Automatic sampling device for chromatographic instrument based on vision and voice recognition

Abstract

The invention discloses an automatic sampling device for a chromatographic instrument based on vision and voice recognition, and belongs to the technical field of chromatographic instruments. Comprising a main control board, and the main control board is connected with a PC end, a visual recognition system, a voice recognition system, a three-axis control module, an injection pump module, a switching valve module and a sample injection valve module. The visual identification system is used for collecting a sample bottle image and identifying the position and the liquid level height of the sample bottle; the voice recognition system is used for recognizing and recording sample information recorded by experimenters through voice; the main control board is used for receiving and processing identification information transmitted by the visual identification system and the voice identification system, controlling the movement of the sample disc and the sampling needle according to the identification information, completing automatic sample injection operation and controlling all the modules to work cooperatively; and the main control board compares the visual identification information with the voice identification information, and performs automatic sample injection through an automatic sample injection method. According to the invention, the input can be ensured to be correct through mutual verification of vision and voice.

Description

Technical Field

[0001] The present invention relates to the technical field of chromatographs, and in particular, to an automatic sampling device for a chromatograph based on vision and speech recognition. Background Art

[0002] Liquid chromatographs and ion chromatographs are commonly used separation and analysis instruments in the field of analytical chemistry, and their sampling systems directly affect the accuracy and repeatability of analysis results. Traditional liquid automatic samplers mostly use mechanical positioning and sensor detection methods, which have the following deficiencies: 1. Limited precision: The mechanical positioning precision is limited by machining and assembly precision, and it is difficult to meet the requirements of high-precision analysis.

[0003] 2. Poor adaptability: It has poor adaptability to sample bottles of different specifications and shapes, and it is necessary to replace fixtures or adjust parameters, which is cumbersome to operate.

[0004] 3. High failure rate: Mechanical components are prone to wear, and sensors are susceptible to environmental interference, resulting in a high failure rate and high maintenance costs.

[0005] 4. Cumbersome operation: Experimental personnel need to input sample information for all sample vials in chromatographic software or automatic sampler control software. Summary of the Invention

[0006] The purpose of the present invention is to overcome the deficiencies of the prior art and provide an automatic sampling device for a chromatograph based on vision and speech recognition.

[0007] The purpose of the present invention is achieved by the following technical solutions: An automatic sampling device for a chromatograph based on vision and speech recognition, including a main control board, the main control board is connected to a PC terminal, a vision recognition system, a speech recognition system, a three-axis control module, an injection pump module, a switching valve module, and a sampling valve module; The visual recognition system includes a camera, a light source, and an image processing unit, which are used to collect images of sample bottles and identify the positions of sample bottles and the liquid levels; the voice recognition system includes a microphone and a voice recognition processing unit, which are used to recognize and record the sample information input by the experimenter through voice; the main control board is used to receive and process the recognition information transmitted by the visual recognition system and the voice recognition system, and control the movement of the sample tray and the sampling needle according to the recognition information to complete the automatic sample injection operation and control the coordinated work of each module; the three-axis control module is used to control the sampling position, and the stepping motor or servo motor is driven by the main control board to accurately move to achieve the positioning of the sampling position and the aspiration of samples; the sample tray is used to place sample bottles and can rotate or translate to achieve the switching of different sample bottles; the sampling needle is used to aspirate samples and can move in the horizontal or vertical direction to achieve positioning; the injection pump module is used to aspirate or discharge the samples in the sample bottles and the cleaning liquid in the cleaning liquid bottle; the switching valve module is used to switch the aspiration and discharge functions of the injection pump; the injection valve module is connected between the infusion pump and the chromatographic column and is used to transfer the sample to the chromatographic system; The main control board compares the visual recognition information and the voice recognition information with each other and performs automatic sample injection through an automatic sample injection method.

[0008] Preferably, a plurality of the cameras and light sources are provided for real-time monitoring of the movement states of all movement modules.

[0009] Preferably, the sample information includes the sample name, sample type, real-time recording information entry time, sample injection volume, and sample injection method.

[0010] Preferably, a cleaning tank is also provided for cleaning the sampling needle. According to the sample information recognized by voice and vision, the corresponding cleaning mode is automatically formed and the corresponding cleaning action is executed.

[0011] Preferably, the sampling needle is made of stainless steel with an inert coating and has holes on the side.

[0012] Preferably, the injection pump module includes a glass syringe and a motor screw drive module.

[0013] Preferably, the injection valve module is an electrically controlled high-pressure two-position six-way valve. When in the sample loading position, the sample is loaded precisely through the movement of the injection pump module; when switched to the injection position, the loaded sample is transferred to the chromatographic system and enters the chromatographic column for separation and subsequent detection.

[0014] Preferably, a cleaning liquid bottle and a waste liquid bottle are also provided; the cleaning liquid bottle is used to store the cleaning liquid, and the cleaning liquid is used to rinse the inner and outer walls of the sampling needle and all connecting pipelines of the device; the waste liquid bottle is used to store all waste liquids generated by the device.

[0015] Preferably, the automatic sampling method includes the following steps: Place the sample vial on the sample tray, input the placed sample information through microphone voice recording, and transmit it to the host computer software after being recognized by the main control board; The vision recognition system monitors the inside of the device cavity in real time and transmits the image information to the host computer software; The host computer software compares the information input by voice with the information of the sample vials on the sample tray recognized by vision. After the comparison is correct, an injection sequence is automatically generated; when an error is found in the comparison, an error prompt message is issued, and an injection sequence is automatically generated after correction; After the host computer software clicks the start acquisition button, the device starts the injection operation; The main control board controls the rotation or translation of the sample tray seat, controls the accurate movement of the sampling needle arm, and the sampling needle pierces into the sample tray to aspirate the sample; at the same time, the camera real-time collects the movement information of all modules and the status of the sample vial to achieve accurate injection; After sampling, the injection valve module is switched to the injection position, and the loaded sample is transferred to the chromatographic system for subsequent separation and detection.

[0016] Preferably, a plurality of the sample trays are provided.

[0017] The beneficial effects of the present invention are: 1) According to the sample information input by the user's voice, it is compared with the image of the sample vial position monitored by vision recognition, and mutual verification can ensure that the input is correct.

[0018] 2) High precision: By using vision recognition technology, accurate positioning of the sample vial position and liquid level height can be achieved, improving the injection accuracy.

[0019] 3) Strong adaptability: It can automatically identify sample vials of different specifications and shapes, without the need to replace fixtures or adjust parameters, and the operation is simple.

[0020] 4) Low failure rate: Reduce the use of mechanical components and sensors, reduce the failure rate, and improve the reliability of the instrument.

[0021] 5) Intelligence: It can realize functions such as automatic identification, automatic positioning, and automatic injection, improving the analysis efficiency.

[0022] 6) Automation: There is no need for the user to manually input a large amount of sample information, and all the sample information to be measured is automatically input through voice recognition.

[0023] 7) High stability: Through real-time monitoring captured by vision, the internal state is detected at any time, the movement position is calibrated in time, and instrument operation errors are prevented from occurring. Description of the Drawings

[0024] Figure 1Schematic block diagram in an embodiment of the present invention; Figure 2 Schematic block diagram in another embodiment of the present invention. Detailed implementation manners

[0025] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts fall within the protection scope of the present invention.

[0026] Refer to Figure 1 - Figure 2 , the present invention provides a technical solution: an automatic sampling device for a chromatograph based on vision and speech recognition, including a main control board, and the main control board is connected to a PC terminal, a vision recognition system, a speech recognition system, a three-axis control module, an injection pump module, a switching valve module, and a sampling valve module; The vision recognition system includes a camera, a light source, and an image processing unit, and is used to collect the image of the sample bottle and identify the position and liquid level height of the sample bottle; the speech recognition system includes a microphone and a speech recognition processing unit, and is used to recognize and record the sample information input by the experimenter through voice; the main control board is used to receive and process the recognition information transmitted by the vision recognition system and the speech recognition system, and control the movement of the sample tray and the sampling needle according to the recognition information to complete the automatic sampling operation, and control the coordinated work of each module; the three-axis control module is used to control the sampling position, and drives a stepping motor or a servo motor through the main control board to accurately move to achieve the positioning of the sampling position and the aspiration of the sample; the sample tray is used to place the sample bottle and can rotate or translate to achieve the switching of different sample bottles; the sampling needle is used to aspirate the sample and can move in the horizontal or vertical direction to achieve positioning; the injection pump module is used to aspirate or discharge the sample in the sample bottle and the cleaning liquid in the cleaning liquid bottle; the switching valve module is used to switch the aspiration and discharge functions of the injection pump; the sampling valve module is connected between the infusion pump and the chromatographic column and is used to transfer the sample to the chromatographic system; The main control board compares the vision recognition information and the speech recognition information with each other and performs automatic sampling through an automatic sampling method.

[0027] In this embodiment, the sample sequence is input through voice recognition; the set sample information is accurately compared through visual recognition information and voice input information to ensure the accuracy of the information of all samples during the test of a large number of samples. All motion modules of the device (including: X-axis - sampling needle arm, Y-axis - sample tray base, Z-axis - sampling needle, injection pump, switching valve, injection valve, etc.) are accurately motion-controlled by the main control board driving a stepper motor or a servo motor. The sample information is input through voice recognition of the microphone and transmitted to the main control board, and the main control board transmits the input information to the PC for input. The visual recognition system monitors the image information inside the cavity of the automatic sampler in real time, compares all the input information of the automatic sampler with the image, monitors the actions of all motion modules of the automatic sampler and the liquid level in the sample bottle in real time, and feeds back the monitored information to the main control board and the PC in real time. The comparison and verification are carried out through the control software of the PC to ensure that all information and motions are correct. The main control board is used for motion control and status monitoring of all modules in the device, collection of voice and visual graphics, control of motion logic, and interaction with the upper computer software of the PC.

[0028] The visual recognition system uses a deep learning algorithm for image recognition. Visual recognition starts from when the sample bottle is placed in the sample tray and is carried out in real time to ensure correct sampling of the device. The sample information input from the voice end is compared with the position of the sample bottle recognized visually to ensure correct placement of the sample and avoid the risk of empty bottles or misplacement. After the combination and comparison of voice and visual recognition, the upper computer software automatically generates a sampling sequence. The visual recognition system monitors the liquid level height in the sample bottle in real time and compares it with the sampling volume of the sample bottle in the sampling sequence to ensure that there is enough sample in the bottle for the number of sampling times and sampling volume edited in the sampling sequence and prevent the phenomenon of aspirating an empty needle. The visual recognition system monitors the operating state of the injection pump in real time and records and feeds back in time when a large section of air bubbles is inhaled or the pump runs dry.

[0029] The visual recognition system monitors the action state of the sampling needle in real time to ensure that each sampling is of the target sample bottle and that each sampling is inserted from the center of the sampling septum of the sample bottle. According to the sampling sequence generated by voice and vision, the sampling system is automatically cleaned according to the sample type and sampling volume. The cleaning action is carried out during the sampling interval time to save analysis time. The visual recognition system monitors the remaining content of the cleaning liquid in the sampling device in real time and gives a prompt when the content is insufficient. The visual recognition system monitors the liquid leakage situation in the sampling device in real time, focuses on monitoring the easy-to-leak points, makes real-time comparison, and gives an alarm or performs corresponding protection actions in time when a liquid leakage phenomenon is found. Sample information can be batch-input when input by voice recognition. For example: when the voice input is "Samples 1 to 75 from the left 1 to the left 75 are unknown samples 1 to 75", 75 sampling tasks of the type "unknown sample" are automatically generated in the sampling list. The visual recognition system can automatically identify the type of sample bottle placed in the sampling device by the user.

[0030] In some embodiments, a plurality of the cameras and light sources are provided for real-time monitoring of the motion states of all motion modules.

[0031] In this embodiment, the vision recognition system includes one or more cameras, one or more light sources, and an image processing unit, which are used to collect sample bottle images, collect information such as the position of the sample bottle and the liquid level height of the sample bottle, and perform real-time monitoring of the motion states of all motion modules. One or more cameras perform real-time monitoring of all the modules to be monitored; the light source is used for internal lighting, and the light source switch can be controlled and the light source brightness can be adjusted according to the image information collected in real time; the image processing unit performs the recognition and processing of the images collected by the cameras, and accurately controls the device through the collected image information.

[0032] In some embodiments, the sample information includes the sample name, sample type, real-time recording information entry time, sample injection volume, and sample injection method.

[0033] In this embodiment, the voice recognition system is used to recognize all the sample information input by the user's voice, including the sample name, sample type, real-time recording information entry time, sample injection volume, and sample injection method (including whether it is full-loop, sampling needle cleaning method, etc.). The input information is transmitted to the main control board for parsing, and is uploaded to the PC-side software to compare and verify the input information with the image information recognized by the camera module.

[0034] In some embodiments, a cleaning tank is further provided for cleaning the sampling needle, and according to the sample information recognized by voice and vision, a corresponding cleaning mode is automatically formed and the corresponding cleaning action is executed.

[0035] In this embodiment, it is not necessary for the user to input different cleaning methods for different samples respectively, which also avoids wasting time on repeated cleaning and improves the test efficiency.

[0036] In some embodiments, the sampling needle is made of stainless steel with an inert coating, and holes are opened on the side of the sampling needle.

[0037] In this embodiment, the sampling needle is used to accurately aspirate the sample, and the sampling needle can move in the horizontal and vertical directions to achieve precise positioning. The sampling needle is made of stainless steel with an inert coating to prevent liquid droplets from hanging and cross-contamination to the greatest extent. The holes on the side of the sampling needle prevent the needle from being blocked.

[0038] In some embodiments, the syringe pump module includes a glass syringe and a motor screw drive module.

[0039] In this embodiment, the syringe pump module is used to accurately control the sample aspiration volume, and the syringe pump module is composed of a high-precision glass syringe and a motor screw drive module.

[0040] In some embodiments, the sample injection valve module is an electronically controlled high-pressure two-position six-way valve. When in the sample loading position, the sample is loaded precisely through the movement of the injection pump module; when switched to the injection position, the loaded sample is transferred to the chromatographic system and enters the chromatographic column for separation and subsequent detection.

[0041] In this embodiment, the sample loading position is the Load position, and the injection position is the Inject position.

[0042] In some embodiments, a cleaning solution bottle and a waste liquid bottle are also provided; the cleaning solution bottle is used to store the cleaning solution, and the cleaning solution is used to rinse the inner and outer walls of the sampling needle and all the connecting pipelines of the device; the waste liquid bottle is used to store all the waste liquids generated by the device.

[0043] In this embodiment, the cleaning solution bottle is used to store the cleaning solution of the automatic sampler. The cleaning solution is used to rinse the inner and outer walls of the sampling needle and all the connecting pipelines of the device to prevent cross-contamination between different samples and achieve accurate sample injection.

[0044] In some embodiments, the automatic sampling method includes the following steps: Place the sample bottle on the sample tray, input the information of the placed sample through microphone voice, and after being recognized by the main control board, transmit it to the host computer software; The vision recognition system monitors the inside of the device cavity in real time and transmits the image information to the host computer software; The host computer software compares the information input by voice with the information of the sample bottles on the sample tray recognized by vision. After the comparison is correct, an injection sequence is automatically generated; when an error is found in the comparison, an error prompt message is sent, and after correction, an injection sequence is automatically generated; After the host computer software clicks the start acquisition button, the device starts the injection operation; The main control board controls the rotation or translation of the sample tray seat and the accurate movement of the sampling needle arm. The sampling needle pierces into the sample tray to aspirate the sample; at the same time, the camera collects the movement information of all modules and the status of the sample bottles in real time to achieve accurate sample injection; After sampling is completed, the sample injection valve module is switched to the injection position to transfer the loaded sample to the chromatographic system for subsequent separation and detection.

[0045] In some embodiments, multiple sample trays are provided.

[0046] The above are only the preferred embodiments of the present invention. It should be understood that the present invention is not limited to the forms disclosed herein, and should not be regarded as excluding other embodiments, but can be used in various other combinations, modifications and environments, and can be changed within the scope of the concept described herein through the above teachings or the techniques or knowledge in related fields. Any changes and modifications made by those skilled in the art without departing from the spirit and scope of the present invention shall fall within the protection scope of the appended claims of the present invention.

Claims

1. An automatic sampling device for a chromatograph based on vision and speech recognition, characterized in that: It includes a main control board, which is connected to a PC, a vision recognition system, a voice recognition system, a three-axis control module, an injection pump module, a switching valve module, and a sampling valve module; The vision recognition system includes a camera, a light source, and an image processing unit, which is used to collect sample bottle images and identify the position and liquid level height of the sample bottle; the voice recognition system includes a microphone and a voice recognition processing unit, which is used to identify and record the sample information input by the experimenter through voice; the main control board is used to receive and process the recognition information transmitted by the vision recognition system and the voice recognition system, and control the movement of the sample tray and the sampling needle according to the recognition information to complete the automatic sampling operation, and control the coordinated work of each module; the three-axis control module is used to control the sampling position, and drive a stepper motor or a servo motor through the main control board to accurately move to achieve sampling position positioning and sample aspiration; the sample tray is used to place sample bottles and can rotate or translate to achieve the switching of different sample bottles; the sampling needle is used to aspirate samples and can move in the horizontal or vertical direction to achieve positioning; the injection pump module is used to aspirate or discharge samples in the sample bottle and cleaning liquid in the cleaning liquid bottle; the switching valve module is used to switch the aspiration and discharge functions of the injection pump; the sampling valve module is connected between the infusion pump and the chromatographic column and is used to transfer the sample to the chromatographic system; The main control board compares the vision recognition information and the voice recognition information with each other and performs automatic sampling through an automatic sampling method.

2. The automatic sampling device for a chromatograph based on visual and speech recognition according to claim 1, characterized in that: A plurality of the cameras and light sources are provided for real-time monitoring of the movement states of all movement modules.

3. The automatic sampling device for a chromatograph based on vision and speech recognition according to claim 1, characterized in that: The sample information includes the sample name, sample type, real-time recording information entry time, sample injection volume, and sample injection method.

4. The automatic sampling device for a chromatograph based on visual and speech recognition according to claim 1, characterized in that: A cleaning tank is also provided for cleaning the sampling needle. According to the sample information recognized by voice and vision, a corresponding cleaning mode is automatically formed and the corresponding cleaning action is executed.

5. The automatic sampling device for a chromatograph based on vision and speech recognition according to claim 1, wherein: The sampling needle is made of stainless steel with an inert coating and has holes on the side.

6. The automatic sampling device for a chromatograph based on visual and speech recognition according to claim 1, characterized in that: The injection pump module includes a glass syringe and a motor screw drive module.

7. The automatic sampling device for a chromatograph based on visual and voice recognition according to claim 1, characterized in that: The sampling valve module is an electrically controlled high-pressure two-position six-way valve. When in the sample loading position, the sample is loaded through the precise movement of the injection pump module; when switched to the injection position, the loaded sample is transferred to the chromatographic system and enters the chromatographic column for separation and subsequent detection.

8. The automatic sampling device for a chromatograph based on vision and speech recognition according to claim 1, characterized in that: A cleaning liquid bottle and a waste liquid bottle are also provided; the cleaning liquid bottle is used to store cleaning liquid, and the cleaning liquid is used to rinse the inner and outer walls of the sampling needle and all connecting pipelines of the device; the waste liquid bottle is used to store all waste liquids generated by the device.

9. The automatic sampling device for a chromatograph based on visual and speech recognition according to any one of claims 1-8, characterized in that: The automatic sampling method includes the following steps: Place the sample bottle on the sample tray, input the sample information of the placed sample through the microphone by voice, and transmit it to the host computer software after being recognized by the main control board; The vision recognition system monitors the inside of the device cavity in real time and transmits the image information to the host computer software; The host computer software compares the information input by voice with the sample bottle information on the sample tray recognized by vision. After the comparison is correct, an injection sequence is automatically generated; when an error is found in the comparison, an error prompt message is sent, and after correction, an injection sequence is automatically generated; After the host computer software clicks the start collection button, the device starts the injection action; The main control board controls the rotation or translation of the sample tray base and the accurate movement of the sampling needle arm. The sampling needle pierces into the sample tray to aspirate the sample. At the same time, the camera real-time collects the movement information of all modules and the status of the sample vial to achieve accurate sample injection. After sampling is completed, the injection valve module is switched to the injection position to transfer the loaded sample to the chromatographic system for subsequent separation and detection.

10. The automatic sampling device for a chromatograph based on visual and voice recognition according to any one of claims 1-8, characterized in that: A plurality of the sample trays are provided.

Images