Milk multi-component online detection device and detection method based on microfluidic technology

Abstract

The invention discloses a milk multi-component online detection device and detection method based on a microfluidic technology, and belongs to the technical field of food detection. The device takes a micro-fluidic chip as a carrier, each micro-fluidic detection channel comprises a sample cell, a reagent sample introduction area, a mixing area, a micro-fluidic mixer with an angular flow channel and a detection color development area, an outlet of the sample cell and an outlet of the corresponding reagent sample introduction area are converged in the mixing area, a sample introduction end of the micro-fluidic mixer is connected with the mixing area, and a sample introduction end of the micro-fluidic mixer is connected with the detection color development area. The sample outlet end is connected with the detection developing area; the optical detection modules are arranged above the detection color development areas and used for collecting images and extracting RGB color characteristic values, and the data processing and control module is electrically connected with the optical detection modules, receives the RGB color characteristic values and converts the RGB color characteristic values into concentration information of corresponding components according to a standard curve. The device can realize simultaneous on-line efficient synchronous detection of two components, and is fast in response, high in precision, and high in practicability and popularization value.

Description

Technical Field

[0001] This invention belongs to the field of food testing technology, and in particular relates to an online detection device and method for multiple components of milk based on microfluidic technology. Background Technology

[0002] In the production, distribution, and quality supervision of milk, protein and lactose content are core indicators for assessing the nutritional value and quality grade of milk. Accurate and efficient detection of these parameters is crucial for ensuring product quality and protecting consumer rights. According to the National Food Safety Standards for the Determination of Protein in Infant Foods and Dairy Products (GB 5413.1-2010) and the National Food Safety Standards for the Determination of Lactose and Sucrose in Infant Foods and Dairy Products (GB 5413.5-2010), the mainstream testing methods in the industry still rely on large-scale laboratory analytical instruments. Protein detection uses the Kjeldahl method as a benchmark, while lactose detection commonly employs high-performance liquid chromatography (HPLC).

[0003] Traditional detection methods have significant limitations: First, the procedures are cumbersome and time-consuming. The Kjeldahl method requires multiple steps, including digestion at 420℃ for 90 minutes, distillation, and titration, with a single sample analysis taking at least 100 minutes. HPLC requires pretreatment steps such as sample dissolution, ultrasonic extraction, and membrane filtration, often taking 2-4 hours in total, which is completely unsuitable for real-time quality control on production lines and shelf-life sampling in supermarkets. Second, the cost and operational barriers are high. The Kjeldahl method costs approximately 6 euros per analysis and requires professional personnel to operate equipment such as fume hoods. The purchase cost of a single HPLC instrument can reach hundreds of thousands of yuan, which is unaffordable for small dairy companies. Third, the detection modes are fragmented. The two components require separate sample preparation and separate analysis, which not only increases sample consumption (more than 2.5g for a single analysis of liquid milk) but also further extends the detection cycle.

[0004] To overcome the limitations of large instruments, miniaturized detection technologies such as colloidal gold and near-infrared spectroscopy have gradually emerged in recent years, but many technical shortcomings still exist. According to documents such as the "Technical Standards for Inspection and Testing of Dairy Products," existing small devices generally face the problem of insufficient mixing efficiency. Most of them adopt simple channel structures and have not optimized the channel design for fluid characteristics, resulting in uneven mixing of milk samples with colorimetric reagents such as Coomassie Brilliant Blue and DNS, and the detection error rate often exceeds 5%.

[0005] As relevant research has pointed out, rapid multi-component detection of complex food matrices requires overcoming three major bottlenecks: mixing efficiency, environmental adaptability, and intelligence. Current milk protein and lactose detection technologies face the dilemma of "insufficient efficiency of large instruments and lack of precision and intelligence in small devices," urgently requiring an integrated solution to achieve the goals of "small sample volume, fast detection speed, high result accuracy, simultaneous detection of two components, and remote data management," meeting the needs of real-time detection in multiple scenarios and driving the upgrade of milk quality testing technology towards on-site, convenient, and intelligent approaches. Summary of the Invention

[0006] To address the aforementioned technical problems and overcome the shortcomings of existing technologies, the primary objective of this invention is to address the issue of low mixing efficiency leading to large detection errors in current milk testing methods. This is achieved by designing a microfluidic mixer with a angularly coupled sawtooth structure to ensure thorough mixing of the milk sample with the protein detection reagent (Coomassie Brilliant Blue reagent) and the lactose detection reagent (DNS reagent). Simultaneously, a temperature control module maintains a stable temperature for the colorimetric reaction, thus resolving the problem of poor detection accuracy caused by uneven mixing and temperature fluctuations, and achieving accurate results for dual-component detection.

[0007] The second objective is to address the pain points of traditional testing methods, which require two sample processing steps, manual data recording, and lack of remote management capabilities. By leveraging a microfluidic chip carrier and combining an OpenMV recognition module, a host computer, and a WiFi module, the system enables "simultaneous detection of protein and lactose with a single sample injection." It also automatically completes RGB value recognition, linear relationship fitting, standard curve plotting, and data upload to the cloud platform, thereby achieving efficient testing processes and intelligent data management.

[0008] The objective of this invention is achieved through the following technical solution: This invention discloses an online multi-component detection device for milk based on microfluidic technology, comprising a microfluidic detection chip integrating two microfluidic detection channels. Each microfluidic detection channel includes a sample cell for accommodating the milk sample to be tested, a reagent injection area, a mixing area, a microfluidic mixer with a angularly coupled sawtooth flow channel, and a detection and color development area. The sample cells of the two detection channels are interconnected through the injection channel, and the outlet of the sample cell of each detection channel merges with the outlet of the corresponding reagent injection area in the mixing area. The injection end of the microfluidic mixer is connected to the mixing area, and the outlet end is connected to the detection and color development area.

[0009] An optical detection module is positioned above each detection color development area to acquire color images of the reaction liquid within the detection color development area and extract RGB color features.

[0010] The data processing and control module is electrically connected to the optical detection module and is used to receive the RGB color feature values ​​and convert them into the concentration information of the corresponding components according to a preset algorithm.

[0011] Furthermore, the angularly coupled sawtooth structure flow channel of the microfluidic mixer is composed of multiple mixing channels arranged periodically along the flow direction, with adjacent mixing channels having bending angles to form a sawtooth structure; each mixing channel has two symmetrical triangular protrusions and bifurcated channels.

[0012] Furthermore, the bending angle α of the hybrid channel is 60°.

[0013] Furthermore, the bending angle β at the corner of the bifurcated channel of the triangular protrusion is 90°, the bending angle γ between the bifurcated channel and the bending channel is 135°, and the bending angle δ between the bending channel and the sample outlet channel of the mixing area or the sample inlet channel of the detection and display area is 135°.

[0014] Furthermore, the width of the two detection channels, excluding the bifurcation channel, is 1mm and the height is 2mm; the width of the bifurcation channel is 0.5mm and the height is 2mm.

[0015] Furthermore, the milk sample pools on the sample injection channels of the two detection channels are connected to the milk inlet; the mixing zone has a height of 3.5 mm and a diameter of 7.4 mm.

[0016] Furthermore, an opening is provided on the microfluidic detection chip in the color development area of ​​the two detection channels. The length L of the opening is 1 / 3 of the length of the microfluidic detection chip, and the distance W from the two sides of the opening to the edge of the two color development areas is 2mm.

[0017] Furthermore, a temperature control module is provided for heating and closed-loop temperature control of the two detection channels of the microfluidic detection chip. The module includes a temperature controller and two heating elements with integrated temperature sensors. The two heating elements cover the bottom of the two detection channels of the microfluidic chip separately. The temperature controller is connected to the data processing and control module. The temperature of the two detection channels is collected in real time by the temperature sensors, and the data processing and control module controls the heating temperature of the two heating elements respectively through the temperature controller.

[0018] Furthermore, the two detection channels are a lactose detection channel and a protein detection channel, respectively. The reagent injection area of ​​the lactose detection channel uses DNS reagent and the temperature is controlled at 95±0.5℃; the reagent injection area of ​​the protein detection channel uses Coomassie Brilliant Blue reagent and the temperature is controlled at 37℃±0.5℃.

[0019] The detection method of the online detection device of the present invention includes the following steps: S1. Sample injection and distribution: The milk sample to be tested is simultaneously distributed to the sample pools of two detection channels through the injection channel; Coomassie Brilliant Blue reagent and DNS reagent are added to the corresponding reagent injection areas respectively; S2. Mixing and colorimetric reaction: The fluid is driven at a flow rate of 3 μL / s to 5.5 μL / s, so that the milk sample and the corresponding reagent in each channel are combined in the mixing area according to the set ratio, and then flow through the microfluidic mixer with the angular coupling sawtooth structure for efficient mixing, and then flow into the lactose detection colorimetric area / protein detection colorimetric area respectively; Under the action of the temperature control module, the temperature of the lactose detection channel is maintained at 95℃±1℃, and the temperature of the protein detection channel is maintained at 37℃±0.5℃; S3. Optical Detection and Data Processing: The optical detection module synchronously acquires color images of two detection and color display areas, extracts RGB feature values, and the data processing and control module calculates and displays the concentrations of protein and lactose according to a preset algorithm.

[0020] The beneficial effects of this invention are as follows: 1. This invention employs a microfluidic chip and designs a simulation-optimized, angle-coupled, sawtooth-shaped microfluidic mixer. This structure enables efficient mixing of milk samples with Coomassie Brilliant Blue and DNS reagents. Combined with an identification module, temperature control module, microcontroller, and host computer, it achieves collaborative detection across modules, improving the uniformity and efficiency of the colorimetric reaction, thereby ensuring the accuracy of the detection results. The entire device has advantages such as small size, simple operation, fast detection speed, and low cost, and can be widely used for rapid on-site testing of milk quality.

[0021] 2. This invention injects milk samples into a microfluidic chip at an optimal mixing efficiency flow rate. Within the reaction channel, a flow rate of 3-5.5 μL / s can achieve a mixing efficiency of over 95%. Coomassie Brilliant Blue reacts with protein molecules to produce a colorimetric reaction, while lactose molecules react with DNS reagent to produce a colorimetric reaction. Different concentrations correspond to different colors, and the color gradually deepens with increasing concentration. The OpenMV recognition module performs image recognition on the developed colors, enabling rapid and accurate acquisition of RGB values, thus achieving simultaneous detection of protein and lactose content.

[0022] 3. The present invention can accurately control the temperature in the two detection channels separately through the temperature control module, providing a stable temperature environment for the colorimetric reaction, avoiding the influence of temperature fluctuations on the reaction results, and further improving the detection accuracy.

[0023] 4. The OpenMV detection module of this invention captures colorimetric images in real time and identifies RGB values. The temperature control module ensures the reaction temperature to improve the accuracy of the results. By setting up a microcontroller and a host computer with an LCD display, the host computer automatically records data, fits linear relationships, plots standard curves, and calculates component content. It can simultaneously output RGB values ​​and content values, facilitating real-time observation of the detection results by operators. Simultaneously, the data is uploaded to the OneNET cloud platform via a WiFi module, facilitating data storage, management, and remote access, thus realizing the intelligent and information-based detection process. This device features dual channels, enabling simultaneous online and efficient synchronous detection of two components. It boasts fast response, high accuracy, and remote data transmission capabilities, making it highly practical and valuable for widespread application. Attached Figure Description

[0024] Figure 1 This is a flowchart illustrating the structure of the present invention.

[0025] Figure 2 This is a schematic diagram of the detection device of the present invention.

[0026] Figure 3 This is a schematic diagram of the microfluidic chip in this invention.

[0027] Figure 4 This is a schematic diagram of an angular buffer channel structure.

[0028] Figure 5 This is a schematic diagram of a serpentine buffer channel structure.

[0029] Figure 6 This is a schematic diagram of a linear buffer channel structure.

[0030] Figure 7 A comparison chart of the mixing efficiency of different buffer channels.

[0031] In the diagram: 1. Microfluidic chip; 2. Lactose reagent inlet; 3. Protein reagent inlet; 4. Sample cell; 5. Lactose microfluidic mixer; 61. Protein microfluidic mixer; 62. Temperature controller; 7. Milk inlet; 8. Protein reagent inlet; 9. Heat sink; 10. LED light source; 11. Milk inlet hose interface; 12. Support platform; 13. Image recognition unit; 14. Lactose mixing area; 15. Protein mixing area; 16. Protein color development area; 17. Lactose color development area; 18. LCD display screen; 19. Serial port; 20. Lid; 21. Box body; 22. Battery; 23. Opening; 24. Detailed Implementation

[0032] To enable those skilled in the art to fully understand the technical solutions of the present invention, the specific structure and operation process of the milk dual-measurement device based on microfluidic technology are described in detail below with reference to embodiments and accompanying drawings. It should be noted that although many specific details are set forth below to ensure a full understanding of the present invention, the present invention can still be implemented using other technical methods different from those described herein; therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below. The technical solutions of some embodiments of the present invention are described below with reference to the accompanying drawings.

[0033] Example: Figures 1-3 As shown in the figure, this embodiment of the invention provides an online multi-component detection device for milk based on microfluidic technology, including a microfluidic chip 1, which integrates two microfluidic detection channels, namely a lactose detection channel and a protein detection channel in this example. Each microfluidic detection channel includes a sample pool 5 for accommodating the milk sample to be tested, a lactose reagent injection area 4 / protein reagent injection area 3, a lactose mixing area 15 / protein mixing area 16, a lactose / protein microfluidic mixer 61 / 62 with a angularly coupled sawtooth-shaped flow channel, and a lactose detection... The colorimetric zone 18 and protein detection colorimetric zone 17 are configured such that the sample cells 5 of the two detection channels are interconnected through the injection channels; the outlet of the sample cell 5 of each detection channel merges with the outlet of the corresponding lactose reagent injection zone 4 / protein reagent injection zone 3 at the lactose mixing zone 15 / protein mixing zone 16; the injection end of the lactose / protein microfluidic mixer 61 / 62 is connected to the lactose / protein mixing zone 15 / 16, and the outlet end is connected to the lactose detection colorimetric zone 18 / protein detection colorimetric zone 17; An optical detection module is positioned above each lactose / protein detection color development area 18 / 17 to acquire color images of the reaction solution within the lactose / protein detection color development area 18 / 17 and extract RGB color feature values. The data processing and control module is electrically connected to the optical detection module and is used to receive the RGB color feature values ​​and convert them into the concentration information of the corresponding components according to its preset algorithm.

[0034] The lactose reagent injection area 4 and the protein reagent injection area 3 are respectively connected to the lactose reagent injection port 2 and the protein reagent injection port 9, and the milk sample pool 5 is connected to the milk injection port 8.

[0035] The microfluidic chip 1 described in this application is the core reaction carrier of the entire detection device. It is placed on the support platform 13 inside the housing 22. A milk sample inlet hose interface 12 is provided on the housing 22, which is connected to the milk sample cell 5 for introducing the milk sample to be tested into the microfluidic chip 1. In this example, the lactose / protein mixing zone 15 / 16 has a height of 3.5 mm and a diameter of 7.4 mm, resulting in good color development.

[0036] The structure of the microfluidic mixer described in this example was obtained through simulation and comparative optimization, which enables efficient mixing of samples and reagents at low flow rates.

[0037] The specific simulation comparison and optimization methods are as follows: The hybrid efficiency of three buffer channel models—angular, serpentine, and straight—was quantitatively compared using finite element simulation software. The angular buffer channel is the angular coupled sawtooth structure described in this example. Figure 3 , 4 As shown in Figure 5, adjacent mixing channels are connected by angular bends; the serpentine buffer channel is a continuous S-shaped channel with angular coupling, as shown in Figure 5; adjacent mixing channels are connected by arc bends; the straight buffer channel is a straight channel with angular coupling, as shown in Figure 5. Figure 6 As shown, the mixing channels are in a straight line configuration.

[0038] All three channels are made of PDMS material and fabricated using soft photolithography. The effective mixing length is uniformly 5 cm, the bifurcation channel has a uniform width of 0.5 mm, the remaining channels have a uniform width of 1 mm, and the height is uniformly 2 mm. The inlet pressure difference is precisely controlled by a micropump. The fluid is set as a milk dilution (viscosity 1.2 mPa·s, flow rate range 2-7 μL / s). The mixing efficiency evaluation index is defined as the reciprocal of the variance of the concentration at the channel outlet section.

[0039] like Figures 4-6 As shown, the color change represents the mixing process: the two fluids tend to become homogeneous: "two colors become one" is the function of the microfluidic mixer: after the sample and reagent come together in the mixing zone, they are divided and folded through the "angular structure buffer channel" to achieve efficient mixing, so that the two fluids corresponding to the two colors gradually merge into a single-color mixture.

[0040] like Figure 7 As shown, the simulation results of the mixing efficiency of the three buffer channels under the condition of a flow rate of 2 μL / s are as follows: The angular buffer channel with the angular coupling sawtooth structure described in this example achieves a mixing efficiency of 0.86, with the best concentration uniformity at the outlet section. The serpentine buffer channel has a mixing efficiency of 0.80, but exhibits local eddy currents. The straight buffer channel has a mixing efficiency of 0.74, but shows obvious concentration stratification. When the flow rate is increased to 7 μL / s: the mixing efficiency of the angular buffer channel increases to 0.99, approaching a complete mixing state; the serpentine buffer channel reaches 0.96, with its streamlined design reducing turbulence losses (such as...). Figure 5 (As shown); the linear channel only increases to 0.89, resulting in limited convective mixing.

[0041] Multiphysics coupling simulation revealed that: Figure 7As shown, the angular buffer channel maintains the highest mixing efficiency throughout the 2-7 μL / s flow rate range. Its 60° bend angle (α) generates the Dean vortex effect, significantly enhancing convective mixing between fluid micro-elements. At the optimal operating condition of 5 μL / s, it is close to saturation, with an outlet concentration variance of only 0.01, which is 40% lower than the serpentine buffer channel and 65% lower than the straight buffer channel. Therefore, this application determines 5 μL / s as the flow rate for detection and identifies the angular buffer channel structure as the optimal microfluidic mixer structure.

[0042] like Figure 3 As shown, the angularly coupled sawtooth structure of the microfluidic mixer consists of multiple mixing channels arranged periodically along the flow direction, with adjacent mixing channels having bending angles to form a sawtooth structure; each mixing channel has two symmetrical triangular protrusions and branching channels.

[0043] In this example, the bending angle α of the mixing channel is 60°. The bending angle β at the corner of the triangular protrusion of the bifurcated channel is 90°, the bending angle γ between the bifurcated channel and the bending channel is 135°, and the bending angle δ between the bending channel and the sample outlet channel / inlet channel of the mixing zone / detection display zone is 135°. By generating a chaotic convection effect, the mixing effect is enhanced, thereby ensuring that the protein and Coomassie Brilliant Blue reagent, and lactose and DNS reagent can react fully in a short time.

[0044] Of the two detection channels, excluding the bifurcation channel 61, the channel width is 1 mm and the channel height is 2 mm; the bifurcation channel has a width of 0.5 mm and a height of 2 mm. The flow rate during detection is 3 μL / s to 5.5 μL / s.

[0045] like Figure 2 As shown, an opening 24 is provided on the microfluidic detection chip 1 between the lactose detection color development area 18 and the protein detection color development area 17 of the two detection channels. The length L of the opening 24 is 1 / 3 of the length of the microfluidic detection chip 1, and the distance W from the two sides of the opening 24 to the edges of the two detection color development areas is 2mm.

[0046] This application also includes a temperature control module for heating and closed-loop temperature control of the two detection channels of the microfluidic chip 1. The module includes a temperature controller 7 and two heating elements with integrated temperature sensors. The two heating elements cover the bottom of the two detection channels of the microfluidic chip 1. The temperature controller 7 is connected to a data processing and control module. The temperature of the two detection channels is collected in real time by the temperature sensors, and the data processing and control module controls the heating temperature of the two heating elements respectively through the temperature controller 7. The temperature of the lactose detection channel is controlled at 95±0.5℃; the temperature of the protein detection channel is controlled at 37℃±0.5℃.

[0047] The heating element can be an existing ceramic heating element; an aluminum heat sink 10 is also provided on the housing 22 to dissipate heat when the temperature is too high, preventing the temperature from exceeding the set range and affecting the detection results. The temperature control module implements closed-loop temperature control through the STM32 microcontroller of the data processing and control module, stabilizing the temperature of the microfluidic chip 1 at the optimal temperature for the colorimetric reaction.

[0048] The optical detection module includes an OpenMV image recognition unit 14 and an LED light source for illuminating it. The OpenMV image recognition unit 14 is positioned above the lactose detection color development area 18 and the protein detection color development area 17 of the two detection channels, and is used to acquire color images after the color development reaction. The LED light source 11 is placed on the side wall of the housing 22 to provide stable illumination and ensure the clarity and consistency of image acquisition. The OpenMV image recognition unit 14 is a high-definition camera connected to the STM32 microcontroller of the data processing and control module. It can extract RGB values ​​from the acquired image, thereby converting the color information into digital signals and transmitting them to the data processing and control module.

[0049] The data processing and control module includes a microcontroller and a host computer that communicates with it. The microcontroller consists of an STM32 microcontroller and a connected LCD display. The STM32 microcontroller is electrically connected to the optical detection module, specifically through a UART serial port 20 on the housing, for receiving the RGB color feature values. The STM32 microcontroller is responsible for two core tasks: first, controlling the reaction temperature while processing and transmitting data; second, analyzing the received RGB color feature values. The host computer compares the RGB value combinations corresponding to different concentrations of standard samples, fits the linear relationship between protein and lactose concentrations and RGB values, and plots standard curves. The microcontroller can directly call the built-in standard curve fitting formula to calculate the concentration information of the corresponding components based on the received RGB values ​​and display it on the LCD display.

[0050] The processed test results are displayed in real time on the LCD screen 19, and simultaneously uploaded to the OneNET cloud platform via the WiFi communication module 15 to achieve remote storage and viewing of the data.

[0051] The housing 22 of the present invention has an openable cover 21 on one side, and a battery 23 is provided inside the housing 22 to connect the data processing and control module, the optical detection module and the temperature control module, and to provide power to them.

[0052] The detection method using the online detection device described in this application includes the following steps: S1. Sample introduction and distribution: The milk sample to be tested is introduced into the milk inlet 8 through the milk sample introduction hose interface 12, and simultaneously distributed to the sample pools 5 of the two detection channels through the sample introduction channel; the lactose detection reagent (DNS reagent) and the protein detection reagent (Coomassie brilliant blue reagent) are added to the corresponding lactose reagent injection area 4 / protein reagent injection area 3 through the lactose reagent injection port 2 / protein reagent injection port 9 respectively; S2. Mixing and colorimetric reaction: The milk sample is injected at a flow rate of 3 μL / s to 5.5 μL / s, so that the milk sample and the corresponding reagent in each detection channel are combined in the lactose mixing zone 15 / protein mixing zone 16 at a ratio of 1:2. After being efficiently mixed by the corresponding angularly coupled sawtooth structure lactose / protein microfluidic mixers 61 / 62, they flow into the lactose detection colorimetric zone 18 / protein detection colorimetric zone 17 respectively. Under the control of the temperature control module, the temperature of the lactose detection channel is maintained at 95℃±1℃, and the temperature of the protein detection channel is maintained at 37℃±0.5℃; as the concentrations of protein and lactose change, the color of the solution after the color reaction gradually deepens.

[0053] S3: Optical Detection and Data Processing: The LED light source 11 is turned on to provide uniform illumination for the protein color development area 17 and the lactose color development area 18. The OpenMV image recognition unit 14 simultaneously acquires images of the color development results of the protein color development area 17 and the lactose color development area 18 and extracts the RGB values. The extracted RGB values ​​are transmitted to the STM32 microcontroller via the UART serial port 20. The STM32 microcontroller compares the received RGB values ​​with the pre-stored standard curve in the microcontroller to calculate the protein concentration in the milk sample to be tested as 2.8g / 100mL-3.8g / 100mL and the lactose concentration as 4.5g / 100mL-5.2g / 100mL.

[0054] Step 4: Display and upload of test results: The STM32 microcontroller sends the calculated protein and lactose concentration values ​​to the LCD display 19 for real-time display. At the same time, it uploads the test data (including RGB values, protein concentration, lactose concentration, and test time) to the OneNET cloud platform via the WiFi communication module to achieve remote monitoring and management of the data.

[0055] The aforementioned device and method enable the synergistic detection of protein and lactose in milk using a microfluidic chip as a carrier. This device boasts high integration, ease of operation, and rapid detection speed, meeting the demands for rapid milk quality testing. Furthermore, the OpenMV module facilitates automated identification of the colorimetric reaction results, while the WiFi module enables wireless transmission of detection data, enhancing the intelligence level of the detection process.

Claims

1. A device for on-line detection of multiple components in milk based on microfluidic technology, characterized in that: The microfluidic detection chip comprises two microfluidic detection channels integrated thereon, each of which comprises a sample pool for containing a milk sample to be detected, a reagent injection area, a mixing area, a microfluidic mixer with an angularly coupled zigzag-shaped flow channel, and a detection chromogenic area, wherein the sample pools of the two detection channels are connected to each other through an injection channel, and the outlet of the sample pool of each detection channel merges with the outlet of the corresponding reagent injection area at the mixing area; the injection end of the microfluidic mixer is connected to the mixing area, and the outlet end is connected to the detection chromogenic area; An optical detection module is arranged above each detection chromogenic area, which is used to collect the color image of the reaction liquid in the detection chromogenic area and extract the RGB color feature; A data processing and control module is electrically connected with the optical detection module, which is used to receive the RGB color feature value and convert it into the concentration information of the corresponding component according to a preset algorithm.

2. The online detection device of claim 1, wherein, The angularly coupled zigzag-shaped flow channel of the microfluidic mixer is composed of a plurality of mixing channels arranged periodically along the flow direction, and adjacent mixing channels have a bending angle, forming a zigzag-shaped structure; each mixing channel has two symmetrical triangular protruding bifurcation channels.

3. The online detection device of claim 2, wherein, The bending angle α of the mixing channel is 60°.

4. The online detection device of claim 2, wherein, The bending angle β of the corner of the triangular protruding bifurcation channel is 90°, the bending angle γ between the bifurcation channel and the bending channel is 135°, and the bending angle δ between the bending channel and the outlet channel of the mixing area or the injection channel of the detection display area is 135°.

5. The online detection device of claim 4, wherein, The channel width of the two detection channels except the bifurcation channel is 1 mm, and the channel height is 2 mm; the width of the bifurcation channel is 0.5 mm; and the height is 2 mm.

6. The online detection device of claim 1, wherein, The milk sample pool and the milk injection port are respectively arranged on the sample injection channel of the two detection channels; the height of the mixing area is 3.5 mm, and the diameter is 7.4 mm.

7. The online detection device of claim 1, wherein, An opening is arranged on the microfluidic detection chip between the two detection channels, and the length L of the opening is 1 / 3 of the length of the microfluidic detection chip, and the distance W between the two sides of the opening and the edges of the two chromogenic areas is 2 mm.

8. The online detection device of claim 1, wherein, A temperature control module is further provided for heating and closed-loop temperature control of the two detection channels of the microfluidic detection chip, which comprises a temperature controller and two heating sheets integrated with temperature sensors, the two heating sheets separately cover the bottoms of the two detection channels of the microfluidic chip, and the temperature controller is connected with the data processing and control module; the temperature of the two detection channels is collected in real time through the temperature sensors, and the data processing and control module controls the heating temperature of the two heating sheets through the temperature controller.

9. The online detection device of claim 8, wherein, The two detection channels are a lactose detection channel and a protein detection channel, wherein the reagent injection area of the lactose detection channel is DNS reagent, and the temperature control is at 95±0.5℃; the reagent injection area of the protein detection channel is Coomassie brilliant blue reagent, and the temperature control is at 37℃±0.5℃.

10. A detection method using the online detection device according to any one of claims 1 to 9, characterized by, The method comprises the following steps: S1. Injection and distribution: the milk sample to be detected is simultaneously distributed to the sample pools of the two detection channels through the injection channel; Coomassie brilliant blue reagent and DNS reagent are respectively added to the corresponding reagent injection area; S2. Mixing and color development reaction: drive the fluid at a flow rate of 3-5.5 μL / s, so that the milk sample in each channel and the corresponding reagent are mixed in the mixing area at a set ratio, then flow through the microfluidic mixer of the angular coupling zigzag structure for efficient mixing, and then flow into the lactose detection color development area / protein detection color development area, respectively; Under the action of the temperature control module, the temperature of the lactose detection channel is maintained at 95℃±1℃, and the temperature of the protein detection channel is maintained at 37℃±0.5℃; S3. Optical detection and data processing: the color images of the two detection color development areas are synchronously collected by the optical detection module, the RGB characteristic values are extracted, and the concentrations of protein and lactose are calculated and displayed according to the preset algorithm by the data processing and control module.

Images