtain multi-source monitoring data; the fire risk prediction module is used for transmitting the multi-source monitoring data to a data centralized control center through a wireless sensing network based on LoRa to predict fire risk and generate a preliminary early warning signal; The rechecking verification module is used for rechecking and verifying the preliminary early warning signal to obtain a rechecking verification result; The grade determining module is used for determining the risk grade according to the fire risk prediction result if the rechecking verification result is the fire confirmation result; and the fire-fighting early warning module is used for triggering a fire-fighting linkage control instruction based on the risk level and issuing the fire-fighting linkage control instruction to the mobile fire-fighting robot for fire-fighting early warning.

Description

Intelligent fire-fighting early warning method and system for booster station Technical Field The application relates to the technical field of fire protection early warning, in particular to an intelligent fire protection early warning method and system for a booster station. Background The booster station is used as a key hub for power transmission, and the internal dense high-voltage electric equipment can generate a complex strong electromagnetic field environment when in operation, so that a serious challenge is formed for the traditional smoke-sensing and temperature-sensing fire disaster detector, the electromagnetic interference is extremely easy to cause equipment misoperation, so that the system frequently generates false alarms, and meanwhile, the shielding effect formed by a large number of metal cabinets and the interference factors such as intermittent heating of the equipment can mask the real early fire disaster characteristics, so that the risk of missing the alarm is also possible. Because the early warning is inaccurate, operation and maintenance personnel have to expend a great deal of effort to carry out on-site recheck, so that the operation and maintenance burden is increased, and the credibility of the intelligent fire-fighting system is seriously weakened. Due to detection failure of hidden dangers such as early overheat and insulation degradation in the electrical equipment, preventive intervention measures cannot be implemented in time, and fire risks are accumulated continuously, so that the operation safety of the booster station is further affected. In summary, in the prior art, because the electromagnetic environment in the booster station is complex, the traditional detector is prone to false alarm and missing alarm, so that fire-fighting early warning is inaccurate, and the operation safety of the booster station is further affected. Disclosure of Invention The application aims to provide an intelligent fire-fighting early warning method and system for a booster station, which are used for solving the technical problems that in the prior art, due to the fact that the electromagnetic environment in the booster station is complex, the traditional detector is prone to false alarm and missing alarm, fire-fighting early warning is inaccurate, and the operation safety of the booster station is further affected. In order to achieve the purpose, the application provides an intelligent fire-fighting early warning method and system for a booster station. The intelligent fire protection early warning method for the booster station is realized through an intelligent fire protection early warning system for the booster station, and comprises the steps of disposing a fixed sensing unit in key equipment of the booster station, establishing a new energy booster station intelligent fire protection network, monitoring equipment states through a preset period to obtain multi-source monitoring data, transmitting the multi-source monitoring data to a data centralized control center through a wireless sensing network based on LoRa to conduct fire risk prediction, generating a preliminary early warning signal, conducting rechecking verification on the preliminary early warning signal to obtain rechecking verification results, determining risk levels according to fire risk prediction results if the rechecking verification results are fire confirmation results, triggering fire protection linkage control instructions based on the risk levels, and sending the fire protection linkage control instructions to a fire protection robot to conduct fire protection early warning. Optionally, the fixed sensing unit comprises a pyrolytic particle detector, a composite electric fire detector, a distributed optical fiber temperature sensing module, and a high-definition visible light and infrared thermal imaging dual-mode cradle head camera. The method comprises the steps of obtaining a synchronous visible light video stream and an infrared thermal imaging image of an early warning area through a fixed sensing unit, carrying out image recognition processing on the basis of the visible light video stream, extracting video image characteristics at least comprising visual characteristics of smoke textures, flame colors and forms and arc flashes, extracting thermal characteristics of a highest temperature point, temperature gradient distribution, an area and a shape of an abnormal high temperature area on the basis of the infrared thermal imaging image, drawing a temperature change trend chart, judging whether fire signs exist on the basis of the video image characteristics and the temperature change trend chart, and if the fire signs exist, checking the fire signs to be confirmed according to a checking verification result. Optionally, if the rechecking verification result is that no fire condition is confirmed, the audible and visual alarm of the corresponding area is released, meanwhile, the reason generated by the