Intelligent micro-fluidic chip grabbing method based on defect priority

Abstract

The invention relates to the technical field of micro-fluidic chip manufacturing, in particular to a micro-fluidic chip intelligent grabbing method based on defect priority. According to the method, accurate defect recognition is achieved through multispectral imaging and deep learning, the grabbing risk is scientifically quantified by adopting a multi-factor risk assessment model, grabbing parameters are adaptively adjusted according to the priority, secondary damage is effectively reduced, and the grabbing efficiency and the production quality are improved. Experiments show that the grabbing success rate can reach 99.2% or above, and the secondary damage rate is lower than 0.5%.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of microfluidic chip manufacturing, and particularly relates to a microfluidic chip intelligent grabbing method based on defect priority. BACKGROUND

[0002] As an important biomedical detection platform, microfluidic chips need to go through multiple precision machining processes in the production process. Due to the fragile material and delicate structure of the chip, cracks, scratches, pollution and other defects are easily generated in the production process. The traditional grabbing system usually uses uniform grabbing parameters, which cannot be differentiated according to the defect condition of the chip, resulting in the same grabbing intensity of the intact chip and the defective chip, which is easy to cause secondary damage or grabbing failure. The existing grabbing system mainly has the following deficiencies: 1) lack of accurate recognition and classification ability of chip defects; 2) unable to evaluate the grabbing risk according to the defect characteristics; 3) fixed grabbing parameters, unable to adaptively adjust; 4) unreasonable grabbing path planning, low production efficiency. Therefore, an intelligent method for recognizing defects, evaluating risks and adaptively adjusting grabbing strategies is urgently needed. SUMMARY

[0003] To achieve the above purpose, the technical scheme adopted by the present application is as follows: A microfluidic chip intelligent grabbing method based on defect priority, the method comprising the following steps: S1, tray positioning step: placing a tray containing several pieces of microfluidic chips on a positioning platform, scanning the identification code on the tray through a sensor, obtaining the tray configuration information, and adjusting and fixing the tray to a preset reference position through the positioning platform based on the tray configuration information; S2, defect detection step: irradiating the tray with a multi-spectral light source, collecting the image of the microfluidic chip through an industrial camera, and analyzing the image using a defect detection model to identify the chip defect type and extract the defect feature parameters; S3, priority evaluation and division: the evaluation system calculates the comprehensive risk value R of each chip according to the defect feature parameters, and divides the chips into four priorities of prohibited grabbing (P0), high risk (P1), medium risk (P2) and low risk (P3) according to the comprehensive risk value R; S4, adaptive grabbing step: the actuator configures different grabbing parameters for chips of different priorities, controls the grabbing mechanism to perform grabbing operation, and monitors the contact force in real time through the sensor during the grabbing process, and stops operation and alarms immediately when the detected contact force exceeds the set threshold.

[0004] The above method is an intelligent grasping scheme integrating machine vision, artificial intelligence and adaptive control. Its core idea is "detecting first and then grasping, grading by quality, and adaptive operation in sequence", aiming to realize efficient, reliable and lossless automation production.

[0005] Preferably, in the priority evaluation and path planning step, the comprehensive risk value R is calculated by the following formula: R = α × R1 + β × R2 + γ × R3, wherein R1 is a grasping risk coefficient for representing the influence degree of defects on the structural integrity of the chip; R2 is a device influence coefficient for representing the potential damage risk of defects on the grasping equipment and surrounding chips; R3 is a detection value coefficient for representing the proportion of the remaining detectable functional area of the chip; α, β, γ are weight coefficients, and the values of α, β, γ are respectively α = 0.5 ± 0.05, β = 0.3 ± 0.05, and γ = 0.2 ± 0.05.

[0006] The model quantifies the risk evaluation in multiple dimensions into a comprehensive index through a weighted linear formula, thereby scientifically determining the processing priority of the chip.

[0007] Preferably, it further comprises a grasping sequence planning unit, which eliminates P0-level chips and plans the grasping sequence in the priority order of P1→P2→P3, while dividing the area based on the spatial distribution of chips of the same priority using a clustering algorithm to generate the optimal grasping path.

[0008] Preferably, in the adaptive grasping step, the grasping parameters include negative pressure value, contact force and buffer stroke; for high-risk priority chips, the grasping strategy adopted is relatively higher negative pressure value, relatively smaller contact force and relatively larger buffer stroke.

[0009] Preferably, in the adaptive grasping step, different grasping strategies are adopted for different priority chips: For P1-level high-risk chips, -55kPa to -60kPa negative pressure value, 0.06N to 0.08N contact force and 3mm to 4mm buffer stroke are adopted; For P2-level medium-risk chips, -45kPa to -50kPa negative pressure value, 0.08N to 0.10N contact force and 2mm to 3mm buffer stroke are adopted; For P3-level low-risk chips, -35kPa to -40kPa negative pressure value, 0.10N to 0.12N contact force and 1mm to 2mm buffer stroke are adopted.

[0010] Preferably, non-risk chips are placed in the middle qualified discharge tray, and risk chips are placed in the unqualified discharge tray.

[0011] Compared with the prior art, the present application has the following beneficial effects: 1. Through multi-spectral imaging and deep learning algorithm, the precise identification and classification of microfluidic chip defects are realized; 2. A multi-factor weighted risk assessment model is used to scientifically quantify the risk of grabbing and provide a basis for grabbing decision; 3. According to the defect priority, the grabbing parameters are self-adaptively adjusted to minimize the secondary damage; 4. Through intelligent path planning, the grabbing efficiency is significantly improved, and the movement distance of the mechanical arm is reduced; 5. Real-time monitoring and abnormal alarm mechanism ensure the safety and reliability of the grabbing process. BRIEF DESCRIPTION OF DRAWINGS

[0012] The drawings described herein are used to provide further understanding of the embodiments of the present application, constitute a part of the present application, and do not constitute a limitation on the embodiments of the present application. In the drawings: Figure 1 A flow chart of a microfluidic chip intelligent grabbing method based on defect priority according to the present application.

[0013] Figure 2 A structure diagram of a microfluidic chip intelligent grabbing system based on defect priority according to the present application.

[0014] Markings in the drawings and corresponding names of parts: 1-tray, 2-positioning platform, 3-multispectral light source, 4-industrial camera, 5-grabbing mechanism, 6-qualified discharge tray, 7-unqualified discharge tray. DETAILED DESCRIPTION

[0015] In order to make the purpose, technical scheme and advantages of the present application more clear and obvious, the present application is further described in detail below in combination with embodiments and drawings. The illustrative embodiments of the present application and their descriptions are only used to explain the present application and do not limit the present application.

[0016] Embodiment: S1, place the tray 1 containing 100 microfluidic chips on the positioning platform 2, scan the identification code on the tray 1 through the sensor, obtain the tray 1 configuration information, and based on the tray 1 configuration information, adjust and fix the tray 1 to the preset reference position through the positioning platform 2; S2, the multispectral light source 3 switches four wavebands in turn to irradiate the tray 1, and three groups of industrial cameras 4 synchronously collect images, and a defect detection model analyzes the images, identifies the chip defect type and extracts the defect feature parameters; S3, the evaluation system calculates the comprehensive risk value R, and divides four priorities: P0 (R≥80), P1 (60≤R<80), P2 (30≤R<60), P3 (R<30); The grasping sequence planning unit eliminates P0-level chips and plans the grasping sequence in the priority order of P1→P2→P3, and at the same time, based on the spatial distribution of chips of the same priority, the clustering algorithm is used for regional division to generate the optimal grasping path; S4: the actuator adopts differentiated grasping parameters according to different priorities: P1 level: negative pressure value -55kPa, contact force 0.06N, buffer stroke 4mm; P2 level: negative pressure value -45kPa, contact force 0.10N, buffer stroke 2.5mm; P3 level: negative pressure value -35kPa, contact force 0.12, buffer stroke 1mm; Finally, the non-risk chips are placed into the qualified discharge tray 6, and the risk chips are placed into the unqualified discharge tray 7.

[0017] Implementation effect: the total time for processing 100 chips is 8 minutes and 25 seconds, the grasping success rate is P1 level 98.3%, P2 level 99.2%, and P3 level 99.8%, and the secondary damage rate is 0.3%.

[0018] The above specific embodiments further illustrate the purpose, technical solutions and beneficial effects of the present application. It should be understood that the above description is only a specific embodiment of the present application and is not used to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A microfluidic chip intelligent grabbing method based on defect priority, characterized in that, The method comprises the following steps: S1, tray positioning step: place a tray (1) containing several microfluidic chips on a positioning platform (2), scan the identification code on the tray through a sensor, obtain the tray configuration information, and adjust and fix the tray to the preset reference position based on the tray configuration information through the positioning platform; S2, defect detection step: irradiate the tray with a multi-spectral light source (3), collect the image of the microfluidic chip through an industrial camera (4), and analyze the image using a defect detection model to identify the chip defect type and extract the defect feature parameters; S3, priority evaluation and division: the evaluation system calculates the comprehensive risk value R of each chip according to the defect feature parameters, and divides the chips into four priorities: prohibited grabbing (P0), high risk (P1), medium risk (P2), and low risk (P3) according to the comprehensive risk value R; S4, adaptive grabbing step: the actuator configures different grabbing parameters for chips of different priorities, controls the grabbing mechanism (5) to perform grabbing operation, and monitors the contact force in real time through the sensor during the grabbing process, and stops operation and alarms immediately when the detected contact force exceeds the set threshold.

2. The microfluidic chip intelligent grabbing method based on defect priority according to claim 1, characterized in that, In step S3, the comprehensive risk value R is calculated by the following formula: R = α × R1 + β × R2 + γ × R3, wherein R1 is a grabbing risk coefficient, which represents the influence degree of defects on the structural integrity of the chip; R2 is a device influence coefficient, which represents the potential damage risk of defects on the grabbing equipment and surrounding chips; R3 is a detection value coefficient, which represents the proportion of the remaining detectable functional area of the chip; α, β, γ are weight coefficients, and their values are α = 0.5 ± 0.05, β = 0.3 ± 0.05, and γ = 0.2 ± 0.

05.

3. The microfluidic chip intelligent grabbing method based on defect priority according to claim 1, characterized in that, In step S3, a grabbing sequence planning unit is also included, which removes P0 level chips and plans the grabbing sequence in the priority order of P1→P2→P3, while dividing the chips of the same priority into regions based on their spatial distribution using a clustering algorithm to generate the optimal grabbing path.

4. The microfluidic chip intelligent grabbing method based on defect priority according to claim 1, characterized in that, In step S4, the grabbing parameters include negative pressure value, contact force and buffer stroke; for high-risk priority chips, the grabbing strategy adopted is: relatively higher negative pressure value, relatively smaller contact force and relatively larger buffer stroke.

5. The microfluidic chip intelligent grabbing method based on defect priority according to claim 1, characterized in that, In step S4, different grabbing strategies are adopted for chips of different priorities: For P1 level high-risk chips, -55kPa to -60kPa negative pressure value, 0.06N to 0.08N contact force and 3mm to 4mm buffer stroke are adopted; For P2 level medium-risk chips, -45kPa to -50kPa negative pressure value, 0.08N to 0.10N contact force and 2mm to 3mm buffer stroke are adopted; For P3 level low-risk chips, -35kPa to -40kPa negative pressure value, 0.10N to 0.12N contact force and 1mm to 2mm buffer stroke are adopted.

6. The microfluidic chip intelligent grabbing method based on defect priority according to claim 1, characterized in that, Also included is the step of placing non-risk chips into the middle qualified discharge tray (6) and placing risk chips into the unqualified discharge tray (7). Also included is the step of placing non-risk chips into the middle qualified discharge tray (6) and placing risk chips into the unqualified discharge tray (7).

Images