A smoke alarm method based on smoke particle trajectory

Abstract

The present application relates to the technical field of fire detection, in particular to a smoke alarm method based on smoke particle motion trajectory, comprising: constructing a three-dimensional flow field distribution map of smoke particles, in view of the cross-sectional area change and curvature change of the current channel at the branch path, taking the predicted smoke particle position as the center, correlating the flow velocity data of each branch path, and obtaining the flow velocity gradient sequence of each branch path; based on the flow velocity gradient sequence of each branch path, when the flow velocity gradient exceeds the flow velocity threshold, updating the smoke particle position in each branch path; trajectory fitting is performed on the smoke particles, and the concentration intensity index and update time of the smoke particles in different branch paths are configured according to the concentration gradient corresponding to the motion trajectory sequence on each branch path; the classification label of each branch path is configured, and hierarchical early warning is carried out according to the flow velocity difference. The accuracy and timeliness of the smoke alarm are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fire detection, in particular to a smoke alarm method based on smoke particle motion trajectory. BACKGROUND

[0002] The existing smoke sensor usually adopts a single-point sensor to collect concentration or flow rate data, and identifies the distribution of smoke according to optical detection, without constructing a global three-dimensional flow field model matching the channel structure, so that the collected smoke concentration is mostly local discrete data, which is difficult to reflect the global flow distribution under smoke distribution and concentration dilution. After a fire accident occurs, it is impossible to complete the smoke alarm processing according to the distribution of smoke flow.

[0003] For example, Chinese Patent Publication No. CN120126271A discloses a smoke sensor labyrinth optimization design and anti-dust pollution method, which includes: detecting the fused particles and water vapor particles of smoke and dust in the air detection area through a signal emitting device, screening out water vapor particles to weaken the smoke alarm caused by false triggering of water vapor particles, and using differential signal processing technology to distinguish and identify the fused particles. At the same time, the fused particles are introduced into the labyrinth measurement cavity to measure the smoke and dust concentration in the labyrinth measurement cavity, thereby enhancing the stability of detection; automatically adjusting the labyrinth channel width in the labyrinth measurement cavity according to the smoke concentration to spatially separate the smoke particles and dust particles, thereby increasing the adaptability of the detector to complex environments; performing fire warning judgment on the smoke concentration in the labyrinth measurement cavity, and starting the cleaning device added in the labyrinth measurement cavity to perform cleaning treatment according to the smoke and dust concentration.

[0004] For example, Chinese Patent Publication No. CN117789394A discloses an early fire smoke detection method based on motion history images, which belongs to the technical field of fire monitoring and alarm. To solve the problem of defects in accuracy, reliability and real-time of fire warning, the accuracy of smoke prediction for fire can be improved through smoke feature extraction and smoke motion analysis, the fire smoke can be accurately identified through analysis of motion patterns and smoke features, the real-time of warning is improved through collection of image sequences, continuous collection of image sequences and real-time analysis can timely discover fire smoke, and the reliability is high. Through analysis of the motion history images of the smoke, the problems of false alarm and missed alarm of the sensor can be avoided, and the scope of application is wide. Through smoke motion analysis and fire smoke detection, different monitoring scenes can be combined with different influencing factors.

[0005] The prior art determines the reflection intensity under the combination of humidity, smoke particles and dust particles by optical refraction, and then determines the smoke particles and smoke concentration in a maze scene; and determines the smoke aggregation condition when a fire occurs by pixel aggregation of an open fire image; but these processing methods ignore the flow field when the smoke diffuses, resulting in low smoke distribution recognition accuracy at some positions, particle tracking trajectory disorder and other conditions in a maze scene due to different distribution of multi-branch flow gradients, which reduces the accuracy of smoke alarm. SUMMARY

[0006] To solve the above technical problems, the technical scheme adopted by the present application is: a smoke alarm method based on smoke particle motion trajectory, comprising: based on the installation coordinates of each sensor in the current channel structure, constructing a three-dimensional flow field distribution diagram of smoke particles.

[0007] For the cross-sectional area change and curvature change of the current channel at the branch path, the flow velocity gradient sequence of each branch path is obtained by associating the flow velocity data of each branch path with the predicted smoke particle position as the center.

[0008] Based on the flow velocity gradient sequence of each branch path, when the flow velocity gradient exceeds the flow velocity threshold, the smoke particle position in each branch path is updated, and the positioning change of the smoke particle on the branch path at each time is output.

[0009] The trajectory fitting of the smoke particles forms the corresponding motion trajectory sequence of the smoke particles, and the concentration intensity index and update time of the smoke particles in different branch paths are configured according to the concentration gradient corresponding to the motion trajectory sequence on each branch path.

[0010] Based on the update time and concentration intensity index of different branch paths, the classification labels of each branch path are configured, and hierarchical early warning is performed according to the flow velocity difference, to obtain the monitoring results of each branch path.

[0011] The beneficial effects of the present application are as follows: firstly, the present application obtains the flow field basic data of each branch path, such as channel type, initial position of smoke particles, air flow rate, cross-sectional area and curvature, synchronizes the data to the sensor coordinates, generates a three-dimensional flow field distribution map through multi-view reconstruction, then determines the particle position deviation by verifying the turbulent effect, takes the branch cross-sectional area and curvature as input geometric parameters, calculates the correlation between the position deviation and the particle position by using the Pearson correlation coefficient, matches the optimal geometric parameters, synchronizes the position deviation to the geometric parameters through particle random arrangement and concentration point superposition processing, and finally fits the flow velocity of each position to construct a flow velocity gradient sequence. The geometric structure and the coordinates of the smoke particles under each branch are determined, the position deviation caused by the turbulent effect is corrected through correlation matching and particle superposition processing, and the flow velocity gradient sequence can better reflect the dynamic distribution of the flow velocity of each branch, so that the constructed three-dimensional flow field can serve as the data benchmark for current smoke tracking, thereby improving the accuracy of flow velocity identification in multiple scenarios.

[0012] Secondly, the present application extracts vorticity and strain rate as flow velocity characteristics, divides the flow velocity rising / dropping double threshold, captures the particle cross-section movement according to the branch geometric shape, triggers position update combined with flow velocity gradient change, to complete the coordinate tracking of smoke particles, and introduces the flow velocity change to avoid the drift and lag of smoke positioning, thereby providing a position update basis for the correlation between subsequent concentration and trajectory.

[0013] Thirdly, the present application timestamps the trajectory points of the motion trajectory sequence and the concentration values, calculates the concentration gradient of each trajectory point and determines the local concentration gradient of the branch, takes the ratio of the maximum value of the local concentration gradient to the pre-warning concentration gradient as a concentration intensity index, correlates the particle update time to determine the data update process in the global scene, then generates a semantic classification label according to the update time and the concentration intensity index, updates the label in real time and judges the change, and finally obtains the output monitoring result. The flow velocity tracing processing of each branch under the position guidance is realized, and the accuracy and timeliness of smoke alarm are improved. BRIEF DESCRIPTION OF DRAWINGS

[0014] The present application will be further described below in combination with the drawings and examples.

[0015] Figure 1 It is a flowchart of a smoke alarm method based on smoke particle motion trajectory.

[0016] Figure 2 It is a flowchart of step S2 of a smoke alarm method based on smoke particle motion trajectory.

[0017] Figure 3 It is a flowchart of step S3 of a smoke alarm method based on smoke particle motion trajectory.

[0018] Figure 4 is a flowchart of step S4 of the smoke alarm method based on the movement trajectory of smoke particles.

[0019] Figure 5 is a flowchart of step S5 of the smoke alarm method based on the movement trajectory of smoke particles. DETAILED DESCRIPTION

[0020] Embodiments of the present application are described in detail below. The embodiments described below are exemplary only, and are not to be construed as limiting the present application. Where specific technical or conditions are not mentioned in the embodiments, the techniques or conditions described in the literature in the art or according to the product instructions are used.

[0021] Reference Figure 1 A smoke alarm method based on the movement trajectory of smoke particles includes: S1, constructing a three-dimensional flow field distribution map of smoke particles based on the installation coordinates of each sensor in the current channel structure.

[0022] S2, for the cross-sectional area change and curvature change of the current channel at the branch path, centering on the predicted smoke particle position, associating the flow rate data of each branch path, and obtaining the flow rate gradient sequence of each branch path.

[0023] S3, based on the flow rate gradient sequence of each branch path, when the flow rate gradient exceeds the flow rate threshold, updating the smoke particle position in each branch path, and outputting the positioning change of the smoke particle on the branch path at each time.

[0024] S4, trajectory fitting is performed on the smoke particles to form a corresponding movement trajectory sequence of the smoke particles, and the concentration intensity index and update time of the smoke particles in different branch paths are configured according to the concentration gradient corresponding to the movement trajectory sequence on each branch path.

[0025] S5, based on the update time and concentration intensity index of different branch paths, configuring the classification label of each branch path, and performing hierarchical early warning according to the flow rate differentiation, and obtaining the monitoring result of each branch path.

[0026] In the current scheme, smoke monitoring is performed for a labyrinth or spiral channel, the smoke diffusion conditions in each branch path or spiral channel during smoke diffusion are determined, and this part of data is synchronized to the three-dimensional flow field distribution, and finally the areas of the smoke spreading path are distinguished, and the position accuracy during smoke alarm is improved.

[0027] A three-dimensional PIV system can be used, which is a fluid mechanics measuring instrument based on the binocular vision principle. On the basis of a two-dimensional PIV system, two sets of digital devices are used to synchronously shoot an experimental region to synthesize a three-dimensional velocity vector field. The system has high-precision algorithm modules, ultra-high spatial resolution cameras, high-speed digital image storage, and other characteristics, and is widely used in the fields of low-speed to high-speed velocity field measurement, industrial fluid mechanics analysis, and micro flow field research. In addition, compared with the two-dimensional system, the system adds the ability to measure the velocity component in the depth direction, which can assist in identifying the flow of smoke particles in multiple directions.

[0028] The selected devices include a laser light source, a high-speed camera array, a tracer particle generator, and a synchronous controller to complete the construction of the three-dimensional flow field. The laser light source is used to illuminate the smoke particles to form a captureable optical signal. The high-speed camera array uses multiple cameras to achieve multi-angle three-dimensional position capture of particles. The tracer particle generator can generate 1-5 μm oil mist particles with good following properties and close to air density. The oil mist particles are released synchronously with the smoke to assist the smoke particles in precise tracking. Through smoke alarm processing of multiple flow field calculations, subsequent smoke concentration at multiple positions can be used to achieve rapid smoke tracking, and then the smoke diffusion tracking process is completed. The synchronous controller is used to synchronize the timestamps of all data.

[0029] Then, the light direction and position information are obtained by single shooting, the particle displacement in the continuous image frame is calculated, the three-dimensional velocity of the particles in the axial, radial, and circumferential directions is measured, and the uniformly distributed transmission laser concentration sensors are arranged at each branch port, pitch change, and inside and outside the pipe cross section to synchronously measure the smoke concentration, thereby realizing the construction of the three-dimensional flow field of the smoke.

[0030] By releasing tracer particles, the three-dimensional flow field raw data (particle position, flow rate), and point concentration data are recorded synchronously, and the three-dimensional flow field distribution map with a timestamp is stored.

[0031] The implementation mode of step S1 includes: S11, obtaining the channel type corresponding to each branch path in the current channel structure, and obtaining the initial position coordinates, air flow rate, cross-sectional area, and curvature of the smoke particles arranged in each branch path to obtain the flow field basic data of each branch path. By uniformly scattering a set of particles in the corresponding branch path, each particle represents a smoke particle and is marked as an initial position set, and then the initial position coordinates are obtained. The air flow rate is obtained by measuring different branch paths.

[0032] The channel type mainly aims at the labyrinth channel and the spiral channel in the current scene; the labyrinth channel refers to a tree-shaped or grid-shaped topological channel composed of multiple straight pipe sections, elbows, branch nodes, confluence nodes and dead ends, and there are a large number of optional branch paths; the spiral channel refers to a coiled channel with an axis extending in an equal-curvature or variable-curvature spiral line, and in the current scheme, the spiral channel belongs to one of the complex structure channels, and the influence of the curvature change of the axis on the smoke movement is similar to that of the multiple branch paths, so the straight line type, straight pipe type, grid type and other different forms of structures of the pipeline are presented according to the structure to indicate the processing mode of the current channel.

[0033] As for the flow rate and direction of the smoke particles, the images obtained continuously are decomposed in three directions of the axial direction, the radial direction and the circumferential direction, and the velocity values in the corresponding directions are obtained.

[0034] S12, synchronizing the flow field basic data to the coordinates of the corresponding sensor, and performing multi-view reconstruction in the coordinates of the sensor to obtain a three-dimensional flow field distribution map.

[0035] After synchronizing the flow field basic data, the data currently collected are synchronized to the coordinates under the corresponding view angle of the sensor based on the three-dimensional coordinates of the sensor configuration, so as to obtain a three-dimensional flow field distribution map.

[0036] In an embodiment of the present application, in step S2, the real-time position of the smoke particles is tracked according to the particle filtering algorithm, and the flow rate gradient in the corresponding branch channel is determined according to the turbulent effect when the air flows, so as to capture the correlation form between the particle movement and the flow rate.

[0037] As for the predicted position of the smoke particles, the initial position coordinates of the smoke particles are updated by the particle filtering method, the particle position is iterated by the particle filtering posterior probability estimation, and the predicted position of the smoke particles is obtained; then, the position deviation generated in different branch paths is obtained by subtracting the currently observed position from the predicted position.

[0038] As shown in FIG. 2, the implementation of step S2 includes the following steps. Figure 2 S21, verifying the turbulent effect of each branch path to determine the position deviation of the smoke particles in different branch paths.

[0039] S22, synchronizing the position deviation to the cross-sectional area and the curvature value of the branch path, and calculating the flow rate of each smoke particle according to the angle of the smoke particle in the axial direction, the radial direction and the circumferential direction.

[0040] S23, fitting the flow rate of the smoke particles at each position to construct a flow rate gradient sequence of each branch path.

[0041] The purpose of verifying the turbulence effect of the branch path is to ensure that the particle is arranged in the region with the most dynamic characteristics of the flow field, that is, after calculating the turbulence intensity value by the turbulence effect, it is determined whether the smoke particle currently tracked is located in the high turbulence area by the value of the turbulence intensity. The turbulence intensity threshold value can be configured, such as 0.15 or the average value of the turbulence intensity measured at the pitch change of the spiral channel or the branch of the labyrinth, and the branch path with a turbulence intensity greater than the turbulence intensity threshold value is regarded as the main verification part at present, and the related position deviation is determined.

[0042] The high turbulence area usually dominates the evolution of the overall flow field, and verifying these areas can improve the prediction accuracy; then the smoke particle located in the high turbulence area is tracked, the position deviation is synchronized to the path attribute of the corresponding branch path, such as the cross-sectional area and the curvature value, and then the flow velocity in the three-dimensional scene is determined according to the corresponding axial, radial and circumferential angles in the three-dimensional direction. After fitting multiple positions, the flow velocity gradient sequence corresponding to the smoke particle in each branch area can be known.

[0043] In step S22, the position deviation is synchronized to the cross-sectional area and the curvature value of the branch path. Since the position deviation is the basis for the current flow velocity calculation, it is necessary to correct or view the position deviation according to the cross-sectional area change and the curvature influence of the spiral channel and the spiral / branch channel in the labyrinth channel, so as to complete the correction of the branch flow velocity and the setting of the gradient sequence.

[0044] The implementation of step S22 further includes: S221, taking the cross-sectional area and the curvature value of the branch path as input, determining the difference of different branch paths.

[0045] S222, the correlation between the difference of the branch path and the position deviation of the smoke particle is obtained, and the cross-sectional area and the curvature value of the branch path are matched according to the correlation.

[0046] When matching the cross-sectional area and the curvature value of the branch path, the implementation further includes: extracting the main branch and the sub-branch of the current branch path, and connecting the branch path to form a connected branch connected with the main branch and the sub-branch.

[0047] For the cross-sectional area and the curvature value of each branch path in the connected branch, the difference value relative to the overall average value of the connected branch is set as the characteristic value corresponding to the cross-sectional area and the curvature value. Here, the difference indicates the difference ratio of the normalized cross-sectional area and curvature value compared with the overall average value, so as to illustrate the influence of different shapes of the connected multiple branches on the transmission of the smoke particle.

[0048] The position deviation and eigenvalue of each branch path in the connected branch are input, the Pearson correlation coefficient is calculated, and the group of cross-sectional area and curvature value with the maximum Pearson correlation coefficient value is taken as the matching result. During the correlation calculation, since there are multiple groups of position deviations of smoke particles in a single branch path, multiple values are extracted for different positions of the single branch path, and if there are multiple smoke particles at a certain cross section, the average value of the smoke particles is selected to complete the data setting. The eigenvalue of the cross-sectional area and curvature value is set according to the sum of the corresponding proportion values of the cross-sectional area and curvature value or the ratio to the total eigenvalue.

[0049] If the Pearson correlation coefficient value of the displacement deviation and the cross-sectional area and curvature value is large, it represents that the branch geometric parameter has a significant influence on the displacement deviation of the smoke particle, otherwise it represents that the branch structure is simple and has a small influence on the movement of the smoke particle.

[0050] S223, after the matching is completed, the smoke particles are randomly arranged at the corresponding branch path, and the positions of the randomly arranged smoke particles are concentrated to determine the smoke particle concentration point.

[0051] After the random distribution arrangement, the K-means clustering algorithm is used to cluster the arranged particles into K clusters, and the multiple cluster centers formed by these clustering clusters can reflect the distribution of the smoke particles at a certain position, so as to find the concentration point of the smoke particle flow, thereby explaining the distribution of the multiple branch paths such as the outer wall of the spiral branch and the elbow of the labyrinth branch.

[0052] S224, according to the center position of the smoke particle concentration point, the smoke particles are superimposed, the positions of the smoke particles after superposition are determined, and the smoke particles are synchronized to the cross-sectional area and curvature value of the branch path; then the flow velocity gradient embodied on the different branch paths can be obtained according to the position change of the smoke particles.

[0053] During the superposition, the cluster cluster of the concentration point is superimposed, the contour coefficient corresponding to each smoke particle is taken as the distance weight, the ratio of the position deviation of each smoke particle to the maximum position deviation is taken as the precision weight, and the product of the precision weight and the distance weight accounts for the ratio of the total precision weight and the distance weight, so as to obtain the weighted average center coordinates by weighted average of the coordinates in the cluster cluster of the concentration point. The center coordinates will represent the concentration of the smoke particles in different branches, and can further reflect the geometric characteristics of different branch paths and reduce the position deviation caused by complex paths.

[0054] In an embodiment of the present application, in step S3, the smoke particles set in the foregoing part are checked, and the flow velocity gradient corresponding to the smoke particles is recorded, so as to prevent the partial flow velocity value from being too large and causing the motion tracking of the smoke particles in the branch path to be distorted.

[0055] As Figure 3 shown, the implementation of step S3 also includes: S31, according to the distribution position of the smoke particles in the branch path, using the flow rate gradient of the smoke particles to obtain the vorticity and strain rate as the flow rate characteristics of the current branch path.

[0056] The currently selected vorticity and strain rate belong to the values obtained by the flow rate gradient under the second order operation, which do not directly depend on the position coordinates and can capture the propagation characteristics of the smoke particles in the three-dimensional flow field.

[0057] The vorticity can be calculated by the central difference method to calculate the partial derivative of the velocity field in three directions, and then the three components are combined into a vorticity vector, and the modulus of the vorticity vector is taken to know the value of the current vorticity.

[0058] The strain is calculated by numerical differentiation to calculate the partial derivative of all velocity components with respect to the coordinates, and after extracting the symmetric part, the average value of the principal strain rate is directly calculated; the high strain area obtained can correspond to the shear layer or the flow acceleration area, etc. The part of the obvious movement of the branch path, and the deformation and diffusion of the smoke particles can be displayed through the strain.

[0059] At the same time, the difference between the square of the vorticity and the square of the strain rate can also be measured by the Q criterion to identify the vortex core structure corresponding to the current smoke particles. When greater than 0, it represents that rotation dominates, and the smoke is easy to form concentrated vortex in the corresponding branch path; less than 0 represents that strain dominates, and the smoke is easy to be stretched and diffused; if equal to 0, it represents that the corresponding branch path is a transition area, and the diffusion form does not exist in the direct dominant type at this position.

[0060] S32, based on the flow rate characteristics of different branch paths, the flow rate threshold is divided into a flow rate rising threshold and a flow rate falling threshold.

[0061] The flow rate threshold will be set according to the low-speed area and the high-speed area of the smoke particles moving in the branch path, such as the flow rate rising threshold is the critical value of the transition from the low-speed area to the high-speed area, and the flow rate falling threshold is the opposite, which is the transition from the high-speed area to the low-speed area; to determine the dynamic characteristics of the acceleration and deceleration of the smoke particles.

[0062] For example, with reference to the average flow rate of the corresponding branch path in the historical data as the reference flow rate, the flow rate rising threshold and the flow rate falling threshold are set in the form of mean ± 3 times the standard deviation, or an empirical coefficient is introduced, and in the form of minimum error sum of squares, 2-4 times the standard deviation of the current flow rate is selected. The multiple of this value is determined based on the minimum error sum of squares.

[0063] S33, according to the geometry of the branch path, determine the motion of the smoke particles in the cross section, when the smoke particles approach the path bifurcation point, record the position of each smoke particle, and update the data at the corresponding branch path when the current smoke particle corresponding flow velocity gradient changes, and take the updated data as the output data.

[0064] At this time, according to the geometry of the branch path, the bifurcation angle, the branch length and the specific situation of the cross section will be recorded, and the motion of the smoke particles reaching the bifurcation point will be recorded, and the data tracking of the smoke particles will be realized.

[0065] When the current smoke particle corresponding flow velocity gradient changes, the implementation manner comprises: S331, if the current flow velocity gradient exceeds the flow velocity rising threshold, the position of the smoke particle in each branch path is updated; the updating manner will adopt Kalman filtering manner, so as to introduce the current laser scattering light flow field to update the position of the smoke particle.

[0066] S332, if the current flow velocity gradient is less than the flow velocity falling threshold, the position of the smoke particle in each branch path is stopped updating.

[0067] S333, if the current flow velocity gradient is between the flow velocity falling threshold and the flow velocity rising threshold, the smoke particle is tracked according to the updating result of the last smoke particle position.

[0068] In step S3, in addition to describing the flow velocity threshold thereof, the gradient change of the smoke particle is also tracked to determine the correlation between the spaces where the different branch paths are located, and the correlation between the paths in the data updating is realized.

[0069] When updating, the velocity direction is mainly corrected, and the coordinates of the smoke particles are updated according to the corrected velocity, so as to complete the record of the smoke coordinates.

[0070] For example, after the particle enters another branch path, the velocity direction needs to be consistent with the direction of the branch path, and part of the main flow velocity component is reserved, assuming that the corrected velocity is: ; wherein, represents the velocity vector of the smoke particle after entering another branch path, represents the velocity vector of the smoke particle in the current branch path; represents the unit direction vector of the branch path, i represents the index number of different branches, which is used to point to other branch paths connected with the current branch path, and the value range is based on the number of other branch paths, for example, the value range of i can be 1, 2; represents the velocity projection term, which indicates the velocity component of the current velocity projection to other branch paths; represents a direction correction coefficient, the value range is 0.7-0.95, and is set based on the angle of the branch path and the inertia of the current smoke particle under the corresponding branch path; represents a main flow velocity vector, and generally represents the average flow velocity upstream of the current branch path; represents a main flow velocity retention coefficient, and is used for reflecting the inertia when the multi-branch path flows, can be set to any value in 0.05-0.3, and is affected by the number of smoke particles arranged on the current branch path; represents a turbulent flow velocity disturbance vector, and the value is the product of the turbulent intensity and a unit random vector, so as to explain the random disturbance caused by the turbulent flow velocity; represents a turbulent disturbance coefficient, the value indicates the influence degree of the turbulent disturbance on the particle velocity, generally takes a value of 0.05-0.2, and the value is selected based on the value of the turbulent intensity.

[0071] At this time, the flow tracking of the branch path at the path intersection can further check the coordinate change of the smoke particle in the three-dimensional flow field; and the updated coordinate can be completed according to the velocity vector and the time at this time, so as to complete the updating of the particle.

[0072] If the updated coordinate exceeds the three-dimensional boundary of the branch path, then the velocity direction needs to be corrected according to the reflection law and the coordinate needs to be updated again, for example ; wherein, represents a boundary normal vector, and finally the setting and tracking of the coordinate change are completed.

[0073] In an embodiment of the present application, in step S4, the trajectory of the smoke particle under the flow velocity gradient and the coordinates corresponding to the smoke concentration are fitted to determine the movement of the smoke particle in the corresponding branch path, and a motion trajectory sequence containing position, concentration, flow velocity and time is formed; then according to the concentration value, a concentration intensity index of the branch path is set, for example, the ratio of the maximum value of the local concentration gradient at the position of the branch path to the pre-alarm concentration gradient is selected as the concentration intensity index of the current branch path; or according to the ratio of the concentration value at the branch path to the pre-alarm concentration value, the concentration distribution of the branch path compared with the pre-alarm is explained, the index will reflect the local concentration gradient of the smoke particle when moving, and then explain the situation that needs to be responded when the smoke is tracked. Under the current set three-dimensional flow field, the concentration gradient can better reflect the diffusion of the smoke in space, and then explain the situation that the fire intensifies or the smoke spreads to the high-risk area.

[0074] As shown in Figure 4 , the implementation mode of step S4 includes: S41, extracting the trajectory points of the motion trajectory sequence, aligning each trajectory point with the time stamp of the concentration value, and then obtaining the concentration value corresponding to the trajectory point of the smoke particle when moving.

[0075] S42, based on the concentration value corresponding to the trajectory point, the concentration gradient corresponding to each trajectory point is calculated, and the concentration gradient of all trajectory points belonging to the same path is regarded as the local concentration gradient corresponding to the branch path; when belonging to the same path, mainly describing the smoke particles fitted in a branch path to explain the local concentration at each trajectory point; at the same time, if the corner position of the branch path is used, the concentration gradient corresponding to the concentration point of the smoke particle at the position can be regarded as the local concentration gradient of the position.

[0076] S43, the ratio of the maximum value of the local concentration gradient of the branch path to the warning concentration gradient is selected as the concentration intensity index of the current branch path; as for the value of the warning concentration gradient, the concentration gradient measured at the corresponding position when the smoke alarm issues a warning is taken as the reference to measure the part of the multi-branch path that can present the smoke alarm under the smoke diffusion.

[0077] S44, record the concentration intensity index at each smoke particle position update time, associate the concentration intensity index with each smoke particle update time, and regard the associated data as the current output data; when recording the update time, mainly according to the change of flow velocity gradient to cause the update of smoke particle coordinates, if the flow velocity gradient changes too small, even through the branch point of the branch path, the position of other branch paths can be obtained by correcting the speed direction and coordinate transformation according to the current flow velocity gradient.

[0078] At the same time, the length of the update time reflects the response sensitivity of the system to the flow field change, such as short update time representing the flow field change is severe, and long update time representing the system only responds to significant flow velocity gradient change to illustrate the flow field change at different branch paths.

[0079] Preferably, when fitting the trajectory of the smoke particle, the method of using the smoke center point in step S2 is mainly used to realize the fitting processing of multiple smoke particles, and according to the Hungarian algorithm, the coordinates of each smoke particle can be converted into a maximum weight matching problem, the probability value of each smoke particle is set according to the posterior probability, and the position of multiple smoke particles in the branch path is fitted, so as to obtain a fitted motion trajectory sequence.

[0080] In one embodiment of the present application, in step S5, each branch path is identified mainly according to the obtained update time and concentration intensity index, and combined with the flow rate gradient recorded on each branch path, the classification label of each branch path is configured in the form of description of various characteristic combinations; then, according to the different smoke trajectories caused by the flow rate difference, the early warning process of the smoke from the starting position, the passing path, the residence time, and the concentration change is determined, and the monitoring processing of each branch path is completed by triggering the early warning at any time.

[0081] As shown in Figure 5 The implementation mode of step S5 includes: S51, using the update time and concentration intensity index of each branch path, setting semantic labels for the branch paths, and obtaining the configured classification labels in the form of combination of semantic labels; the semantic labels corresponding to the concentration intensity index can include labels representing the concentration diffusion related to the branch path, such as high concentration and rapid accumulation, high concentration and slow accumulation, and low concentration and slow diffusion, and the update time is divided into long update time and short update time to explain the flow rate gradient mutation at the junction of different branch paths; the flow rate gradient is then described using high flow rate gradient and low flow rate gradient, and then whether each branch path is complex or its shape is described, thereby completing the configuration of the classification label.

[0082] S52, traversing each branch path and updating the classification label of each branch path in real time; judging the change of the classification label before and after the update, if the change occurs after the update, the differences of the update time, the concentration intensity index and the flow rate gradient are confirmed respectively, and the differences of the indexes are used for hierarchical early warning.

[0083] When the classification label changes, it indicates that the flow at the current position changes significantly, for example, the flow rate increases to cause the particles to be carried rapidly, the path update frequency increases; the concentration gradient drives the particles to diffuse to the low concentration area, the path update direction changes; the flow rate and the concentration jointly act, the path presents a spiral or a bifurcation, etc., causing a significant difference in the current update time description, which needs to be handled in time; at the same time, when the classification label on each path changes, it will also correspond to abnormal events such as leakage, release, blockage and obstacles, and the specific situation of the smoke particles needs to be further explained.

[0084] When the hierarchical early warning is performed according to the differences of the indexes, the implementation mode includes: determining the part belonging to the anomaly according to the difference values of the current update time, the concentration intensity index and the flow rate gradient.

[0085] If it is a single index anomaly, the difference value of the anomaly is output, and the coordinate corresponding to the difference value is regarded as the monitoring result of the output.

[0086] If it is a multi-index anomaly, the combination of abnormal indexes is determined as the current abnormal scene, and the abnormal scene is taken as the output monitoring result.

[0087] When a single-index anomaly occurs, for example, for the update time, the average of the update times of all branch paths globally is taken as the reference time. When the update time in any branch path changes by more than 20%, it is considered that the update time is abnormal, indicating that the corresponding part will have a flow rate or concentration anomaly, which needs to be checked in time.

[0088] The concentration intensity value is also set to a reference value according to the global average value. However, the concentration intensity value will be identified for any part with a change rate exceeding the reference value by 50% to determine the part where pollution rapidly accumulates, without the need to verify the absolute increase or decrease in concentration.

[0089] Since the flow rate in the branch path is limited based on the rising and falling thresholds and is updated, the flow rate will be set to the average flow rate of the corresponding geometric feature according to the geometric characteristics of the current branch path, and the part exceeding the reference value by 30% will be identified for early warning to determine the corresponding early warning of flow field blockage or leakage. Thus, the single-index anomaly judgment of the three indexes is completed.

[0090] If it belongs to a multi-index anomaly, the difference between the three indexes before and after the update is introduced, and the difference is matched according to the numerical interval prepared in advance to determine the current abnormal scene, and the abnormal scene is taken as the monitoring result of the corresponding branch path.

[0091] S53, if no change occurs after the update, the current classification label is taken as the basis for hierarchical early warning, and the data after early warning is taken as the monitoring result of each branch path.

[0092] When no change occurs, it indicates that the current three-dimensional flow field is in a relatively stable state, and the high concentration gradient part and the smoke accumulation part indicated in the current classification label can be directly used for early warning to complete the early warning processing of each branch area.

[0093] Although the embodiments of the present application have been shown and described above, it should be understood that the above embodiments are exemplary and should not be construed as limiting the present application. Those skilled in the art can make changes, modifications, replacements and variations to the above embodiments within the scope of the present application, which are still covered by the protection scope of the present application.

Claims

1. A smoke alarm method based on the trajectory of smoke particles, characterized in that, include: Based on the installation coordinates of each sensor within the current channel structure, a three-dimensional flow field distribution map of smoke particles is constructed. Based on the changes in cross-sectional area and curvature of the current channel at the branch paths, and taking the predicted smoke particle positions as the center, the velocity data of each branch path are correlated to obtain the velocity gradient sequence of each branch path. Based on the velocity gradient sequence of each branch path, when the velocity gradient exceeds the velocity threshold, the position of smoke particles in each branch path is updated, and the positional change of smoke particles on the branch path at each time point is output. Trajectory fitting is performed on smoke particles to form a sequence of motion trajectories corresponding to the smoke particles. Based on the concentration gradient corresponding to the motion trajectory sequence on each branch path, the concentration intensity index and update time of smoke particles on different branch paths are configured. Based on the update time and concentration intensity indicators of different branch paths, classification labels are configured for each branch path, and graded early warnings are carried out according to the differences in flow velocity to obtain the monitoring results of each branch path.

2. The smoke alarm method based on the movement trajectory of smoke particles according to claim 1, characterized in that, The methods for implementing three-dimensional flow field distribution maps include: Obtain the channel type corresponding to each branch path within the current channel structure, and set the initial position coordinates, air velocity, cross-sectional area, and curvature of the smoke particles in each branch path to obtain the basic flow field data for each branch path; The basic flow field data is synchronized to the coordinates of the corresponding sensor, and multi-view reconstruction is performed using the sensor coordinates to obtain a three-dimensional flow field distribution map.

3. The smoke alarm method based on the motion trajectory of smoke particles according to claim 1, characterized in that, The implementation methods of flow velocity gradient sequences include: Turbulence effects were verified in each branch path to determine the positional deviation of smoke particles in different branch paths. The positional deviation is synchronized to the cross-sectional area and curvature values ​​of the branch path, and the flow velocity corresponding to each smoke particle is calculated according to the axial, radial and circumferential angles of the smoke particles. The velocity of smoke particles at various locations is fitted to construct the velocity gradient sequence for each branch path.

4. The smoke alarm method based on the motion trajectory of smoke particles according to claim 3, characterized in that, When synchronizing the positional deviation to the cross-sectional area and curvature values ​​of the branch path, the implementation methods also include: Using the cross-sectional area and curvature values ​​of the branch paths as input, the differences between different branch paths are determined; The correlation between the differences in branch paths and the positional deviation of smoke particles is determined, and the cross-sectional area and curvature values ​​of the branch paths are matched with the correlation. After matching is completed, smoke particles are randomly deployed at the corresponding branch paths, and the locations of the randomly deployed smoke particles are concentrated to determine the concentration point of the smoke particles. Based on the center position of the smoke particle concentration point, the smoke particles are superimposed to determine the position of the corresponding smoke particles after superposition, and these smoke particles are synchronized to the cross-sectional area and curvature value of the branch path.

5. A smoke alarm method based on the trajectory of smoke particles according to claim 4, characterized in that, The implementation methods for matching the cross-sectional area and curvature values ​​of branch paths also include: Extract the main branch and sub-branch of the current branch path, and combine the branch paths into connected branches that link the main branch and sub-branch; For each branch path in the connected branches, the characteristic value corresponding to the cross-sectional area and curvature value is set based on the difference relative to the overall mean of the connected branches; The positional deviation and eigenvalue of each branch path in the connected branches are input, and the Pearson correlation coefficient is used to calculate the result. The set of cross-sectional area and curvature values ​​with the largest Pearson correlation coefficient is used as the output matching result.

6. A smoke alarm method based on the trajectory of smoke particles according to claim 1, characterized in that, The implementation of updating the smoke particle positions within each branch path also includes: Based on the distribution of smoke particles in the branch path, the vorticity and strain rate are obtained using the velocity gradient of the smoke particles, which are used as the velocity characteristics of the current branch path. Based on the flow velocity characteristics of different branch paths, the flow velocity threshold is divided into a flow velocity increase threshold and a flow velocity decrease threshold. Based on the geometry of the branch path, the anomaly of the smoke particles in the cross section is determined. When the smoke particles approach the path bifurcation point, the position of each smoke particle is recorded. When the velocity gradient corresponding to the current smoke particle changes, the data at the corresponding branch path is updated, and the updated data is used as the output data.

7. A smoke alarm method based on the trajectory of smoke particles according to claim 6, characterized in that, When the velocity gradient corresponding to the current smoke particles changes, the implementation methods include: If the current velocity gradient exceeds the velocity rise threshold, the positions of smoke particles in each branch path will be updated. The update method will use Kalman filtering to introduce the optical flow field scattered by the current laser to update the positions of smoke particles. If the current velocity gradient is less than the velocity decrease threshold, then stop updating the position of smoke particles in each branch path; If the current velocity gradient is between the velocity decrease threshold and the velocity increase threshold, the smoke particles are tracked based on the previous update of the smoke particle position.

8. A smoke alarm method based on the trajectory of smoke particles according to claim 1, characterized in that, When configuring the concentration intensity index and update time of smoke particles in different branch paths, the implementation methods also include: The trajectory points of the motion trajectory sequence are extracted, and each trajectory point is aligned with the concentration value by timestamp, so as to obtain the concentration value corresponding to the trajectory point of the smoke particles when they are moving. Based on the concentration values ​​corresponding to the trajectory points, the concentration gradient corresponding to each trajectory point is calculated, and the concentration gradient of all trajectory points belonging to the same path is regarded as the local concentration gradient under the corresponding branch path. The ratio of the maximum local concentration gradient of the branch path to the warning concentration gradient is selected as the concentration intensity index of the current branch path. Record the concentration intensity index at each smoke particle location update, correlate the concentration intensity index with the update time of each smoke particle, and use the correlated data as the current output data.

9. A smoke alarm method based on the trajectory of smoke particles according to claim 1, characterized in that, When obtaining the monitoring results for each branch path, the implementation methods include: Using the update time and concentration intensity indicators of each branch path, semantic labels are set for the branch paths, and the configured classification labels are obtained in the form of a combination of semantic labels. Traverse each branch path and update the classification labels of each branch path in real time; determine the changes in the classification labels before and after the update; if the changes occur after the update, confirm the differences in update time, concentration intensity index and flow velocity gradient respectively, and conduct graded early warning based on the differences in each index. If no changes occur after the update, the current category label will be used as the basis for the hierarchical warning, and the data after the warning will be regarded as the monitoring results of each branch path.

10. A smoke alarm method based on the motion trajectory of smoke particles according to claim 9, characterized in that, When implementing tiered early warning based on differences in various indicators, the methods include: The portion belonging to the anomaly is determined based on the difference between the current update time, concentration intensity index, and flow velocity gradient. If a single indicator is abnormal, the output will be the difference value of the abnormality, and the coordinate corresponding to the difference value will be regarded as the output monitoring result. If multiple indicators are abnormal, the current abnormal scenario is determined by the combination of abnormal indicators, and the abnormal scenario is used as the output monitoring result.

Images