A multi-modal partition induction precision trapping method based on airflow field reconstruction

Abstract

This application provides a multimodal zoned induction and precise trapping method based on airflow field reconstruction, relating to the field of pest control technology. The method includes constructing a peripheral light-odor composite attraction zone and pre-optimizing the odor concentration using AI; forming a circumferential swirling locking zone using an annular guide hood and monitoring the residence time in real time; increasing the fan speed to form an axial laminar flow trapping zone after target detection and using a trajectory prediction algorithm; periodically and intelligently executing reverse micro-blowing to break up the stacked structure of the sticky traps; dynamically adjusting the light-attracting and airflow modes according to environmental parameters and handling external interference. This application achieves high trapping accuracy and low energy consumption.

Description

Technical Field

[0001] This invention relates to the field of pest control technology, and in particular to a multimodal zoning-based precise trapping method based on airflow field reconstruction. Background Technology

[0002] Existing sticky insect traps and suction-type mosquito killers typically use ultraviolet light sources to attract insects, combined with fan suction or sticky traps to achieve insect control. However, they suffer from the following technical problems: the flight trajectory after attraction is uncontrollable, leading to secondary escapes by insects; the fan only serves a suction function and does not participate in flight path control; insects tend to pile up on the surface of the sticky trap, reducing the effective adhesion area; and the light-attracting, airflow, and odor modules operate in fixed modes without being adjusted according to insect behavior.

[0003] To address the aforementioned problems, there is an urgent need for a trapping method that can actively control the flight trajectory of insects, improve capture efficiency, and reduce the probability of escape. Summary of the Invention

[0004] To address the aforementioned issues, this application provides a multimodal partitioned induction method for precise trapping based on airflow field reconstruction.

[0005] This application provides a multimodal partitioned induction and precise trapping method based on airflow field reconstruction, which adopts the following technical solution: A multimodal partitioned induction method for precise trapping based on airflow field reconstruction includes the following steps: S1. Construct a peripheral light-odor composite attraction zone and pre-optimize the odor concentration using AI; S2. A circumferential vortex locking zone is formed by an annular guide shroud, and the residence time is monitored in real time; S3. After detecting the target, increase the fan speed to form an axial laminar flow capture zone and use a trajectory prediction algorithm; S4. Periodic and intelligent reverse microblowing to break up the stacked structure of the adhesive traps; S5. Dynamically adjust the light-inducing and airflow modes according to environmental parameters, and handle external interference.

[0006] Optionally, the radius R of the peripheral light-odor composite attraction zone is 1-3m and dynamically adjusted according to the environment, and the odor release concentration is pre-optimized based on historical capture data using an AI algorithm.

[0007] Optionally, the airflow velocity in the circumferential swirling locking zone is 0.5-2.0 m / s and is adaptively adjusted according to the size of the insect; multi-sensor fusion is used to monitor the residence time in real time, and the swirling intensity is finely adjusted if the threshold is exceeded.

[0008] Optionally, the fan speed is increased to 2000-3000 rpm, and a threshold adaptive mechanism is introduced to reduce trigger sensitivity in high-density insect swarms.

[0009] Optionally, the reverse micro-blowing time is 0.3-1.5 seconds, supports intelligent triggering based on stacking thresholds of image sensors, and combines vibration feedback to adapt the intensity, performing a brief forward purging after the reverse blowing.

[0010] In summary, this application includes at least one of the following beneficial technical effects: The swirling locking zone significantly prolongs insect dwell time, reducing the probability of escape; phased airflow control combined with trajectory prediction improves capture accuracy; and the intelligent reverse micro-blowing mechanism enhances the utilization rate of the sticky trap and extends equipment life. AI-assisted adaptive adjustment reduces energy consumption and is optimized for different insect species; anomaly handling enhances the ability to resist natural wind interference in outdoor environments. Attached Figure Description

[0011] Figure 1 This is the flowchart of this method. Detailed Implementation

[0012] The terminology used in the following embodiments is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification and appended claims of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to also include expressions such as “one or more,” unless the context clearly indicates otherwise. It should also be understood that in the following embodiments of this application, “at least one” and “one or more” refer to one, two, or more than two. The term “and / or” is used to describe the relationship between related objects, indicating that three relationships may exist; for example, A and / or B can indicate: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character “ / ” generally indicates that the preceding and following related objects are in an “or” relationship.

[0013] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0014] The following is in conjunction with the appendix Figure 1 The present invention will be described in further detail below.

[0015] This application discloses a multimodal partitioned induction and precise trapping method based on airflow field reconstruction, referring to... Figure 1 This includes the following steps: S1. Construct an outer diffusion attraction field, activate the ring-shaped ultraviolet LED light source and biomimetic odor release module to form a light-odor composite attraction zone with a radius of R around the device. R can be dynamically adjusted to 1-3m according to the environment. The fan runs at the first low speed of 800-1500rpm to form a low-speed outward airflow field. At the same time, integrate AI algorithm to analyze historical captured data and pre-optimize the odor release concentration to enhance the initial attraction targeting. The AI ​​algorithm is a multilayer perceptron (MLP) neural network model, with the following specific structure: Input layer: 5 nodes, the input feature vector is the historical capture frequency, ambient temperature, relative humidity, current time period, and proportion of previous odor concentration; Hidden layer 1: 16 nodes, ReLU activation function; Hidden layer 2: 8 nodes, ReLU activation function; Output layer: CO2 ratio, lactate ratio and release interval in seconds are three nodes. The Sigmoid activation function is normalized to [0,1] and then mapped to the actual range. The training process employs supervised learning, using no fewer than 1000 historical datasets; data preprocessing includes Min-Max normalization; the loss function is mean squared error (MSE); the optimizer is Adam, with an initial learning rate of 0.001, and training for 200 epochs; online fine-tuning is performed hourly using newly captured data.

[0016] S2. A circumferential swirling locking zone is formed. Through the annular guide shroud and its inner wall spiral guide rib structure, the airflow generated by the fan forms a circumferential swirling zone around the air inlet. The airflow velocity is controlled at 0.5-2.0m / s to accommodate different insect sizes. After the insect enters, its flight posture is disturbed and it is trapped in the swirling zone. Infrared, photoelectric and acoustic multi-sensor fusion is used to monitor the residence time in real time. If the threshold is exceeded, the swirling intensity is finely adjusted to prevent energy waste or insect fatigue and escape.

[0017] S3. Target-triggered axial capture: When an infrared, photoelectric, or AI vision sensor detects a flying target entering the vortex locking zone, the fan speed is increased to the second high speed of 2000-3000 rpm, and a prediction algorithm is used to predict the insect's trajectory; a directional laminar accelerated airflow is formed along the axis of the device, with a gradual flow velocity curve to avoid sudden changes that could lead to escape; the insect is sucked into the sticky trap or electric shock zone. At the same time, an adaptive threshold is introduced: if the insect population density is high, the trigger sensitivity is reduced to process in batches and reduce response delay. The trajectory prediction algorithm uses a standard Kalman filter, and its implementation is as follows: State vector X = [x, y, v] x , vᵧ]ᵀ(unit: mm, mm / s; The state transition matrix F = [[1, 0, Δt, 0], [0, 1, 0, Δt], [0, 0, 1, 0], [0,0, 0, 1]], where Δt = 0.1s sampling period; The process noise covariance Q = diag(0.01, 0.01, 0.1, 0.1); The measurement matrix H = [[1,0,0,0], [0,1,0,0]] measures only the position, and the velocity is estimated by sensor fusion. The measurement noise covariance R = diag(5, 5) represents the pixel-to-mm conversion error.

[0018] Algorithm steps: Time update: X{k|k-1} = FX{k-1|k-1}; P{k|k-1} = FP{k-1|k-1} Fᵀ + Q; Measurement update: Kalman gain Kk = P{k|k-1} Hᵀ (HP{k|k-1} Hᵀ + R)^(-1); X{k|k} =X{k|k-1} + Kk (Z_k - HX{k|k-1}), where Z_k is the real-time position measurement of the sensor; P{k|k} = (I - Kk H)P{k|k-1}; Trajectory extrapolation: Using the updated state, predict forward N=5 steps to obtain the future position sequence.

[0019] Step S4: Periodic and intelligent reverse micro-blowing to prevent stacking. Every preset time interval T, or when the image sensor detects a stacking threshold, the control system briefly reverses the fan for 0.3-1.5 seconds, creating a reverse airflow. This disturbs the loosely stacked insects on the adhesive plate surface, causing the partially adhered insects to shift, roll, or slightly detach, thus breaking up the localized stacking structure and restoring some effective adhesion area. Immediately after the reverse micro-blowing ends, a brief forward micro-blowing for 0.2-0.5 seconds at a wind speed of 0.5 m / s is executed to blow the loosened residue into the collection tank below or out of the device, avoiding secondary pollution. For weak stacking, a single low-intensity reverse blow and vibration are used. In cases of severe stacking: use pulsed multi-round, 2-3 times, with a 1-second interval between each time, combined with vibration feedback and real-time monitoring by an acceleration sensor. If the amplitude decay is less than 20%, extend the interval by 0.2 seconds.

[0020] Step S5: Rhythm Coordination and Anomaly Handling. Based on data from temperature and humidity sensors, capture frequency, and wind speed sensors, automatically adjust the UV LED flash frequency, fan speed curve, and odor release interval. If external interference is detected, such as natural wind >2m / s, temporarily switch to high vortex mode or pause induction to maintain stability. Periodically self-check sensor accuracy; if the deviation >5%, trigger calibration.

[0021] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. Any equivalent modifications or variations made by those skilled in the art based on the content disclosed in the present invention should be included within the scope of protection set forth in the claims.

Claims

1. A multimodal partitioned induction and precise trapping method based on airflow field reconstruction, characterized in that, Includes the following steps: S1. Construct a peripheral light-odor composite attraction zone and pre-optimize the odor concentration using AI; S2. A circumferential vortex locking zone is formed by an annular guide shroud, and the residence time is monitored in real time; S3. After detecting the target, increase the fan speed to form an axial laminar flow capture zone and use a trajectory prediction algorithm; S4. Periodic and intelligent reverse microblowing to break up the stacked structure of the adhesive traps; S5. Dynamically adjust the light-inducing and airflow modes according to environmental parameters, and handle external interference.

2. The multimodal partitioned induction and precise trapping method based on airflow field reconstruction according to claim 1, characterized in that: In step S1, the radius R of the peripheral light-odor composite attraction zone is 1-3m and is dynamically adjusted according to the environment; the odor release concentration is pre-optimized based on historical capture data using an AI algorithm.

3. The multimodal partitioned induction and precise trapping method based on airflow field reconstruction according to claim 1, characterized in that: In step S2, the airflow velocity in the circumferential swirling locking zone is 0.5-2.0 m / s and is adaptively adjusted according to the size of the insect. Multi-sensor fusion is used to monitor the residence time in real time, and the swirling intensity is finely adjusted if the threshold is exceeded.

4. The multimodal partitioned induction and precise trapping method based on airflow field reconstruction according to claim 1, characterized in that: In step S3, the fan speed is increased to 2000-3000 rpm, and a threshold adaptive mechanism is introduced to reduce the trigger sensitivity when there is a high density of insects.

5. The multimodal partitioned induction and precise trapping method based on airflow field reconstruction according to claim 1, characterized in that: In step S4, the reverse micro-blowing time is 0.3-1.5 seconds, supporting intelligent triggering of stacking threshold based on image sensor, and combined with vibration feedback to adapt the intensity. A brief forward purging is performed after backflushing.

Images