Chemical experiment decision-making generation method and device

Abstract

Disclosed in the present invention are a chemical experiment decision-making generation method and device. The method comprises: receiving a target chemical formula input by a user; generating a target chemical synthetic route on the basis of the target chemical formula; and splitting the target chemical synthetic route into experimental operation flows according to an operation sequence, and generating corresponding control instructions to execute the experimental operation flows, wherein the experimental operation flows comprise operation steps, and operation information in the operation steps, the operation information comprises raw material information, reaction conditions, and operation actions. By means of the present invention, a user only needs to input chemical molecular formulas to be synthesized, and a robot can determine an optimal chemical synthetic route on the basis of said chemical molecular formulas input by the user, and split the optimal chemical synthetic route into experimental operation flows for execution. In this way, decision-making and operation instructions of the whole experiment are independently generated without manual participation, thereby providing decision-making support for independent completion of the whole chemical experiment by the robot.

Description

A method and apparatus for generating chemical experimental decisions Technical Field

[0001] This invention relates to the field of automated operation of chemical experiments, and more specifically, to a method, apparatus, robot, and computer-readable medium for generating chemical experiment decisions. Background Technology

[0002] Currently, most traditional laboratories require researchers to perform experiments for extended periods, and most experiments are highly repetitive, making manual experimentation inefficient and costly. With the advancement of science and technology, the application of robots in chemical laboratories has become a popular area of ​​robotics application, and the automation of chemical laboratories is becoming increasingly important in scientific research and industrial applications.

[0003] Currently, the application of robots in chemical experiments is limited to a single experimental step or a single operational step (such as weighing or filtering). The decision-making regarding the entire experimental route and the judgment of experimental steps still rely on human staff. Therefore, there is an urgent need to design a method that can make decisions and judgments throughout the entire operational process of a chemical experiment, thereby automating the entire process from design to execution. Technical issues

[0004] In view of the above, the present invention provides a method, apparatus, robot and computer-readable medium for generating chemical experimental decisions, in order to at least partially solve at least one of the above-mentioned technical problems. Technical solutions

[0005] To address the aforementioned technical problems, the first aspect of this invention proposes a method for generating chemical experimental decisions, the method comprising:

[0006] Receive the target chemical formula input by the user;

[0007] Generate the target chemical synthesis route based on the target chemical formula;

[0008] The target chemical synthesis route is broken down into experimental operation procedures according to the operation sequence, and corresponding control instructions are generated to execute the experimental operation procedures. The experimental operation procedures include: operation steps and operation information in each operation step. The operation information includes: raw material information, reaction conditions and operation actions.

[0009] According to a preferred embodiment of the present invention, the method further includes:

[0010] Real-time monitoring of environmental information, and output of feedback information based on the environmental information;

[0011] And / or,

[0012] receiving spectrum data of reactants in different reaction nodes, determining and outputting spectrum analysis results according to the target chemical synthesis route and the spectrum data;

[0013] and,

[0014] According to at least one of the feedback information, the spectrum analysis results, and the user input experimental information, the abnormality in the experimental process is investigated, and the abnormality reason and the abnormality solution are output.

[0015] According to a preferred embodiment of the present application, the target chemical synthesis route is generated according to the target chemical formula, which comprises:

[0016] The target chemical formula is input into a first model, and a plurality of alternative chemical synthesis routes are output according to the ranking.

[0017] According to the similarity of each alternative chemical synthesis route and the reference reaction, and the preset reaction index, the target chemical synthesis route is selected from the plurality of alternative chemical synthesis routes.

[0018] According to a preferred embodiment of the present application, the target chemical synthesis route is split into an experimental operation process according to the operation sequence, which comprises:

[0019] The target chemical synthesis route is input into a second model, and the required raw materials for reaction, the reference reaction, and the experimental operation process of the reference reaction are output.

[0020] According to the experimental operation process of the reference reaction and the required raw materials for reaction, the raw material information and the experimental operation process of the target chemical synthesis route are generated.

[0021] The experimental operation process is generated according to the raw material information and the experimental operation process of the target chemical synthesis route.

[0022] According to a preferred embodiment of the present application, the environment information is monitored in real time, and the feedback information is output according to the environment information, which comprises:

[0023] At least one of image information, audio information, and physical parameter information in the environment is collected.

[0024] The feedback information is output according to the collected at least one information.

[0025] According to a preferred embodiment of the present application, the image information in the environment is collected, the image information is analyzed, and the target experimental material information is obtained by comparing the analysis result with the experimental material information; the region where the target experimental material information is located is output as the region where the material corresponding to the image information is located.

[0026] Or,

[0027] Collect at least one physical parameter information in the environment, compare the physical parameter information with the set physical parameter information, and determine and output whether the physical parameter is abnormal based on the comparison result;

[0028] or,

[0029] Real-time acquisition of image and / or audio information, analysis of the image and / or audio information, and output of analysis results.

[0030] According to a preferred embodiment of the present invention, the feedback information is environmental anomaly information, and at least one of the environmental anomaly information, the spectrum analysis results, and the experimental information input by the user is input into the anomaly identification model, and the anomaly cause and anomaly solution are output.

[0031] To address the aforementioned technical problems, a second aspect of the present invention provides a chemical experiment decision generation device, the device comprising:

[0032] The receiving module is used to receive the target chemical formula input by the user;

[0033] The first generation module is used to generate a target chemical synthesis route based on the target chemical formula;

[0034] The second generation module is used to break down the target chemical synthesis route into experimental operation procedures according to the operation sequence, and generate corresponding control instructions to execute the experimental operation procedures; the experimental operation procedures include: operation steps and operation information in each operation step, the operation information including: raw material information, reaction conditions and operation actions.

[0035] To address the aforementioned technical problems, a third aspect of the present invention provides a chemical experiment decision generation robot, comprising:

[0036] Processor; and

[0037] A memory that stores computer-executable instructions, which, when executed, cause the processor to perform the methods described above.

[0038] To address the aforementioned technical problems, a fourth aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores one or more programs that, when executed by a processor, implement the aforementioned method. Beneficial effects

[0039] The application can receive a target chemical molecular formula input by a user, identify the target molecular structure according to the target chemical molecular formula, generate a target chemical synthesis route according to the target molecular structure, split the chemical synthesis route into an experimental operation process, and generate corresponding control instructions, so that each experimental step in the experimental operation process is executed according to the operation sequence, and the entire experiment is completed. In this way, the user only needs to input a chemical molecular formula to be synthesized, and the robot can determine the optimal chemical synthesis route according to the chemical molecular formula input by the user, and split the optimal chemical synthesis route into an experimental operation process for execution. Thus, the decision and operation instructions of the entire experiment are independently generated without human intervention, providing decision support for the robot to independently complete the entire chemical experiment. Compared with the prior art, the application has at least the following beneficial effects:

[0040] 1. The optimal chemical synthesis route can be generated according to the molecular structure formula to complete the decision of the experimental route.

[0041] 2. The chemical synthesis route can be split into multiple operation steps according to the operation sequence to complete the decision of the experimental operation process.

[0042] 3. The environment information is monitored in real time, and feedback information is output according to the environment information; the physical perception and understanding of the surrounding environment are realized, so as to provide environmental reference information for the chemical experiment operation process and assist in completing the execution or abnormal judgment of the experiment operation.

[0043] 4. The spectrum analysis result is determined and output according to the target chemical synthesis route and the spectrum data of the reactants in different reaction nodes, and automatic spectrum analysis is realized.

[0044] 5. At least one of the environment feedback information, the spectrum analysis result, and the experimental information input by the user can be used to troubleshoot the abnormality in the experiment process, and the abnormality reason and the abnormality solution are output. BRIEF DESCRIPTION OF DRAWINGS

[0045] In order to make the technical problems solved by the application, the technical means adopted and the technical effects obtained more clear, the specific embodiments of the application will be described in detail below with reference to the drawings. However, it should be declared that the drawings described below are only the drawings of exemplary embodiments of the application, and those skilled in the art can obtain the drawings of other embodiments from these drawings without creating any creative labor.

[0046] Fig. 1 is a flowchart of a chemical experiment decision generation method according to the application;

[0047] Fig. 2 is a raw material information diagram of a target chemical synthesis route according to an embodiment of the application;

[0048] Fig. 3 is a schematic diagram of an experimental operation process according to an embodiment of the application;

[0049] Fig. 4a-4d are original spectra before reaction starts in an embodiment of the present application;

[0050] Fig. 5a-5b are spectra after reaction starts in an embodiment of the present application;

[0051] Fig. 6 is a structural framework diagram of a chemical experiment decision generation device in the present application;

[0052] Fig. 7 is a structural diagram of an example embodiment of a chemical experiment decision generation robot in the present application;

[0053] Fig. 8 is a schematic diagram of an embodiment of a computer readable medium in the present application. Best Mode for Carrying Out the Invention

[0054] An embodiment of the present application provides a chemical experiment decision generation method, which comprises:

[0055] Receiving a target chemical formula input by a user;

[0056] Generating a target chemical synthesis route according to the target chemical formula;

[0057] Splitting the target chemical synthesis route into an experimental operation flow according to an operation sequence, and generating corresponding control instructions to execute the experimental operation flow; the experimental operation flow comprises operation steps and operation information in each operation step, and the operation information comprises raw material information, reaction conditions and operation actions. Embodiments of the present application

[0058] Exemplary embodiments of the present application will now be described more fully hereinafter with reference to the accompanying drawings, in which exemplary embodiments can be embodied in various specific forms. It should be understood that the exemplary embodiments described herein are not limited to the ones set forth herein but encompass all modifications falling within the scope of the invention. Rather, these exemplary embodiments are provided so that the inventive concept can be fully conveyed to those skilled in the art.

[0059] Structures, properties, effects or other features described in a certain specific embodiment in accordance with the technical concept of the present application can be combined with one or more other embodiments in any suitable manner.

[0060] In the introduction of specific embodiments, the detailed description of structures, properties, effects or other features is to enable a full understanding of the embodiments by those skilled in the art. However, it does not exclude that those skilled in the art can implement the present application without the above-mentioned structures, properties, effects or other features in specific cases.

[0061] The flowchart in the drawing is only an example of flow demonstration, and does not represent that all the contents, operations and steps in the flowchart must be included in the scheme of the present application, nor does it represent that the execution order shown in the drawing must be executed. For example, some operations / steps in the flowchart can be decomposed, some operations / steps can be combined or partially combined, etc. Without departing from the inventive concept of the present application, the execution order shown in the flowchart can be changed according to the actual situation.

[0062] The block diagram in the drawing generally represents functional entities, which do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0063] The same reference signs in the various drawings represent the same or similar elements, components or parts, and thus the repeated description of the same or similar elements, components or parts can be omitted hereinafter. It should also be understood that although the first, second, third, etc. denotative adjectives may be used herein to describe various devices, elements, components or parts, these devices, elements, components or parts should not be limited by these adjectives. That is, these adjectives are only used to distinguish one from another. For example, a first device can also be referred to as a second device without departing from the essential technical solution of the present application. In addition, the terms "and / or", "and / or" mean all combinations of one or more of the listed items.

[0064] Fig. 1 is a flowchart of a chemical experiment decision generation method provided by the present application; as shown in Fig. 1, the method comprises:

[0065] S1, receiving a target chemical formula input by a user;

[0066] The target chemical formula can be a chemical molecular formula, a chemical molecular structural formula, a chemical molecular SMILES, a CAS number, etc. of a chemical experiment to be synthesized. The present application supports the user to input the target chemical formula through text, image and voice.

[0067] S2, generating a target chemical synthesis route according to the target chemical formula;

[0068] Before this step, a plurality of chemical formulas (chemical molecular formula, chemical molecular structural formula, chemical molecular SMILES, CAS number, etc.) and corresponding chemical synthesis routes can be collected as a training set to train the first model, so that the trained first model can output a plurality of chemical synthesis routes that can generate the chemical formula according to the chemical formula. The chemical synthesis route can be a chemical reaction formula. This step can include:

[0069] S21, input the target chemical formula into the first model, and output multiple alternative chemical synthesis routes according to the ranking;

[0070] S22, select a target chemical synthesis route from the multiple alternative chemical synthesis routes according to the similarity between each alternative chemical synthesis route and the reference reaction, and a preset reaction index.

[0071] In step S21, the ranking can be performed according to a preset index, such as the frequency of use of the synthesis route, the time required for the synthesis route, etc.

[0072] The following takes the synthesis of (2-fluoro-6-methylphenyl) tert-butyl carbamate molecule as an example to explain steps S1-S2 in detail:

[0073] [According to Rule 26, corrected on 29.10.2024] In step S1, the user can input the target molecular structure Or enter the SMILES of the molecule: CC1=CC=CC(F)=C1NC(=O)OC(C)(C)C, or draw the formula of the molecule on paper, show it to the external camera to obtain the picture and identify the formula of the molecule;

[0074] [According to Rule 26, corrected on 29.10.2024] In step S21, input the obtained target molecular structure Or the SMILES of the molecule: CC1=CC=CC(F)=C1NC(=O)OC(C)(C)C into the first model, and the first model outputs 207 feasible alternative chemical synthesis routes according to the chemical reaction rules and reaction mechanism, of which the top 3 routes are:

[0075] [According to Rule 26, corrected on 29.10.2024] Route 1:

[0076] [According to Rule 26, corrected on 29.10.2024] CC1=CC=CC(F)=C1N >> CC1=CC=CC(F)=C1NC(=O)OC(C)(C)C

[0077] [According to Rule 26, corrected on 29.10.2024] Route 2:

[0078] [According to Rule 26, corrected on 29.10.2024] CC(C)(C)OC(=O)NC1=CC=CC=C1F >> CC1=CC=CC(F)=C1NC(=O)OC(C)(C)C

[0079] [According to Rule 26, corrected on 29.10.2024] Route 3:

[0080] [According to Rule 26, correct on 29.10.2024] CC1 = CC = CC (F) = C1Br >> CC1 = CC = CC (F) = C1NC (= O) OC (C) (C) C

[0081] [According to Rule 26, correct on 29.10.2024] In step S22, the reference reaction is a reaction selected from the first model training data according to the reaction scheme, which has a similarity greater than a threshold value, and can include reactions recorded in literature and existing patents. The preset reaction indicators can include reaction conditions, reaction yield, etc. The reaction yield is obtained by counting the actual yield of the reference reaction.

[0082] [According to Rule 26, correct on 29.10.2024] Then in step S22, the similarity of each candidate chemical reaction route to the reference reaction is calculated. Among them: similar reaction sites (same reaction mechanism), and the structure of the reactant is as similar as possible (the properties of the substances are close). The reaction similarity can be first generated by a neural network to generate a high-dimensional abstract representation of the reaction, and then the similarity is calculated.

[0083] [According to Rule 26, correct on 29.10.2024] Continuing the above example, because route 1 has the highest similarity to the reference reaction recorded in the literature and patents in the first model training data, the reaction conditions are normal temperature and the reaction yield is high, therefore, route 1 is recommended as the target synthesis route of the chemical formula, and the complete synthesis route of route 1 is:

[0084] [According to Rule 26, correct on 29.10.2024] CC (C) (C) OC (= O) OC (= O) OC (C) (C) C. CC1 = CC = CC (F) = C1N >> CC1 = CC = CC (F) = C1NC (= O) OC (C) (C) C

[0085] In another embodiment, different types of chemical formulas (chemical molecular formula, chemical molecular structure formula, chemical molecular SMILES, CAS number, etc.) input by the user can be converted into the same type of chemical formula in step S1; correspondingly, in step S2, the type of chemical formula and the corresponding chemical synthesis route are pre-collected as a training set to train the first model, so that the trained first model can output multiple chemical synthesis routes that can generate the chemical formula according to the type of chemical formula.

[0086] Preferably, the different types of chemical formula input by the user are converted into SMILES, which is more convenient for computer processing than other types of chemical formula. Then the chemical formula in SMILES and the corresponding chemical synthesis route can be collected in advance as a training set to train the first model, and the trained first model can output a plurality of chemical synthesis routes in SMILES format that can generate the chemical formula according to the chemical formula in SMILES. At the same time, the chemical synthesis route in SMILES format can be converted into any other type (such as: chemical formula, chemical molecular structure formula, chemical molecular SMILES, CAS number, etc.) of chemical synthesis route according to the user's preference and output to the user interface. For example, the chemical synthesis route in SMILES format can be converted into any other type of chemical synthesis route through the mapping relationship between SMILES and other types of molecular formula. The following will be described by taking the conversion of different types of chemical formula input by the user in different ways into SMILES as an example.

[0087] In one example of step S1, handwriting input is supported, so the user can input the target chemical molecular structure formula in the input area through an input device (such as: a stylus, a keyboard, etc.) or directly by hand. This step can obtain the target SMILES corresponding to the target chemical molecular structure formula by querying the molecular structure library, wherein the molecular structure library pre-stores SMILES and the corresponding molecular structure formula.

[0088] In another example of step S1, picture input is supported, so the user can place a picture (image) of the target chemical molecular structure formula in the image recognition area to input the target chemical molecular structure formula. This step can identify the target chemical molecular structure formula through a visual recognition algorithm, and obtain the target SMILES corresponding to the target chemical molecular structure formula by querying the molecular structure library.

[0089] In yet another example of step S1, voice input is supported, so the user can read out the name of the target chemical molecular formula to input the target chemical molecular formula in the form of voice. This step can identify the text corresponding to the target chemical molecular formula through a voice recognition algorithm, and convert it into the target SMILES.

[0090] S3, splitting the target chemical synthesis route into an experimental operation process according to the operation sequence, and generating corresponding control instructions to execute the experimental operation process;

[0091] This step designs a corresponding experimental operation process according to the target chemical synthesis route determined in step S2, and generates control instructions to realize the control of the specific experimental operation process. The experimental operation process includes operation steps and operation information in each operation step, and the operation information includes raw material information, reaction conditions, reference reactions, and operation actions.

[0092] For example, the splitting of the target chemical synthesis route into experimental operation procedures according to the operation sequence can include:

[0093] S31, inputting the target chemical synthesis route into a second model to output the required raw materials for the reaction, reference reactions, and experimental operation procedures of the reference reactions;

[0094] Wherein, the experimental operation procedure can be text information containing the operation sequence, operation steps, reaction conditions, and operation actions. Before this step, the second model can be trained in advance by collecting existing chemical synthesis routes and corresponding raw materials, reference reactions, and experimental operation procedures of the reference reactions.

[0095] S32, generating raw material information and experimental operation procedures of the target chemical synthesis route according to the experimental operation procedures of the reference reactions and the required raw materials for the reaction;

[0096] Wherein, the raw material information can include the types of raw materials and the amount of each type. Before this step, the reaction generation model can be trained in advance by collecting the experimental operation procedures of the reference reactions, the required raw materials for the reaction, and the corresponding raw material information and experimental operation procedures. The raw material information and experimental operation procedures of the target chemical synthesis route are obtained through the reaction generation model.

[0097] S33, generating experimental operation procedures according to the raw material information and experimental operation procedures of the target chemical synthesis route.

[0098] For example, the experimental operation procedures can be output in the form of an experimental flowchart according to the raw material information and experimental operation procedures of the target chemical synthesis route. Before this step, the splitting model can be trained in advance by collecting existing raw material information, experimental operation procedures, and corresponding experimental flowcharts. The experimental operation procedures are obtained through the splitting model.

[0099] The following is an example of step S2, still taking the preparation of (2-fluoro-6-methylphenyl) tert-butyl carbamate molecule as an example to illustrate step S3 in detail.

[0100] In step S31, route 1 in step S22 is input into the second model to output the following information:

[0101] (1) Required raw materials for the reaction:

[0102] (2) Reference reactions:

[0103] CC(C)(C)OC(=O)OC(=O)OC(C)(C)C.CC1=CC=C(C(Br)=C1N) > CC1=CC=C(C(Br)=C1N)

[0104] (3) Experimental operation process of the reference reaction: N-tert-butyloxycarbonylation (Boc protection) of amine: At room temperature (30-35 °C), water (2.5 mL) was mixed with 2-bromo-6-methylaniline (0.235 g, 2.5 mmol) by stirring, then (Boc)20 (0.60 g, 2.75 mmol, 1.1 equivalent) was added to form transparent liquid drops. After stirring the reaction mixture, white emulsion appeared on the surface of the reaction vessel wall, accompanied by slow bubbling. When white solid precipitated, it indicated that the reaction was completed (monitored by thin layer chromatography and infrared spectroscopy after 30 minutes of reaction), and the supernatant was discarded. Water (2 times x 5 mL) was added to the solid residue, and the mixture was stirred for a few minutes. After each time of adding water, the water was discarded. The residual solid was dried under vacuum to obtain the target product 0.46 g as a white solid with a yield of 95%.

[0105] (4-bromo-2-methylphenyl) tert-butylcarbamate: off-white solid, yield 90%, melting point: 97 °C. 1H NMR (300 MHz; CDC13) δ: 1.51 (s, 9H), 2.21 (s, 3H), 6.21 (bs, 1H), 7.27-7.30 (m, 2H), 7.72 (d, J = 8.05 Hz, 1H). MS (EI): m / z 286 (M+).

[0106] In step S32, the experimental operation process of the reference reaction in step S31 and the raw materials required for the reaction are input into the reaction generation model to obtain the raw material information and the experimental operation process of Route 1 as follows:

[0107] The raw material information is shown in FIG. 2.

[0108] (2) Experimental operation process: N-tert-butyloxycarbonylation (Boc protection) of amine: At room temperature (30-35 °C), water (2.5 mL) was mixed with 2-bromo-6-methylaniline (0.235 g, 2.5 mmol) by stirring, then (Boc)20 (0.60 g, 2.75 mmol, 1.1 equivalent) was added to form transparent liquid drops. After stirring the reaction mixture, white emulsion appeared on the surface of the reaction vessel wall, accompanied by slow bubbling. When white solid precipitated, it indicated that the reaction was completed (monitored by thin layer chromatography and infrared spectroscopy after 30 minutes of reaction), and the supernatant was discarded. Water (2 times x 5 mL) was added to the solid residue, and the mixture was stirred for a few minutes. After each time of adding water, the water was discarded. The residual solid was dried under vacuum to obtain the target product 0.46 g as a white solid with a yield of 95%.

[0109] In step S33, the raw material information in step S32 and the experimental operation process are input into the splitting model, and the experimental operation process of route 1 is obtained as shown in FIG. 3.

[0110] Through the above steps S1-S3, the user only needs to input the chemical molecular formula to be synthesized, and the present application can determine the optimal chemical synthesis route according to the user input chemical molecular formula, and split the optimal chemical synthesis route into the experimental operation process for execution. Thus, the decision and operation instructions of the entire experiment are independently generated, without human intervention, providing decision support for the robot to independently complete the entire chemical experiment. Further, the present application also supports environmental perception feedback, which can realize physical perception and understanding of the surrounding environment, thereby providing environmental reference information for the chemical experiment operation process to assist in completing the execution or abnormal judgment of the experiment operation. The method can further include:

[0111] S4, real-time monitoring of environmental information, and outputting feedback information according to the environmental information;

[0112] In this embodiment, the environmental information can be at least one of image information, audio information, and physical parameter information in the surrounding environment; the physical parameter information can be temperature, humidity, pressure, pH value, etc. This step can collect one or more of image information, audio information, and physical parameter information in the environment, and output feedback information according to the collected information. For example, historical image information, audio information, and physical parameter information in the surrounding environment can be pre-collected as training data to train a feedback model, and the feedback information can be output by the feedback model.

[0113] In this embodiment, the feedback information is different according to the different monitored environmental information. In one example, the image information in the environment can be collected, and the feedback model analyzes the image information and compares the analysis result with the experimental material information to obtain target experimental material information; the region where the target experimental material information is located is output as the region where the material corresponding to the image information is located, thereby assisting in positioning the material during the experimental operation process. Wherein: the image information can be a label image on a material bottle, which can be a label code, a two-dimensional code, etc. The experimental material information and its region can be pre-stored in an experimental material library. For example, in one experimental scenario, the user tells the robot to start the experiment according to the experimental operation process output in step S3, the robot goes to the material area, captures the two-dimensional code label image of the material bottle through the external camera, analyzes the two-dimensional code label in the image, and compares it with the existing experimental material information in the experimental physical library to obtain the target experimental material information; and the region where the target experimental material information is located is output as the region where the material corresponding to the image information is located.

[0114] In another example, at least one physical parameter information in the environment can be collected, the physical parameter information is compared with the set physical parameter information, and whether the physical parameter is abnormal is judged and output according to the comparison result; so as to identify the physical parameter abnormality in the experiment process. Wherein: the set physical parameter information can be the standard value of the physical parameter in the experiment process. For example, in an experimental scene, the real-time temperature in the reaction container is monitored by a thermocouple, and if it is judged that the current monitoring temperature deviates from the set reaction temperature, the temperature abnormality is output, and further, the temperature abnormality can be sent to the subsequent abnormal situation investigation to provide analysis basis for the abnormal situation investigation.

[0115] In yet another example, image information and / or audio information can be collected, the image information and / or the audio information is analyzed, and the analysis result is output to realize interaction with the user in the experiment process. Wherein: the analysis of the image information can be: using a visual recognition algorithm to identify the target in the image, and the analysis of the audio information can be: using a large language model to process the user input audio. The analysis of the image information and the audio information can be: comprehensive processing of the image recognition result and the audio processing result, wherein: the image recognition result is the output result after analyzing and understanding the current image, and the large language model generates a reply output based on the output result after analyzing and understanding the current image and the user input audio. For example, in an experimental scene, a user draws the molecular structure formula of (2-fluoro-6-methylphenyl) tert-butyl carbamate on a white paper and places it under an external camera, and asks the robot in voice: can you help me synthesize 1g of this molecule? The robot obtains the image captured by the camera, identifies the molecular structure of the molecule through a visual recognition algorithm, and asks the user in voice whether it is the molecule The robot generates a target chemical synthesis route and an experimental operation process through steps S2 and S3 after the user confirms, and informs the user of the design result through voice.

[0116] Based on the above steps S1-S3, during the reaction process, the spectrum data of the reactants can be analyzed to generate analysis results for reference by the experiment personnel. Therefore, the method can further include:

[0117] S5, receiving spectrum data of reactants in different reaction nodes, determining and outputting spectrum analysis results according to the target chemical synthesis route and the spectrum data;

[0118] Wherein: different reaction nodes can be set according to test needs, such as: 5-8min after the reaction starts, 1h after the reaction, etc., which is not limited by the present application.

[0119] The spectrum data can include at least one of liquid chromatography-mass spectrometry (LC-MS), infrared spectrum (IR), ultraviolet-visible spectrum (UV-Vis). Wherein: the LC-MS spectrum includes: liquid chromatography data file of retention time and absorption signal intensity, and mass / charge ratio (m / z) and corresponding signal intensity mass spectrometry data file (such as mzML, mzXML, mgf, etc.), IR spectrum includes: data file of wave number (cm-1) and absorption intensity, and UV-Vis spectrum includes: data file of absorption or transmission light intensity and wavelength. The spectrum analysis result can include: qualitative analysis of substances existing in the reaction intermediate sample, and quantitative analysis of absolute or relative concentration of each compound in the reaction intermediate sample.

[0120] Before this step, historical chemical synthesis routes, spectrum data and corresponding spectrum analysis results can be used as a training set to train a spectrum analysis model. Then, the target chemical synthesis route and spectrum data are input into the spectrum analysis model to output the spectrum analysis result. Further, the spectrum analysis result can be output to a display for a worker to view.

[0121] The following reaction is performed using 1-benzyl-1H-pyrrole-2,5-dione and ethyl diazoacetate

[0122] The following reaction is performed using 1-benzyl-1H-pyrrole-2,5-dione and ethyl diazoacetate

[0123] According to the experimental operation process in step S3, the current chemical reaction information is obtained through the data interface. The chemical reaction information can include: the reaction formula corresponding to the current reaction (i.e. reactant and product information), and all material information added in the reaction system.

[0124] Before the reaction starts, the raw materials 1-benzyl-1H-pyrrole-2,5-dione and ethyl diazoacetate are detected by LC-MS, and the detected original spectrum as shown in FIGS. 4a-4d is uploaded or sent to the robot, and the robot receives the initial original spectrum of the raw materials.

[0125] After the reaction starts, a small amount of reactant sample is taken for LC-MS detection during the reaction, and the spectrum as shown in FIGS. 5a-5b is obtained and uploaded or sent to the robot, and the robot receives the spectrum in the reaction.

[0126] The robot inputs the received original spectrum, the spectrum in the reaction, and the target chemical synthesis route in step S2 into the spectrum analysis model, and outputs the following results:

[0127] 1. There are three chemical substances in the current reaction system: 1-benzyl-1H-pyrrole-2,5-dione, ethyl diazoacetate and 5-benzyl-4,6-dioxo-1,3a,4,5,6,6a-hexahydro pyrrolo[3,4-c]pyrazole-3-carboxylate;

[0128] 2. The three substances have absorption under ultraviolet and mass spectrometry detection environment;

[0129] 3. The first two substances are raw materials, and the third substance is a product;

[0130] 4. The concentration ratio of the three substances is about 3.43%, 89.41%, and 6.36% (the absorption of related substances in this reaction at 254 nm is stronger)

[0131] Further, based on the above spectrum analysis results, the reaction suggestion can be given: the reactants are still residual, and it is suggested to continue the reaction.

[0132] Based on the above steps S1-S5, during the reaction process, various parameters of the reaction process can be monitored in real time to identify abnormal situations or unsatisfactory reactions in the chemical reaction process, analyze possible reasons, and propose solutions. Among them: the monitored parameters can be at least one of the feedback information in step S4, the spectrum analysis result in step S5, or the user input experimental information, and the method can further include:

[0133] S6, according to at least one of the feedback information, the spectrum analysis result, the user input experimental information, the abnormality in the experimental process is investigated, and the abnormal reason and the abnormal solution are output.

[0134] In this embodiment, the user input experimental information can be the problem description or fault phenomenon in the reaction process input by the user through voice, text, etc. Illustratively, the abnormal reason can be: insufficient purity of raw materials, unsuitable reaction conditions, equipment failure, occurrence of side reactions, etc.; the abnormal solution can be to replace the synthesis route, adjust the reaction conditions, replace the reagent, clean the equipment, etc.

[0135] Specifically, the feedback information can be that in step S4, by collecting at least one physical parameter information in the environment, comparing the physical parameter information with the set physical parameter information, judging and outputting whether the physical parameter is abnormal according to the comparison result; the physical parameter abnormality identified in the experimental process is the environmental abnormality information.

[0136] Before this step, at least one of historical environmental abnormal information, atlas analysis result, and user input experiment information, and corresponding abnormal reason and abnormal solution can be used as a training set to train an abnormal analysis model, and then at least one of real-time collected environmental abnormal information, atlas analysis result, and user input experiment information is input into the abnormal analysis model, and the abnormal reason and the abnormal solution are output.

[0137] The step S6 is described in detail below through a specific example.

[0138] For example, in the process of the following chemical reaction,

[0139] The physical parameter information (temperature value, pressure value) is monitored in real time through the step S4, and the compound composition in the reaction system is obtained through real-time monitoring and analysis of the step S5.

[0140] The spectrum analysis result in the reaction process shows that the following substances not in the expectation appear in the sample through the step S5:

[0141] The spectrum analysis result is input into the abnormal analysis model, and the output is that the reaction has an abnormal situation, and the reason can be that the raw material is impure and contains impurities which leads to the following side reactions:

[0142] And the abnormal solution output is to replace the raw material with higher purity.

[0143] In addition to the above various embodiments, the present application can also provide an intuitive and friendly user interface, support interaction with users through voice, images and video based on the user interface, facilitate user problem input, and output of various analysis results in the present solution.

[0144] It should be noted that the training of the model provided in each step of the present application can be based on a pre-trained model with some general and broad capabilities, and the pre-trained model is further trained on its corresponding training set to adjust the pre-trained model parameters to obtain a model applicable to a specific field in each step, so as to better complete the scene and task faced by the model. In addition, the model can be optimized and trained by continuously collecting various data (such as user feedback and experiment results, etc.) in the experiment process to improve the accuracy of the model.

[0145] Based on the above chemical experiment decision generation method, the present application also provides a chemical experiment decision generation device, and FIG. 6 is a structural framework schematic diagram of a chemical experiment decision generation device according to the present application, as shown in FIG. 6, the device comprises:

[0146] The receiving module 61 is configured to receive a target chemical formula input by a user.

[0147] The first generating module 62 is configured to generate a target chemical synthesis route according to the target chemical formula.

[0148] The second generating module 63 is configured to split the target chemical synthesis route into an experimental operation process according to an operation sequence, and generate a corresponding control instruction to execute the experimental operation process; the experimental operation process includes operation steps and operation information in each operation step, and the operation information includes raw material information, reaction conditions, reference reactions, and operation actions.

[0149] Further, the device further includes:

[0150] The feedback module is configured to monitor environmental information in real time, and output feedback information according to the environmental information.

[0151] And / or,

[0152] The spectrum analysis module is configured to receive spectrum data of reaction intermediate samples in different reaction nodes, determine and output spectrum analysis results according to the target chemical synthesis route and the spectrum data.

[0153] And,

[0154] The anomaly analysis module is configured to troubleshoot anomalies in an experimental process according to at least one of the feedback information, the spectrum analysis results, and user-input experimental information, and output anomaly causes and anomaly solutions.

[0155] In a specific embodiment, the first generating module 62 includes:

[0156] The first input module is configured to input the target chemical formula into a first model, and output a plurality of alternative chemical synthesis routes according to a ranking.

[0157] The screening module is configured to select a target chemical synthesis route from the plurality of alternative chemical synthesis routes according to a similarity between each alternative chemical synthesis route and a reference reaction, and a preset reaction index.

[0158] The second generating module 63 includes:

[0159] The second input module is configured to input the target chemical synthesis route into a second model, and output raw materials required for a reaction, reference reactions, and experimental operation processes of the reference reactions.

[0160] The first sub-generating module is configured to generate raw material information and experimental operation processes of the target chemical synthesis route according to the experimental operation processes of the reference reactions and the raw materials required for the reaction.

[0161] The second sub-generation module is configured to generate an experimental operation process according to the raw material information of the target chemical synthesis route and an experimental operation process.

[0162] The feedback module is specifically configured to collect at least one of image information, audio information, and physical parameter information in the environment, and output feedback information according to the collected at least one information.

[0163] In a specific embodiment, the feedback module collects image information in the environment, analyzes the image information, and compares the analysis result with experimental material information to obtain target experimental material information; and outputs a region where the target experimental material information is located as a region where a material corresponding to the image information is located.

[0164] Alternatively,

[0165] The feedback module collects at least one physical parameter information in the environment, compares the physical parameter information with set physical parameter information, and judges and outputs whether the physical parameter is abnormal according to the comparison result.

[0166] Alternatively,

[0167] The feedback module collects image information and / or audio information in real time, analyzes the image information and / or the audio information, and outputs an analysis result.

[0168] In a specific embodiment, the feedback information is environment abnormal information, the anomaly analysis module inputs at least one of the environment abnormal information, a spectrum analysis result, and user input experimental information into an anomaly recognition model, and outputs an anomaly cause and an anomaly solution.

[0169] Those skilled in the art can understand that each module in the above-mentioned device embodiments can be distributed in the device as described, or can be changed accordingly and distributed in one or more devices different from the above-mentioned embodiments. The modules of the above-mentioned embodiments can be combined into one module, or can be further split into multiple sub-modules.

[0170] The chemical experiment decision-making robot embodiment of the present application is described below. The robot can be regarded as an embodiment of the entity form of the above-mentioned method and device embodiments of the present application. For the details described in the robot embodiment of the present application, it should be regarded as a supplement to the above-mentioned method or device embodiments; for the details not disclosed in the robot embodiment of the present application, reference can be made to the above-mentioned method or device embodiments.

[0171] FIG. 7 is a structural block diagram of an exemplary embodiment of a robot according to the present application. The robot shown in FIG. 7 is only an example and should not limit the function and use range of the embodiments of the present application in any way.

[0172] As shown in FIG. 7, the robot 400 of this example embodiment is represented in the form of a general purpose data processing device. The components of the robot 400 can include, but are not limited to, at least one processing unit 410, at least one storage unit 420, a bus 430 that connects the various electronic device components, including the storage unit 420 and the processing unit 410, a display unit 440, and the like.

[0173] The storage unit 420 has stored therein computer readable programs, which can be source or read-only program codes. The programs can be executed by the processing unit 410 such that the processing unit 410 performs the steps of various embodiments of the present application. For example, the processing unit 410 can perform the steps shown in FIG. 1.

[0174] The storage unit 420 can include a readable medium in the form of a volatile storage unit, such as a random access memory (RAM) 4201 and / or a cache memory unit 4202, and can further include a read-only memory (ROM) 4203. The storage unit 420 can also include programs / utilities 4204 having a set (at least one) of program modules 4205, including but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which can include implementation of a network environment, alone or in some combination.

[0175] The bus 430 can be representative of one or more of several types of bus structures, including a storage unit bus or storage unit controller, a peripheral bus, a graphics acceleration port, a processing unit bus, or a local bus using any of a variety of bus structures.

[0176] The robot 400 can also communicate with one or more external devices 300 such as a keyboard, a display, a network device, a Bluetooth device, and so on, such that a user can interact with the robot 400 via the external devices 300 and / or such that the robot 400 can communicate with one or more other data processing devices, such as a router, a modem, and so on. Such communication can be via the input / output (I / O) interface 450, and can also be via a network adapter 460 to one or more networks, such as a local area network (LAN), a wide area network (WAN), and / or the public network, such as the Internet. The network adapter 460 can communicate to the other modules of the robot 400 via the bus 430. It should be appreciated that although not shown in FIG. 7, other hardware and / or software modules can be used in the robot 400, including but not limited to, microcode, device drivers, redundant processing units, external disk drive arrays, RAID electronic devices, tape drives, and data backup storage electronic devices, and the like.

[0177] FIG. 8 is a schematic diagram of an embodiment of a computer readable medium of the present application. As shown in FIG. 8, the computer program can be stored on one or more computer readable media. The computer readable media can be a readable signal medium or a readable storage medium. The readable storage medium, for example, can be, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include an electrical connection having one or more wires, a portable disc, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. When the computer program is executed by one or more data processing devices, the computer readable medium enables the above-described method of the present application, i.e., receiving a target chemical formula input by a user; generating a target chemical synthesis route according to the target chemical formula; splitting the target chemical synthesis route into an experimental operation flow according to an operation sequence, and generating corresponding control instructions to execute the experimental operation flow; the experimental operation flow includes operation steps and operation information in each operation step, and the operation information includes raw material information, reaction conditions, and operation actions.

[0178] From the above description of the embodiments, those skilled in the art can easily understand that the exemplary embodiments described in the present application can be implemented by software, or by software in combination with necessary hardware. Therefore, the technical solutions according to the embodiments of the present application can be embodied in the form of a software product, which can be stored in a computer readable storage medium (which can be a CD-ROM, an U disk, a mobile hard disk, etc.) or a network, and includes a number of instructions to enable a data processing device (which can be a personal computer, a server, or a network device, etc.) to perform the above-described method according to the present application.

[0179] The computer readable storage medium can include a data signal carried in a baseband or as part of a carrier wave propagating through the transmission medium, and bearing the readable program code. Such a propagated data signal can take on many forms, including but not limited to electro-magnetic, optical, or any suitable combination thereof. The computer readable storage medium can also be any medium that can be read by a readable medium other than the readable storage medium, which can send, propagate or transmit the program for use by or in connection with an instruction execution electronic device, apparatus or device. The program code contained on the readable storage medium can be transmitted by any suitable medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination thereof.

[0180] The program code may be executed by one or more programmable processing devices, which can include microprocessors, digital signal processors (DSPs), central processing units (CPUs), graphics processing units (GPUs) or field programmable gate arrays (FPGAs).

[0181] In summary, the present application can be implemented by a computer program method, device, electronic device or computer readable medium. Some or all of the functions of the present application can be implemented in practice using a general-purpose data processing device such as a microprocessor or a digital signal processor (DSP).

[0182] The above-described specific embodiments have further detailed the purposes, technical solutions and beneficial effects of the present application. It should be understood that the present application is not inherently related to any specific computer, virtual device or electronic device, and various general-purpose devices can implement the present application. The above-described specific embodiments are merely examples of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for generating a chemical experiment decision, characterized by, The method comprises: receiving a target chemical formula input by a user; generating a target chemical synthesis route according to the target chemical formula; splitting the target chemical synthesis route into an experimental operation process according to an operation sequence, and generating corresponding control instructions to execute the experimental operation process; the experimental operation process comprises operation steps and operation information in each operation step, and the operation information comprises raw material information, reaction conditions, and operation actions.

2. The method of claim 1, wherein, The method further comprises: monitoring environmental information in real time and outputting feedback information according to the environmental information; and / or, receiving spectral data of reactants in different reaction nodes, determining and outputting spectral analysis results according to the target chemical synthesis route and the spectral data; and, checking an abnormality in an experimental process according to at least one of the feedback information, the spectral analysis results, and experimental information input by the user, and outputting an abnormality cause and an abnormality solution.

3. The method of claim 1, wherein, The generation of the target chemical synthesis route according to the target chemical formula comprises: inputting the target chemical formula into a first model, and outputting a plurality of alternative chemical synthesis routes according to a ranking; selecting a target chemical synthesis route from the plurality of alternative chemical synthesis routes according to a similarity of each alternative chemical synthesis route to a reference reaction and a preset reaction index.

4. The method of claim 1, wherein, The splitting of the target chemical synthesis route into an experimental operation process according to an operation sequence comprises: inputting the target chemical synthesis route into a second model, outputting a required raw material for a reaction, a reference reaction, and an experimental operation process of the reference reaction; generating raw material information and an experimental operation process of the target chemical synthesis route according to the experimental operation process of the reference reaction and the required raw material for the reaction; generating an experimental operation process according to the raw material information and the experimental operation process of the target chemical synthesis route.

5. The method of claim 2, wherein, The real-time monitoring of environmental information and the output of feedback information according to the environmental information comprise: collecting at least one of image information, audio information, and physical parameter information in an environment; outputting feedback information according to the collected at least one information.

6. The method of claim 5, wherein: image information in an environment is collected, the image information is analyzed, and target experimental material information is obtained by comparing an analysis result with experimental material information; a region where the target experimental material information is located is output as a region where material corresponding to the image information is located; or, at least one physical parameter information in an environment is collected, the physical parameter information is compared with set physical parameter information, and whether a physical parameter is abnormal is determined and output according to a comparison result; or, image information and / or audio information is collected in real time, the image information and / or the audio information is analyzed, and an analysis result is output.

7. The method of claim 2, wherein, The feedback information is environmental abnormality information, at least one of the environmental abnormality information, spectral analysis results, and experimental information input by a user is input into an abnormality recognition model, and an abnormality cause and an abnormality solution are output.

8. A chemical experiment decision generation device characterized by comprising: The device comprises: a receiving module configured to receive a target chemical formula input by a user; a first generation module configured to generate a target chemical synthesis route according to the target chemical formula; A second generation module is configured to split the target chemical synthesis route into an experimental operation flow according to an operation sequence, and generate corresponding control instructions to execute the experimental operation flow; the experimental operation flow includes operation steps and operation information in each operation step, and the operation information includes raw material information, reaction conditions, and operation actions.

9. A chemical experiment decision-making generation robot, characterized by, comprising: a processor; and a memory storing computer-executable instructions that, when executed, cause the processor to perform the method of any of claims 1-7.

10. A computer readable storage medium, wherein, The computer-readable storage medium stores one or more programs, which when executed by a processor, implement the method of any of claims 1-7.

Images