WO2022141234A1 - Method for establishing experimental workflow model - Google Patents

Abstract

A method for establishing an experimental workflow model, comprising: analyzing and extracting various steps from the beginning to the end of an experiment, dependency relationships between the steps, and input and output information between the steps; extracting an operation involved in each step according to various steps, the dependency relationships between the steps, and the input and output information between the steps; according to various steps, the dependency relationships between the steps, the input and output between the steps, and information of the operations, converting an experiment process into process stages, converting the process stages into process operations, and converting the process operations into process actions; and matching the process actions with an action set. Said method converts an experiment process into process stages, converts the process stages into process operations, and converts the process operations into process actions. In a process control model, key actions are defined, and in an actual experimental design process, only experiment operations involved in an experiment process need to be concerned, and there is no need to concern about the details of experimental equipment involved in underlying specific experiments.

Description

Experimental workflow model establishment method technical field

The invention relates to experiment automation, in particular to a method for establishing an experiment workflow model.

Background technique

With the rapid development of information technology, facing the growing competitive pressure brought by the digital economy era, enterprises in the traditional manufacturing industry have also begun to try to use information technology to empower business processes, especially in some repetitive labor-intensive industries. In business scenarios, enterprises expect to use information technology to automatically operate the whole or part of the business process, the process of information, data or tasks among multiple participants according to certain predefined rules.

According to the definition of Workflow Management Coalition (WfMC, Workflow Management Coalition), Workflow technology is the core of realizing business process modeling, business process simulation, business process optimization, business process management and integration, and finally realizing business process automation. technology. In 1995, WfMC released the workflow reference model and released the first workflow process description language WPDL (Workflow Process Definition Language). WPDL describes the entities contained in the workflow process process definition, the attributes of the entities, and the relationship between entities and entities, and how to merge these process definitions into process models and common data. In May 2001, WfMC released a workflow process definition specification based on XML (eXtensible Markup Language) - XML Process Definition Language XPDL (XML Process Definition Language). XPDL maps WPDL on the basis of XML description format, and carries out appropriate Add and modify.

Since the workflow or business rules in the field of research experiments need to be changed frequently, such a workflow should not only satisfy the continuous and efficient trial and error of scientific researchers, but also meet the requirements of accurate and convenient definition of production and development processes in actual production scenarios. . WPDL and XPDL are both process metamodels based on the same workflow, and they unify the modeling methods of various businesses in the real world. And provide reference and reference for the modeling of specific business. However, in the actual landing process, it often fails to meet the application requirements.

SUMMARY OF THE INVENTION

Based on this, it is necessary to provide an experimental workflow model establishment method that can improve the experimental efficiency.

A method for establishing an experimental workflow model, comprising:

Extraction steps: analyze and extract the steps from the beginning to the end of the experiment, the dependencies between the steps, and the input and output information between the steps;

Extraction operations: Extract the operations involved in each step according to each step, the dependencies between steps, and the input and output information between steps;

Transformation: According to each step, the dependencies between steps, the input and output between steps, and the operation information, the experimental process is transformed into a process segment, a process segment is converted into a process operation, and a process operation is converted into a process action according to the process operation;

Match: Match process actions to action sets.

In a preferred embodiment, the method further includes: expanding: if new process actions are involved, expanding into the action set.

In a preferred embodiment, the experiment includes: a solubility judgment test. The process of converting the solubility judgment test process into a solubility judgment test includes: an experiment preparation stage, a reaction stage, a dissolution judgment stage, and a result storage stage. The solubility judgment stage The process operations transformed from the process operations in the experimental preparation phase of the experiment include: test tube rack preparation operations, solution preparation operations, and solute preparation operations, and the process operations transformed into the reaction phase of the solubility judgment experiment include: solute addition operations, solution addition operations 2. The test tube rack oscillating operation, the dissolution determination stage of the solubility determination experiment includes: dissolution analysis operation, and the process operation of the result storage stage of the determination solubility experiment includes: moving the test tube rack to the storage area; The process action of the test tube rack preparation operation includes: taking the test tube rack, the process action of the solution preparation operation for judging the solubility experiment includes: taking the solution to the preparation area, and the process action of the solute preparation operation for judging the solubility experiment includes: taking the solute to the preparation area. Preparation area; the process action of adding a solute in the solubility judgment experiment includes: adding a solute, the process action of adding a solution in the solubility judgment experiment includes: adding a solution, and the process action of the test tube rack shaking operation in the solubility judgment experiment Including: shaking, the process action of the dissolution analysis operation for judging the solubility experiment includes: dissolution analysis, and the process action of moving the test tube rack to the storage area in the solubility judgment experiment includes: moving the final material.

In a preferred embodiment, the criterion for judging the dissolving in the dissolving analysis is: adding a set amount of pure solvent into the test tube, gradually accelerating the oscillation from a low speed, and oscillating at each speed for 10 s until the sensor reads Until the standard deviation is greater than 10, take the maximum oscillation speed v of which the standard deviation of the sensor reading is less than 10 as the speed of collecting sensor data when judging the dissolution; Let stand for 10s, then oscillate at speed v for 10s, collect sensor readings and standard deviation, and add 100 μl of solvent one by one until the standard deviation of sensor readings is less than 10, and it is judged that the solution is clear.

In a preferred embodiment, the standard for judging the dissolving in the dissolving analysis is as follows: add 2 ml of pure solvent into the test tube, gradually accelerate the oscillation from 5% of the maximum speed, and accelerate by 10% each time. Oscillate for 10 s until the standard deviation of the sensor reading is greater than 10. Take the maximum oscillation speed v with the standard deviation of the sensor reading less than 10 as the speed of collecting sensor data when judging the dissolution and clearing; after adding solid and liquid to the test tube, use the 70% of the maximum speed is shaken at high speed to fully dissolve the solid, then stand for 10s, then shake at speed v for 10s, collect the sensor readings and standard deviation, and add 100 μl of solvent one by one until the standard deviation of the sensor readings is less than 10, and it is judged that the solution is clear. .

In a preferred embodiment, the experiment includes: a crystallization experiment by the elution method, and the process sections transformed into the crystallization experiment process by the elution method include: an experimental preparation stage, a preparation stage, a powder processing stage, and an analysis stage. The elution method The transformation of the experimental preparation stage of the crystallization experiment into process operations includes: solution operation and solute operation. The process actions in the experimental preparation stage of the elution method crystallization experiment include: taking the solute, taking the solution, and taking consumables. The elution method crystallization experiment The transformation of the preparation stage into the process operation includes: sample preparation operation, shaking operation, and dissolving and clearing judgment operation, and the process actions of the preparation stage of the crystallization experiment of the dissolution method include: adding solute, adding solution, shaking, controlling temperature, separating liquid, Dissolving and clearing, the transformation of the powder processing stage of the elution method crystallization experiment into process operations includes: suction filtration operation, powder scraping operation, and vacuum drying operation, and the process actions of the powder treatment stage of the elution method crystallization experiment include: opening and closing the lid , suction filtration, scraping powder, temperature control, vacuum drying, powder storage, the analysis stage of the crystallization experiment of the dissolution method is transformed into a process operation including: XRD (X-ray diffraction X-ray diffractometer technology) operation, the dissolution The process actions of the XRD operation in the crystallization experiment include: opening the equipment, waiting, and monitoring.

In a preferred embodiment, the experiment includes: a synthesis experiment, and the process stages transformed from the synthesis experiment process include: an experiment preparation stage, a preparation stage, and an analysis stage, and the transformation of the experiment preparation stage of the synthesis experiment into a process operation includes: a solution operation, solute operation, the process actions in the experimental preparation stage of the synthesis experiment include: taking the solute, taking the solution, and taking consumables, and the preparation stage of the synthesis experiment is transformed into process operations including: sample preparation operation, shaking operation, extraction, rotation Steaming and HPLC preparation, the process actions in the preparation stage of the synthesis experiment include: adding solute, adding solution, shaking, controlling temperature, separating liquid, and extracting, and the analysis stage of the synthesis experiment is transformed into process operations including: HPLC operation, all The process actions of the HPLC operation of the synthesis experiment include: turning on the equipment, waiting, and monitoring.

In a preferred embodiment, the experiment includes: polyvinyl chloride production, the conversion of the polyvinyl chloride production process into a process section includes: polymerization, recovery, and drying, and the process operation of the polyvinyl chloride production includes: preparing the reactor operation, The charging operation and the reaction operation, the process actions of the reaction operation for the production of polyvinyl chloride include: charging, heating, and holding.

In a preferred embodiment, before the matching step, it includes: action set induction: according to the action set required for experiment automation at model transformation and disassembly, the model transformation and disassembly include: natural language experiment description transformation, action set In summary; the action set includes: adding, collecting layers, concentrating, recrystallization, degassing, vacuum drying, drying with desiccant, extraction, filtration, preparing solution, microwave, separation, acid-base, phase separation, purification, quenching One or more of , reflux, set temperature, sonication, stirring, waiting, flushing, using other steps, invalid operation, no operation.

In a preferred embodiment, the matching further includes:

Workflow transformation: transform process actions into experimental automation workflows described in XML;

Submit: Submit the experiment automation workflow to the workflow execution engine.

The above-mentioned experimental workflow model establishment method converts the experimental process into a process segment, converts the process segment into a process operation, and converts the process operation into a process according to each step, the dependency between the steps, the input and output between the steps, and the operation information. Action, in the process control model, defines the key action (Action), in the actual experimental design process of the experimenter, only need to care about the experimental operation involved in the experimental process, and do not need to specifically care about the underlying specific experiments Details of the experimental equipment involved. When a new experiment requires new experimental equipment, it is only necessary to abstract the characteristic capabilities of the experimental equipment into actions that can be selected by the experimenter, so as to realize the flexible expansion of the industrial experimental process, and integrate the experimental process data with the business. Data decoupling enhances the scalability of the workflow.

Description of drawings

1 is a partial flowchart of a method for establishing an experimental workflow model according to an embodiment of the present invention;

Fig. 2 is a partial flow chart of an experiment for determining solubility according to an embodiment of the present invention;

3 is a partial schematic diagram of the disassembly of an experiment automation workflow according to an embodiment of the present invention;

4 is a partial schematic diagram of an action set according to an embodiment of the present invention;

Fig. 5 is the partial flow chart of the crystallization experiment of the elution method of one embodiment of the present invention;

6 is a partial schematic diagram of the transformation of the elution crystallization experiment according to an embodiment of the present invention;

7 is a partial schematic diagram of the transformation of the synthesis experiment of an embodiment of the present invention.

Detailed ways

As shown in FIG. 1 , a method for establishing an experimental workflow model according to an embodiment of the present invention includes:

Step S101, extraction step: analyzing and extracting each step from the beginning to the end of the experiment, the dependencies between the steps, and the input and output information between the steps;

Step S103, extracting operations: extracting operations involved in each step according to each step, the dependencies between the steps, and the input and output information between the steps;

Step S105, conversion: according to each step, the dependency between the steps, the input and output between the steps, and the operation information, the experimental process is converted into a process segment, a process segment is converted into a process operation, and a process operation is converted into a process action according to the process operation;

Step S107, matching: matching the process action with the action set.

In step S101 of this embodiment, the extraction step includes: grammatical analysis of articles, documents, and Excel tables to extract information such as each step from the beginning to the end of the experiment, the laziness relationship between steps, and the input and output between steps.

The method for establishing an experimental workflow model in this embodiment further includes: expanding: if a new process action is involved, expanding into an action set.

Before the matching step, action set induction is included: according to the action set required for experiment automation at model transformation and disassembly, the model transformation and disassembly include: natural language experiment description transformation and action set induction.

Action set includes: adding, collecting layers, concentrating, recrystallization, degassing, vacuum drying, drying with desiccant, extraction, filtration, preparing solution, microwave, separation, acid-base, phase separation, purification, quenching, refluxing, setting One or more of setting temperature, sonication, stirring, waiting, rinsing, using other steps, ineffective operation, no operation, etc.

The experimental process of natural language experiment description conversion is transformed through model, and the process action set (Action List) required for experiment automation is disassembled, as well as information such as steps, dependencies between steps, input and output between steps, etc. These steps And operation actions can be mapped to process segments, process operations, and process actions in the process model.

After the matching step in this embodiment, it also includes:

Workflow transformation: transform process actions into experimental automation workflows described in XML;

Submit: Submit the experiment automation workflow to the workflow execution engine.

The natural language experiment description transformation in this embodiment takes the solubility judgment experiment as an example.

As shown in Figure 2, determine the solubility experiment:

(1) The robotic arm clamps a test tube rack with 12 test tubes and places it on a desktop balance. The balance is reset to zero after it is stable;

(2) The manipulator hoisted the funnel with the solid powder, and added 20 mg of solid to each of the 12 test tubes;

(3) The robotic arm grabs the test tube rack that has been filled with solid powder, moves it to the shaker with temperature control, and adds 2ml of solvent to each test tube (the type of solvent for each test tube is different, for one solute, it needs to test 40-50 solvent);

(4) (Dissolved clear judgment standard: 2ml of pure solvent is added to the test tube, gradually accelerate from low speed, oscillate at each speed for 10s, until the standard deviation of the sensor reading is greater than 10, take the maximum standard deviation of the sensor reading less than 10 The oscillation speed v is used as the speed of collecting sensor data when judging the dissolving and clearing; after adding solid and liquid to the test tube, it will first shake at high speed to fully dissolve the solid, then stand for 10s, and then oscillate at speed v for 10s to collect sensor readings and standards If the difference is poor, add 100 μl of solvent one by one until the standard deviation of the sensor reading is less than 10, and it is judged that the solution is clear)

(5) Output the experimental temperature, the added solute, the total amount of the solvent, and the calculated solubility;

(6) For the solvent that is completely dissolved after adding 2ml of solvent, it is necessary to re-judge the solubility, repeat steps 1-5, increase the amount of the solute weighed by 20mg each time, and add 2ml of solvent until the determination of the solubility of the solute in various solvents up to solubility;

(7) If the total amount of added solvent reaches the upper limit of 5ml, but still does not reach the clear solvent, steps 1-5 need to be repeated, and the amount of solute weighed is reduced to 10mg until the upper limit of 5ml solvent is reached again. If it dissolves clearly, record the solubility of the solute in the solvent <2mg/ml;

(8) Output solubility table

(9) After all solubility judgments are completed, the solutes are recovered uniformly

For the 1-9 steps of solubility judgment, an experimental flow chart or a process flow chart can be drawn by the experimental business personnel or technical personnel. Figure 2 is a flow chart of the experiment. In the flow chart, each node involves the experimental operation actions described in the natural language experiment flow. For the scenario of batch experimental process verification, if the business personnel manually formulate the flowchart, the efficiency cannot meet the needs of rapid automated experiments. In this scenario, NLP (Natural Language Processing) can be used, and NLP can be used for unstructured The experimental process description language is converted into structured data that the machine can understand.

The action set of this embodiment is summarized. After the above experimental process is disassembled through the experimental automation workflow in Fig. 3, the final disassembled granularity is Action. After reaching a certain granularity, these experiments are formed by a set of actions (Action) through different combinations and sorting to form a certain experimental reaction route. The action set is shown in Figure 4 (action set for experimental automation). At the same time, for the scenario of batch experimental process verification, the use of natural language processing NLP can directly obtain the structured reaction route experimental action set from the unstructured experimental process description language.

As shown in FIG. 3 , the process section in which the solubility judgment experiment process is transformed into the solubility judgment experiment includes: the experiment preparation stage, the reaction stage, the dissolution judgment stage, and the result storage stage.

The process operations transformed into the process operations in the experimental preparation stage of the solubility judgment experiment include: test tube rack preparation operations, solution preparation operations, and solute preparation operations. The process operations transformed into the reaction stage of the solubility judgment experiment include: adding a solute, adding a solution, and shaking a test tube rack. The dissolving and clearing judgment stage of the solubility judgment experiment includes: the dissolving and clearing analysis operation. The process operation in the storage stage of the result of the judgment of the solubility experiment includes: moving the test tube rack to the storage area.

The process actions of the test tube rack preparation operation for judging the solubility experiment include: taking the test tube rack. The process actions of the solution preparation operation of the solubility judgment experiment include: taking the solution to the preparation area. The process actions of the solute preparation operation for judging the solubility experiment include: taking the solute to the preparation area. The process action of judging the adding solute operation of the solubility experiment includes: adding solute, the process action of the adding solution operation of judging the solubility experiment includes: adding solution, and the process action of judging the test tube rack shaking operation of the solubility experiment includes: shaking, so The process action of the dissolution analysis operation in the solubility judgment experiment includes: dissolution analysis, and the process action of moving the test tube rack to the storage area in the solubility judgment experiment includes: moving the final material.

Further, in the dissolving and clearing analysis of the present embodiment, the criterion for judging the dissolving clearness is: adding a set amount such as 2ml of pure solvent to the test tube, and gradually accelerating the oscillation from a low speed (such as 20% lower than the maximum speed of the equipment). , oscillate for 10s at each speed until the standard deviation of the sensor reading is greater than 10, take the maximum oscillation speed v with the standard deviation of the sensor reading less than 10 as the speed of collecting sensor data when the subsequent judgment is dissolved; add solid and liquid to the test tube After that, it will firstly vibrate at high speed (for example, higher than 50% of the maximum speed of the device) to fully dissolve the solid, then stand for 10s, then oscillate at speed v for 10s, collect sensor readings and standard deviation, and add 100 μl of solvent one by one until the sensor When the standard deviation of the readings is less than 10, it is judged that the solution is clear.

Preferably, the standard for judging the dissolving clarity in the dissolving and clearing analysis is: add 2ml of pure solvent to the test tube, gradually accelerate the oscillation from 5% of the maximum speed, each acceleration is 10%, and oscillate at each speed for 10s until the sensor Until the standard deviation of the readings is greater than 10, take the maximum oscillation speed v with the standard deviation of the sensor readings less than 10 as the speed of collecting sensor data in the subsequent judgment of dissolving; after adding solid and liquid to the test tube, first use 70% of the maximum speed for High-speed shaking makes the solid fully dissolved, then stand for 10s, then shake at speed v for 10s, collect sensor readings and standard deviation, and add 100 μl of solvent successively until the standard deviation of sensor readings is less than 10, and it is judged that the solution is clear.

The experimental procedures are obtained through a lot of attempts, and the so-called experimental SOP (Experimental Standard Operating Procedure) is formed. The process model in the industry is to classify, divide and sort on the existing SOP. For example, the solubility test is divided into preparation stage, reaction stage, dissolution judgment stage, and result storage stage according to the main stages. The reaction stage can be further divided into solute addition operation, solution addition operation, temperature control operation, test tube rack shaking operation, etc. For example, the solute addition operation includes the solute addition operation. Got the "Add +XX solute" action. The final mapping of this action from the action set is the "Add" action. The final module involved in the Add action is the weighing module, and the weighing module is composed of mechanical arms, powder buckets, and balance equipment, and is finally completed through these devices. "Add" action operation. The experimental designer does not need to care about the equipment involved and the operation of the equipment with each other.

In step S105 of this embodiment, in the transformation step, the process in the process model definition, the process consists of one or a group of process segments, and these process segments form an ordered set, which can be either serial or parallel, or serial and both. A process segment is a part of a process that normally operates independently of other process segments. Typically it will cause a series of planned physical or chemical changes in the experimental material being processed.

For example, the typical process sections in the PVC production process are:

Polymerization: polymerization of vinyl chloride monomers into polyvinyl chloride;

Recovery: recovery of remaining vinyl chloride monomer;

Drying: dry polyvinyl chloride.

Process operations in the process model definition, each process segment consists of one or a group of ordered process operations. Process operations represent the main processing activities. Process operations often cause chemical or physical changes to the material being processed. Typical process operations for the polymerization of vinyl chloride monomers to polyvinyl chloride are:

Prepare the reactor: Evacuate the oxygen in the reactor;

Loading: add demineralized water and surfactant;

Reaction: Add vinyl chloride monomer and catalyst, heat to 550C^-60°C and keep at this temperature until the reactor pressure drops.

Process actions in the process model definition, each process operation may be subdivided into one or a group of ordered process actions, the processing actions required by these process actions to perform the process action are smaller processing activities that are combined into process actions . The typical process actions of the reaction process operation of the PVC production process are:

Feeding: add the required amount of catalyst to the reactor;

Feeding: adding the required amount of vinyl chloride monomer to the reactor;

Heating: heat the contents of the reactor to 550°C-60°C;

Hold: Maintain reactor contents temperature at 550C-600C until reactor pressure drops.

Further, the experiments in this embodiment include: crystallization experiments by dissolution method.

As shown in Figure 5, the steps of the dissolution method experiment are as follows:

(1) According to the solubility table, calculate the mass M that is 103% times the amount of the solute required for 2ml of solvent (the laser sensor has a minimum liquid level requirement, here takes 2ml as an example) for each solvent (to ensure the subsequent judgment of the solubility When , add 100μl once to achieve clear solution)

(2) Weighing

2.1 The robotic arm clamps the test tube rack with 12 test tubes and places it on the desktop balance, and the balance is reset to zero after it is stable;

2.2 The manipulator hoisted the funnel with solid powder, and added solids of mass M into each of the 12 test tubes;

2.3 The robotic arm grabs the test tube rack that has been filled with solid powder and moves it to a temperature-controlled shaker.

(3) Judgment of pipetting and dissolving

3.1 Lift the pipette with the robotic arm, add 2ml of solvent to the test tube (12 test tubes with different reagents), oscillate at high speed for 3 minutes to fully dissolve the solid, use a laser sensor to determine whether it is dissolved, if not, proceed to the next step ;

3.2 Add 100 μl of the corresponding solvent to each test tube, and use the laser sensor to determine whether it is dissolved. Stop adding solvent to the test tubes that have been dissolved, and continue to add 100 μl of solvent to the undissolved test tubes until it is judged that all 12 test tubes are dissolved. until.

(4) Add anti-solvent

4.1 Lift the pipette with the robotic arm, and add 100 μl of anti-solvent to the test tube one by one (the anti-solvent corresponding to each test tube is not necessarily the same) until the laser sensor judges that there is solid precipitation or it is added to 3 times of the good solvent;

4.2 The robotic arm transfers the test tube rack to the position where the test tube rack is stored, and continues to control the temperature and stir for a certain period of time.

(5) Suction filtration and drying

5.1 The robotic arm transfers the test tube rack to the suction filtration position for suction filtration;

5.2 The robotic arm transfers the test tube rack to a dry position for drying.

The specific implementation steps are as follows:

Extraction steps: analyze and extract the steps from the beginning to the end of the experiment, the dependencies between the steps, and the input and output information between the steps. As shown in Figure 5, the following experimental flow chart is drawn according to the experimental description of the dissolution method.

Extraction operations: Extract the operations involved in each step according to each step, the dependencies between steps, and the input and output information between steps;

Transformation: According to each step, the dependencies between steps, the input and output between steps, and the operation information, the experimental process is transformed into a process segment, a process segment is converted into a process operation, and a process operation is converted into a process action according to the process operation. Combined with the experimental description and the experimental flow chart, referring to Figure 4 (action set of experimental automation), the experimental actions involved in the current experimental process and the experimental entities involved are summarized.

Action Name	solvent
Add to	2ml
Add to	500ul
Suction filtration
dry

Match: Match process actions to action sets.

Workflow transformation: transform process actions into experimental automation workflows described in XML; part of the code is as follows:

Submit: Submit the experiment automation workflow to the workflow execution engine.

As shown in FIG. 6 , further, the transformation of the elution crystallization experimental process in this embodiment is as follows; the process sections transformed into the elution crystallization experimental process include: experimental preparation stage, preparation stage, powder processing stage, and analysis stage. The experimental preparation stage of the elution crystallization experiment is transformed into process operations including: solution operation and solute operation. The process actions in the experimental preparation stage of the elution crystallization experiment include: taking the solute, taking the solution, and taking the consumables.

The transformation of the preparation stage of the crystallization experiment by the dissolution method into the process operation includes: sample preparation operation, shaking operation, and dissolution judgment operation. The process actions in the preparation stage of the elution crystallization experiment include: adding solute, adding solution, shaking, controlling temperature, separating liquid, and dissolving.

The transformation of the powder processing stage into the process operation in the crystallization experiment of the elution method includes: suction filtration operation, powder scraping operation, and vacuum drying operation. The process actions of the powder processing stage of the elution crystallization experiment include: opening and closing the lid, suction filtration, scraping the powder, controlling the temperature, vacuum drying, and powder storage.

The analysis stage of the elution crystallization experiment is transformed into the process operation including: XRD (X-ray diffraction X-ray diffractometer technology) operation. The process actions of the XRD operation in the elution crystallization experiment include: opening the equipment, waiting, and monitoring.

Further, the experiments in this embodiment include: synthesis experiments. As shown in FIG. 7 , the process stages transformed into the synthetic experiment process include: the experimental preparation stage, the preparation stage, and the analysis stage. The experimental preparation stage of synthesis experiment is transformed into process operation including: solution operation, solute operation. The process actions in the experimental preparation stage of the synthesis experiment include: taking the solute, taking the solution, and taking the consumables. The transformation of the preparation stage of the synthesis experiment into the process operation includes: sample preparation operation, shaking operation, extraction, rotary evaporation, and HPLC preparation. The process actions in the preparation stage of the synthesis experiment include: adding solute, adding solution, shaking, controlling temperature, separating liquid and extracting. The analytical phase of the synthesis experiment was transformed into process operations including: HPLC operations. The process actions of the HPLC (High Performance Liquid Chromatography) operation of the synthesis experiment include: opening the equipment, waiting, and monitoring.

The experimental automation workflow model method of the present invention is based on the XPDL standard proposed by WfMC, combined with the characteristics of the experimental automation business process, and proposes an XML-based workflow process model to describe the field, and realizes experimental design and experiment by embedding the process model. The phase separation of the equipment control program enhances the ability to expand the workflow design of experimental automation, and is successfully applied to current crystallization and synthesis experimental automation projects.

Compared with XPDL, the process definition language is improved, and the core process control model is introduced to make the experimental process definition more standardized.

In this process control model, we define the key actions (Action). In the actual experimental design process, the experimenter only needs to care about the experimental operations involved in the experimental process, and does not need to specifically care about the details of the experimental equipment involved in the underlying specific experiments. When a new experiment requires new experimental equipment, it is only necessary to abstract the characteristic capabilities of the experimental equipment into actions that can be selected by the experimenter, so as to realize the flexible expansion of the industrial experimental process. Decoupling the experimental process data from the business data enhances the scalability of the workflow.

The transformation of natural language experiment description through model transfer, process model combination, XML description, etc. is more conducive to the batch execution of workflow engines.

The experimental automation workflow model method of the present invention decouples the operation of various complicated experimental equipment by the experimental personnel, and concentrates on the experimental design research. Combined with industrial automation, the concurrency of experiments is greatly improved, the human-intensive bottleneck in the field of experiments is broken, and industrial robots are used to replace experimental operators. After increasing the concurrency of experiments, the trial and error of experiments can be accelerated, and the research and development efficiency of drug research and development, biosynthesis and other fields can be improved.

Taking the above ideal embodiments according to the present application as inspiration, and through the above descriptions, relevant personnel can make various changes and modifications without departing from the technical idea of the present application. The technical scope of the present application is not limited to the content in the description, and the technical scope must be determined according to the scope of the claims.

As will be appreciated by those skilled in the art, the embodiments of the present application may be provided as a method, a system, or a computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present application. It will be understood that each flow and/or block in the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to the processor of a general purpose computer, special purpose computer, embedded processor or other programmable data processing device to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing device produce Means for implementing the functions specified in a flow or flow of a flowchart and/or a block or blocks of a block diagram.

These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory result in an article of manufacture comprising instruction means, the instructions The apparatus implements the functions specified in the flow or flow of the flowcharts and/or the block or blocks of the block diagrams.

These computer program instructions can also be loaded on a computer or other programmable data processing device to cause a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process such that The instructions provide steps for implementing the functions specified in the flow or blocks of the flowcharts and/or the block or blocks of the block diagrams.

Claims (10)

1. A method for establishing an experimental workflow model, comprising:

   Extraction steps: analyze and extract the steps from the beginning to the end of the experiment, the dependencies between the steps, and the input and output information between the steps;

   Extraction operations: Extract the operations involved in each step according to each step, the dependencies between steps, and the input and output information between steps;

   Transformation: According to each step, the dependencies between steps, the input and output between steps, and the operation information, the experimental process is transformed into a process segment, a process segment is converted into a process operation, and a process operation is converted into a process action according to the process operation;

   Match: Match process actions to action sets.
2. The method for establishing an experimental workflow model according to claim 1, further comprising: expanding: if a new process action is involved, expanding it into an action set.
3. The method for establishing an experimental workflow model according to claim 1, wherein the experiment includes: a solubility judgment experiment, and the process section in which the solubility judgment experiment process is transformed into a solubility judgment experiment includes: an experiment preparation stage, a reaction stage, a dissolution In the judgment stage and the result storage stage, the process operations in the experimental preparation stage of the judging solubility experiment are converted into process operations including: test tube rack preparation operations, solution preparation operations, and solute preparation operations, and the reaction stage in the judging solubility experiment is transformed into The process operations include: adding a solute, adding a solution, and shaking the test tube rack. The dissolving determination stage of the solubility determination experiment includes: a dissolving analysis operation, and the process operation in the storage phase of the results of the solubility determination experiment includes: moving the test tube The process action of the test tube rack preparation operation of the solubility judgment experiment includes: taking the test tube rack, and the process action of the solution preparation operation of the solubility judgment experiment includes: taking the solution to the preparation area, and the process action of the solubility judgment experiment The process action of the solute preparation operation includes: taking the solute to the preparation area; the process action of the solute addition operation of the solubility judgment experiment includes: adding a solute, and the process action of the solution addition operation of the solubility judgment experiment includes: adding a solution, the The process action of the shaking operation of the test tube rack for judging the solubility experiment includes: shaking, the process action for judging the dissolving analysis operation of the solubility experiment includes: dissolving analysis, and the process action for moving the test tube rack to the storage area in the judging solubility experiment includes: : Move the final material.
4. The method for establishing an experimental workflow model according to claim 2, wherein the criterion for judging the dissolving in the dissolving analysis is: adding a set amount of pure solvent into the test tube, and gradually accelerating the oscillation from a low speed, Oscillate for 10 s at each speed until the standard deviation of the sensor reading is greater than 10, and take the maximum oscillation speed v with the standard deviation of the sensor reading less than 10 as the speed of collecting sensor data when judging the dissolution; after adding solid and liquid to the test tube , it will first shake at high speed to fully dissolve the solid, then stand for 10s, then shake at speed v for 10s, collect the sensor readings and standard deviation, and add 100 μl of solvent one by one until the standard deviation of the sensor readings is less than 10, and it is judged that the solution is clear.
5. The method for establishing an experimental workflow model according to claim 2, wherein the criterion for judging the dissolving in the dissolving analysis is: adding 2 ml of pure solvent into the test tube, and gradually accelerating from 5% of the maximum speed Oscillation, accelerating by 10% each time, oscillating at each speed for 10s, until the standard deviation of the sensor reading is greater than 10, take the maximum oscillation speed v with the standard deviation of the sensor reading less than 10 as the speed of collecting sensor data when the subsequent judgment is dissolved; After the solid and liquid are added to the test tube, first use 70% of the maximum speed for high-speed shaking to fully dissolve the solid, then stand for 10s, then shake at speed v for 10s, collect sensor readings and standard deviation, and add 100 μl of solvent one by one until the sensor When the standard deviation of the readings is less than 10, it is judged that the solution is clear.
6. The method for establishing an experimental workflow model according to claim 1, wherein the experiment comprises: a crystallization experiment by a dissolution method, and the process section transformed into a crystallization experiment process by the dissolution method includes: an experimental preparation stage, a preparation stage, a powder In the processing stage and the analysis stage, the experimental preparation stage of the crystallization experiment by the elution method is transformed into process operations including: solution operation and solute operation, and the process actions in the experimental preparation stage of the elution crystallization experiment include: taking the solute, taking the solution , take the consumables, the preparation stage of the elution method crystallization experiment is transformed into process operations including: sample preparation operation, vibration operation, and dissolution determination operation, and the process actions of the preparation stage of the elution method crystallization experiment include: adding solute, Adding solution, shaking, controlling temperature, separating liquid, and dissolving, the powder treatment stage of the elution method crystallization experiment is converted into process operations including: suction filtration operation, powder scraping operation, vacuum drying operation, the elution method crystallization experiment The process actions of the powder processing stage include: opening and closing the lid, suction filtration, scraping powder, temperature control, vacuum drying, powder storage, and the transformation of the analysis stage of the elution crystallization experiment into process operations includes: XRD operation, the elution process The process actions of the XRD operation in the crystallization experiment include: opening the equipment, waiting, and monitoring.
7. The method for establishing an experimental workflow model according to claim 1, wherein the experiment comprises: a synthesis experiment, and a process segment transformed into a synthesis experiment process comprises: an experiment preparation stage, a preparation stage, and an analysis stage, and the synthesis experiment The transformation of the experimental preparation stage into process operations includes: solution operation and solute operation, and the process actions in the experimental preparation stage of the synthesis experiment include: taking solute, taking solution, and taking consumables, and the transformation of the preparation stage of the synthesis experiment into process operations includes: : sample preparation operation, shaking operation, extraction, rotary evaporation, HPLC preparation, the process actions in the preparation stage of the synthesis experiment include: adding solute, adding solution, shaking, controlling temperature, liquid separation, extraction, analysis of the synthesis experiment The transformation of stages into process operations includes: HPLC operations, and the process actions of the HPLC operations in the synthesis experiment include: opening equipment, waiting, and monitoring.
8. The method for establishing an experimental workflow model according to any one of claims 1 to 7, wherein the experiment includes: polyvinyl chloride production, and the conversion of the polyvinyl chloride production process into a process section includes: polymerization, recovery, drying, The process operation of the polyvinyl chloride production includes: preparing the reactor operation, charging operation, and reaction operation, and the process action of the reaction operation of the polyvinyl chloride production includes: feeding, heating, and holding.
9. The method for establishing an experimental workflow model according to any one of claims 1 to 7, characterized in that, before the matching step, the step includes: action set induction: according to the action set required for experiment automation at model transformation and disassembly, all The model transformation and disassembly include: natural language experiment description transformation, action set induction; the action set includes: adding, collecting layers, concentrating, recrystallization, degassing, vacuum drying, desiccant drying, extraction, filtration, and preparation of solutions , microwave, separation, acid-base, phase separation, purification, quenching, reflux, set temperature, ultrasonic, stirring, waiting, washing, using other steps, invalid operation, no operation one or more.
10. The method for establishing an experimental workflow model according to any one of claims 1 to 7, wherein after the matching, the method further comprises:

    Workflow transformation: transform process actions into experimental automation workflows described in XML;

    Submit: Submit the experiment automation workflow to the workflow execution engine.

Images