Multi-spectral beam anti-interference smoke detector

Abstract

The present invention belongs to the technical field of dust and smoke discrimination detection. Specifically, a multi-spectral beam anti-interference smoke detector is provided, which includes: by using the dual-transmission and single-reception mode in combination with a smoke compensation recognition algorithm to distinguish the characteristics of smoke and dust, different treatments are carried out for different characteristics, which improves the reliability of the product and the economic benefits of product maintenance. At the same time, through the optimization of the maze structure, false alarms caused by dust are reduced; according to the light changes in different scenarios, the optimal spectral band combination is automatically selected to enhance the anti-interference ability and detection accuracy, and enhance the adaptability of the detector to different scenarios. At the same time, by using a multi-spectral beam smoke detector to detect fire smoke, when abnormal changes are detected in multiple spectral bands simultaneously and the change trend conforms to the fire characteristics, the detector will issue an alarm, avoiding false alarm problems caused by single-spectral beams, thereby enhancing the anti-interference ability and detection reliability of the detector.

Description

Technical Field

[0001] The present invention belongs to the technical field of dust and smoke discrimination detection, and relates to a multi-spectral beam anti-interference smoke detector. Background Art

[0002] Fire is a disaster that seriously threatens life and property safety. Timely and accurately detecting fire smoke is crucial for preventing and controlling fires. In various places, such as commercial buildings, industrial factories, and residences, reliable fire detection equipment is required to ensure safety. Traditional smoke detectors can play a role in fire warning to a certain extent, but with the increasingly complex environment and the continuous improvement of the requirements for fire detection accuracy, their limitations have gradually emerged.

[0003] There are various interference factors in the modern environment, such as dust, fog, and different lighting conditions. These factors will affect the normal operation of smoke detectors, resulting in false alarms or missed alarms. For example, in industrial production workshops, a large amount of dust will cause traditional detectors to misjudge as smoke, thus issuing unnecessary alarms; in places with different light intensities and spectral distributions, single-spectral detectors may not be able to accurately identify fire smoke, reducing the reliability of detection.

[0004] In order to cope with the challenges of complex environments and improve the accuracy and reliability of fire detection, it is necessary to continuously develop and improve smoke detection technology. The multi-spectral beam smoke detection technology has emerged. It uses the information of multiple spectral bands for comprehensive analysis, can more accurately identify fire smoke, reduce the influence of interference factors, and provides a more effective solution for fire detection.

[0005] In the prior art, there are already some related solutions involving dust and smoke discrimination detection. For example, the patent with the Chinese patent publication number CN118840820A discloses a multi-parameter double-smoke-chamber smoke detector system. There are two smoke chambers, smoke chamber A and smoke chamber B, on its upper cover. By using emission tubes and receiving tubes with different spectra to obtain a variety of parameter information, the detection accuracy is improved. On one side of smoke chamber A and smoke chamber B, there is a non-smoke baffle that effectively blocks non-smoke particles entering the smoke chamber from different directions but allows true smoke to pass through unobstructed. Both smoke chamber A and smoke chamber B adopt a trapezoidal design with a top diameter smaller than the bottom diameter, and the smoke inlets of smoke chamber A and smoke chamber B are inclined; the double-smoke-chamber structure, combined with the non-smoke baffle and the inclined surface design, effectively reduces false alarms and missed alarms. The detector synchronously controls and coordinates two sets of transmitting and receiving modules, analyzes and processes the mutual relationship between various parameters, and matches with a database of known smoke characteristics to determine whether there is a fire, improving the anti-false-alarm ability.

[0006] Another patent with the Chinese patent publication number CN114463925A discloses a fire detection alarm, which includes an upper housing, a lower housing, a mounting plate and a smoke detector. A plurality of ventilation grooves are provided on the lower housing. The upper housing is fixedly connected to the lower housing. The smoke detector is fixed on the mounting plate, and the mounting plate is fixedly installed between the upper housing and the lower housing. A channel is provided above the smoke detector, and components are provided at the ends of each channel. The present invention uses the differences in the scattering intensities of particles with different particle size distribution ranges and the differences in the forward and backward scattering angles of red and blue light with different frequencies and wavelengths of the spectrum to qualitatively identify the types of smoke, so as to achieve the purpose of early fire warning and monitoring.

[0007] Although the above solution proposes some solutions for distinguishing dust and smoke detection, there are still some limitations, specifically: at the first level, it is difficult to automatically adjust the detection parameters according to the light changes in different scenarios. Under different light intensities and spectral distributions, the performance of the detector may be greatly affected, and it is impossible to ensure accurate detection of fire smoke in various scenarios. For example, in outdoor strong light or indoor weak light environments, the sensitivity and accuracy of the detector may decrease, which cannot meet the requirements of actual applications.

[0008] On another level, only the spectra of different frequencies and wavelengths of red and blue light are used for detection, and the spectral dimension is relatively single. In a complex environment, it may be more susceptible to interference substances with similar red and blue light scattering characteristics due to the lack of multi-dimensional spectral information, resulting in misjudgment. Summary of the Invention

[0009] In view of this, to solve the problems raised in the above background technology, a multi-spectral beam anti-interference smoke detector is proposed.

[0010] The object of the present invention can be achieved through the following technical solutions: The present invention provides a multi-spectral beam anti-interference smoke detector, including: a smoke mode discrimination unit, which judges the pollution compensation mode and smoke mode of the smoke scene through maze structure optimization and software compensation algorithm.

[0011] An adaptive spectral fire warning unit, in the smoke mode, automatically selects the optimal spectral band combination according to the light changes in different scenarios, detects the fire characteristics according to the optimal spectral band combination, and issues a fire alarm.

[0012] A dust characteristic supervision unit, in the pollution compensation mode, monitors the maze dust pollution level and uses the Internet of Things platform for unified early warning management.

[0013] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The present invention can automatically select the optimal spectral band combination according to the light changes in different scenarios, thereby improving the anti-interference ability and detection accuracy, and enhancing the adaptability of the detector to different scenarios. At the same time, by using multiple spectral bands to detect fire smoke, when abnormal changes are detected simultaneously in multiple spectral bands and the change trend conforms to the characteristics of a fire, the detector will issue an alarm, avoiding false alarms caused by a single spectral beam, thereby enhancing the anti-interference ability and detection reliability of the detector.

[0014] (2) The present invention uses a dual-transmission and single-reception mode in combination with a smoke compensation recognition algorithm to distinguish the characteristics of smoke and dust, and different treatments for different characteristics improve the product reliability and the economic benefits of product maintenance.

[0015] (3) Through the optimization of the maze structure, the present invention reduces false alarms caused by dust. At the same time, since the pollution compensation has an upper limit, after the dust capacity becomes larger, it is equivalent to delaying the compensation, thereby making the maze life longer. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0017] Figure 1 It is a schematic diagram of the unit connection of the present invention.

[0018] Figure 2 It is a schematic diagram of the maze structure of the present invention.

[0019] Figure 3 It is a schematic diagram of the pollution compensation mode and the corresponding judgment process of the smoke mode in the smoke scenario of the present invention.

[0020] Reference numerals: 01, front shell; 02, bottom shell; 03, red light emitting tube; 04, infrared light emitting tube; 05, receiving tube. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0022] Please refer to Figure 1As shown in the figure, the present invention provides a multi-spectral beam anti-interference smoke detector, which includes a smoke mode discrimination unit, an adaptive spectral fire warning unit, and a dust characteristic supervision unit. The smoke mode discrimination unit, the adaptive spectral fire warning unit, and the dust characteristic supervision unit are connected in sequence.

[0023] Please refer to Figure 2 As shown in the figure, the smoke mode discrimination unit determines the pollution compensation mode and the smoke mode of the smoke scene through the maze structure optimization combined with the software compensation algorithm.

[0024] The maze structure is designed for the outer shell or channel around the sensor, which is used to optimize the light scattering effect to better detect smoke.

[0025] In an embodiment, the software compensation algorithm is specifically: using the dual-transmitter and single-receiver mode combined with the smoke compensation recognition algorithm to determine the pollution compensation mode and the smoke mode of the smoke scene.

[0026] The dual-transmitter and single-receiver mode includes two transmitter tubes, namely LED1 and LED2, and also includes a receiver tube 05, namely LED3.

[0027] Among them, LED1 is an IR transmitter tube, that is, infrared transmitter tube 04; LED2 is an R transmitter tube, that is, red light transmitter tube 03; a receiver tube 05 can use a broadband infrared receiver tube with broadband receiving characteristics to receive infrared signals of different frequencies.

[0028] Smoke particles with different particle sizes and compositions have different scattering and absorption characteristics for infrared light of different frequencies. When the infrared transmitter tube 04 emits infrared light of different frequencies into the smoke scene, the smoke will cause the infrared light to scatter and absorb, resulting in changes in the light intensity and frequency components received by the receiver tube 05. The receiver tube 05 receives and converts these changed optical signals to obtain the particulate matter state of the smoke scene.

[0029] In a further embodiment, the maze structure optimization is specifically: as Figure 2 shown, triangular stripe traps are added to the inner bottom surface of the face shell 01 that cooperates with the maze bottom shell 02. The stripe traps are vertically distributed with the infrared transmitter tube 04, that is, IR. The stripe traps can enhance the anti-dust accumulation ability, effectively improve the service life of the maze, and reduce false alarms of the product; at the same time, the triangular stripe traps make the bottom surface a rough surface, and the light of the transmitter tube is scattered more evenly. A concave trap is adopted in the intersection area of the three lamp tubes of the maze bottom shell to form a maze signal sensitive area.

[0030] The present invention optimizes the maze (smoke sensor) structure, reduces false alarms caused by dust. At the same time, since the pollution compensation has an upper limit, after the dust capacity becomes larger, it is equivalent to delaying the compensation, thereby making the maze life longer.

[0031] Please refer to Figure 3 As shown, in a further embodiment, the pollution compensation mode and the smoke mode for judging the smoke scene are optimized through the maze structure in cooperation with the software compensation algorithm, and the content includes: S01. Obtain the power-on state of the smoke alarm, including the power-on state and the normal standby state.

[0032] The power-on state means that the smoke detector detects smoke or dust components, and then starts to power on to detect the fire smoke.

[0033] S02. Under the power-on state, LED1 and LED2 simultaneously accelerate the acquisition of smoke signals.

[0034] S02-1. When the maze structure does not recognize the smoke alarm or sensor failure, judge whether the continuous detection times of LED1 and LED2 exceed the preset times.

[0035] S02-1.1. When the continuous detection times of LED1 and LED2 exceed the preset times, judge whether the fluctuation value of the continuous acquisition signals of LED1 and LED2 in the preset times is less than or equal to the preset fluctuation value. When it is less than or equal to the preset fluctuation value, it is judged as the pollution compensation mode.

[0036] In the pollution compensation mode, the average values of all sampling signals of LED1 and LED2 within the preset times are respectively taken as the clean compensation values of LED1 and LED2, and then enter the next round of compensation.

[0037] S02-1.2. When the fluctuation value of the continuous acquisition signals of LED1 and LED2 in the preset times exceeds the preset fluctuation value, it is judged as the smoke mode. At the same time, judge whether the signal fluctuation value of LED1 or LED2 during the preset sampling process exceeds the preset fluctuation value. If it does not exceed, clear the continuous sampling times of LED1 and LED2 to 0.

[0038] S02-2. If a smoke alarm occurs, no clean value compensation is performed during the alarm period, and a new round of clean value compensation starts after the alarm ends.

[0039] If a maze (sensor) failure occurs, such as the value of LED1 or LED2 collected is lower than the preset failure value, it is judged as a maze failure. Whether it meets the pollution compensation conditions or not, the sampling values received during the failure period are not used to update the clean values of LED1 or LED2. After the failure is eliminated, a new round of clean value compensation starts.

[0040] S03. In the normal standby state, do not accelerate the detection of smoke signal acquisition.

[0041] In a further embodiment, the preset number in the continuous detection times of the LED1 and LED2 exceeding the preset number is moderately adjusted differently according to the actual smoke scenario, and the content includes: identifying the corresponding size difference characteristics of smoke particles and dust particles through the built-in sensor in the maze structure, and determining the content ratio of smoke particles and dust particles.

[0042] Light sources such as the infrared emitter 04 arranged in the maze structure emit light of different wavelengths. Particles of different diameters have different degrees of scattering or absorption of light of different wavelengths. For example, due to different compositions, shapes, and size distributions, the scattering intensities of smoke particles and dust particles for red light and infrared light may be different. By analyzing the scattering signal intensities of light of different wavelengths received by the receiving tube 05, it is possible to indirectly distinguish whether the particles entering the maze structure are smoke particles or dust particles, and determine their respective content ratios according to the relationship between the signal intensity and the particle concentration.

[0043] The light source of the smoke scenario is detected by integrating sensors of different spectral bands in the detector to generate signal intensities of different spectral bands, and the light source complexity is determined by calculating the standard deviation of the signal intensities of different spectral bands. The larger the standard deviation, the greater the difference in signal intensities of each band, and the higher the light source complexity. For example, for a set of signal intensity data from different spectral bands, calculate its standard deviation. If the standard deviation is large, it indicates that the light signal intensities of these bands are more dispersed, and the light source may contain a variety of spectral components with different intensities, and the complexity is high.

[0044] The sensors integrating different spectral bands are sensors for receiving and detecting light signals of specific wavelengths.

[0045] The light sources of different spectral bands include visible light, infrared light, ultraviolet light, etc. For example, in some complex environmental monitoring scenarios, there may be a mixture of light generated by multiple different light-emitting mechanisms or multiple different light-emitting substances, reflecting the complexity of the light source composition. For example, in a smoke scenario, if in addition to the infrared light emitted by the infrared emitter 04, there are also scattered or reflected lights from other light sources in the surrounding environment such as natural light and lamp light, then the spectral bands received by the receiving tube 05 during smoke detection will increase, resulting in a reduction in detection accuracy due to the influence of other light sources.

[0046] Based on the empirical formula, according to the content ratio P of the smoke particles and the light source complexity L, determine the preset number of times for accelerating the smoke signal acquisition at the smoke alarm point , which is , where a, b, and c are weight coefficients, which can be adjusted and optimized through experiments or actual application data. For example, after multiple tests, it is found that when a = 2, b = 1, and c = 5, it can better adapt to most scenarios.

[0047] Specifically, for the case where the proportion of smoke particles exceeds the proportion of dust particles, the alarm point and the alarm detection count times remain unchanged, and normal alarm is carried out; for the case where the proportion of dust particles exceeds the proportion of smoke particles, the alarm point (the cleanliness values of the updated LED1 and LED2) is increased to improve the product's ability to adapt to environmental changes and enhance the product's dust resistance.

[0048] In a further embodiment, the content of S02-1.1 includes: after the product is powered on, LED1 emits light wave 1 once every 1 second, and at the same time, LED2 also emits light wave 2 once every 1 second, and light wave 1 and light wave 2 have different wavelengths, which are the sampling wavelengths.

[0049] Taking the continuous detection of LED1 and LED2 for a corresponding preset number of times as the total accelerated acquisition period, and dividing the total accelerated acquisition period into a first compensation period and a second compensation period. The first compensation period and the second compensation period respectively contain corresponding sampling times, both sampling once per second.

[0050] If the signal fluctuations of LED1 and LED2 in the first compensation period do not exceed the preset fluctuation value, then the average value of the corresponding acquisition signals of all acquisition times in the first compensation period is used as the new cleanliness value of LED1 or LED2 to replace the original cleanliness value of LED1 or LED2.

[0051] After the first compensation period, it enters the second compensation period, and after the total accelerated acquisition period, it enters normal standby.

[0052] If during normal standby, it is detected that the fluctuation value of the continuous acquisition signals of LED1 and LED2 in the preset number of times is less than or equal to the preset fluctuation value, then it enters the pollution compensation mode, and at the same time, the average value of all samplings of LED1 and LED2 in the first compensation period under the normal standby acquisition period is updated and used as the new cleanliness value of LED1 or LED2.

[0053] Exemplarily, both LED1 and LED2 emit 30 times, that is, the total accelerated acquisition period is 30 samplings. Among them, the first round of compensation totals 26 samplings, sampling once per second; if the signal fluctuations of LED1 and LED2 do not exceed the preset fluctuation value (WAV_LMT), then the average value of the signals of these 26 samplings is used as the new cleanliness value of LED1 or LED2 to replace the original cleanliness value of LED1 or LED2. Specifically, after averaging the 26 sampled signal values of LED1 collected and recorded, the average value is used as the new cleanliness value of LED1; after averaging the 26 sampled signal values of LED2 collected and recorded, the average value is used as the new cleanliness value of LED2.

[0054] After the first round of compensation sampling is performed 26 times, there are 4 times of accelerated sampling, also sampling once every 1 second, thus forming a 30-second sampling period with the previous 26 samplings. It has no other function. Among them, the first round of compensation sampling and the 4 times of accelerated sampling are both defined as accelerated acquisitions. The reason for defining them as accelerated is that the sampling interval time will be longer during standby.

[0055] After 30 times of accelerated acquisitions, it enters normal standby. During normal standby, the acquisition periods are as follows: For LED1, it samples once every 8 seconds, and for LED2, it samples once every 40 seconds, which can save power without affecting smoke detection.

[0056] If during normal standby, it meets the condition that "the fluctuation values of the continuously acquired signals of LED1 and LED2 in the preset number of times are less than or equal to the preset fluctuation value", then it enters the pollution compensation mode, and the average values of the 26 samplings of LED1 and the average values of the 26 samplings of LED2 in the normal standby acquisition period will be updated and used as the new clean values of LED1 or LED2.

[0057] In a further embodiment, the simultaneous accelerated acquisition of smoke signals by LED1 and LED2 includes two conditions for entering the accelerated acquisition mode: One situation: During the corresponding sampling process of the total accelerated acquisition period, if the signal fluctuation value of LED1 or LED2 exceeds the preset fluctuation value during a certain sampling process, then enter the accelerated acquisition mode: Action 1, clear the continuous sampling counts of LED1 and LED2 to 0 and start re-sampling from the first count, and at the same time clear the sampled values of LED1 and LED2 held in the previous sampling; Action 2, then perform accelerated sampling, with both LED1 and LED2 sampling once every 1 second, and re-sample according to the corresponding number of samplings in the first compensation period, and then update the clean values of LED1 and LED2.

[0058] Another situation: During normal standby, if the signal fluctuation value of LED1 or LED2 exceeds the preset fluctuation value during a certain sampling process, then enter the accelerated acquisition mode: Action 1, clear the continuous sampling counts of LED1 and LED2 to 0 and start re-sampling from the first count, and at the same time clear the sampled values of LED1 and LED2 held in the previous sampling; Action 2, then perform accelerated sampling, with both LED1 and LED2 sampling once every 1 second, and re-sample according to the corresponding number of samplings in the first compensation period, and then update the clean values of LED1 and LED2.

[0059] Exemplarily, during the 7th implementation sampling, if the signal fluctuation value of LED1 or LED2 exceeds the preset fluctuation value during this sampling process, the accelerated acquisition mode is entered. The 8 data (sampling values of LED1 and LED2) recorded in the previous 8 samplings will be discarded, and the sampling count will be reset. For the subsequent 22 accelerated samplings, sampling is also performed once every 1 second, thus forming a 30-second sampling period with the previous 8 samplings to make the sampling periods consistent. Of course, if during any of the subsequent 22 accelerated samplings, a sampling meets the condition for entering the accelerated acquisition mode, the accelerated acquisition mode will also be entered.

[0060] The present invention uses a dual-transmission single-reception mode in combination with a smoke compensation recognition algorithm to distinguish the characteristics of smoke and dust, and performs different treatments on different characteristics, thereby improving the product reliability and the economic benefits of product maintenance.

[0061] In the smoke mode, the adaptive spectral fire warning unit automatically selects the optimal spectral band combination according to the light changes in different scenarios to enhance the adaptability of the detector to different scenarios, and detects the fire characteristics according to the optimal spectral band combination to issue a fire alarm.

[0062] In the implementation, the content of automatically selecting the optimal spectral band combination according to the light changes in different scenarios in the smoke mode includes: listing each spectral band category, detecting the corresponding signal intensity of each spectral band category in the smoke scenario, and forming a spectral feature vector. Each spectral band category includes categories such as visible light, infrared light, and ultraviolet light.

[0063] Arrange the signal intensities of each spectral band collected in a certain order to form a spectral feature vector. For example, if the signal intensities of 10 different spectral bands are measured, then the spectral feature vector is a 10-dimensional vector, and each element of the vector corresponds to the signal intensity value of a spectral band.

[0064] Establish a smoke component feature library, obtain the transmission characteristics of smoke particles in the smoke scenario for each spectral band, and generate a smoke particle characteristic vector in the smoke scenario.

[0065] Specifically, in different fire scenarios (such as wood combustion, plastic combustion, grease combustion, etc.), industrial production scenarios (such as smog generated from smelting, chemical industry, etc.), and natural environment scenarios (such as forest fires, volcanic eruptions, etc.), professional smoke collection equipment is used to collect smoke samples. During the collection process, attention should be paid to recording the source of the samples, collection time, environmental conditions (temperature, humidity, air pressure, etc.). Spectral analysis is performed on the collected smoke samples using equipment such as spectrometers to measure the transmitted light intensity of each smoke sample in different spectral frequency bands (such as ultraviolet light, visible light, infrared light, and other sub-frequency bands). The spectrometer should have high resolution and a wide spectral range to ensure accurate acquisition of spectral information in each frequency band. At the same time, appropriate measurement parameters (such as integration time, number of scans, etc.) are set to ensure the accuracy and reliability of the measurement results.

[0066] According to the source and composition of the smoke samples, each smoke particle category is classified and sorted. For example, smoke can be classified into organic smoke (such as smoke generated from wood and plastic combustion), inorganic smoke (such as smoke generated from metal smelting), etc. The transmission characteristics (such as mean transmittance, standard deviation, absorption peak position, etc.) of each smoke particle category in different spectral frequency bands are used as characteristic information and stored in the smoke composition characteristic library. At the same time, information such as the source of the sample, collection time, and environmental conditions is associated to facilitate subsequent query and analysis.

[0067] Select the transmission characteristics that are representative and discriminatory for the smoke particle categories from the smoke composition characteristic library. For example, select the transmittance in the spectral frequency band with a large difference between different smoke categories as the characteristic.

[0068] Arrange the selected characteristics in a certain order to construct a smoke particle composition characteristic vector. Each element of the vector corresponds to a specific transmission characteristic, and the dimension of the vector depends on the number of selected characteristics. For example, if the transmittance of 10 spectral frequency bands is selected as the characteristic, the dimension of the smoke particle composition characteristic vector is 10.

[0069] Regarding the transmission characteristics, smoke will significantly reduce the transmittance of visible light, but has less impact on the transmittance of certain near-infrared bands.

[0070] By analyzing the correlation between the smoke particle characteristic vector and the spectral characteristic vector, each spectral frequency band for effective identification is screened to determine the optimal spectral frequency band combination.

[0071] Specifically: Calculate the correlation coefficient between the spectral characteristic vector and the smoke particle composition characteristic vector. For example, use methods such as the Pearson correlation coefficient to measure the degree of association between the signal intensity of each spectral frequency band and the smoke particle composition characteristics. Screen out the spectral frequency bands with a correlation coefficient higher than the preset correlation coefficient as the spectral frequency bands for effective identification because they can better reflect the composition information of the smoke particles.

[0072] According to the light changes in different scenarios, automatically select the optimal spectral band combination, which can improve the anti-interference ability and detection accuracy, and enhance the adaptability of the detector to different scenarios.

[0073] In a further embodiment, detect fire characteristics according to the optimal spectral band combination and issue a fire alarm. The content is as follows: Detect fire smoke. When abnormal changes are simultaneously detected in multiple spectral bands belonging to the optimal spectral band combination and the change trend conforms to fire characteristics, the detector issues an alarm. In this way, the false alarm problem caused by a single spectral beam can be avoided, thereby enhancing the anti-interference ability and detection reliability of the detector, and it is particularly suitable for installation in places such as computer rooms and power distribution rooms where extremely high precision is required for smoke detection.

[0074] The specific situations where the detected abnormal changes and the change trend conform to fire characteristics include: the signal intensity of some spectral bands suddenly increases, new spectral components are generated, or the proportion of the original spectral components changes.

[0075] For example, in the infrared spectral region, due to the thermal radiation generated by a fire, the intensity of infrared light in the corresponding frequency band will increase significantly because high-temperature objects in the fire can radiate a large amount of infrared light. At the same time, in some frequency bands of the visible light, the light signal intensity may decrease due to the scattering and absorption of light by smoke. For example, the tiny particles in the smoke will scatter and absorb visible light of some wavelengths, making the intensity of the visible light of this wavelength received by the detector weaker.

[0076] Under normal environmental conditions, the spectral distribution of the light source is relatively stable, but when a fire occurs, new spectral components are generated or the proportion of the original spectral components changes. For example, some specific gaseous molecules are generated during the combustion process, and these molecules will emit characteristic spectra, enabling the detector to detect signals in new spectral bands.

[0077] The dust characteristic supervision unit monitors the maze dust pollution level in the pollution compensation mode and uses the Internet of Things platform for unified early warning management.

[0078] In an embodiment, the monitoring of the maze dust pollution level in the pollution compensation mode and the use of the Internet of Things platform for unified early warning management include: determining the maze pollution level by detecting the reflection characteristics of the corresponding light surface at the bottom of the maze structure and the content ratio of dust particles.

[0079] Different pollution levels are classified according to the degree of change in the specular reflection characteristics and the proportion of dust particles. For example, the pollution levels can be divided into three levels: mild, moderate, and severe. In the case of mild pollution, there is a slight change in the specular reflection characteristics, such as the decrease in the intensity of the reflected light within a certain range (e.g., 5% - 15%), and the proportion of dust particles is relatively low (e.g., 10% - 30%); in the case of moderate pollution, the decrease in the intensity of the reflected light is more obvious (15% - 30%), and the proportion of dust particles increases (30% - 50%); in the case of severe pollution, the intensity of the reflected light drops significantly (exceeding 30%), and the proportion of dust particles is very high (exceeding 50%). These specific numerical ranges need to be determined based on the actual equipment performance, usage environment, and a large amount of experimental data to ensure the scientificity and rationality of the pollution level classification.

[0080] When the pollution level reaches the preset pollution level threshold, a warning for pollution accumulation in the maze structure is issued, and the alarm signal is fed back to the equipment maintenance and management terminal through the Internet of Things platform.

[0081] The maze structure plays a crucial role in devices such as smoke detectors. It can effectively distinguish smoke from other interfering factors and ensure the accurate detection of the target by the device. When the pollution accumulates to a certain extent, it will affect the characteristics of light scattering, reflection, etc., and interfere with the judgment of the detector on smoke. Timely warning allows maintenance personnel to carry out cleaning and maintenance before the equipment performance is severely affected, thereby ensuring the continuous normal operation of the equipment.

[0082] The above content is only an example and explanation of the concept of the present invention. Those skilled in the art of this technology can make various modifications or supplements to the described specific embodiments or use similar methods for substitution, as long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they should fall within the protection scope of the present invention.

Claims

1. Multi-spectral beam anti-interference smoke detector, characterized in that, Including: Smoke A mode discrimination unit that determines the pollution compensation mode and smoke mode of the smoke scene through maze structure optimization combined with a software compensation algorithm; An adaptive spectral fire warning unit that, in the smoke mode, automatically selects the optimal spectral band combination according to the light changes in different scenes, detects fire characteristics based on the optimal spectral band combination, and issues a fire alarm; A dust characteristic supervision unit that monitors the maze dust pollution level in the pollution compensation mode and uses the Internet of Things platform for unified warning management; In the smoke mode, automatically selecting the optimal spectral band combination according to the light changes in different scenes, the content includes: Listing each spectral band category, detecting the corresponding signal intensity of each spectral band category in the smoke scene, and forming a spectral feature vector; Establishing a smoke component feature library, obtaining the transmission characteristics of smoke particles in the smoke scene for each spectral band, and generating a smoke particle characteristic vector in the smoke scene; By analyzing the correlation between the smoke particle characteristic vector and the spectral feature vector, screening out the spectral bands that can be effectively identified, and determining the optimal spectral band combination.

2. The multi-spectral beam anti-interference smoke detector according to claim 1, characterized in that, The software compensation algorithm is specifically: using the dual-transmitter and single-receiver mode combined with the smoke compensation recognition algorithm to determine the pollution compensation mode and smoke mode of the smoke scene; The dual-transmitter and single-receiver mode includes two transmitter tubes, namely LED1 and LED2, and also includes a receiver tube, namely LED3; Among them, LED1 is an IR transmitter tube, that is, an infrared transmitter tube; LED2 is an R transmitter tube, that is, a red light transmitter tube.

3. The multi-spectral beam anti-interference smoke detector according to claim 2, characterized in that The optimization of the maze structure is specifically: Adding triangular stripe traps on the inner bottom surface of the maze face shell, and the stripe traps are vertically distributed with the infrared transmitter tube IR; Adopting a concave trap in the intersection area of the three lamp tubes at the bottom of the maze to form a maze signal sensitive area.

4. The multi-spectral beam anti-interference smoke detector according to any one of claims 1-3, characterized in that, Determining the pollution compensation mode and smoke mode of the smoke scene through maze structure optimization combined with a software compensation algorithm, the content includes: S01. Obtain the power-on state of the smoke alarm, including the power-on state and the normal standby state; the power-on state means that the smoke detector detects smoke or dust components, and then starts to power on to detect the fire smoke; S02. In the power-on state, LED1 and LED2 simultaneously accelerate the smoke signal acquisition; S02-1. When the maze structure does not recognize a smoke alarm or a sensor failure, determine whether the continuous detection times of LED1, LED2, and LED3 exceed a preset number of times; S02-1.

1. When the continuous detection times of LED1, LED2, and LED3 exceed the preset number of times, determine whether the fluctuation value of the continuously collected signals of LED1, LED2, and LED3 in the preset number of times is less than or equal to the preset fluctuation value. When it is less than or equal to the preset fluctuation value, it is determined as the pollution compensation mode; In the pollution compensation mode, take the average value of all sampling signals of LED1 and LED2 within the preset number of times as the clean compensation values of LED1 and LED2, and then enter the next round of compensation; S02-1.

2. When the fluctuation values of the continuously collected signals of LED1, LED2, and LED3 exceed the preset fluctuation value within the preset number of times, it is determined as the smoke mode. At the same time, it is judged whether the signal fluctuation value of LED1 or LED2 during the preset sampling process exceeds the preset fluctuation value. If it exceeds, the continuous sampling times of LED1, LED2, and LED3 are all cleared to 0; S02-2. If a smoke alarm occurs, no cleanliness value compensation is performed during the alarm period, and a new round of cleanliness value compensation starts after the alarm ends; If a maze fault occurs, regardless of whether it meets the pollution compensation conditions, the sampled values received during the fault period are not used to update the cleanliness values of LED1 or LED2. After the fault is eliminated, a new round of cleanliness value compensation starts; S03. In the normal standby state, the detection of smoke signal collection is not accelerated.

5. The multi-spectral beam anti-interference smoke detector according to claim 4, characterized in that, The preset number of times for the continuous detection times of LED1, LED2, and LED3 to exceed the preset number of times is moderately adjusted differently according to the actual smoke scenario. The content includes: Identify the corresponding size difference characteristics of smoke particles and dust particles through the built-in sensor in the maze structure, and determine the content ratio of dust particles and smoke particles; Detect the light source of the smoke scenario through sensors integrating different spectral bands in the detector to generate signal intensities of different spectral bands, and determine the light source complexity by calculating the standard deviation of the signal intensities of different spectral bands; Based on the empirical formula, according to the content ratio of the smoke particles and the light source complexity, determine the preset number of times for accelerating the smoke signal collection at the smoke alarm point.

6. The multi-spectral beam anti-interference smoke detector according to claim 4, wherein The content of S02-1.1 includes: After the product is powered on, LED1 emits light wave 1 once every 1 second, and at the same time, LED2 also emits light wave 2 once every 1 second, and light wave 1 and light wave 2 have different wavelengths; Taking the continuous detection of LED1 and LED2 for the corresponding preset number of times as the total acceleration acquisition period, and dividing the total acceleration acquisition period into a first compensation period and a second compensation period. The first compensation period and the second compensation period respectively contain the corresponding number of sampling times, both sampling once per second; If the signal fluctuations of LED1 and LED2 in the first compensation period do not exceed the preset fluctuation value, then take the average value of the collected signals corresponding to all the sampling times in the first compensation period as the new cleanliness value of LED1 or LED2, replacing the original cleanliness value of LED1 or LED2; After the first compensation period, enter the second compensation period, and enter the normal standby after the total acceleration acquisition period; If during normal standby, it is detected that the fluctuation values of the continuously collected signals of LED1 and LED2 within the preset number of times are less than or equal to the preset fluctuation value, then enter the pollution compensation mode, and at the same time update the average value of all the samplings of LED1 and LED2 in the first compensation period under the normal standby acquisition period as the new cleanliness value of LED1 or LED2.

7. The multi-spectral beam anti-interference smoke detector according to claim 6, characterized in that, The simultaneous acceleration of smoke signal collection by LED1 and LED2 includes two conditions for entering the acceleration acquisition mode: One situation: during the corresponding sampling process of the total acceleration acquisition period, if the signal fluctuation value of LED1 or LED2 exceeds the preset fluctuation value during a certain sampling process, then enter the acceleration acquisition mode; Another situation: During normal standby, if the signal fluctuation value of LED1 or LED2 exceeds the preset fluctuation value during a certain sampling process, the accelerated acquisition mode is entered.

8. The multi-spectral beam anti-interference smoke detector according to claim 1, characterized in that, Detecting fire characteristics according to the optimal spectral band combination and issuing a fire alarm, which includes: Detecting fire smoke. When abnormal changes are simultaneously detected in multiple spectral bands belonging to the optimal spectral band combination and the change trend conforms to the fire characteristics, the detector issues an alarm.

9. The multi-spectral beam anti-interference smoke detector according to claim 5, characterized in that, Monitoring the maze dust pollution level in the pollution compensation mode and using the Internet of Things platform for unified early warning management, which includes: Determining the maze pollution level by detecting the reflection characteristics of the corresponding smooth surface at the bottom of the maze structure and the content ratio of dust particles; When the pollution level reaches the preset pollution level threshold, a warning of maze structure pollution accumulation is issued, and the alarm signal is fed back to the equipment maintenance management end through the Internet of Things platform.

Images