Search

KR-102964303-B1 - Fire detection and prevention system of building automatic control panel using artificial neural network

KR102964303B1KR 102964303 B1KR102964303 B1KR 102964303B1KR-102964303-B1

Abstract

The present invention relates to a fire detection and prevention system for a building automatic control panel using an artificial neural network, and more specifically, to a fire detection and prevention system for a building automatic control panel using an artificial neural network that can detect fire precursors or the occurrence of a fire at an early stage in a building automatic control panel installed in an automatic control system that automatically controls power used in a building, thereby preventing the spread of fire, minimizing economic losses, and preventing damage caused by fire.

Inventors

  • 신여름

Assignees

  • 주식회사 동천기공

Dates

Publication Date
20260513
Application Date
20250724

Claims (3)

  1. In a fire detection and prevention system that detects a fire inside a Direct Digital Control (DDC) panel by detecting and monitoring multiple different areas within the DDC panel connected to multiple control facilities, A heat detection unit (10) that detects a thermal reaction in one or more of the multiple different areas within the above building automatic control panel; A sensor module (20) for detecting one or more of smoke, humidity, dust, and sound in one or more of the plurality of different areas; An analysis module (30) including a storage unit (32) that analyzes whether a fire has occurred inside the control panel by analyzing the detection values received from the heat detection unit and the sensor module using an artificial neural network (ANN), and patterns the smell that causes the fire and the smell that occurs during a fire and stores them as reference values; and It includes an alarm means (40) that receives a signal from the above analysis module and displays a fire precursor phenomenon or fire occurrence externally through sound or light; The sensor module (20) above includes, It includes an odor detection sensor (21) for detecting an odor caused by overheating of an electrical component in one or more of the plurality of different regions, and Air is collected from one or more of the areas within a building automatic control panel divided into multiple areas, and is configured to exclude false detection by cross-verifying the detection values detected by each odor detection sensor (21) at mutually spaced locations. The reference value stored in the above storage unit (32) is a pattern of the gas component ratio contained in the air during a fire precursor phenomenon or fire occurrence, and the gas component ratio contained in the air during the fire precursor and the initial stage of fire occurrence is patterned, and the gas component ratio is analyzed multiple times during the patterning process, and the minimum value obtained during the analysis is set as a reference pattern for fire precursor judgment and stored. It includes an artificial intelligence sensor (22) that detects the thermal response of terminals and cables connected to air conditioners, water tanks, pumps, drainage tanks, and fans installed in the building, and performs self-verification, self-adaptation, and self-identification. The above artificial intelligence sensor (22) is, The sensed value is transmitted to the central control server (CP) through the wired/wireless communication unit (50), and the central control server (CP) transmits the sensed value detected through the artificial intelligence sensor (22) to the previously stored administrator terminal. The above analysis module (30) is, A detection and amplification unit (31) that converts the detection value received from the heat detection unit and the sensor module into data to detect a signal and amplifies the detected signal, and A location determination unit (33) that receives an electrical signal from the detection and amplification unit and compares the received electrical signal with a reference value pattern stored in the storage unit, and determines the location where a fire-causing odor or a fire-generating odor occurs using a restricted Boltzmann machine (RBM) model among artificial neural networks, and A fire detection and prevention system for a building automatic control panel using an artificial neural network, characterized by including a level judgment unit (34) that receives an electrical signal from the detection and amplification unit and determines whether a fire has occurred and the fire level in comparison with a reference value stored in the storage unit.
  2. delete
  3. delete

Description

Fire detection and prevention system of building automatic control panel using artificial neural network The present invention relates to a fire detection and prevention system for a building automatic control panel using an artificial neural network, and more specifically, to a fire detection and prevention system for a building automatic control panel using an artificial neural network that can detect fire precursors or the occurrence of a fire at an early stage in a building automatic control panel installed in an automatic control system that automatically controls power used in a building, thereby preventing the spread of fire, minimizing economic losses, and preventing damage caused by fire. Generally, a Building Automation System (BAS) is a system that enhances the efficiency of building operations by centrally integrating and controlling various facility systems within a building, such as heating, cooling, ventilation, lighting, and power. Through this, it reduces energy consumption and lowers labor operating costs while maintaining a comfortable environment. If arcs or cable short circuits occur in the electrical equipment or cables constituting these building automatic control systems and maintenance is not performed promptly, it will subsequently lead to flashovers, electric shocks, and electrical fires. However, only circuit breakers and electronic overcurrent relays (EOCR) are applied as means to detect various phenomena such as current pulses, heat, light, electromagnetic waves, and sound in electrical equipment or cables constituting building automatic control systems. Meanwhile, Direct Digital Control (DDC) is a building automation control system that uses a digital computer to directly control final control elements (e.g., valves, dampers, air conditioners, water tanks, pumps, drainage tanks, fans, etc.), and enables precise and efficient control by processing control signals digitally instead of using the existing analog method. These building automation control panels offer advantages such as precise temperature and humidity control, energy savings, ease of maintenance, system performance optimization, automation, and improved efficiency, and are being applied to small buildings as well as large ones. There are two main factors that can cause fires in the above building automatic control panel: one is heat generation due to overcurrent, and the other is heat generation due to voltage drop. In addition, even if no overcurrent flows, the inside of the building automatic control panel becomes severely overheated locally due to an increase in control resistance or deterioration of the relay, and this situation can occur even within the normal current range, leading to frequent cases where the circuit breaker fails to detect it and results in a fire. The existing fire prevention device proposed to prevent the above-mentioned problems is equipped with temperature sensors and fire sensors on the main components constituting the interior of a building automatic control panel, and is configured to detect a fire through rapid temperature changes via the temperature sensor or the detection of flames or smoke via the fire sensor, thereby shutting off the circuit breaker of the building automatic control panel. However, this existing fire prevention device frequently results in complete destruction because it is difficult to extinguish the fire in the control panel by the time a fire-related signal is detected by the sensor, as the fire has already progressed to some extent. Additionally, there is a problem with a high frequency of error signals caused by increased sensor sensitivity. And there was a problem where fuses installed to cut off overcurrents frequently melted due to fire. FIG. 1 is a block diagram illustrating a fire detection and prevention system for a building automatic control panel using an artificial neural network according to the present invention. FIG. 2 is a flowchart illustrating the operation sequence of a fire detection and prevention system for a building automatic control panel using an artificial neural network according to the present invention. FIG. 3 is an exemplary diagram illustrating an embodiment of a building automatic control panel constituting the present invention. In addition to the above objectives, other objectives and features of the present invention will become apparent through the description of embodiments with reference to the accompanying drawings. Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as generally understood by those skilled in the art to which the present invention pertains. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an ideal or overly formal sense unless explicitly defined in this application. Hereinafter, a preferred