Automatic experiment control system and control method

Abstract

The invention provides an automatic experiment control system and control method. The system comprises a man-machine interaction module; the system comprises a system main control module and a system working module, the system working module comprises a water-free and oxygen-free working module, a non-inert working module and an experiment result analysis module, and the water-free and oxygen-free working module, the non-inert working module and the experiment result analysis module respectively comprise a plurality of sub-working modules. The automatic control system and the control method provided by the invention can be oriented to different types of chemical experiments, and process reconstruction and recombination of each working unit can be realized according to experiment requirements, so that automatic experiment requirements of different types of experiments can be met; the invention provides a qualitative and quantitative fusion process level control method, the effectiveness of the experiment result can be effectively predicted by finely monitoring and controlling the process, so that the invalid experiment is judged and terminated in advance, the consumable consumption is effectively reduced, the invalid experiment time is shortened, and the experiment efficiency is improved.

Description

Automatic experimental control system and control method Technical Field

[0001] This invention relates to an experimental control system and control method, and more particularly to an automatic experimental control system and control method. Background Technology

[0002] Significant progress has been made in the innovation of chemical products in the field of organic synthesis, but the level of automation in laboratory research has lagged behind. To date, despite the introduction of the Chemputer automated synthesis system based on continuous flow reactions and the realization of automated new material synthesis systems based on AGVs and six-axis robotic arms, the full-process automation of organic synthesis still faces many challenges. Existing automated systems still need improvement in terms of efficiency, accuracy, scope, and stability, especially in areas such as solid weighing under anhydrous and oxygen-free conditions, automated post-processing, and high-throughput detection.

[0003] Furthermore, current automated chemical experimental platforms are limited by the concept of traditional automated production. The automated chemical synthesis robots designed are generally fixed-process, which cannot be flexibly configured and have poor scalability, and cannot meet the needs of different types of chemical synthesis experiments. In addition, there is a lack of monitoring and control of the experimental process and corresponding result prediction, and a lack of foresight for experimental failures. This leads to high costs due to wasted experimental consumables, as well as wasted experimental time, resulting in low experimental efficiency. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to overcome the defects in the prior art and provide a chemical synthesis experimental control system and control method that can realize automatic reconfiguration.

[0005] The present invention solves the above-mentioned technical problems through the following technical solution: an automatic experimental control system, comprising:

[0006] The human-computer interaction module is used for setting experimental parameters, issuing control commands, collecting experimental data, monitoring experimental status, and predicting experimental results.

[0007] The system's main control module is used to enable real-time communication between the human-machine interaction module and the IoT gateway, ensuring reliable data upload and download.

[0008] The system also includes a working module, comprising an anhydrous and oxygen-free working module, a non-inert working module, and an experimental result analysis module. Each of the anhydrous and oxygen-free working module, the non-inert working module, and the experimental result analysis module includes multiple sub-working modules for performing various experimental procedures. The sub-working modules collect multiple experimental parameters from the experimental procedures and send them to the system main control module and the human-machine interaction module via the Internet of Things. Each sub-working module includes a distributed processing center.

[0009] Preferably, the sub-modules of the anhydrous and oxygen-free working module include a reagent bottle capping unit, a reagent storage unit, a solid sample dispensing unit, a liquid sample dispensing unit, a reaction bottle capping unit, an experimental reaction unit, and a material transfer unit.

[0010] Preferably, the sub-modules of the non-inert working module include a reagent bottle capping unit, a reagent storage unit, a material storage unit, a solid sample dispensing unit, a liquid sample dispensing unit, a reaction bottle capping unit, an experimental reaction unit, and a material transfer unit.

[0011] Preferably, the sub-modules of the experimental results analysis module include an extraction unit, a filtration unit, and an analysis unit.

[0012] In another aspect, this invention discloses an automatic experimental control method, applied to an automatic experimental control system as described in any of the above-mentioned methods, comprising:

[0013] Based on the aforementioned automatic experimental control system, a Bayesian network model is constructed, and the Bayesian network adopts an expert system approach.

[0014] Qualitative and quantitative analyses of various experimental parameters during the experiment were performed based on the Bayesian network model.

[0015] In response to the prediction results of the Bayesian network model, the experimental process is controlled to be terminated in a timely manner.

[0016] Preferably, the construction of a Bayesian network model based on the automatic experimental control system specifically includes:

[0017] Node mapping, mapped to test nodes, functional failure nodes, failure mode nodes, and component nodes;

[0018] The CF layer Bayesian network is constructed, which includes component nodes and failure mode nodes. The component nodes are the parent nodes, pointing to the failure mode nodes, and the failure mode nodes are the child nodes.

[0019] The FM layer Bayesian network is constructed, which includes failure mode nodes and functional failure nodes. The failure mode nodes are used as parent nodes, pointing to the functional failure nodes they affect, and the functional failure nodes are child nodes.

[0020] The Bayesian network of the MT layer is constructed. This part of the network contains functional failure nodes and test nodes. The functional failure nodes in the FM layer are the parent nodes, pointing to the test nodes they affect, and these test nodes are the child nodes.

[0021] Network aggregation establishes a three-level, four-layer Bayesian network structure by connecting the same nodes in the CF, FM, and MT layers, thereby realizing the construction of the overall Bayesian network model.

[0022] Preferably, the node mapping step specifically includes,

[0023] The experiment is mapped as a test ontology to a test node, and the experiment includes multiple experimental procedures.

[0024] The failure entity belonging to functional failure is mapped to the functional failure node. The failure entity belonging to functional failure refers to the experimental procedure that causes functional failure based on the failure mode.

[0025] The failure ontology belonging to the failure mode type is mapped to the failure mode node. The failure ontology belonging to the failure mode type is the experimental failure factor, corresponding to the experimental data collected in each unit.

[0026] Each unit body is selected and mapped as a component node. The unit body is a number of sub-working modules of the anhydrous and oxygen-free working module, the non-inert working module, and the experimental result analysis module.

[0027] In another aspect, the present invention provides an automatic experimental control device, comprising,

[0028] The network construction module is used to construct a Bayesian network model based on the automatic experimental control system, wherein the Bayesian network adopts an expert system approach.

[0029] The analysis module is used to perform qualitative and quantitative analysis of various experimental parameters during the experiment based on the Bayesian network model.

[0030] The program control module is used to control the timely termination of the experimental process in response to the prediction results of the Bayesian network model.

[0031] In another aspect, the present invention provides an electronic device including a memory, a processor, and a computer program stored in the memory and for running on the processor, wherein the processor executes the computer program to implement the automatic experimental control method described in any one of the above descriptions.

[0032] In another aspect, the present invention provides a computer program product comprising a computer program that, when executed by a processor, implements the automatic experimental control method as described in any one of the above descriptions.

[0033] Based on common knowledge in the field, the above-mentioned preferred conditions can be combined arbitrarily to obtain various preferred embodiments of the present invention.

[0034] The positive and progressive effects of this invention are as follows: The automatic control system and control method provided by this invention can be adapted to different types of chemical experiments, and can realize the process-based reconfiguration and reorganization of each working unit according to the experimental requirements to meet the automation experimental needs of different types of experiments; it provides a process-level control method that integrates qualitative and quantitative analysis, which can effectively predict the validity of experimental results through refined monitoring and control of the process, thereby determining and terminating failed experiments in advance, effectively reducing the consumption of consumables and reducing the time of invalid experiments, thereby improving experimental efficiency. Attached Figure Description

[0035] Figure 1 is a framework diagram of an automatic experimental control system provided in Embodiment 1 of the present invention;

[0036] Figure 2 is a flowchart of an automatic experimental control method provided in Embodiment 2 of the present invention;

[0037] Figure 3 is a flowchart illustrating step S10 in the method provided in Embodiment 2 of the present invention;

[0038] Figure 4 is a schematic diagram of the Bayesian network model provided in Embodiment 2 of the present invention;

[0039] Figure 5 is a schematic diagram of the automatic experimental control device provided in Embodiment 3 of the present invention;

[0040] Figure 6 is a schematic diagram of the electronic device provided in Embodiment 4 of the present invention. Detailed Implementation

[0041] The present invention will be further illustrated by way of embodiments below, but the present invention is not limited to the scope of the embodiments described herein.

[0042] In this embodiment of the invention, prefixes such as "first" and "second" are used merely to distinguish different descriptive objects and do not limit the position, order, priority, quantity, or content of the described objects. The use of ordinal numbers and other prefixes to distinguish descriptive objects in this embodiment of the invention does not constitute a limitation on the described objects; the description of the described objects is given in the context of the embodiments, and the use of such prefixes should not constitute unnecessary restrictions. Furthermore, in the description of this embodiment, unless otherwise stated, "multiple" means two or more.

[0043] In the embodiments of the present invention, the collection, storage, use, processing, transmission, provision and disclosure of user personal information comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0044] Example 1

[0045] Figure 1 is a framework diagram of an automatic experimental control system provided by an exemplary embodiment of the present invention.

[0046] The human-computer interaction module is used for setting experimental parameters, issuing control commands, collecting experimental data, monitoring experimental status, and predicting experimental results. Preferably, the human-computer interaction module is located in the cloud. The module adopts a modular programming approach, allowing users to select the necessary modules from the system and configure the experimental workflow according to their experimental needs, enabling reconfigurable types of experiments. This module supports IoT communication and connects to the system's main control module using the MQTT protocol.

[0047] The system main control module is used to realize real-time communication between the human-machine interaction module and the Internet of Things gateway to ensure reliable data upload and download; preferably, the system main control module is a control system based on a programmable logic controller.

[0048] The system's working modules include an anhydrous and oxygen-free working module, a non-inert working module, and an experimental result analysis module. Each of these modules includes multiple sub-modules for performing various experimental procedures. The sub-modules collect multiple experimental parameters from these procedures and transmit them to the system's main control module and human-computer interaction module via the Internet of Things. Each sub-module also includes a distributed processing center.

[0049] The anhydrous and oxygen-free working module, the non-inert working module, and the multiple sub-working modules of the experimental result analysis module can be added or removed as needed for the experimental task, and can be combined arbitrarily by those skilled in the art.

[0050] The control mode of the waterless and oxygen-free working module is configured with a central edge processing center, and each sub-working module is configured with a distributed processing center; the collected data includes environmental data, and in particular, the environmental data of the non-inertial working module includes water and oxygen data.

[0051] The centralized edge processing center serves as the central processing center for this module. On one hand, it interfaces with the main control module to receive relevant instructions, configure each work unit according to the user's experimental plan, and distribute the configurations to the distributed processing centers of each unit, thereby meeting the requirements for reconfigurable experiments. On the other hand, it interfaces with the distributed processing centers configured in each unit within the module to receive the process variables and working status monitoring results of each unit and feed them back to the centralized edge processing center of this module to predict experimental results and achieve process-level control.

[0052] The sub-modules of the anhydrous and oxygen-free working module include a reagent bottle capping unit, a reagent storage unit, a solid sample dispensing unit, a liquid sample dispensing unit, a reaction bottle capping unit, an experimental reaction unit, and a material transfer unit.

[0053] The reagent bottle cap opening and closing unit of the anhydrous and oxygen-free working module includes a six-axis robot, an electric gripper, related fixing devices, and a distributed processing center. The corresponding collected experimental data includes: motor rotation data, robot axis motion data, gripper travel data, force feedback data, etc. The reagent bottle cap opening and closing unit is implemented using a six-axis robot and a rotating mechanism. The distributed processing center of this unit receives system commands and controls the relevant action units (six-axis robot, electric gripper, and related fixing devices) to complete the commanded actions. The six-axis robot gripper uses an electric gripper with force feedback, which can effectively control the cap clamping force and ensure the cap opening and closing effect. When liquid reagent enters this module, this unit completes the cap opening; after the liquid sample addition unit completes the sample addition, this unit completes the cap closing.

[0054] The reagent storage unit of the anhydrous and oxygen-free working module is a rotary intelligent reagent storage unit, which receives system instructions from a distributed processing center. The rotary intelligent reagent storage unit includes a rotary positioner and a reagent storage unit. The reagent storage unit is identified through an RFID identification system; the collected experimental data includes: the rotation angle of the positioner, the storage location status of the reagent storage unit, and the ID data read by the RFID system. The distributed processing center of this unit receives system instructions, and the rotary positioner controls the rotation of the reagent storage unit, allowing for precise control of the rotation angle; simultaneously, the RFID identification system ensures that the robot can reliably grasp the required reagents.

[0055] The solid sample dispensing unit of the anhydrous and oxygen-free working module includes a solid sample dispensing robot and a high-precision solid weighing balance. The solid sample dispensing robot comprises a six-axis robot, an electric gripper, and a distributed processing center. The corresponding collected experimental data includes: motion data of each axis of the solid sample dispensing robot, weight of the dispensing solid, and static charge detection data of the balance. The distributed processing center of this unit receives system instructions, first controlling the solid sample dispensing robot to pick up the required solid reagent from the reagent storage unit, then delivering it to the balance's dispensing area, and then issuing a dispensing command to the balance. The balance is equipped with an anti-static device to precisely control the amount of solid reagent added. After dispensing is completed, feedback is sent to the control center.

[0056] The liquid sampling unit of the anhydrous and oxygen-free working module includes a liquid sampling robot and an automatic pipette. The liquid sampling robot comprises a six-axis robot, a motorized gripper, and a distributed processing center. The corresponding collected experimental data includes: motion data of each axis of the solid sampling robot, gripper travel data, force feedback data, and automatic pipette data. The distributed processing center of this unit receives system instructions, first calculates the required amount of liquid reagent to be added, then controls the liquid sampling robot to draw the required reagent from the reagent bottle, and then controls the pipette to add the sample to one or more reaction flasks. The automatic pipette has liquid level detection and negative pressure sensing functions to ensure precise control of the amount of liquid reagent added. After the sampling is completed, feedback is sent to the control center.

[0057] The reaction flask opening unit of the anhydrous and oxygen-free working module includes a six-axis robot, a locking mechanism, and a distributed processing center. The corresponding collected experimental data includes: motion data of each axis of the robot, travel data of the locking mechanism, etc. The distributed processing center of this unit receives system instructions, and after completing the pipetting of all liquid reagents for this reaction, the six-axis robot, in conjunction with the locking mechanism, completes the capping of the reaction flask.

[0058] The experimental reaction unit of the anhydrous and oxygen-free working module includes a six-axis robot, a magnetic stirrer, and a distributed processing center. The collected experimental data includes: motion data of each axis of the robot, stirring speed of the magnetic stirrer, heating temperature, and other data. The distributed processing center receives system instructions, transports the capped reaction flask to the magnetic stirrer, and controls the reaction process parameters such as heating and stirring according to the experimental parameter settings. Simultaneously, the distributed processing center collects parameters from corresponding sensors for process control and self-regulation applications.

[0059] The material transfer unit of the anhydrous and oxygen-free working module includes a six-axis robot, a segmented multi-point conveyor belt, and a distributed processing center. The corresponding collected experimental data includes: motion data of each robot axis, position data of the segmented multi-point conveyor belt, etc. Due to space constraints, the material transfer unit in the anhydrous and oxygen-free working module uses a segmented multi-point conveyor belt, enabling multi-directional material flow between working units. This is a key component for achieving reconfigurable experimental reactions in the anhydrous and oxygen-free working module. The distributed processing center of this unit receives instructions from the module's central edge processing center and performs material transfer between units according to the user's experimental configuration. Simultaneously, it receives information from the distributed processing centers of each unit in the module and transfers materials requiring adjustment during the reaction process.

[0060] The non-inertial working module employs distributed control. Each unit in this module is equipped with a distributed processing center, connected to the system's main control module. Unlike the anhydrous and oxygen-free working module, the non-inertial working module does not have a dedicated central edge processing center; instead, the main control module handles the relevant computational tasks. In this module, the main control module divides tasks based on the experimental requirements received from the human-computer interaction module. If an anhydrous and oxygen-free reaction is required, it activates the working units in the anhydrous and oxygen-free working module (activating some or all units depending on the experimental needs); otherwise, it activates the non-inertial working units to carry out the relevant reaction process actions. Similar to the anhydrous and oxygen-free working module, the main control module configures each working unit within the module and distributes the configurations to the distributed processing centers of each unit, thus meeting the requirements for reconfigurable experiments. Furthermore, it interfaces with the distributed processing centers configured in each unit within the module, receiving the process quantities and operational status monitoring results of each unit and feeding them back to the module's central edge processing center for experimental result prediction, achieving process-level control.

[0061] The sub-modules of the non-inert working module also include a reagent bottle capping unit, a reagent storage unit, a solid sample dispensing unit, a liquid sample dispensing unit, a reaction bottle capping unit, an experimental reaction unit, and a material transfer unit, as well as a material storage unit.

[0062] The reagent bottle opening and closing unit of the non-inertial working module includes a six-axis robot, an electric gripper, related fixing devices, and a distributed processing center. The corresponding collected experimental data includes: motor rotation data, robot axis motion data, gripper travel data, force feedback data, etc. The distributed processing center of this unit receives system commands and controls the relevant action units (six-axis robot, electric gripper, and related fixing devices) to complete the commanded actions. The six-axis robot gripper uses an electric gripper with force feedback, which can effectively control the bottle cap clamping force and ensure the opening and closing effect. When liquid reagent enters this module, this unit completes the opening; after the liquid dispensing unit completes the dispensing, this unit completes the closing.

[0063] The reagent storage unit of the non-inertial working module is a drawer-type intelligent reagent storage unit, and it receives system instructions from the distributed processing center.

[0064] The rotating intelligent reagent storage unit includes a movable reagent bottle tray, a fixed reagent loading rack, and a six-axis robot. The reagent storage unit stores reagents using electronic tags and precise robot positioning, with dual error correction. The corresponding collected experimental data includes: robot axis motion data, gripper travel data, force feedback data, tray motor motion data, and electronic tag recognition data.

[0065] The reagent library is designed with 6 layers, each with 180 storage locations. It adopts a 15×12 (column×row) partitioning method, and its total storage capacity can reach 1080, allowing for multiple backups of the same reagent.

[0066] The reagent storage unit provides two movable reagent bottle trays, each with 10 positions, arranged in a 2×5 (column × row) layout. Each retrieval capacity is 20 different reagents. The unit also has a fixed reagent loading rack with 20 positions, arranged in a 4×5 (column × row) layout. The reagent storage unit uses electronic tags (containing chemical information such as reagent name, CAS number, concentration, volume, solvent, and shelf life) to identify reagent types. It employs a dual error correction mechanism for reagent storage: using electronic tags and precise robot positioning for storage, with double error correction. Reagent inventory is dynamically updated based on usage. Reagent safety management includes: hazardous chemical use approval, usage records, and hazard warnings (including temperature and humidity sensors); a built-in MSDS database (for safety information query and display); automatic bottle retrieval and placement; retrieving 10 reagent bottles in less than 6 minutes; and a reproducibility rate of less than 0.1 mm for 100 samplings.

[0067] The drawer-type automatic reagent dispenser stores reagent bottles, with each drawer capable of storing 280 bottles. The reagent dispenser control program receives a call command and automatically pops out the corresponding drawer based on the number. The robot determines the location of the reagent bottle based on the number and retrieves it from the automatic drawer. An electronic tag is compared; if the reagent bottle is correct, it is placed on the reagent bottle temporary storage rack; otherwise, an alarm is triggered. The control system records all operations.

[0068] The material storage unit of the non-inertial working module is an automated design of a telescopic mechanism, receiving system instructions from a distributed processing center. This unit stores materials using electronic tags and precise robot positioning, with dual error correction. The corresponding collected experimental data includes: telescopic motor motion data, electronic tag recognition data, etc.

[0069] The material storage area is designed with three layers, storing: 20 trays of tippers, 4 trays of sample dispensing heads, 8 trays of injection and filtration tools, 8 trays of reaction flasks, and 2 trays of waste containers. The material storage area uses electronic tags (containing reagent name, CAS number, name, quantity, etc.) to identify reagent types. The material storage area employs a dual error correction mechanism for reagent storage: using electronic tags and precise robot positioning to store materials, providing double error correction.

[0070] The solid sample dispensing unit of the non-inert working module includes a solid sample dispensing robot and a high-precision solid weighing balance. The solid sample dispensing robot comprises a six-axis robot, an electric gripper, a three-dimensional solid reagent storage area, and a distributed processing center. The corresponding collected experimental data includes: motion data of each axis of the solid sample dispensing robot, weight of the dispensing solid, and static charge detection data of the balance. The distributed processing center of this unit receives system instructions, first controlling the solid sample dispensing robot to pick up the required solid reagent from the reagent storage unit, then delivering it to the balance's dispensing area, and then issuing a dispensing command to the balance. The balance is equipped with an anti-static device to precisely control the amount of solid reagent added. After dispensing, feedback is sent to the control center. Unlike the anhydrous and oxygen-free working module, this unit has a three-dimensional solid reagent storage area, using a 4×16 (column × row) partition, capable of storing 64 kinds of solid reagents, fully meeting experimental needs.

[0071] The liquid sampling unit of the non-inertial working module includes a liquid sampling robot and an automated pipette. The liquid sampling robot comprises a six-axis robot, a motorized gripper, and a distributed processing center. The corresponding collected experimental data includes: motion data of each axis of the solid sampling robot, gripper travel data, force feedback data, and automated pipette data. The distributed processing center of this unit receives system instructions, first calculates the required amount of liquid reagent to be added, then controls the liquid sampling robot to draw the required reagent from the reagent bottle, and then controls the pipette to add the sample to one or more reaction flasks. The automated pipette has liquid level detection and negative pressure sensing functions to ensure precise control of the amount of liquid reagent added. After the sampling is completed, feedback is sent to the control center.

[0072] The reaction flask opening unit of the non-inertial working module includes a six-axis robot, a locking mechanism, and a distributed processing center; the corresponding collected experimental data includes: motion data of each axis of the robot, travel data of the locking mechanism, etc. The distributed processing center of this unit receives system instructions, and after completing the pipetting of all liquid reagents for this reaction, the six-axis robot, in conjunction with the locking mechanism, completes the capping of the reaction flask.

[0073] The experimental reaction unit of the non-inertial working module includes a six-axis robot, a magnetic stirrer, and a distributed processing center. The corresponding collected experimental data includes: motion data of each axis of the robot, stirring speed of the magnetic stirrer, heating temperature, and other data. The distributed processing center of this unit receives system instructions, transports the capped reaction flask to the magnetic stirrer, and controls the reaction process parameters such as heating and stirring according to the experimental parameter settings. Simultaneously, the distributed processing center collects parameters from corresponding sensors for process control and self-regulation applications.

[0074] The material transfer unit of the non-inertial working module is implemented by a composite robot. The composite robot includes a six-axis robot, an automated guided vehicle (AGV), and a distributed processing center; the corresponding collected experimental data includes: robot axis motion data, AGV position data, LiDAR scan data, gripper stroke data, force feedback data, etc.

[0075] As a key component enabling reconfigurable experimental responses in the non-inertial working module, the composite robot employs wireless communication to interact with the system's main control module and other working units within the non-inertial working module. It is equipped with a safety-scanning laser sensor for SLAM scanning (simultaneously performing localization and mapping) to adapt to different experimental scenarios. Additionally, it features a rear sonar for detecting obstacles behind it and a front-lower laser sensor for detecting low-lying objects in front, comprehensively ensuring the composite robot's safe operation during transport. Furthermore, the composite robot is also equipped with a 3D vision camera for high-precision positioning during experimental execution, meeting the requirements of experimental actions.

[0076] The experimental results analysis module includes an extraction unit, a filtration unit, and a results analysis unit. The corresponding collected data includes: extraction process detection data, working time data, filtration process detection data, working time data, and data on the actions of the results analysis unit. This is used to meet different post-experiment processing needs, thereby effectively ensuring the flexibility and reconfigurability of the experiment.

[0077] Similar to the non-lazy working modules, the experimental results analysis module also employs distributed control. Each unit in this module is equipped with a distributed processing center connected to the system's main control module. There is no dedicated central edge processing center in this module; the main control module handles the relevant computational tasks. Within this module, the main control module configures the experimental results analysis tasks based on the experimental scheme requirements received from the human-computer interaction module, selecting and combining corresponding extraction, filtering, and analysis units to meet the requirements of different experimental results analyses and achieve experimental reconfigurability. The experimental analysis results undergo standardized data processing by the distributed processing center of the analysis unit and are then uploaded to the main control module, completing the entire experimental process.

[0078] The extraction unit includes a centrifuge, a six-axis robot, an automated guided vehicle (AGV), and a distributed processing center. The extraction unit utilizes a centrifuge. The distributed processing center receives instructions from the system's main control module and, after the composite robot places the capped reaction flask in position, uses high-speed centrifugation to extract the reagents after the reaction.

[0079] The filtration unit includes a six-axis robot, a filtration auxiliary device, and a distributed processing center. The distributed processing center receives instructions from the system's main control module, employs a triple filtration method, and injects the filtered reagent bottle into the analysis bottle; then, the six-axis robot places it into the results analysis unit.

[0080] The results analysis unit includes a six-axis robot, analytical instruments, a data processing device, and a distributed processing center. The analytical instruments include various general-purpose experimental result analysis instruments such as chromatographs and mass spectrometers. The data processing device is used to perform data standardization processing, and the data is then uploaded from the distributed processing center of this unit to the human-computer interaction module via the system's main control module to provide the user with the final experimental results.

[0081] The automatic experimental control system provided in this embodiment can be adapted to different types of chemical experiments. It can reconfigure and reorganize the process of each working unit according to the experimental requirements to meet the automation needs of different types of experiments. It provides a process-level control method that integrates qualitative and quantitative methods. Through fine monitoring and control of the process, it can effectively predict the validity of experimental results, thereby determining and terminating failed experiments in advance. This effectively reduces the consumption of consumables and the time spent on ineffective experiments, thereby improving experimental efficiency.

[0082] Example 2

[0083] Figure 2 shows a flowchart of the automatic experimental control method provided by the present invention. This method is applied to the control system of the above embodiment and specifically includes the following steps:

[0084] S10. Based on the automatic experimental control system, a Bayesian network model is constructed, and the Bayesian network adopts an expert system approach.

[0085] S20, Based on the Bayesian network model, perform qualitative and quantitative analysis on each experimental parameter during the experiment;

[0086] S30, in response to the prediction results of the Bayesian network model, controls the timely termination of the experimental process.

[0087] Step S10, which involves constructing a Bayesian network model based on the automatic experimental control system, specifically includes the following steps, as shown in Figures 3 and 4.

[0088] S11, Node Mapping, mapped to test nodes, functional failure nodes, failure mode nodes, and component nodes;

[0089] The experiment is mapped as a test ontology to a test node, and the experiment includes multiple experimental procedures.

[0090] The failure ontology belonging to functional failure is mapped to functional failure node. The failure ontology belonging to functional failure refers to the experimental procedure that leads to functional failure based on the failure mode. Preferably, in this embodiment, it refers to 8 categories such as solid sample addition, liquid sample addition, sealing, reaction, post-processing, and result analysis.

[0091] The failure ontology belonging to the failure mode type is mapped to the failure mode node. The failure ontology belonging to the failure mode type is the experimental failure factor, specifically corresponding to the experimental data collected in each unit.

[0092] Each unit body is selected and mapped as a component node. The unit body is a number of sub-working modules of the anhydrous and oxygen-free working module, the non-inert working module, and the experimental result analysis module.

[0093] S12, CF layer Bayesian network construction, this part of the network contains component nodes and failure mode nodes, with component nodes as parent nodes pointing to failure mode nodes, and failure mode nodes as child nodes.

[0094] S13, FM layer Bayesian network construction. This part of the network includes failure mode nodes and functional failure nodes. The failure mode node is the parent node, pointing to the functional failure node it affects. The functional failure node is the child node. This step is the most important step in realizing the leap from system hardware to failure testing.

[0095] S14, MT layer Bayesian network construction. This part of the network includes functional failure nodes and test nodes. The functional failure nodes in the FM layer are the parent nodes, pointing to the test nodes they affect, and these test nodes are the child nodes.

[0096] S15, Network Aggregation: A three-level, four-layer Bayesian network structure is established by connecting identical nodes in the CF, FM, and MT layers, thus constructing the overall Bayesian network diagnostic model. The Bayesian network model designed using this method includes four types of nodes: test, functional failure, failure mode, and component. Specifically, test nodes are denoted as T, functional failure nodes as M, failure mode nodes as F, and component nodes as C.

[0097] In a complex chemical experiment, each step inevitably introduces process errors. Although the error of each individual step will not cause the overall failure of the experiment, the cumulative effect of multiple steps often leads to the failure of the experiment before the reaction has even begun. When conducting experiments, personnel will eliminate such cases based on experience. However, current automated experiments lack relevant judgment mechanisms and will mechanically follow the steps to carry out experiments that have already failed, resulting in a waste of experimental consumables and experimental time.

[0098] To address this issue, this invention introduces a Bayesian network to deduce and predict experimental results based on the experimental process state, thereby enabling timely termination of failed experimental processes.

[0099] A Bayesian network is a probabilistic reasoning graph model that combines qualitative and quantitative methods, consisting of a network topology and network parameters. The network topology represents qualitative knowledge and is formally a Directed Acyclic Graph (DAG). Nodes in the DAG represent random variables in the problem; for failure identification, these variables typically represent failure symptoms, system structure, failure modes, etc. Directed edges represent dependencies between nodes, reflecting the impact of system failures and their propagation paths. Network parameters, on the other hand, represent quantitative knowledge, using probabilistic representations to characterize system uncertainty. These parameters mainly include marginal probabilities and conditional probabilities. Marginal probabilities reflect the likelihood of a node being in a certain state, while conditional probabilities reflect the degree of correlation between nodes.

[0100] Establishing a complete Bayesian network model requires estimating the network parameters. In this design, an expert system approach is adopted, whereby chemists specializing in this type of experiment provide empirical reference values ​​to assign parameters to the Bayesian network, thereby enabling the prediction of experimental failures.

[0101] Specifically, as a concrete example, using a Bayesian network model as the diagnostic basis, the abnormal test item records of the functional test are input, and the corresponding prediction results are obtained, as shown in Table 1. If the input abnormal test item is the Ullmann-Ma reaction test (T5), the prediction result suspects that experimental steps C6, C7, C8, C11, C12, and C13 have failed, and gives the probability of failure, which is consistent with the actual failure situation.

[0102] Table 1:

[0103]

[0104] The automatic control method provided in this embodiment can be applied to different types of chemical experiments. It can reconfigure and reorganize the process of each working unit according to the experimental requirements to meet the automation needs of different types of experiments. It provides a process-level control method that integrates qualitative and quantitative methods. Through refined monitoring and control of the process, it can effectively predict the validity of experimental results, thereby determining and terminating failed experiments in advance. This effectively reduces the consumption of consumables and the time spent on ineffective experiments, thereby improving experimental efficiency.

[0105] Example 3

[0106] Corresponding to the aforementioned embodiments of the automatic experimental control method, the present invention also provides embodiments of the automatic experimental control device.

[0107] Figure 5 is a schematic diagram of an automatic experimental control device provided in an exemplary embodiment of the present invention. The device includes:

[0108] Network construction module 1 is used to construct a Bayesian network model based on the automatic experimental control system, wherein the Bayesian network adopts an expert system approach.

[0109] Analysis module 2 is used to perform qualitative and quantitative analysis on various experimental parameters during the experiment based on the Bayesian network model.

[0110] The program control module 3 is used to control the timely termination of the experimental process in response to the prediction results of the Bayesian network model.

[0111] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of the present invention according to actual needs.

[0112] Example 4

[0113] Figure 6 is a schematic diagram of an electronic device according to an example embodiment of the present invention. The electronic device includes a memory, a processor, and a computer program stored in the memory and used to run on the processor. When the processor executes the computer program, it implements the automatic experimental control method described in any of the above embodiments. The electronic device 60 shown in Figure 6 is merely an example and should not impose any limitation on the function and scope of use of the embodiments of the present invention.

[0114] As shown in Figure 6, the electronic device 60 can be manifested as a general-purpose computing device, such as a server device. The components of the electronic device 60 may include, but are not limited to: at least one processor 61, at least one memory 62, and a bus 63 connecting different system components (including memory 62 and processor 61).

[0115] Bus 63 includes a data bus, an address bus, and a control bus.

[0116] The memory 62 may include volatile memory, such as random access memory (RAM) 621 and / or cache memory 622, and may further include read-only memory (ROM) 623.

[0117] The memory 62 may also include a program tool 625 (or utility) having a set (at least one) program module 624, such program module 624 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment.

[0118] The processor 61 executes various functional applications and data processing by running computer programs stored in the memory 62, such as the automated experimental control method provided in any of the above embodiments.

[0119] Electronic device 60 can also communicate with one or more external devices 64 (e.g., keyboard, pointing device, etc.). This communication can be performed through input / output (I / O) interface 65. Furthermore, electronic device 60 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public network, such as the Internet) via network adapter 66. As shown, network adapter 66 communicates with other modules of electronic device 60 via bus 63. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with electronic device 60, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, and data backup storage systems.

[0120] It should be noted that although several units / modules or sub-units / modules of the electronic device have been mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to embodiments of the invention, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided and embodied by multiple units / modules.

[0121] Example 5

[0122] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the automatic experimental control method provided in any of the above embodiments.

[0123] The readable storage medium may be more specifically adopted, including but not limited to: portable disk, hard disk, random access memory, read-only memory, erasable programmable read-only memory, optical storage device, magnetic storage device, or any suitable combination thereof.

[0124] Example 6

[0125] This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the automatic experimental control method described in any of the above embodiments.

[0126] The program code for executing the computer program product of the present invention can be written in any combination of one or more programming languages. The program code can be executed entirely on the user device, partially on the user device, as a standalone software package, partially on the user device and partially on a remote device, or entirely on a remote device.

[0127] While specific embodiments of the present invention have been described above, those skilled in the art should understand that these are merely illustrative examples, and the scope of protection of the invention is defined by the appended claims. Those skilled in the art can make various changes or modifications to these embodiments without departing from the principles and essence of the present invention, but all such changes and modifications fall within the scope of protection of the present invention.

Claims

1. An automatic experimental control system, characterized in that, include: The human-computer interaction module is used for setting experimental parameters, issuing control commands, collecting experimental data, monitoring experimental status, and predicting experimental results. The system main control module is used to realize real-time communication between the human-machine interaction module and the IoT gateway to ensure reliable data upload and download; and the system working module includes an anhydrous and oxygen-free working module, a non-inertial working module, and an experimental result analysis module. The anhydrous and oxygen-free working module, the non-inertial working module, and the experimental result analysis module all include multiple sub-working modules for performing various experimental procedures and collecting multiple experimental parameters in the experimental procedures and sending them to the system main control module and the human-machine interaction module through the IoT. The sub-working modules include a distributed processing center.

2. The automatic experimental control system as described in claim 1, characterized in that, The sub-modules of the anhydrous and oxygen-free working module include a reagent bottle capping unit, a reagent storage unit, a solid sample dispensing unit, a liquid sample dispensing unit, a reaction bottle capping unit, an experimental reaction unit, and a material transfer unit.

3. The automatic experimental control system as described in claim 1, characterized in that, The sub-modules of the non-inert working module include a reagent bottle capping unit, a reagent storage unit, a material storage unit, a solid sample dispensing unit, a liquid sample dispensing unit, a reaction bottle capping unit, an experimental reaction unit, and a material transfer unit.

4. The automatic experimental control system as described in claim 1, characterized in that, The sub-modules of the experimental results analysis module include an extraction unit, a filtration unit, and an analysis unit.

5. An automatic experimental control method, characterized in that, An automatic experimental control system as described in any one of claims 1-4 includes: constructing a Bayesian network model based on the automatic experimental control system, wherein the Bayesian network adopts an expert system approach; performing qualitative and quantitative analysis on various experimental parameters during the experimental process based on the Bayesian network model; and controlling the timely termination of the experimental process in response to the prediction results of the Bayesian network model.

6. The automatic experimental control method as described in claim 5, characterized in that, The construction of the Bayesian network model based on the automatic experimental control system specifically includes: node mapping, which maps to test nodes, functional failure nodes, failure mode nodes, and component nodes; and the construction of the CF layer Bayesian network, which includes component nodes and failure mode nodes, with component nodes as parent nodes pointing to failure mode nodes, and failure mode nodes as child nodes. The FM layer Bayesian network is constructed, which includes failure mode nodes and functional failure nodes. The failure mode nodes are used as parent nodes, pointing to the functional failure nodes they affect, and the functional failure nodes are child nodes. The Bayesian network of the MT layer is constructed. This part of the network contains functional failure nodes and test nodes. The functional failure nodes in the FM layer are the parent nodes, pointing to the test nodes they affect, and these test nodes are the child nodes. Network aggregation establishes a three-level, four-layer Bayesian network structure by connecting the same nodes in the CF, FM, and MT layers, thereby realizing the construction of the overall Bayesian network model.

7. The automatic experimental control method as described in claim 6, characterized in that, The node mapping step specifically includes mapping the experiment as a test ontology to a test node, wherein the experiment includes multiple experimental procedures; and mapping the failure ontology belonging to functional failure to a functional failure node, wherein the failure ontology belonging to functional failure refers to the experimental procedure that causes functional failure based on the failure mode. The failure ontology belonging to the failure mode type is mapped to the failure mode node. The failure ontology belonging to the failure mode type is the experimental failure factor, corresponding to the experimental data collected in each unit. Each unit ontology is selected and mapped to the component node. The unit ontology is multiple sub-working modules of the anhydrous and oxygen-free working module, the non-inert working module, and the experimental result analysis module.

8. An automatic experimental control device, characterized in that, It includes: a network construction module for constructing a Bayesian network model based on the automatic experimental control system, wherein the Bayesian network adopts an expert system approach; an analysis module for performing qualitative and quantitative analysis on various experimental parameters during the experimental process based on the Bayesian network model; and a program control module for controlling the timely termination of the experimental process in response to the prediction results of the Bayesian network model.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and for running on the processor, characterized in that, When the processor executes the computer program, it implements the automatic experimental control method according to any one of claims 5-7.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the automatic experimental control method as described in any one of claims 5-7.

Images