Multispectral light beam anti-interference smoke detector

Abstract

The invention belongs to the technical field of dust smoke distinguishing detection, and particularly discloses and provides a multispectral light beam anti-interference smoke detector, which is characterized in that smoke and dust characteristics are distinguished by matching a double-transmitting single-receiving mode with a smoke compensation recognition algorithm, and different characteristics are processed differently, so that the product reliability is improved, the economic benefit of product maintenance is improved, and the product quality is improved. Meanwhile, by optimizing the labyrinth structure, false alarms caused by dust are reduced; according to light changes of different scenes, the optimal spectrum frequency band combination is automatically selected, the anti-interference capability and the detection precision are improved, the adaptability of the detector to different scenes is enhanced, and meanwhile, the fire smoke is detected by adopting the multi-spectrum light beam smoke detector, so that the detection efficiency is improved. When a plurality of spectrum frequency bands simultaneously detect abnormal changes and the change trend accords with fire characteristics, the detector gives an alarm, so that the problem of false alarm caused by a single spectrum light beam is avoided, and the anti-interference capability and the detection reliability of the detector are enhanced.

Description

Technical Field

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

[0002] Fire is a disaster that seriously threatens the safety of life and property. Timely and accurate detection of fire smoke is crucial to the prevention and control of fire. In various places, such as commercial buildings, industrial plants, and residential buildings, reliable fire detection equipment is needed to ensure safety. Traditional smoke detectors can play a role in fire warning to a certain extent, but with the increasing complexity of the environment and the continuous improvement of the requirements for fire detection accuracy, their limitations are gradually emerging.

[0003] There are various interference factors in the modern environment, such as dust, fog, different lighting conditions, etc. 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 can cause traditional detectors to misjudge it as smoke, thereby issuing unnecessary alarms; in places with different light intensities and spectral distributions, single-spectrum detectors may not be able to accurately identify fire smoke, reducing the reliability of detection.

[0004] In order to meet 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. Multi-spectral beam smoke detection technology has emerged. It uses information from multiple spectral bands for comprehensive analysis, can more accurately identify fire smoke, reduce the impact of interference factors, and provide a more effective solution for fire detection.

[0005] In the prior art, there are some solutions related to the detection of dust and smoke. For example, the Chinese patent publication number CN118840820A discloses a multi-parameter dual smoke chamber smoke detector system, which has two smoke chambers, smoke chamber A and smoke chamber B, on its upper cover. By using transmitting tubes and receiving tubes with different spectra to obtain multiple parameter information, the detection accuracy is improved. One side of the smoke chamber A and the smoke chamber B is provided with a non-smoke shielding plate that effectively blocks non-smoke particles entering the smoke chamber from different directions but allows real smoke to flow freely. The smoke chamber A and the smoke chamber B are both designed in a trapezoidal shape with a top diameter smaller than the bottom diameter of the smoke chamber, and the smoke inlets of the smoke chamber A and the smoke chamber B are inclined. The dual smoke chamber structure is combined with the non-smoke shielding plate and the inclined design to effectively reduce false alarms and missed alarms. The detector analyzes and processes the relationship between the parameters by synchronously controlling and coordinating the two sets of transmitting and receiving modules, and matches them with the database of known smoke characteristics to determine whether there is a fire, thereby improving the ability to prevent false alarms.

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

[0007] Although the above schemes have proposed some solutions for distinguishing dust and smoke detection, there are still some limitations. Specifically, it is difficult to automatically adjust the detection parameters according to the light changes in different scenes. Under different light intensities and spectral distributions, the performance of the detector may be greatly affected, and it is impossible to guarantee accurate detection of fire smoke in various scenes. For example, in strong light outdoors or weak light indoors, the sensitivity and accuracy of the detector may decrease, which cannot meet the needs of practical applications.

[0008] On the other hand, only using red and blue light with different frequencies and wavelengths for detection, the spectral dimension is relatively single. In complex environments, due to the lack of multi-dimensional spectral information, it may be more susceptible to interference substances with similar red and blue light scattering characteristics, leading to misjudgment. Summary of the invention

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

[0010] The purpose 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 pattern discrimination unit, which judges the pollution compensation mode and smoke mode of the smoke scene through maze structure optimization and software compensation algorithm.

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

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

[0013] Compared with the prior art, the present invention has the following beneficial effects: (1) The present invention can automatically select the optimal spectral frequency band combination according to the light changes in different scenes, thereby improving the anti-interference ability and detection accuracy, and enhancing the adaptability of the detector to different scenes. At the same time, by using multiple spectral frequency bands to detect fire smoke, the detector will only sound an alarm when multiple spectral frequency bands detect abnormal changes at the same time and the change trend conforms to the fire characteristics, avoiding the false alarm problem 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-transmitter and single-receiver mode with a smoke compensation recognition algorithm to distinguish the characteristics of smoke and dust, and performs different treatments on different characteristics to improve product reliability and the economic benefits of product maintenance.

[0015] (3) The present invention reduces false alarms caused by dust by optimizing the maze structure. At the same time, since pollution compensation has an upper limit, when the dust holding capacity increases, compensation can be delayed, thereby extending the life of the maze. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0017] Figure 1 It is a schematic diagram of 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 flow of the smoke mode in the smoke scene of the present invention.

[0020] Figure numerals: 01, surface shell, 02, bottom shell, 03, red light emitting tube, 04, infrared emitting tube, 05, receiving tube. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] See also Figure 1As shown, the present invention provides a multi-spectral beam anti-interference smoke detector, including a smoke pattern discrimination unit, an adaptive spectrum fire warning unit and a dust characteristic supervision unit. The smoke pattern discrimination unit, the adaptive spectrum fire warning unit and the dust characteristic supervision unit are connected in sequence.

[0023] See also Figure 2 As shown, the smoke mode determination unit determines the pollution compensation mode and smoke mode of the smoke scene by optimizing the maze structure and coordinating the software compensation algorithm.

[0024] A maze is a housing or channel design around the sensor that optimizes light scattering for better smoke detection.

[0025] In an implementation manner, the software compensation algorithm specifically includes: using a dual-transmitting and single-receiving mode in combination with a smoke compensation recognition algorithm to determine a pollution compensation mode and a smoke mode of a smoke scene.

[0026] The dual-transmit and single-receive mode includes two transmitting tubes, namely LED1 and LED2, and also includes a receiving tube 05, namely LED3.

[0027] Among them, LED1 is an IR transmitting tube, that is, an infrared transmitting tube 04; LED2 is an R transmitting tube, that is, a red light transmitting tube 03; a receiving tube 05 can use a broadband infrared receiving tube, which has a broadband receiving characteristic and is used to receive infrared signals of different frequencies.

[0028] Smoke particles of different sizes and compositions have different scattering and absorption characteristics for infrared light of different frequencies. When infrared light of different frequencies emitted by infrared transmitting tube 04 enters the smoke scene, the smoke will scatter and absorb the infrared light, causing the light intensity and frequency components received by receiving tube 05 to change. Receiver tube 05 receives and converts these changing light signals to obtain the particle state of the smoke scene.

[0029] In a further embodiment, the maze structure is optimized, specifically: Figure 2 As shown, a triangular stripe trap is added to the bottom surface of the face shell 01 that matches the labyrinth bottom shell 02. The stripe trap is vertically distributed with the infrared emitting tube 04, namely IR. The stripe trap can enhance the ability to resist dust accumulation, effectively increase the service life of the labyrinth, and reduce product false alarms. At the same time, the triangular stripe trap makes the bottom surface rough, and the light of the emitting tube is scattered more evenly. A concave trap is used at the intersection of the three lamp tubes in the labyrinth bottom shell to form a labyrinth signal sensitive area.

[0030] The present invention reduces false alarms caused by dust by optimizing the structure of the maze (smoke sensor). At the same time, since pollution compensation has an upper limit, when the dust holding capacity increases, compensation can be delayed, thereby extending the life of the maze.

[0031] See also Figure 3 As shown, in a further embodiment, the pollution compensation mode and smoke mode of the smoke scene are determined by maze structure optimization in conjunction with a software compensation algorithm, including: S01, obtaining 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 fire smoke.

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

[0034] S02-1. When the maze structure does not identify a smoke alarm or a sensor failure, determine whether the number of consecutive detections of LED1 and LED2 exceeds a preset number.

[0035] S02-1.1. When the number of consecutive detections of LED1 and LED2 exceeds the preset number, determine whether the fluctuation value of the consecutively collected signals of LED1 and LED2 in the preset number of times is less than or equal to the preset fluctuation value. If it is less than or equal to the preset fluctuation value, determine it 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 taken as the clean compensation values ​​of LED1 and LED2, and then the next round of compensation is entered.

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

[0038] S02-2. If a smoke alarm occurs, no cleanliness value compensation will be performed during the alarm period, and a new round of cleanliness value compensation will begin after the alarm ends.

[0039] If a maze (sensor) fault occurs, such as the collected LED1 value or LED2 value is lower than the preset fault value, it is judged as a maze fault. Regardless of whether the pollution compensation conditions are met, the sampled values ​​received during the fault period are not used to update the clean values ​​of LED1 or LED2. A new round of clean value compensation will start after the fault is eliminated.

[0040] S03. In normal standby mode, smoke signal collection is not accelerated.

[0041] In a further embodiment, whether the number of consecutive detections of LED1 and LED2 exceeds a preset number is adjusted appropriately according to the actual smoke scene, including: identifying the corresponding size difference characteristics of smoke particles and dust particles through the built-in sensor of the maze structure, and determining the content ratio of smoke particles and dust particles.

[0042] The light sources such as the infrared transmitting tube 04 set 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, smoke particles and dust particles may have different scattering intensities for red light and infrared light due to different compositions, shapes and size distributions. By analyzing the scattered signal intensity 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 the proportion of each content based on the relationship between the signal intensity and the particle concentration.

[0043] By integrating sensors with different spectral frequency bands in the detector to detect the light source of the smoke scene, the signal strength of different spectral frequency bands is generated, and the light source complexity is determined by calculating the standard deviation of the signal strength of different spectral frequency bands. The larger the standard deviation, the greater the difference in signal strength between each frequency band, and the higher the complexity of the light source. For example, for a set of signal strength data from different spectral frequency bands, calculate its standard deviation. If the standard deviation is large, it means that the light signal intensity distribution of these frequency bands is relatively dispersed, and the light source may contain multiple spectral components of different intensities, which is more complex.

[0044] The sensor integrating different spectral frequency bands is a sensor used for receiving and detecting light signals of specific wavelengths.

[0045] The light sources of different spectral frequency bands include visible light, infrared light, ultraviolet light, etc. For example, in some complex environment monitoring scenes, mixed light produced by multiple different luminous mechanisms or multiple different luminous substances may be included at the same time, reflecting the complexity of the light source components. For example, in a smoke scene, if the light source is not only the infrared light emitted by the infrared transmitting tube 04, but also other light sources in the surrounding environment such as scattered or reflected light from natural light, lamplight, etc., then the spectral frequency bands received by the receiving tube 05 during smoke detection will increase, which will be affected by other light sources and reduce the detection accuracy.

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

[0047] Specifically, when the proportion of smoke particles exceeds the proportion of dust particles, the alarm point and the alarm detection count will not change, and a normal alarm will be given; when the proportion of dust particles exceeds the proportion of smoke particles, the alarm point (updated cleanliness values ​​of LED1 and LED2) will be increased to improve the product's ability to adapt to environmental changes and to improve the product's ability to resist dust.

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

[0049] The total accelerated acquisition period is defined as the preset number of consecutive detections of LED1 and LED2, and is divided into a first compensation period and a second compensation period. The first compensation period and the second compensation period each include a corresponding number of samplings, both of which are sampled once per second.

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

[0051] After the first compensation cycle, the second compensation cycle is entered, and after the total accelerated acquisition cycle, the normal standby mode is entered.

[0052] If in normal standby mode, it is detected that the fluctuation value of the continuous acquisition signal of LED1 and LED2 in the preset number of times is less than or equal to the preset fluctuation value, the pollution compensation mode is entered, and the average value of all corresponding samples of LED1 and LED2 in the first compensation cycle under the normal standby acquisition cycle is updated and used as the new clean value of LED1 or LED2.

[0053] Exemplarily, LED1 and LED2 are both emitted 30 times, that is, the total accelerated acquisition cycle is 30 samples, of which the first round of compensation is sampled 26 times in total, and the sample is sampled once per second; if the signal fluctuations of LED1 and LED2 do not exceed the preset fluctuation value (WAV_LMT), the average value of the 26 collected signals is used as the new clean value of LED1 or LED2, replacing the original clean value of LED1 or LED2. Specifically, after averaging the 26 collected and recorded LED1 sampling signal values, the average value is used as the new clean value of LED1; after averaging the 26 collected and recorded LED2 sampling signal values, the average value is used as the new clean value of LED2.

[0054] After the first round of 26 compensation samplings, there are 4 accelerated samplings, also sampled once every 1 second, thus forming a 30-second sampling cycle with the previous 26 samplings, and there is no other effect. Among them, the first round of compensation sampling and the 4 accelerated samplings are defined as accelerated acquisition. The reason for being defined as acceleration is that the sampling interval will be longer during standby.

[0055] After 30 times of accelerated collection, it enters normal standby mode. During normal standby mode, the collection cycle is: LED1 is sampled once every 8 seconds, and LED2 is sampled once every 40 seconds, which can save power consumption without affecting smoke detection.

[0056] If in normal standby mode, the condition that "the fluctuation value of the LED1 and LED2 continuously collected signals in the preset number of times is less than or equal to the preset fluctuation value" is met, the pollution compensation mode will be entered, and the average value of 26 LED1 samplings and 26 LED2 samplings in the normal standby collection cycle will be updated and used as the new clean value of LED1 or LED2.

[0057] In a further embodiment, the simultaneous accelerated smoke signal acquisition of LED1 and LED2 includes two conditions for entering the accelerated acquisition mode: one situation: during the sampling process corresponding to the total accelerated acquisition cycle, if the LED1 or LED2 signal fluctuation value exceeds the preset fluctuation value during a certain sampling process, the accelerated acquisition mode is entered: Action 1, the number of consecutive sampling times of LED1 and LED2 are cleared to 0 and the sampling is re-counted from the first time, and the sampling values ​​of LED1 and LED2 previously sampled and held are cleared to zero; Action 2, subsequent accelerated sampling, LED1 and LED2 are sampled once every 1 second, and are re-sampled according to the sampling times corresponding to the first compensation cycle, and then the clean values ​​of LED1 and LED2 are updated.

[0058] Another situation: in normal standby mode, if the LED1 or LED2 signal fluctuation value exceeds the preset fluctuation value during a certain sampling process, the accelerated acquisition mode is entered: Action 1, the continuous sampling times of LED1 and LED2 are cleared to 0 and the sampling is re-counted from the first time, and the sampling values ​​of LED1 and LED2 previously sampled and held are cleared to zero; Action 2, the sampling is then accelerated, LED1 and LED2 are sampled once every 1 second, and re-sampled according to the corresponding sampling times of the first compensation cycle, and then the clean values ​​of LED1 and LED2 are updated.

[0059] For example, when the 7th sampling is performed, if the fluctuation value of the LED1 or LED2 signal exceeds the preset fluctuation value during the sampling process, the accelerated acquisition mode will be entered, and the 8 data (the sampling values ​​of LED1 and LED2) recorded in the first 8 sampling records will be discarded, and the sampling times will be reset. The next 22 accelerated samplings are also sampled once every 1 second, thus forming a 30-second sampling cycle with the previous 8 samplings, making the sampling cycle consistent; of course, if one of the 22 accelerated samplings meets the conditions for entering the accelerated acquisition mode, the accelerated acquisition mode will also be entered.

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

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

[0062] In an implementation manner, in the smoke mode, the optimal spectral frequency band combination is automatically selected according to the light changes in different scenes, including: listing each spectral frequency band category, detecting the corresponding signal intensity of each spectral frequency band category in the smoke scene, and forming a spectral feature vector, wherein each spectral frequency band category includes visible light, infrared light, ultraviolet light and other categories.

[0063] The signal strengths of each spectral frequency band collected are arranged in a certain order to form a spectral feature vector. For example, if the signal strengths of 10 different spectral frequency bands are measured, the spectral feature vector is a 10-dimensional vector, and each element of the vector corresponds to the signal strength value of a spectral frequency band.

[0064] A smoke component feature library is established to obtain the transmission characteristics of smoke particles in smoke scenes for each spectral band, and generate smoke particle feature vectors in smoke scenes.

[0065] Specifically, in different fire scenes (such as wood burning, plastic burning, grease burning, etc.), industrial production scenes (such as smoke generated by smelting, chemical industry, etc.) and natural environment scenes (such as forest fires, volcanic eruptions, etc.), use professional smoke collection equipment to collect smoke samples. During the collection process, pay attention to recording the source of the samples, collection time, and environmental conditions (temperature, humidity, air pressure, etc.). Use spectrometers and other equipment to perform spectral analysis on the collected smoke samples and measure the transmitted light intensity of each smoke sample in different spectral bands (ultraviolet light, visible light, infrared light and other subdivided bands). The spectrometer should have high resolution and a wide spectral range to ensure that the spectral information of each band can be accurately obtained. At the same time, set appropriate measurement parameters (such as integration time, number of scans, etc.) to ensure the accuracy and reliability of the measurement results.

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

[0067] Select a transmittance characteristic feature that is representative and discriminative of the smoke particle category from the smoke component feature library. For example, select the transmittance of a spectral frequency band that has a large difference between different smoke categories as a feature.

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

[0069] Regarding the transmittance characteristics, smoke significantly reduces the transmittance of visible light, but has little effect on the transmittance in some near-infrared bands.

[0070] By analyzing the correlation between the smoke particle characteristic vector and the spectral characteristic vector, the effectively identified spectral frequency bands are screened and the optimal spectral frequency band combination is determined.

[0071] Specifically, the correlation coefficient between the spectral feature vector and the smoke particle component characteristic vector is calculated, for example, the correlation degree between the signal strength of each spectral frequency band and the smoke particle component characteristics is measured using methods such as the Pearson correlation coefficient. Spectral frequency bands with correlation coefficients higher than the preset correlation coefficient are selected as effective spectral frequency bands for identification, because they can better reflect the composition information of smoke particles.

[0072] Automatically selecting the optimal spectral frequency band combination based on the changes in light in different scenes can improve anti-interference ability and detection accuracy, and enhance the adaptability of the detector to different scenes.

[0073] In a further embodiment, the fire characteristics are detected according to the optimal spectral frequency band combination, and a fire alarm is issued. The content is: fire smoke is detected, and when multiple spectral frequency bands belonging to the optimal spectral frequency band combination detect abnormal changes at the same time and the change trend is consistent with the fire characteristics, the detector issues an alarm. In this way, the false alarm problem caused by a single spectral light beam can be avoided, thereby enhancing the anti-interference ability and detection reliability of the detector. It is particularly suitable for installation in places such as computer rooms and distribution rooms that have extremely high requirements for smoke detection accuracy.

[0074] The specific circumstances in which abnormal changes are detected and the trend of changes conforms to the characteristics of fire include: the signal intensity of certain spectral frequency bands suddenly increases, new spectral components are generated, or the ratio of original spectral components changes.

[0075] For example, in the infrared spectrum, due to the heat radiation generated by the fire, the intensity of the infrared light in the corresponding frequency band will increase significantly, because the high-temperature objects in the fire can radiate a large amount of infrared light. At the same time, in some visible light bands, 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 some wavelengths of visible light, making the intensity of the visible light of that wavelength received by the detector weaker.

[0076] Under normal circumstances, the spectral distribution of the light source is relatively stable, but when a fire occurs, new spectral components will be generated or the proportion of the original spectral components will change. For example, some specific gaseous molecules will be generated during the combustion process, and these molecules will emit characteristic spectra, allowing the detector to detect new spectral frequency band signals.

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

[0078] In an implementation manner, the dust pollution level of the maze is monitored in the pollution compensation mode, and unified early warning management is performed using the Internet of Things platform, including: determining the maze pollution level by detecting the corresponding light surface reflection characteristics at the bottom of the maze structure and the content ratio of dust particles.

[0079] Different pollution levels are divided according to the degree of change in the light surface reflection characteristics and the proportion of dust particles. For example, the pollution level can be divided into three levels: mild, moderate and severe. In mild pollution, the light surface reflection characteristics change slightly, such as the reflected light intensity decreases within a certain range (such as 5%-15%), and the dust particle content is relatively low (such as 10%-30%); in moderate pollution, the reflected light intensity decreases more significantly (15%-30%), and the dust particle content increases (30%-50%); in severe pollution, the reflected light intensity decreases significantly (more than 30%), and the dust particle content is very high (more than 50%). These specific numerical ranges need to be determined based on the actual equipment performance, use 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 maze structure pollution accumulation warning is issued, and the alarm signal is fed back to the equipment maintenance management end through the Internet of Things platform.

[0081] The maze structure plays a key role in devices such as smoke detectors. It can effectively distinguish smoke from other interference factors and ensure that the device accurately detects the target. When pollution accumulates to a certain extent, it will affect the scattering and reflection characteristics of light, interfering with the detector's judgment of smoke. Timely warning allows maintenance personnel to clean and maintain the equipment before the performance is seriously affected, thereby ensuring the continued normal operation of the equipment.

[0082] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they shall all fall within the protection scope of the present invention.

Claims

1. Multi-spectral beam anti-interference smoke detector, characterized in that: include: smoke The mode discrimination unit determines the pollution compensation mode and smoke mode of the smoke scene through maze structure optimization and software compensation algorithm; The adaptive spectrum fire warning unit, in smoke mode, automatically selects the optimal spectrum frequency band combination according to the light changes in different scenes, detects fire characteristics based on the optimal spectrum frequency band combination, and issues a fire alarm; The dust characteristic monitoring unit monitors the dust pollution level of the maze in the pollution compensation mode and uses the Internet of Things platform for unified early warning management.

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

3. The multi-spectral beam anti-interference smoke detector according to claim 2, characterized in that: The maze structure optimization is specifically as follows: Add a triangular stripe trap on the inner bottom surface of the maze shell, and the stripe trap is vertically distributed with the infrared transmitting tube IR; A concave trap is used at the intersection of the three light tubes at the bottom shell 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 to 3, characterized in that: The pollution compensation mode and smoke mode of the smoke scene are determined by the maze structure optimization and software compensation algorithm, including: S01, obtaining the power-on status of the smoke alarm, including the power-on status and the normal standby status; S02, in the power-on state, LED1 and LED2 simultaneously accelerate the collection of smoke signals; S02-1, when the maze structure does not identify the smoke alarm or sensor failure, determine whether the number of consecutive detections of LED1 and LED2 exceeds the preset number; S02-1.

1. When the number of consecutive detections of LED1 and LED2 exceeds the preset number, it is determined whether 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, and when it is less than or equal to the preset fluctuation value, it is determined to be in the pollution compensation mode; In the pollution compensation mode, the average values ​​of all sampling signals within the preset times of LED1 and LED2 are taken as the clean compensation values ​​of LED1 and LED2, and then the next round of compensation is started; S02-1.

2. When the fluctuation value of the LED1 and LED2 continuous sampling signals does not exceed the preset fluctuation value in the preset number of times, it is determined to be the smoke mode. At the same time, it is determined whether the fluctuation value of the LED1 or LED2 signal exceeds the preset fluctuation value during the preset sampling process. If it does not exceed the preset fluctuation value, the number of continuous sampling of LED1 and LED2 is cleared to 0; S02-2. If a smoke alarm occurs, no cleanliness value compensation will be performed during the alarm period, and a new round of cleanliness value compensation will begin after the alarm ends; If a maze fault occurs, regardless of whether the pollution compensation conditions are met, the sampled values ​​received during the fault period are not used to update the clean values ​​of LED1 or LED2. A new round of clean value compensation begins after the fault is eliminated; S03. In normal standby mode, smoke signal collection is not accelerated.

5. The multi-spectral beam anti-interference smoke detector according to claim 4, characterized in that: Whether the number of consecutive detections of LED1 and LED2 exceeds the preset number is appropriately adjusted according to the actual smoke scene, including: The built-in sensor in the maze structure identifies the size differences of smoke particles and dust particles, and determines the proportion of dust particles and smoke particles; By integrating sensors with different spectral frequency bands in the detector, the light source of the smoke scene is detected, the signal strength of different spectral frequency bands is generated, and the light source complexity is determined by calculating the standard deviation of the signal strength of different spectral frequency bands; Based on the empirical formula, according to the content ratio of the smoke particles and the complexity of the light source, the preset number of times for accelerating the collection of smoke signals at the smoke alarm point is determined.

6. The multi-spectral beam anti-interference smoke detector according to claim 4, characterized in that: The contents of S02-1.1 include: After the product is powered on, LED1 emits light wave 1 once every 1 second, and LED2 also emits light wave 2 once every 1 second, and light wave 1 and light wave 2 have different wavelengths; The total accelerated acquisition period is defined as the number of consecutive detections of LED1 and LED2 corresponding to the preset number of times, and the total accelerated acquisition period is divided into a first compensation period and a second compensation period, wherein the first compensation period and the second compensation period respectively include corresponding sampling times, both of which are sampled once per second; If the signal fluctuations of LED1 and LED2 in the first compensation cycle do not exceed the preset fluctuation value, the average value of the corresponding collected signals of all the collection times in the first compensation cycle is used as the new clean value of LED1 or LED2 to replace the original clean value of LED1 or LED2; After the first compensation cycle, the second compensation cycle is entered, and after the total accelerated acquisition cycle, the normal standby mode is entered; If in normal standby mode, it is detected that the fluctuation value of the continuous acquisition signal of LED1 and LED2 in the preset number of times is less than or equal to the preset fluctuation value, the pollution compensation mode is entered, and the average value of all corresponding samples of LED1 and LED2 in the first compensation cycle under the normal standby acquisition cycle is updated and used as the new clean value of LED1 or LED2.

7. The multi-spectral beam anti-interference smoke detector according to claim 6, characterized in that: The LED1 and LED2 simultaneously accelerate the smoke signal acquisition and include two conditions for entering the accelerated acquisition mode: One situation: In the sampling process corresponding to the total accelerated acquisition cycle, if the LED1 or LED2 signal fluctuation value exceeds the preset fluctuation value in a certain sampling process, the accelerated acquisition mode is entered: Action 1, the continuous sampling times of LED1 and LED2 are cleared to 0 and the sampling is re-counted from the first time, and the sampling values ​​of LED1 and LED2 previously sampled and held are cleared to zero; Action 2, the sampling is then accelerated, and the sampling is re-sampled according to the sampling times corresponding to the first compensation cycle, and then the clean values ​​of LED1 and LED2 are updated; Another situation: in normal standby mode, if the LED1 or LED2 signal fluctuation value exceeds the preset fluctuation value during a certain sampling process, the accelerated acquisition mode is entered: Action 1, the continuous sampling times of LED1 and LED2 are cleared to 0 and the sampling is re-counted from the first time, and the sampling values ​​of LED1 and LED2 previously sampled and held are cleared to zero; Action 2, the sampling is then accelerated, and the sampling is re-sampled according to the corresponding sampling times of the first compensation cycle, and then the clean values ​​of LED1 and LED2 are updated.

8. The multi-spectral beam anti-interference smoke detector according to claim 1, characterized in that: In the smoke mode, the optimal spectral frequency band combination is automatically selected according to the light changes in different scenes, including: List the categories of each spectral frequency band, detect the corresponding signal strength of each spectral frequency band category in the smoke scene, and form a spectral feature vector; Establish a smoke component feature library, obtain the transmission characteristics of smoke particles in the smoke scene to each spectral band, and generate the smoke particle feature vector in the smoke scene; By analyzing the correlation between the smoke particle characteristic vector and the spectral characteristic vector, the effectively identified spectral frequency bands are screened and the optimal spectral frequency band combination is determined.

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

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

Images