A workflow task processing system to assist high-voltage electrolyte experiments

Abstract

The embodiment of the present invention relates to a workflow task processing system for assisting high-voltage electrolyte experiments, the system comprising: a workflow task scheduling module, a workflow task pool, a workflow task database, a task execution module, a molecular dynamics simulation interface library, a quantum chemical calculation interface library, and an AI model interface library; wherein the molecular dynamics simulation interface library is connected to the functional interface of the molecular dynamics simulation software, the quantum chemical calculation interface library is connected to the functional interface of the quantum chemical calculation software, and the AI ​​model interface library is connected to the functional interface of the AI ​​model. The workflow task scheduling module completes three types of auxiliary task processing by calling the task execution module: high-throughput screening tasks, mechanism prediction tasks, and performance prediction tasks. Using this system as an auxiliary means for high-voltage electrolyte experiments can achieve the purpose of reducing the number of experimental trial and error, improving experimental accuracy, shortening the experimental cycle, and reducing experimental costs.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a workflow task processing system for assisting high-voltage electrolyte experiments. Background Art

[0002] The traditional approach to developing high-voltage electrolytes is experimental. Researchers first specify an initial experimental plan, synthesize the electrolyte based on it, and use a series of experimental tests to observe the electrolyte's physical and chemical phenomena. Quantum chemical calculations are then used to interpret the experimental results and optimize the experimental plan. These steps are then repeated based on the optimized plan, repeating the process until the final experimental results meet the preset specifications. Due to factors such as limited experimental resources, this experimental approach has long development cycles and high costs, which have not been effectively addressed. Summary of the Invention

[0003] The purpose of the present invention is to address the defects of the existing technology and provide a workflow task processing system for assisting high-voltage electrolyte experiments, including: a workflow task scheduling module, a workflow task pool, a workflow task database, a task execution module, a molecular dynamics simulation interface library, a quantum chemistry calculation interface library and an AI model interface library; among them, the molecular dynamics simulation interface library can be connected to the functional interface of any molecular dynamics simulation software, the quantum chemistry calculation interface library can be connected to the functional interface of any quantum chemistry calculation software, and the AI ​​model interface library can be connected to the functional interface of any AI model. The workflow task scheduling module completes three types of auxiliary task processing by calling the task execution module: high-throughput screening tasks, mechanism prediction tasks and performance prediction tasks. If this system is applied to the experimental development of high-voltage electrolytes, it can complete the screening of large quantities of electrolytes for high-voltage electrolyte experiments by executing high-throughput screening tasks. After obtaining a batch of electrode + electrolyte systems, it can perform mechanism prediction tasks to predict the molecular-level reaction mechanism, interfacial reaction mechanism, and bipolar SEI / CEI film formation mechanism of each electrode + electrolyte system. It can also complete the performance prediction work of batch electrolytes for high-voltage electrolyte experiments by executing performance prediction tasks. Using this system as an auxiliary means for high-voltage electrolyte experiments can achieve the goals of reducing the number of experimental trial and error, improving experimental accuracy, shortening the experimental cycle, and reducing experimental costs.

[0004] To achieve the above objectives, an embodiment of the present invention provides a workflow task processing system for assisting high-voltage electrolyte experiments, the system comprising: a workflow task scheduling module, a workflow task pool, a workflow task database, a task execution module, a molecular dynamics simulation interface library, a quantum chemistry calculation interface library, and an AI model interface library;

[0005] The workflow task scheduling module is connected to the workflow task pool, the workflow task database and the task execution module respectively; the workflow task scheduling module is used to receive the first task file sent by the high-voltage electrolyte experiment party; and initialize a corresponding task data storage area in the workflow task pool as the first data area; and perform sub-data area planning on the first data area according to the first task file; and call the task execution module to perform task execution processing on the first task file; and after confirming that the task execution processing is completed, perform task report preparation processing according to the first task file and the first data area to generate a corresponding first report file and send it back to the high-voltage electrolyte experiment party; and the first task file and the first report file form a corresponding task backup record and store it in the workflow task database; and delete the first data area;

[0006] The workflow task pool is used to store a plurality of task data storage areas;

[0007] The workflow task database is used to store a plurality of task backup records;

[0008] The task execution module is respectively connected to the molecular dynamics simulation interface library, the quantum chemical calculation interface library and the AI ​​model interface library; the task execution module includes a first scheduling unit, a first scheduling task pool and a first prediction unit set; the first scheduling unit is respectively connected to the workflow task scheduling module, the first scheduling task pool and the first prediction unit set; the first scheduling task pool is connected to the first prediction unit set; the first prediction unit set includes a basic physical property prediction unit, a transport property prediction unit, an electrochemical property prediction unit, a molecular reaction mechanism prediction unit, an interface reaction mechanism prediction unit, an SEI / CEI film formation mechanism prediction unit and a battery performance prediction unit; each prediction unit of the first prediction unit set is connected to one or all of the three interface libraries: the molecular dynamics simulation interface library, the quantum chemical calculation interface library and the AI ​​model interface library;

[0009] The first scheduling unit is used to initialize a corresponding data storage area in the first scheduling task pool as the second data area when receiving any first task execution file sent by the workflow task scheduling module, and allocate a unique data area identifier to the current second data area as the corresponding second data area identifier, and set a return data area and a task status data initialized to an unfinished state in the second data area; and use the prediction unit in the first prediction unit set that matches the first prediction unit type of the first task execution file as the corresponding first prediction unit; and send the second data area identifier and the first prediction input data of the first task execution file to the first prediction unit; and regularly Identify whether the task status data in the second data area is in a completed state; if so, extract the data in the return data area of ​​the second data area and send it back to the workflow task scheduling module as the corresponding first execution result file, and delete the second data area at the end of the send-back; the first task execution file includes the first prediction unit type and the first prediction input data; the first prediction unit type includes a basic physical property prediction type, a transport property prediction type, an electrochemical property prediction type, a molecular reaction mechanism prediction type, an interface reaction mechanism prediction type, an SEI / CEI film formation mechanism prediction type, and a battery performance prediction type; the task status data includes an unfinished state and a completed state;

[0010] Each prediction unit in the first prediction unit set is used to locally preset a corresponding prediction processing flow, and the prediction processing flow calls the interfaces of the three interface libraries; each prediction unit in the first prediction unit set is also used to, when receiving the second data area identifier and the first prediction input data sent by the first scheduling unit, perform corresponding property, mechanism or performance prediction processing according to the first prediction input data by the locally preset prediction processing flow to obtain corresponding first prediction output data, and use the second data area corresponding to the second data area identifier in the first scheduling task pool as the corresponding current data area, and store the first prediction output data in the return data area of ​​the current data area, and update the task status data in the current data area to a completed state;

[0011] The molecular dynamics simulation interface library includes a plurality of first simulation interfaces;

[0012] The quantum chemical calculation interface library includes a plurality of first calculation interfaces;

[0013] The AI ​​model interface library includes multiple first AI model interfaces.

[0014] Preferably, the first task file includes a first task type, first task data and a first task configuration;

[0015] The first task types include high-throughput screening type, mechanism prediction type and performance prediction type;

[0016] When the first task type is a high-throughput screening type, the corresponding first task data is a first electrolyte data set to be screened; the first electrolyte data set includes one or more first electrolyte data; the first electrolyte data includes a first component data set and a first component ratio; the first component data set includes multiple first component data, the first component data includes a first component type, a first component name and a first component SMILES sequence, the first component type includes an electrolyte salt, a solvent and an additive; the first component ratio is the ratio of the electrolyte salt, the solvent and the additive of the corresponding electrolyte, and the first component ratio = electrolyte salt ratio: solvent ratio: additive ratio;

[0017] When the first task type is a high-throughput screening type, the corresponding first task is configured as a multi-level screening indicator set, consisting of a primary physical property indicator set, a secondary transport property indicator set, and a tertiary electrochemical property indicator set; the primary physical property indicator set includes a viscosity indicator, a dielectric constant indicator, a melting point temperature indicator, and a boiling point temperature indicator; the secondary transport property indicator set includes a diffusion coefficient indicator, a conductivity indicator, and a migration number indicator; the tertiary electrochemical property indicator set includes an oxidation potential indicator, a reduction potential indicator, and an electrochemical window indicator;

[0018] When the first task type is a mechanism prediction type, the corresponding first task data is a first electrode + electrolyte system set to be predicted; the first electrode + electrolyte system set includes one or more first electrode + electrolyte systems; the first electrode + electrolyte system is an interface model system formed by the fusion of an electrode plate model system and an electrolyte model system, and the first electrode + electrolyte system includes multiple first atoms, multiple first connecting bonds and multiple first clusters; the atomic properties of the first atom include atomic identification, atomic element type, atomic charge, atomic size, atomic coordinates and the cluster identification of the atom; the bond properties of the first connecting bond include connecting bond identification, connecting bond type and the atom pair identification group corresponding to the connecting bond; the first cluster includes cluster identification and cluster type;

[0019] When the first task type is a mechanism prediction type, the corresponding first task is configured as a multi-class prediction state machine, which consists of a molecular reaction mechanism prediction state bit, an interface reaction mechanism prediction state bit, and an SEI / CEI film formation mechanism prediction state bit, and all three state bits include two state configuration values: an activation state and an inactivation state; and the state configuration value of at least one of the three state bits is an activation state;

[0020] When the first task type is a performance prediction type, the corresponding first task data is a second electrolyte data set to be predicted; the second electrolyte data set includes one or more second electrolyte data; the second electrolyte data includes a second component data set and a second component ratio; the second component data set includes a plurality of second component data, the second component data includes a second component type, a second component name, and a second component SMILES sequence, the second component type includes an electrolyte salt, a solvent, and an additive; the second component ratio is the ratio of the electrolyte salt, the solvent, and the additive of the corresponding electrolyte, and the second component ratio = electrolyte salt ratio: solvent ratio: additive ratio;

[0021] When the first task type is a performance prediction type, the corresponding first task configuration is empty;

[0022] Each of the first simulation interfaces is a pre-specified function call interface of molecular dynamics simulation software; each of the first calculation interfaces is a pre-specified function call interface of quantum chemistry calculation software; each of the first AI model interfaces is a pre-specified function call interface of an AI model.

[0023] Preferably, the workflow task scheduling module is specifically configured to identify the first task type of the first task file when planning sub-data areas of the first data area according to the first task file;

[0024] When the first task type is a high-throughput screening type, the number of the first electrolyte data in the first electrolyte data set of the first task file is counted to obtain the corresponding total number of first electrolytes; a sub-data area of ​​the total number of the first electrolytes is created in the first data area and recorded as the corresponding first electrolyte data area, and a high-throughput screening task output data area is created; and three sub-data areas are further created in each of the first electrolyte data areas, namely, a primary physical property prediction data area, a secondary transport property prediction data area, and a tertiary electrochemical property prediction data area; the first electrolyte data areas correspond one-to-one to the first electrolyte data;

[0025] When the first task type is a mechanism prediction type, the number of the first electrode + electrolyte systems in the first electrode + electrolyte system set of the first task file is counted to obtain the corresponding total number of first systems; and a sub-data area of ​​the total number of the first systems is created in the first data area and recorded as the corresponding first system data area; and the three status bits of the first task configuration of the first task file are identified, if the molecular reaction mechanism prediction status bit is in an activated state, a corresponding molecular reaction mechanism prediction data area is created in each of the first system data areas, if the interface reaction mechanism prediction status bit is in an activated state, a corresponding interface reaction mechanism prediction data area is created in each of the first system data areas, and if the SEI / CEI film formation mechanism prediction status bit is in an activated state, a corresponding SEI / CEI film formation mechanism prediction data area is created in each of the first system data areas; the first system data area corresponds one-to-one to the first electrode + electrolyte system;

[0026] When the first task type is a performance prediction type, the number of the second electrolyte data in the second electrolyte data set of the first task file is counted to obtain the corresponding total number of second electrolytes; and a sub-data area of ​​the total number of the second electrolytes is created in the first data area and recorded as the corresponding second electrolyte data area; the second electrolyte data corresponds one-to-one to the second electrolyte data area.

[0027] Preferably, the workflow task scheduling module is specifically configured to identify the first task type of the first task file when the task execution module is called to perform task execution processing on the first task file;

[0028] When the first task type is a high-throughput screening type, calling the task execution module and the first data area to perform high-throughput screening task execution processing according to the first task file; and confirming the completion of the current high-throughput screening task execution processing when the current high-throughput screening task execution processing is completed;

[0029] When the first task type is a mechanism prediction type, calling the task execution module and the first data area to perform mechanism prediction task execution processing according to the first task file; and confirming the completion of the current task execution processing when the current mechanism prediction task execution processing ends;

[0030] When the first task type is a performance prediction type, the task execution module and the first data area are called to perform performance prediction task execution processing according to the first task file; and the task execution processing is confirmed to be completed when the performance prediction task execution processing ends.

[0031] Furthermore, the workflow task scheduling module is specifically configured to extract each first electrolyte data of the first electrolyte data set of the first task file as a corresponding first prediction input data when the task execution module and the first data area are called to perform high-throughput screening task execution processing according to the first task file, and set a corresponding first prediction unit type as a basic physical property prediction type, and form a corresponding first task execution file by the first prediction unit type corresponding to each first electrolyte data and the first prediction input data; and send each first task execution file obtained this time to the first scheduling unit of the task execution module;

[0032] and receiving the first execution result files corresponding to the first task execution files returned by the first scheduling unit; and each time a first execution result file is received, taking the first physical property prediction data area of ​​the first electrolyte data area corresponding to the first electrolyte data corresponding to the current first execution result file in the first data area as the corresponding current data area, and extracting the corresponding viscosity prediction data, dielectric constant prediction data, melting point temperature prediction data, and boiling point temperature prediction data from the current first execution result file and storing them in the current data area;

[0033] After confirming that the first-level physical property prediction data areas of all the first electrolyte data areas in the first data area are stored in the corresponding execution result file, all prediction data of each of the first-level physical property prediction data areas are verified based on the first-level physical property indicator set of the first task configuration of the first task file to obtain a corresponding first verification result; and the first electrolyte data corresponding to the first-level physical property prediction data area for which the first verification result is that it meets the standard is recorded as the corresponding first-level qualified electrolyte data; and a corresponding first-level qualified electrolyte data set is formed by all the obtained first-level qualified electrolyte data; the first verification result includes meeting the standard and not meeting the standard;

[0034] and extracting each first-level qualified electrolyte data from the first-level qualified electrolyte data set as a corresponding first prediction input data, setting a corresponding first prediction unit type as a transport property prediction type, and forming a corresponding first task execution file from the first prediction unit type and the first prediction input data corresponding to each first-level qualified electrolyte data; and sending each first task execution file obtained this time to the first scheduling unit of the task execution module;

[0035] and receiving the first execution result files corresponding to the first task execution files returned by the first scheduling unit; and upon receiving each first execution result file, taking the secondary transport property prediction data area of ​​the first electrolyte data area corresponding to the first electrolyte data corresponding to the current first execution result file in the first data area as the corresponding current data area, and extracting the corresponding diffusion coefficient prediction data, conductivity prediction data, and migration number prediction data from the current first execution result file and storing them in the current data area;

[0036] After confirming that the secondary transport property prediction data areas of the first electrolyte data area corresponding to all the first-level qualified electrolyte data in the first data area are stored in the corresponding execution result file, all prediction data of each of the second-level transport property prediction data areas are verified based on the second-level transport property indicator set configured in the first task of the first task file to obtain a corresponding second verification result; and the first-level qualified electrolyte data corresponding to the second-level qualified electrolyte data area whose second verification result is qualified is recorded as the corresponding second-level qualified electrolyte data; and a corresponding second-level qualified electrolyte data set is formed by all the obtained second-level qualified electrolyte data; the second verification result includes qualified and unqualified results;

[0037] and extracting each of the secondary qualified electrolyte data from the secondary qualified electrolyte data set as a corresponding first prediction input data, setting a corresponding first prediction unit type as an electrochemical property prediction type, and forming a corresponding first task execution file from the first prediction unit type and the first prediction input data corresponding to each of the secondary qualified electrolyte data; and sending each of the first task execution files obtained this time to the first scheduling unit of the task execution module;

[0038] and receiving the first execution result files corresponding to the first task execution files returned by the first scheduling unit; and upon receiving each first execution result file, taking the three-level electrochemical property prediction data area of ​​the first electrolyte data area corresponding to the first electrolyte data corresponding to the current first execution result file in the first data area as the corresponding current data area, and extracting the corresponding oxidation potential prediction data, reduction potential prediction data, and electrochemical window prediction data from the current first execution result file and storing them in the current data area;

[0039] After confirming that the third-level electrochemical property prediction data area of ​​the first electrolyte data area corresponding to all the second-level qualified electrolyte data in the first data area are stored in the corresponding execution result file, all the predicted data of each third-level electrochemical property prediction data area are verified based on the third-level electrochemical property indicator set of the first task configuration of the first task file to obtain a corresponding third verification result; and the second-level qualified electrolyte data corresponding to the third-level electrochemical property prediction data area whose third verification result is qualified is recorded as the corresponding third-level qualified electrolyte data; and a corresponding third-level qualified electrolyte data set is formed by all the obtained third-level qualified electrolyte data; the third verification result includes qualified and unqualified;

[0040] The obtained three-level qualified electrolyte data set is stored in the high-throughput screening task output data area of ​​the first data area; and when the data is successfully stored, it is confirmed that the execution of this high-throughput screening task is completed.

[0041] Further preferably, the workflow task scheduling module is specifically used to extract the viscosity prediction data, dielectric constant prediction data, melting point temperature prediction data and boiling point temperature prediction data of the current first-level physical property prediction data area as the corresponding current viscosity, current dielectric constant, current melting point temperature and current boiling point temperature when the first-level physical property indicator set configured by the first task based on the first task file verifies all the prediction data of each first-level physical property prediction data area to obtain the corresponding first verification result; and when the current viscosity is matched with the viscosity indicator of the first-level physical property indicator set according to the preset viscosity matching rule, and the current dielectric constant is matched with the dielectric constant indicator of the first-level physical property indicator set according to the preset dielectric constant matching rule. When the current melting point temperature matches the melting point temperature index of the first-level physical property index set according to the preset melting point temperature matching rule, and the current boiling point temperature matches the boiling point temperature index of the first-level physical property index set according to the preset boiling point temperature matching rule, the first verification result is set to be up to standard; and when the current viscosity does not match the viscosity index of the first-level physical property index set according to the viscosity matching rule, or the current dielectric constant does not match the dielectric constant index according to the dielectric constant matching rule, or the current melting point temperature does not match the melting point temperature index according to the melting point temperature matching rule, or the current boiling point temperature does not match the boiling point temperature index according to the boiling point temperature matching rule, the first verification result is set to be unsatisfactory;

[0042] The workflow task scheduling module is specifically configured to extract the diffusion coefficient prediction data, conductivity prediction data and migration number prediction data of the current secondary transport property prediction data area as the corresponding current diffusion coefficient, current conductivity and current migration number when the secondary transport property indicator set configured by the first task file based on the first task file verifies all the prediction data of each secondary transport property prediction data area to obtain a corresponding second verification result; and extract the diffusion coefficient prediction data, conductivity prediction data and migration number prediction data of the current secondary transport property prediction data area as the corresponding current diffusion coefficient, current conductivity and current migration number; and when the current diffusion coefficient matches the diffusion coefficient indicator of the secondary transport property indicator set according to a preset diffusion coefficient matching rule, and the When the current conductivity matches the conductivity index of the secondary transport property index set according to the preset conductivity matching rule, and the current mobility number matches the mobility number index of the secondary transport property index set according to the preset mobility number matching rule, the second verification result is set as meeting the standard; and when the current diffusion coefficient does not match the diffusion coefficient index according to the diffusion coefficient matching rule, or the current conductivity does not match the conductivity index according to the conductivity matching rule, or the current mobility number does not match the mobility number index according to the mobility number matching rule, the second verification result is set as failing to meet the standard;

[0043] The workflow task scheduling module is specifically used to, when the three-level electrochemical property indicator set configured by the first task based on the first task file verifies all the prediction data of each three-level electrochemical property prediction data area to obtain a corresponding third verification result, extract the oxidation potential prediction data, reduction potential prediction data and electrochemical window prediction data of the current three-level electrochemical property prediction data area as the corresponding current oxidation potential, current reduction potential and current electrochemical window; and when the current oxidation potential is matched with the oxidation potential indicator of the three-level electrochemical property indicator set according to a preset oxidation potential matching rule, and the current reduction potential is matched with the three-level electrochemical property indicator set according to a preset reduction potential matching rule. When the reduction potential index of the chemical property index set matches, and the current electrochemical window matches the electrochemical window index of the three-level electrochemical property index set according to the preset electrochemical window matching rules, the third verification result is set to be up to standard; and when the current oxidation potential does not match the oxidation potential index of the three-level electrochemical property index set according to the oxidation potential matching rules, or the current reduction potential does not match the reduction potential index of the three-level electrochemical property index set according to the reduction potential matching rules, or the current electrochemical window does not match the electrochemical window index of the three-level electrochemical property index set according to the electrochemical window matching rules, the third verification result is set to be unsatisfactory.

[0044] Furthermore, the workflow task scheduling module is specifically configured to identify the molecular reaction mechanism prediction status bit, the interface reaction mechanism prediction status bit, and the SEI / CEI film formation mechanism prediction status bit of the first task configuration of the first task file when the task execution module is called and the first data area performs the mechanism prediction task execution processing according to the first task file;

[0045] When the molecular reaction mechanism prediction state bit is in an activated state, one or more molecular structures in each of the first electrode + electrolyte systems of the first electrode + electrolyte system set of the first task file are extracted to form a corresponding first electrolyte molecule set, and each of the first electrolyte molecule sets is used as a corresponding first prediction input data, and a corresponding first prediction unit type is set as a molecular reaction mechanism prediction type, and a corresponding first task execution file is formed by the first prediction unit type and the first prediction input data corresponding to each of the first electrode + electrolyte systems; and each of the first task execution files obtained this time is sent to the first scheduling unit of the task execution module; and the first execution result file corresponding to each of the first task execution files this time is received returned by the first scheduling unit; and each time a first execution result file is received, the molecular reaction mechanism prediction data area of ​​the first system data area corresponding to the first electrode + electrolyte system corresponding to the current first execution result file in the first data area is used as the corresponding current data area, and the current first execution result file is stored in the current data area;

[0046] When the interface reaction mechanism prediction state bit is in the activated state, each of the first electrode + electrolyte systems of the first electrode + electrolyte system set of the first task file is extracted as a corresponding first prediction input data, and a corresponding first prediction unit type is set as the interface reaction mechanism prediction type, and the first prediction unit type and the first prediction input data corresponding to each of the first electrode + electrolyte systems form a corresponding first task execution file; and each of the first task execution files obtained this time is sent to the first scheduling unit of the task execution module; and the first execution result file corresponding to each of the first task execution files this time returned by the first scheduling unit is received; and each time a first execution result file is received, the interface reaction mechanism prediction data area of ​​the first system data area corresponding to the first electrode + electrolyte system corresponding to the current first execution result file in the first data area is used as the corresponding current data area, and the current first execution result file is stored in the current data area;

[0047] When the SEI / CEI film formation mechanism prediction status bit is in an activated state, each of the first electrode + electrolyte system of the first electrode + electrolyte system set of the first task file is extracted as a corresponding first prediction input data, and a corresponding first prediction unit type is set as the SEI / CEI film formation mechanism prediction type, and the first prediction unit type and the first prediction input data corresponding to each of the first electrode + electrolyte systems form a corresponding first task execution file; and each of the first task execution files obtained this time is sent to the first scheduling unit of the task execution module; and the first execution result file corresponding to each of the first task execution files this time returned by the first scheduling unit is received; and each time a first execution result file is received, the SEI / CEI film formation mechanism prediction data area of ​​the first system data area corresponding to the first electrode + electrolyte system corresponding to the current first execution result file in the first data area is used as the corresponding current data area, and the current first execution result file is stored in the current data area;

[0048] After all execution result files of all the first electrode+electrolyte systems are stored, it is confirmed that the execution of this mechanism prediction task is completed.

[0049] Furthermore, the workflow task scheduling module is specifically configured to extract each second electrolyte data of the second electrolyte data set of the first task file as a corresponding first prediction input data when the task execution module and the first data area are called to perform the performance prediction task execution processing according to the first task file, and set a corresponding first prediction unit type as a battery performance prediction type, and form a corresponding first task execution file by the first prediction unit type corresponding to each first electrolyte data and the first prediction input data; and send each first task execution file obtained this time to the first scheduling unit of the task execution module;

[0050] and receiving the first execution result files corresponding to the first task execution files returned by the first scheduling unit; and upon receiving each first execution result file, taking the second electrolyte data area corresponding to the second electrolyte data corresponding to the current first execution result file in the first data area as the corresponding current data area, and storing the current first execution result file in the current data area;

[0051] After the corresponding execution result files are stored in all the second electrolyte data areas in the first data area, it is confirmed that the execution processing of this performance prediction task is completed.

[0052] Preferably, the workflow task scheduling module is specifically configured to identify the first task type of the first task file when performing task report preparation processing according to the first task file and the first data area to generate a corresponding first report file and send it back to the high-voltage electrolyte experiment party;

[0053] When the first task type is a high-throughput screening type, the stored data in the high-throughput screening task output data area of ​​the first data area is extracted as the corresponding first screening electrolyte set; and the first report file corresponding to the first screening electrolyte set is composed and sent back to the high-voltage electrolyte experiment party;

[0054] When the first task type is a mechanism prediction type, the stored data of each first system data area in the first data area is extracted as a corresponding first system mechanism prediction sub-report, and all the obtained first system mechanism prediction sub-reports are combined to form a corresponding first report file and sent back to the high-voltage electrolyte experiment party;

[0055] When the first task type is a performance prediction type, the stored data of each of the second electrolyte data areas in the first data area are extracted as the corresponding first electrolyte performance prediction sub-report, and the corresponding first report file is composed of all the first electrolyte performance prediction sub-reports obtained and sent back to the high-voltage electrolyte experiment party.

[0056] An embodiment of the present invention provides a workflow task processing system for assisting high-voltage electrolyte experiments, including: a workflow task scheduling module, a workflow task pool, a workflow task database, a task execution module, a molecular dynamics simulation interface library, a quantum chemistry calculation interface library, and an AI model interface library; wherein, the molecular dynamics simulation interface library can be connected to the functional interface of any molecular dynamics simulation software, the quantum chemistry calculation interface library can be connected to the functional interface of any quantum chemistry calculation software, and the AI ​​model interface library can be connected to the functional interface of any AI model. The workflow task scheduling module completes three types of auxiliary task processing by calling the task execution module: high-throughput screening tasks, mechanism prediction tasks, and performance prediction tasks. If this system is applied to the experimental development of high-voltage electrolytes, it can complete the screening of large quantities of electrolytes for high-voltage electrolyte experiments by executing high-throughput screening tasks. After obtaining a batch of electrode + electrolyte systems, it can perform mechanism prediction tasks to predict the molecular-level reaction mechanism, interfacial reaction mechanism, and bipolar SEI / CEI film formation mechanism of each electrode + electrolyte system. It can also complete the performance prediction work of batch electrolytes for high-voltage electrolyte experiments by executing performance prediction tasks. Using this system as an auxiliary means for high-voltage electrolyte experiments reduces the number of experimental trial and error, improves experimental accuracy, shortens the experimental cycle, and reduces experimental costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 A module structure diagram of a workflow task processing system for assisting high-voltage electrolyte experiments provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0058] To make the objectives, technical solutions, and advantages of the present invention more apparent, the present invention will be further described in detail below with reference to the accompanying drawings. It should be understood that the embodiments described herein are merely some, rather than all, of the present invention. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without inventive effort are intended to fall within the scope of protection of the present invention.

[0059] The embodiment of the present invention provides a workflow task processing system to assist high-voltage electrolyte experiments, such as Figure 1 As shown in the module structure diagram of a workflow task processing system for assisting high-voltage electrolyte experiments provided by an embodiment of the present invention, the workflow task processing system 1 mainly includes: a workflow task scheduling module 11, a workflow task pool 12, a workflow task database 13, a task execution module 14, a molecular dynamics simulation interface library 15, a quantum chemical calculation interface library 16 and an AI model interface library 17.

[0060] (1) Workflow task scheduling module 11:

[0061] The workflow task scheduling module 11 is connected to the workflow task pool 12 , the workflow task database 13 and the task execution module 14 respectively.

[0062] The workflow task scheduling module 11 is used to receive the first task file sent by the high-voltage electrolyte experiment party 2; and initialize a corresponding task data storage area in the workflow task pool 12 as the first data area; and plan the sub-data area of ​​the first data area according to the first task file; and call the task execution module 14 to perform task execution processing on the first task file; and after confirming that the task execution processing is completed, perform task report preparation processing according to the first task file and the first data area to generate a corresponding first report file and send it back to the high-voltage electrolyte experiment party 2; and the first task file and the first report file form a corresponding task backup record and store it in the workflow task database 13; and delete the first data area.

[0063] Here, the first task file of the embodiment of the present invention includes a first task type, first task data and a first task configuration; the first task type includes a high-throughput screening type, a mechanism prediction type and a performance prediction type;

[0064] 1) When the first task type is high-throughput screening:

[0065] 1a. The corresponding first task data is a first electrolyte data set to be screened; the first electrolyte data set includes one or more first electrolyte data; the first electrolyte data includes a first component data set and a first component ratio; the first component data set includes multiple first component data, the first component data includes a first component type, a first component name, and a first component SMILES sequence, and the first component type includes an electrolyte salt, a solvent, and an additive; the first component ratio is the ratio of the electrolyte salt, the solvent, and the additive of the corresponding electrolyte, and the first component ratio = electrolyte salt ratio: solvent ratio: additive ratio;

[0066] 1b. The corresponding first task is configured as a multi-level screening index set, consisting of a primary physical property index set, a secondary transport property index set, and a tertiary electrochemical property index set; the primary physical property index set includes viscosity index, dielectric constant index, melting point temperature index, and boiling point temperature index; the secondary transport property index set includes diffusion coefficient index, conductivity index, and migration number index; the tertiary electrochemical property index set includes oxidation potential index, reduction potential index, and electrochemical window index;

[0067] 2) When the first task type is mechanism prediction:

[0068] 2a. The corresponding first task data is a first electrode + electrolyte system set to be predicted; the first electrode + electrolyte system set includes one or more first electrode + electrolyte systems; the first electrode + electrolyte system is an interface model system formed by the fusion of an electrode plate model system and an electrolyte model system, and the first electrode + electrolyte system includes multiple first atoms, multiple first connecting bonds, and multiple first clusters; the atomic properties of the first atom include an atom identifier, an atomic element type, an atomic charge, an atomic size, an atomic coordinate, and an identifier of the cluster in which the atom is located; the bond properties of the first connecting bond include a connecting bond identifier, a connecting bond type, and an atom pair identifier group corresponding to the connecting bond; the first cluster includes a cluster identifier and a cluster type;

[0069] 2b. The corresponding first task is configured as a multi-class prediction state machine, consisting of a molecular reaction mechanism prediction state bit, an interface reaction mechanism prediction state bit, and an SEI / CEI film formation mechanism prediction state bit. Each of the three state bits includes two state configuration values: an activation state and an inactivation state. At least one of the three state bits has an activation state configuration value.

[0070] 3) When the first task type is performance prediction:

[0071] 3a. The corresponding first task data is a second electrolyte data set to be predicted; the second electrolyte data set includes one or more second electrolyte data; the second electrolyte data includes a second component data set and a second component ratio; the second component data set includes multiple second component data, the second component data includes a second component type, a second component name, and a second component SMILES sequence, and the second component type includes an electrolyte salt, a solvent, and an additive; the second component ratio is the ratio of the electrolyte salt, the solvent, and the additive of the corresponding electrolyte, and the second component ratio = electrolyte salt ratio: solvent ratio: additive ratio;

[0072] 3b. The corresponding first task configuration is empty.

[0073] In one implementation of the embodiment of the present invention, the workflow task scheduling module 11 is specifically configured to: when planning sub-data areas for the first data area according to the first task file:

[0074] Step A1, identifying a first task type of a first task file;

[0075] Step A2: When the first task type is a high-throughput screening type, the number of first electrolyte data in the first electrolyte data set of the first task file is counted to obtain the corresponding total number of first electrolytes; a sub-data area for the total number of first electrolytes is created in the first data area and recorded as the corresponding first electrolyte data area, and a high-throughput screening task output data area is created; and three sub-data areas are further created in each first electrolyte data area, namely, a primary physical property prediction data area, a secondary transport property prediction data area, and a tertiary electrochemical property prediction data area;

[0076] Here, the first electrolyte data area corresponds one-to-one to the first electrolyte data;

[0077] Step A2, when the first task type is a mechanism prediction type, the number of first electrode + electrolyte systems in the first electrode + electrolyte system set of the first task file is counted to obtain the corresponding total number of first systems; and a sub-data area of ​​the total number of first systems is created in the first data area and recorded as the corresponding first system data area; and the three status bits of the first task configuration of the first task file are identified. If the molecular reaction mechanism prediction status bit is in an activated state, a corresponding molecular reaction mechanism prediction data area is created in each first system data area; if the interface reaction mechanism prediction status bit is in an activated state, a corresponding interface reaction mechanism prediction data area is created in each first system data area; if the SEI / CEI film formation mechanism prediction status bit is in an activated state, a corresponding SEI / CEI film formation mechanism prediction data area is created in each first system data area;

[0078] Here, the first system data area corresponds one-to-one to the first electrode + electrolyte system;

[0079] Step A3: When the first task type is a performance prediction type, the number of second electrolyte data in the second electrolyte data set of the first task file is counted to obtain a corresponding total number of second electrolytes; and a sub-data area for the total number of second electrolytes is created in the first data area and recorded as the corresponding second electrolyte data area;

[0080] Here, the second electrolyte data corresponds to the second electrolyte data area one-to-one.

[0081] In another implementation of the embodiment of the present invention, the workflow task scheduling module 11 is specifically configured to, when calling the task execution module 14 to perform task execution processing on the first task file:

[0082] Step B1, identifying the first task type of the first task file;

[0083] Step B2: When the first task type is a high-throughput screening type, the task execution module 14 and the first data area are called to perform high-throughput screening task execution processing according to the first task file; and when the high-throughput screening task execution processing is completed, the task execution processing is confirmed to be completed;

[0084] Step B3: When the first task type is a mechanism prediction type, the task execution module 14 and the first data area are called to perform the mechanism prediction task execution process according to the first task file; and when the current mechanism prediction task execution process is completed, the task execution process is confirmed to be completed;

[0085] Step B4, when the first task type is a performance prediction type, call the task execution module 14 and the first data area to perform performance prediction task execution processing according to the first task file; and confirm the completion of the current task execution processing when the current performance prediction task execution processing ends.

[0086] In another implementation of the embodiment of the present invention, in the above step B2, the workflow task scheduling module 11 is specifically configured to, when calling the task execution module 14 and the first data area to perform high-throughput screening task execution processing according to the first task file:

[0087] Step B21, extracting each first electrolyte data of the first electrolyte data set of the first task file as a corresponding first prediction input data, and setting a corresponding first prediction unit type as a basic physical property prediction type, and forming a corresponding first task execution file by the first prediction unit type and the first prediction input data corresponding to each first electrolyte data; and sending each first task execution file obtained this time to the first scheduling unit 141 of the task execution module 14; and receiving the first execution result file corresponding to each first task execution file returned by the first scheduling unit 141; and when each first execution result file is received, taking the primary physical property prediction data area of ​​the first electrolyte data area corresponding to the first electrolyte data corresponding to the current first execution result file in the first data area as the corresponding current data area, and extracting the corresponding viscosity prediction data, dielectric constant prediction data, melting point temperature prediction data and boiling point temperature prediction data from the current first execution result file and storing them in the current data area;

[0088] Step B22: After confirming that the first-level physical property prediction data areas of all first electrolyte data areas in the first data area have been stored in the corresponding execution result files, all prediction data of each first-level physical property prediction data area are verified based on the first-level physical property indicator set configured in the first task of the first task file to obtain a corresponding first verification result; and the first electrolyte data corresponding to the first-level physical property prediction data area for which the first verification result is that it meets the standard is recorded as the corresponding first-level qualified electrolyte data; and a corresponding first-level qualified electrolyte data set is formed by all the obtained first-level qualified electrolyte data;

[0089] Among them, the first verification results include meeting the standards and failing to meet the standards;

[0090] Here, in another implementation of the embodiment of the present invention, the workflow task scheduling module 11 is specifically used to extract the viscosity prediction data, dielectric constant prediction data, melting point temperature prediction data and boiling point temperature prediction data of the current first-level physical property prediction data area as the corresponding current viscosity, current dielectric constant, current melting point temperature and current boiling point temperature when verifying all the prediction data of each first-level physical property prediction data area based on the first-level physical property indicator set configured by the first task file to obtain the corresponding first verification result; and when the current viscosity is matched with the viscosity indicator of the first-level physical property indicator set according to the preset viscosity matching rule, and the current dielectric constant is matched with the first-level physical property indicator set according to the preset dielectric constant matching rule. When the dielectric constant index of the primary physical property index set matches, and the current melting point temperature matches the melting point temperature index of the primary physical property index set according to the preset melting point temperature matching rule, and the current boiling point temperature matches the boiling point temperature index of the primary physical property index set according to the preset boiling point temperature matching rule, the first verification result is set to be up to standard; and when the current viscosity does not match the viscosity index of the primary physical property index set according to the viscosity matching rule, or the current dielectric constant does not match the dielectric constant index according to the dielectric constant matching rule, or the current melting point temperature does not match the melting point temperature index according to the melting point temperature matching rule, or the current boiling point temperature does not match the boiling point temperature index according to the boiling point temperature matching rule, the first verification result is set to be unsatisfactory;

[0091] Here, the preset viscosity matching rule is a preset matching rule for comparing with the corresponding viscosity index, and the comparison relationship includes greater than, less than, equal to, greater than or equal to, and less than or equal to; for example, if the viscosity matching rule is preset to be less than the viscosity index, then when the current viscosity is matched with the viscosity index of the primary physical property index set according to the viscosity matching rule, the current viscosity should be less than the corresponding viscosity index; conversely, if the current viscosity is greater than or equal to the corresponding viscosity index, it means that the current viscosity does not match the viscosity index of the primary physical property index set according to the viscosity matching rule;

[0092] Similarly, the dielectric constant matching rule is a pre-set matching rule for comparison with the corresponding dielectric constant index, and the comparison relationship includes greater than, less than, equal to, greater than or equal to, and less than or equal to; the melting point temperature matching rule is a pre-set matching rule for comparison with the corresponding melting point temperature index, and the comparison relationship includes greater than, less than, equal to, greater than or equal to, and less than or equal to; the boiling point temperature matching rule is a pre-set matching rule for comparison with the corresponding boiling point temperature index, and the comparison relationship includes greater than, less than, equal to, greater than or equal to, and less than or equal to;

[0093] Step B23, extract each first-level qualified electrolyte data of the first-level qualified electrolyte data set as a corresponding first prediction input data, set a corresponding first prediction unit type as a transport property prediction type, and form a corresponding first task execution file by the first prediction unit type and the first prediction input data corresponding to each first-level qualified electrolyte data; and send each first task execution file obtained this time to the first scheduling unit 141 of the task execution module 14; and receive the first execution result file corresponding to each first task execution file returned by the first scheduling unit 141; and when each first execution result file is received, use the secondary transport property prediction data area of ​​the first electrolyte data area corresponding to the first electrolyte data corresponding to the current first execution result file in the first data area as the corresponding current data area, and extract the corresponding diffusion coefficient prediction data, conductivity prediction data and migration number prediction data from the current first execution result file and store them in the current data area;

[0094] Step B24: After confirming that the secondary transport property prediction data areas of the first electrolyte data area corresponding to all the first-level qualified electrolyte data in the first data area are stored in the corresponding execution result file, all the prediction data in each second-level transport property prediction data area are verified based on the second-level transport property indicator set configured in the first task of the first task file to obtain a corresponding second verification result; and the first-level qualified electrolyte data corresponding to the second-level qualified electrolyte data area for which the second verification result is qualified is recorded as the corresponding second-level qualified electrolyte data; and a corresponding second-level qualified electrolyte data set is formed by all the obtained second-level qualified electrolyte data;

[0095] Among them, the second verification results include meeting the standards and failing to meet the standards;

[0096] Here, in another implementation of the embodiment of the present invention, the workflow task scheduling module 11 is specifically configured to, when verifying all the prediction data of each secondary transport property prediction data area based on the secondary transport property indicator set of the first task configuration of the first task file to obtain a corresponding second verification result, extract the diffusion coefficient prediction data, the conductivity prediction data, and the migration number prediction data of the current secondary transport property prediction data area as the corresponding current diffusion coefficient, current conductivity, and current migration number; and when the current diffusion coefficient matches the diffusion coefficient indicator of the secondary transport property indicator set according to a preset diffusion coefficient matching rule, and the current conductivity matches the conductivity indicator of the secondary transport property indicator set according to a preset conductivity matching rule, and the current migration number matches the migration number indicator of the secondary transport property indicator set according to a preset migration number matching rule, set the second verification result as met; and when the current diffusion coefficient does not match the diffusion coefficient indicator according to the diffusion coefficient matching rule, or the current conductivity does not match the conductivity indicator according to the conductivity matching rule, or the current migration number does not match the migration number indicator according to the migration number matching rule, set the second verification result as unmet;

[0097] Here, similar to the aforementioned viscosity matching rule, dielectric constant matching rule, melting point temperature matching rule, and boiling point temperature matching rule: the diffusion coefficient matching rule is a pre-set matching rule for comparison with the corresponding diffusion coefficient index, and the comparison relationship includes greater than, less than, equal to, greater than or equal to, and less than or equal to; the conductivity matching rule is a pre-set matching rule for comparison with the corresponding conductivity index, and the comparison relationship includes greater than, less than, equal to, greater than or equal to, and less than or equal to; the migration number matching rule is a pre-set matching rule for comparison with the corresponding migration number index, and the comparison relationship includes greater than, less than, equal to, greater than or equal to, and less than or equal to;

[0098] Step B25, extract each secondary qualified electrolyte data of the secondary qualified electrolyte data set as a corresponding first prediction input data, set a corresponding first prediction unit type as an electrochemical property prediction type, and form a corresponding first task execution file by the first prediction unit type and the first prediction input data corresponding to each secondary qualified electrolyte data; and send each first task execution file obtained this time to the first scheduling unit 141 of the task execution module 14; and receive the first execution result file corresponding to each first task execution file returned by the first scheduling unit 141; and when each first execution result file is received, use the third electrochemical property prediction data area of ​​the first electrolyte data area corresponding to the first electrolyte data corresponding to the current first execution result file in the first data area as the corresponding current data area, and extract the corresponding oxidation potential prediction data, reduction potential prediction data and electrochemical window prediction data from the current first execution result file and store them in the current data area;

[0099] Step B26, after confirming that the third-level electrochemical property prediction data area of ​​the first electrolyte data area corresponding to all the second-level qualified electrolyte data in the first data area are stored in the corresponding execution result file, all the predicted data in each third-level electrochemical property prediction data area are verified based on the third-level electrochemical property indicator set configured in the first task of the first task file to obtain a corresponding third verification result; and the second-level qualified electrolyte data corresponding to the third-level electrochemical property prediction data area for which the third verification result is qualified is recorded as the corresponding third-level qualified electrolyte data; and a corresponding third-level qualified electrolyte data set is formed by all the obtained third-level qualified electrolyte data;

[0100] Among them, the third verification results include meeting the standards and failing to meet the standards;

[0101] Here, in another implementation of an embodiment of the present invention, the workflow task scheduling module 11 is specifically used to extract the oxidation potential prediction data, reduction potential prediction data and electrochemical window prediction data of the current three-level electrochemical property prediction data area as the corresponding current oxidation potential, current reduction potential and current electrochemical window when verifying all the prediction data of each three-level electrochemical property prediction data area based on the three-level electrochemical property indicator set configured by the first task file to obtain the corresponding third verification result; and when the current oxidation potential matches the oxidation potential indicator of the three-level electrochemical property indicator set according to the preset oxidation potential matching rule, and the current reduction potential matches the oxidation potential indicator of the three-level electrochemical property indicator set, and the current reduction potential matches the oxidation potential indicator of the three-level electrochemical property indicator set according to the preset oxidation potential matching rule ... When the current electrochemical window matches the electrochemical window index of the three-level electrochemical property index set according to the preset reduction potential matching rule, and the current electrochemical window matches the electrochemical window index of the three-level electrochemical property index set according to the preset electrochemical window matching rule, the third verification result is set to be up to standard; and when the current oxidation potential does not match the oxidation potential index of the three-level electrochemical property index set according to the oxidation potential matching rule, or the current reduction potential does not match the reduction potential index of the three-level electrochemical property index set according to the reduction potential matching rule, or the current electrochemical window does not match the electrochemical window index of the three-level electrochemical property index set according to the electrochemical window matching rule, the third verification result is set to be unsatisfactory;

[0102] Here, similar to the aforementioned viscosity matching rule, dielectric constant matching rule, melting point temperature matching rule, boiling point temperature matching rule, diffusion coefficient matching rule, conductivity matching rule, and migration number matching rule: the oxidation potential matching rule is a pre-set matching rule for comparison with the corresponding oxidation potential index, and the comparison relationship includes greater than, less than, equal to, greater than or equal to, and less than or equal to; the reduction potential matching rule is a pre-set matching rule for comparison with the corresponding reduction potential index, and the comparison relationship includes greater than, less than, equal to, greater than or equal to, and less than or equal to; the electrochemical window matching rule is a pre-set matching rule for comparison with the corresponding electrochemical window index, and the comparison relationship includes greater than, less than, equal to, greater than or equal to, and less than or equal to;

[0103] Step B27, and store the obtained three-level qualified electrolyte data set into the high-throughput screening task output data area of ​​the first data area; and when the data is successfully stored, confirm that the execution of this high-throughput screening task is completed.

[0104] In another implementation of the embodiment of the present invention, in the aforementioned step B3, the workflow task scheduling module 11 is specifically configured to, when calling the task execution module 14 and the first data area to perform the mechanism prediction task execution processing according to the first task file:

[0105] Step B31, identifying the molecular reaction mechanism prediction status bit, the interface reaction mechanism prediction status bit, and the SEI / CEI film formation mechanism prediction status bit of the first task configuration of the first task file;

[0106] Step B32, when the molecular reaction mechanism prediction state bit is in the activated state, one or more types of molecular structures in each first electrode + electrolyte system of the first electrode + electrolyte system set of the first task file are extracted to form a corresponding first electrolyte molecule set, and each first electrolyte molecule set is used as a corresponding first prediction input data, and a corresponding first prediction unit type is set as a molecular reaction mechanism prediction type, and a corresponding first task execution file is formed by the first prediction unit type and the first prediction input data corresponding to each first electrode + electrolyte system; and each first task execution file obtained this time is sent to the first scheduling unit 141 of the task execution module 14; and the first execution result file corresponding to each first task execution file this time is received from the first scheduling unit 141; and each time a first execution result file is received, the molecular reaction mechanism prediction data area of ​​the first system data area corresponding to the first electrode + electrolyte system corresponding to the current first execution result file in the first data area is used as the corresponding current data area, and the current first execution result file is stored in the current data area;

[0107] Step B33, when the interface reaction mechanism prediction state bit is in the activated state, extract each first electrode + electrolyte system of the first electrode + electrolyte system set of the first task file as a corresponding first prediction input data, and set a corresponding first prediction unit type as the interface reaction mechanism prediction type, and form a corresponding first task execution file by the first prediction unit type and the first prediction input data corresponding to each first electrode + electrolyte system; and send each first task execution file obtained this time to the first scheduling unit 141 of the task execution module 14; and receive the first execution result file corresponding to each first task execution file returned by the first scheduling unit 141; and when each first execution result file is received, use the interface reaction mechanism prediction data area of ​​the first system data area corresponding to the first electrode + electrolyte system corresponding to the current first execution result file in the first data area as the corresponding current data area, and store the current first execution result file in the current data area;

[0108] Step B34, when the SEI / CEI film formation mechanism prediction state bit is in the activated state, each first electrode + electrolyte system of the first electrode + electrolyte system set of the first task file is extracted as a corresponding first prediction input data, and a corresponding first prediction unit type is set as the SEI / CEI film formation mechanism prediction type, and the first prediction unit type and the first prediction input data corresponding to each first electrode + electrolyte system form a corresponding first task execution file; and each first task execution file obtained this time is sent to the first scheduling unit 141 of the task execution module 14; and the first execution result file corresponding to each first task execution file this time is received from the first scheduling unit 141; and each time a first execution result file is received, the SEI / CEI film formation mechanism prediction data area of ​​the first system data area corresponding to the first electrode + electrolyte system corresponding to the current first execution result file in the first data area is used as the corresponding current data area, and the current first execution result file is stored in the current data area;

[0109] In step B35 , after all execution result files of all first electrode+electrolyte systems are stored, it is confirmed that the execution of this mechanism prediction task is completed.

[0110] In another implementation of the embodiment of the present invention, in the aforementioned step B4, the workflow task scheduling module 11 is specifically configured to, when calling the task execution module 14 and the first data area to perform the performance prediction task execution processing according to the first task file:

[0111] Step B41: extract each second electrolyte data from the second electrolyte data set of the first task file as a corresponding first prediction input data, set a corresponding first prediction unit type as a battery performance prediction type, and form a corresponding first task execution file from the first prediction unit type and the first prediction input data corresponding to each first electrolyte data; and send each first task execution file obtained this time to the first scheduling unit 141 of the task execution module 14;

[0112] Step B42, receiving the first execution result files corresponding to the first task execution files returned by the first scheduling unit 141; and each time a first execution result file is received, taking the second electrolyte data area corresponding to the second electrolyte data corresponding to the current first execution result file in the first data area as the corresponding current data area, and storing the current first execution result file in the current data area;

[0113] In step B43, after all the second electrolyte data areas in the first data area have stored the corresponding execution result files, it is confirmed that the execution processing of this performance prediction task is completed.

[0114] In another implementation of the embodiment of the present invention, the workflow task scheduling module 11 is specifically configured to perform task report preparation processing according to the first task file and the first data area to generate a corresponding first report file and send it back to the high-voltage electrolyte experiment party 2:

[0115] Step C1, identifying a first task type of a first task file;

[0116] Step C2: When the first task type is a high-throughput screening type, the stored data in the high-throughput screening task output data area of ​​the first data area is extracted as the corresponding first screening electrolyte set; and a corresponding first report file is formed from the first screening electrolyte set and sent back to the high-voltage electrolyte experiment party 2;

[0117] Step C3: When the first task type is a mechanism prediction type, the stored data in each first system data area in the first data area is extracted as a corresponding first system mechanism prediction sub-report, and all the obtained first system mechanism prediction sub-reports are combined to form a corresponding first report file and sent back to the high-voltage electrolyte experiment party 2;

[0118] Step C4, when the first task type is a performance prediction type, the stored data of each second electrolyte data area in the first data area is extracted as the corresponding first electrolyte performance prediction sub-report, and the corresponding first report file is composed of all the obtained first electrolyte performance prediction sub-reports and sent back to the high-voltage electrolyte experiment party 2.

[0119] (2) Workflow task pool 12:

[0120] The workflow task pool 12 is used to store multiple task data storage areas.

[0121] (3) Workflow task database 13:

[0122] The workflow task database 13 is used to store multiple task backup records.

[0123] (IV) Task execution module 14:

[0124] The task execution module 14 is connected to the molecular dynamics simulation interface library 15, the quantum chemistry calculation interface library 16 and the AI ​​model interface library 17 respectively.

[0125] The task execution module 14 includes a first scheduling unit 141, a first scheduling task pool 142 and a first prediction unit set 143; the first scheduling unit 141 is connected to the workflow task scheduling module 11, the first scheduling task pool 142 and the first prediction unit set 143 respectively; the first scheduling task pool 142 is connected to the first prediction unit set 143; the first prediction unit set 143 includes a basic physical property prediction unit 1431, a transport property prediction unit 1432, an electrochemical property prediction unit 1433, a molecular reaction mechanism prediction unit 1434, an interface reaction mechanism prediction unit 1435, an SEI / CEI film formation mechanism prediction unit 1436 and a battery performance prediction unit 1437; each prediction unit of the first prediction unit set 143 is connected to one or all of the three interface libraries: the molecular dynamics simulation interface library 15, the quantum chemical calculation interface library 16 and the AI ​​model interface library 17.

[0126] The first scheduling unit 141 is used to initialize a corresponding data storage area in the first scheduling task pool 142 as the second data area when receiving any first task execution file sent by the workflow task scheduling module 11, and allocate a unique data area identifier to the current second data area as the corresponding second data area identifier, and set a return data area and a task status data initialized to an unfinished state in the second data area; and use the prediction unit in the first prediction unit set 143 that matches the first prediction unit type of the first task execution file as the corresponding first prediction unit; and send the second data area identifier and the first prediction input data of the first task execution file to the first prediction unit; and regularly identify whether the task status data in the second data area is in a completed state. If so, extract the data in the return data area of ​​the second data area as the corresponding first execution result file and send it back to the workflow task scheduling module 11, and delete the second data area at the end of the send-back;

[0127] Among them, the first task execution file includes the first prediction unit type and the first prediction input data; the first prediction unit type includes the basic physical property prediction type, the transport property prediction type, the electrochemical property prediction type, the molecular reaction mechanism prediction type, the interface reaction mechanism prediction type, the SEI / CEI film formation mechanism prediction type and the battery performance prediction type; the task status data includes the unfinished status and the completed status.

[0128] Each prediction unit in the first prediction unit set 143 is used to preset a corresponding prediction processing flow locally, and the prediction processing flow calls the interfaces of the three interface libraries.

[0129] Each prediction unit in the first prediction unit set 143 is also used to, when receiving the second data area identifier and the first prediction input data sent by the first scheduling unit 141, perform corresponding property, mechanism or performance prediction processing according to the first prediction input data by a locally preset prediction processing flow to obtain corresponding first prediction output data, and use the second data area corresponding to the second data area identifier in the first scheduling task pool 142 as the corresponding current data area, store the first prediction output data in the return data area of ​​the current data area, and update the task status data in the current data area to a completed state.

[0130] Here, the prediction processing flow preset locally in the basic physical property prediction unit 1431 in the first prediction unit set 143 is used to perform corresponding basic physical property prediction processing according to the first prediction input data to obtain corresponding first prediction output data. The basic physical properties here include viscosity, dielectric constant, melting point temperature and boiling point temperature.

[0131] The transport property prediction unit 1432 in the first prediction unit set 143 has a locally preset prediction processing flow for performing corresponding transport property prediction processing based on the first prediction input data to obtain corresponding first prediction output data. The transport properties here include diffusion coefficient, conductivity, and transference number.

[0132] The locally preset prediction processing flow of the electrochemical property prediction unit 1433 in the first prediction unit set 143 is used to perform corresponding electrochemical property prediction processing according to the first prediction input data to obtain corresponding first prediction output data. The electrochemical properties here include oxidation potential, reduction potential and electrochemical window.

[0133] The locally preset prediction processing flow of the molecular reaction mechanism prediction unit 1434 in the first prediction unit set 143 is used to perform corresponding molecular reaction mechanism prediction processing based on the first prediction input data to obtain corresponding first prediction output data. The molecular reaction mechanism here includes a reaction path, reaction products, and reaction barriers.

[0134] The interface reaction mechanism prediction unit 1435 in the first prediction unit set 143 has a locally preset prediction processing flow for performing corresponding interface reaction mechanism prediction processing based on the first prediction input data to obtain corresponding first prediction output data. The interface reaction mechanism here includes the reaction grid network of the positive and negative electrode interfaces.

[0135] The SEI / CEI film formation mechanism prediction unit 1436 in the first prediction unit set 143 has a locally pre-configured prediction processing flow for performing a corresponding SEI / CEI film formation mechanism prediction process based on the first prediction input data to obtain corresponding first prediction output data. The SEI / CEI film formation mechanism herein includes film thickness, density, dissolution voltage / current / temperature, ionic / electronic conductivity, and the like.

[0136] The battery performance prediction unit 1437 in the first prediction unit set 143 has a locally pre-configured prediction processing flow for performing corresponding battery performance prediction processing based on the first prediction input data to obtain corresponding first prediction output data. Battery performance here includes battery coulombic efficiency, high and low temperature capacity, rate capability, DC internal resistance (DCR), and electrochemical impedance spectroscopy (EIS).

[0137] In another implementation of the present invention, the prediction processing flow pre-installed in the basic physical property prediction unit 1431 is specifically configured to extract the first component SMILES sequence of each component and the first component ratio of all components from the first electrolyte data of the first prediction input data to form a corresponding first input tensor when performing corresponding basic physical property prediction processing based on the first prediction input data to obtain corresponding first prediction output data; and send the first input tensor to a pre-designated first AI model interface 171 in the AI ​​model interface library 17 for corresponding property prediction processing to obtain a corresponding first prediction vector; and extract the corresponding viscosity prediction data, dielectric constant prediction data, melting point temperature prediction data, and boiling point temperature prediction data from the first prediction vector to form the corresponding first prediction output data. As previously mentioned, the first electrolyte data here includes: a first component data set {first component data} and a first component ratio; wherein the first component data = first component type + first component name + first component SMILES sequence; the first component type includes electrolyte salt, solvent, and additive.

[0138] In another implementation of an embodiment of the present invention, the locally preset prediction processing flow of the transport property prediction unit 1432 is specifically used to extract the first component SMILES sequence of each component and the first component ratio of all components from the first electrolyte data of the first prediction input data to perform molecular system modeling to obtain the corresponding first prediction system when performing corresponding transport property prediction processing according to the first prediction input data to obtain corresponding first prediction output data; and send the first prediction system to a pre-designated first simulation interface 151 in the molecular dynamics simulation interface library 15 to perform corresponding molecular dynamics simulation to obtain a corresponding first simulation trajectory file; and send the first simulation trajectory file to a pre-designated first calculation interface 161 in the quantum chemical calculation interface library 16 to perform corresponding quantum chemical calculation and obtain corresponding diffusion coefficient prediction data, conductivity prediction data and migration number prediction data through calculation to form the corresponding first prediction output data.

[0139] In another implementation of an embodiment of the present invention, the locally preset prediction processing flow of the electrochemical property prediction unit 1433 is specifically used to extract the first component SMILES sequence of each component and the first component ratio of all components from the first electrolyte data of the first prediction input data to perform molecular system modeling to obtain a corresponding second prediction system when performing corresponding electrochemical property prediction processing according to the first prediction input data to obtain corresponding first prediction output data; and send the second prediction system to a pre-designated first simulation interface 151 in the molecular dynamics simulation interface library 15 to perform corresponding molecular dynamics simulation to obtain a corresponding second simulation trajectory file; and send the second simulation trajectory file to a pre-designated first calculation interface 161 in the quantum chemical calculation interface library 16 to perform corresponding quantum chemical calculation and obtain corresponding oxidation potential prediction data, reduction potential prediction data and electrochemical window prediction data through calculation to form the corresponding first prediction output data.

[0140] In another implementation of the embodiment of the present invention, the prediction processing flow preset locally in the molecular reaction mechanism prediction unit 1434 is specifically used to, when performing corresponding molecular reaction mechanism prediction processing according to the first prediction input data to obtain corresponding first prediction output data, send the first electrolyte molecule set of the first prediction input data to a pre-designated first calculation interface 161 in the quantum chemical calculation interface library 16 for molecular configuration optimization; and send the optimized first electrolyte molecule set to a pre-designated first simulation interface 151 in the molecular dynamics simulation interface library 15 for molecular dynamics simulation under the reaction force field, and terminate the simulation when no new product objects are generated in the simulation to obtain a corresponding third simulation trajectory file; and send the third simulation trajectory file to a pre-designated first AI model interface 171 in the AI ​​model interface library 17 for reaction product identification, and perform reaction path identification based on the temporal relationship of each identified product, and send the third simulation trajectory file to a pre-designated first calculation interface 161 in the quantum chemical calculation interface library 16 for corresponding quantum chemical calculation and obtain the corresponding reaction barrier through calculation, and the obtained reaction products, reaction path and reaction barrier constitute the corresponding first prediction output data.

[0141] In another implementation of the embodiment of the present invention, the prediction processing flow preset locally in the interface reaction mechanism prediction unit 1435 is specifically used to send the first electrode + electrolyte system of the first prediction input data to a pre-designated first calculation interface 161 in the quantum chemical calculation interface library 16 for configuration optimization when the corresponding interface reaction mechanism prediction processing is performed according to the first prediction input data to obtain the corresponding first prediction output data; and based on the deep potential energy molecular dynamics theory, the optimized first electrode + electrolyte system is predicted to have positive and negative electrode reaction grid networks, and the corresponding first prediction output data is composed of the obtained positive and negative electrode reaction grid networks.

[0142] In another implementation of an embodiment of the present invention, the prediction processing flow preset locally in the SEI / CEI film formation mechanism prediction unit 1436 is specifically used to send the first electrode + electrolyte system of the first prediction input data to a pre-designated first calculation interface 161 in the quantum chemical calculation interface library 16 for configuration optimization when the corresponding SEI / CEI film formation mechanism prediction processing is performed according to the first prediction input data to obtain the corresponding first prediction output data; and construct the positive and negative electrode reaction grid networks of the optimized first electrode + electrolyte system based on the deep potential energy molecular dynamics theory; and predict the film thickness, density, dissolution voltage / current / temperature, and ion / electronic conductivity of the CEI film and SEI film generated on the positive and negative electrodes based on the constructed positive and negative electrode reaction grid networks based on the kinetic Monte Carlo (KMC) method, and the obtained prediction data constitute the corresponding first prediction output data.

[0143] In another implementation of an embodiment of the present invention, the prediction processing flow preset locally by the battery performance prediction unit 1437 is specifically used to extract the second component SMILES sequence of each component and the second component ratio of all components from the second electrolyte data of the first prediction input data to perform molecular system modeling to obtain a corresponding third prediction system when performing corresponding battery performance prediction processing according to the first prediction input data to obtain corresponding first prediction output data; and perform tensor conversion on the third prediction system to obtain a corresponding second input tensor; and send the second input tensor to a pre-designated first AI model interface 171 in the AI ​​model interface library 17 to perform corresponding basic physical property prediction processing to obtain a corresponding first prediction tensor; and then send the first prediction tensor to a pre-designated first AI model interface 171 in the AI ​​model interface library 17 to perform corresponding battery performance prediction processing to obtain a corresponding second prediction vector; and extract the corresponding battery coulombic efficiency prediction data, high and low temperature capacity prediction data, rate performance prediction data, DC resistance prediction data, and AC impedance prediction data from the second prediction vector to form the corresponding first prediction output data. As can be seen from the previous text, the second electrolyte data here includes: a second component data set {second component data} and a second component ratio; wherein, the second component data = second component type + second component name + second component SMILES sequence; the second component type includes electrolyte salt, solvent and additive.

[0144] (V) Molecular dynamics simulation interface library 15:

[0145] The molecular dynamics simulation interface library 15 includes a plurality of first simulation interfaces 151 .

[0146] Here, each first simulation interface 151 is a function calling interface of a pre-specified molecular dynamics simulation software; for example, a function calling interface of software such as LAMMPS, AMBER, CHARMM, GROMACS, and OpenMM.

[0147] (6) Quantum Chemical Computation Interface Library 16:

[0148] The quantum chemistry calculation interface library 16 includes a plurality of first calculation interfaces 161 .

[0149] Here, each first calculation interface 161 is a pre-specified function call interface of quantum chemical calculation software; for example, a function call interface of quantum chemical calculation software that implements HF / 6-311G basis set level calculation of the Hartree-Fock method, a function call interface of quantum chemical calculation software that implements B3LYP / 6-311G basis set level calculation of the density functional method, a function call interface of quantum chemical calculation software that implements MP2 / 6-311G basis set level calculation of the perturbation theory method, a function call interface of quantum chemical calculation software that implements CCSD(T) / 6-311G basis set level calculation of the coupled cluster method, and the like.

[0150] (VII) AI model interface library 17:

[0151] The AI ​​model interface library 17 includes a plurality of first AI model interfaces 171 .

[0152] Here, each first AI model interface 171 is a function calling interface of a pre-specified AI model; for example, a function calling interface of a property and performance prediction model implemented with the Uni-mol model or the Uni-QSAR model as a reference; a function calling interface of a product identification model implemented with the Uni-mol model, the Uni-QSAR model, the convolutional neural network model or the graph neural network model as a reference.

[0153] It should be noted that the division of the modules of the above system is merely a division of logical functions. In actual implementation, they can be fully or partially integrated into a single physical entity or physically separated. Furthermore, these modules can be implemented entirely in software called by a processing element; or entirely in hardware; or some modules can be implemented in software called by a processing element, while others can be implemented in hardware. For example, the workflow task scheduling module can be a separate processing element, or it can be integrated into a chip of the above-mentioned device. Furthermore, it can be stored in the memory of the above-mentioned device in the form of program code, called by a processing element of the above-mentioned system to perform the functions of the above-mentioned module. The implementation of other modules is similar. Furthermore, these modules can be fully or partially integrated together, or implemented independently. The processing element described here can be an integrated circuit with signal processing capabilities. During implementation, the various method steps of the aforementioned method or the various module processing steps of the aforementioned system can be completed by hardware integrated logic circuits in the processor element or by software instructions.

[0154] For example, these modules of the above system can be one or more integrated circuits configured to implement the aforementioned method, such as one or more application-specific integrated circuits (ASICs), one or more digital signal processors (DSPs), or one or more field programmable gate arrays (FPGAs). For another example, when a module of the above system is implemented by scheduling program code through a processing element, the processing element can be a general-purpose processor, such as a central processing unit (CPU) or other processor that can call program code. For another example, these modules can be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0155] In the above embodiments, all or part of the embodiments may be implemented using software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments may be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the above method embodiments are generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The above-mentioned computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the above-mentioned computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, Bluetooth, microwave, etc.) means. The above-mentioned computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media. The above-mentioned available medium can be a magnetic medium (such as a floppy disk, hard disk, tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid state disk (SSD)), etc.

[0156] An embodiment of the present invention provides a workflow task processing system for assisting high-voltage electrolyte experiments, including: a workflow task scheduling module, a workflow task pool, a workflow task database, a task execution module, a molecular dynamics simulation interface library, a quantum chemistry calculation interface library, and an AI model interface library; wherein, the molecular dynamics simulation interface library can be connected to the functional interface of any molecular dynamics simulation software, the quantum chemistry calculation interface library can be connected to the functional interface of any quantum chemistry calculation software, and the AI ​​model interface library can be connected to the functional interface of any AI model. The workflow task scheduling module completes three types of auxiliary task processing by calling the task execution module: high-throughput screening tasks, mechanism prediction tasks, and performance prediction tasks. If this system is applied to the experimental development of high-voltage electrolytes, it can complete the screening of large quantities of electrolytes for high-voltage electrolyte experiments by executing high-throughput screening tasks. After obtaining a batch of electrode + electrolyte systems, it can perform mechanism prediction tasks to predict the molecular-level reaction mechanism, interfacial reaction mechanism, and bipolar SEI / CEI film formation mechanism of each electrode + electrolyte system. It can also complete the performance prediction work of batch electrolytes for high-voltage electrolyte experiments by executing performance prediction tasks. Using this system as an auxiliary means for high-voltage electrolyte experiments reduces the number of experimental trial and error, improves experimental accuracy, shortens the experimental cycle, and reduces experimental costs.

[0157] Professionals should also be further aware that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0158] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein may be implemented using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0159] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A workflow task processing system for assisting high-voltage electrolyte experiments, characterized in that: The system includes: a workflow task scheduling module, a workflow task pool, a workflow task database, a task execution module, a molecular dynamics simulation interface library, a quantum chemistry calculation interface library and an AI model interface library; The workflow task scheduling module is connected to the workflow task pool, the workflow task database and the task execution module respectively; the workflow task scheduling module is used to receive the first task file sent by the high-voltage electrolyte experiment party; and initialize a corresponding task data storage area in the workflow task pool as the first data area; and perform sub-data area planning on the first data area according to the first task file; and call the task execution module to perform task execution processing on the first task file; and after confirming that the task execution processing is completed, perform task report preparation processing according to the first task file and the first data area to generate a corresponding first report file and send it back to the high-voltage electrolyte experiment party; and the first task file and the first report file form a corresponding task backup record and store it in the workflow task database; and delete the first data area; The workflow task pool is used to store a plurality of task data storage areas; The workflow task database is used to store a plurality of task backup records; The task execution module is respectively connected to the molecular dynamics simulation interface library, the quantum chemical calculation interface library and the AI ​​model interface library; the task execution module includes a first scheduling unit, a first scheduling task pool and a first prediction unit set; the first scheduling unit is respectively connected to the workflow task scheduling module, the first scheduling task pool and the first prediction unit set; the first scheduling task pool is connected to the first prediction unit set; the first prediction unit set includes a basic physical property prediction unit, a transport property prediction unit, an electrochemical property prediction unit, a molecular reaction mechanism prediction unit, an interface reaction mechanism prediction unit, an SEI / CEI film formation mechanism prediction unit and a battery performance prediction unit; each prediction unit of the first prediction unit set is connected to one or all of the three interface libraries: the molecular dynamics simulation interface library, the quantum chemical calculation interface library and the AI ​​model interface library; The first scheduling unit is used to initialize a corresponding data storage area in the first scheduling task pool as the second data area when receiving any first task execution file sent by the workflow task scheduling module, and assign a unique data area identifier to the current second data area as the corresponding second data area identifier, and set a return data area and a task status data initialized to an unfinished state in the second data area; and concentrate the prediction units that match the first prediction unit type of the first task execution file in the first prediction unit as the corresponding first prediction unit; and send the second data area identifier and the first prediction input data of the first task execution file to the first prediction unit; and regularly identify whether the task status data in the second data area is in a completed state. If so, extract the data in the return data area of ​​the second data area as the corresponding first execution result file and send it back to the workflow task scheduling module, and delete the second data area at the end of the send-back; the first task execution file includes the first prediction unit type and the first prediction input data; the first prediction unit type includes basic physical property prediction type, transport property prediction type, electrochemical property prediction type, molecular reaction mechanism prediction type, interface reaction mechanism prediction type, S E I / CE I film formation mechanism prediction type and battery performance prediction type; the task status data includes an unfinished state and a completed state; Each prediction unit in the first prediction unit set is used to locally preset a corresponding prediction processing flow, and the prediction processing flow calls the interfaces of the three interface libraries; each prediction unit in the first prediction unit set is also used to, when receiving the second data area identifier and the first prediction input data sent by the first scheduling unit, perform corresponding property, mechanism or performance prediction processing according to the first prediction input data by the locally preset prediction processing flow to obtain corresponding first prediction output data, and use the second data area corresponding to the second data area identifier in the first scheduling task pool as the corresponding current data area, and store the first prediction output data in the return data area of ​​the current data area, and update the task status data in the current data area to a completed state; The molecular dynamics simulation interface library includes a plurality of first simulation interfaces; The quantum chemical calculation interface library includes a plurality of first calculation interfaces; The AI ​​model interface library includes multiple first AI model interfaces; Wherein, each of the first simulation interfaces is a pre-specified function call interface of a molecular dynamics simulation software; each of the first calculation interfaces is a pre-specified function call interface of a quantum chemistry calculation software; each of the first AI model interfaces is a pre-specified function call interface of an AI model; The first simulation interface at least includes a function call interface of LAMMPS, AMBER, CHARMM, GROMACS, and OpenMM software; The first calculation interface at least includes a quantum chemical calculation software function call interface for implementing HF / 6-311G basis set level calculations of the Hartree-Fock method, a quantum chemical calculation software function call interface for implementing B3LYP / 6-311G basis set level calculations of the density functional method, a quantum chemical calculation software function call interface for implementing MP2 / 6-311G basis set level calculations of the perturbation theory method, and a quantum chemical calculation software function call interface for implementing CCSD(T) / 6-311G basis set level calculations of the coupled cluster method; The first AI model interface at least includes a function call interface for a property and performance prediction model based on the Uni-mol model or the Uni-QSAR model, and a function call interface for a product identification model based on the Uni-mol model, the Uni-QSAR model, the convolutional neural network model, or the graph neural network model; The prediction processing flow preset locally in the basic property prediction unit, when performing corresponding basic property prediction processing according to the first prediction input data to obtain corresponding first prediction output data, extracts the first component SMILES sequence of each component and the first component ratio of all components from the first electrolyte data of the first prediction input data to form a corresponding first input tensor; and sends the first input tensor to a pre-specified first AI model interface to perform corresponding property prediction processing to obtain a corresponding first prediction vector; and extracts corresponding viscosity prediction data, dielectric constant prediction data, melting point temperature prediction data, and boiling point temperature prediction data from the first prediction vector to form the corresponding first prediction output data; The prediction processing flow preset locally in the molecular reaction mechanism prediction unit is specifically used for sending the first electrolyte molecule set of the first prediction input data to a pre-designated first computing interface for molecular configuration optimization when the corresponding molecular reaction mechanism prediction processing is performed according to the first prediction input data to obtain the corresponding first prediction output data; and sending the optimized first electrolyte molecule set to a pre-designated first simulation interface for molecular dynamics simulation under the reaction force field and terminating the simulation when no new product objects are generated in the simulation to obtain the corresponding third simulation trajectory file; and sending the third simulation trajectory file to a pre-designated first AI model interface for reaction product identification, and performing reaction path identification based on the temporal relationship of each identified product, and sending the third simulation trajectory file to a pre-designated first computing interface for corresponding quantum chemical calculation and obtaining the corresponding reaction barrier through calculation, and the obtained reaction products, reaction path and reaction barrier constitute the corresponding first prediction output data; The interface reaction mechanism prediction unit locally presets a prediction processing flow, when performing corresponding interface reaction mechanism prediction processing according to the first prediction input data to obtain corresponding first prediction output data, sending the first electrode + electrolyte system of the first prediction input data to a pre-specified first calculation interface for configuration optimization; and performing positive and negative electrode reaction grid network prediction on the optimized first electrode + electrolyte system based on deep potential molecular dynamics theory, and forming the corresponding first prediction output data from the obtained positive and negative electrode reaction grid networks; The prediction processing flow preset locally in the battery performance prediction unit, when performing corresponding battery performance prediction processing according to the first prediction input data to obtain the corresponding first prediction output data, extracts the second component SMILES sequence of each component and the second component ratio of all components from the second electrolyte data of the first prediction input data to perform molecular system modeling to obtain the corresponding third prediction system; and performs tensor conversion on the third prediction system to obtain the corresponding second input tensor; and sends the second input tensor to a pre-designated first AI model interface to perform corresponding basic physical property prediction processing to obtain the corresponding first prediction tensor; and then sends the first prediction tensor to a pre-designated first AI model interface to perform corresponding battery performance prediction processing to obtain the corresponding second prediction vector; and extracts the corresponding battery coulombic efficiency prediction data, high and low temperature capacity prediction data, rate performance prediction data, DC resistance prediction data, and AC impedance prediction data from the second prediction vector to form the corresponding first prediction output data.

2. The workflow task processing system for assisting high-voltage electrolyte experiments according to claim 1 is characterized in that: The first task file includes a first task type, first task data and a first task configuration; The first task types include high-throughput screening type, mechanism prediction type and performance prediction type; When the first task type is a high-throughput screening type, the corresponding first task data is a first electrolyte data set to be screened; the first electrolyte data set includes one or more first electrolyte data; the first electrolyte data includes a first component data set and a first component ratio; the first component data set includes multiple first component data, the first component data includes a first component type, a first component name and a first component SMILES sequence, the first component type includes an electrolyte salt, a solvent and an additive; the first component ratio is the ratio of the electrolyte salt, the solvent and the additive of the corresponding electrolyte, and the first component ratio = electrolyte salt ratio: solvent ratio: additive ratio; When the first task type is a high-throughput screening type, the corresponding first task is configured as a multi-level screening indicator set, consisting of a primary physical property indicator set, a secondary transport property indicator set, and a tertiary electrochemical property indicator set; the primary physical property indicator set includes a viscosity indicator, a dielectric constant indicator, a melting point temperature indicator, and a boiling point temperature indicator; the secondary transport property indicator set includes a diffusion coefficient indicator, a conductivity indicator, and a migration number indicator; the tertiary electrochemical property indicator set includes an oxidation potential indicator, a reduction potential indicator, and an electrochemical window indicator; When the first task type is a mechanism prediction type, the corresponding first task data is a first electrode + electrolyte system set to be predicted; the first electrode + electrolyte system set includes one or more first electrode + electrolyte systems; The first electrode + electrolyte system is an interface model system formed by the fusion of an electrode plate model system and an electrolyte model system, and the first electrode + electrolyte system includes a plurality of first atoms, a plurality of first connecting bonds, and a plurality of first clusters; the atomic properties of the first atoms include an atomic identifier, an atomic element type, an atomic charge, an atomic size, an atomic coordinate, and an identifier of a cluster in which the atoms are located; the bond properties of the first connecting bonds include a connecting bond identifier, a connecting bond type, and an atom pair identifier group corresponding to the connecting bond; the first cluster includes a cluster identifier and a cluster type; When the first task type is a mechanism prediction type, the corresponding first task is configured as a multi-class prediction state machine, which consists of a molecular reaction mechanism prediction state bit, an interface reaction mechanism prediction state bit, and an SEI / CEI film formation mechanism prediction state bit, and all three state bits include two state configuration values: an activation state and an inactivation state; and the state configuration value of at least one of the three state bits is an activation state; When the first task type is a performance prediction type, the corresponding first task data is a set of second electrolyte data to be predicted; The second electrolyte data set includes one or more second electrolyte data; the second electrolyte data includes a second component data set and a second component ratio; the second component data set includes a plurality of second component data, the second component data includes a second component type, a second component name, and a second component SMILES sequence, the second component type includes an electrolyte salt, a solvent, and an additive; the second component ratio is the ratio of the electrolyte salt, the solvent, and the additive of the corresponding electrolyte, and the second component ratio = electrolyte salt ratio: solvent ratio: additive ratio; When the first task type is a performance prediction type, the corresponding first task configuration is empty.

3. The workflow task processing system for assisting high-voltage electrolyte experiments according to claim 2 is characterized in that: The workflow task scheduling module is specifically configured to identify the first task type of the first task file when planning sub-data areas of the first data area according to the first task file; When the first task type is a high-throughput screening type, counting the number of the first electrolyte data in the first electrolyte data set of the first task file to obtain a corresponding total number of first electrolytes; and creating a sub-data area for the total number of the first electrolytes in the first data area as the corresponding first electrolyte data area, and creating a high-throughput screening task output data area; and further creating three sub-data areas in each of the first electrolyte data areas, namely a primary physical property prediction data area, a secondary transport property prediction data area, and a tertiary electrochemical property prediction data area; the first electrolyte data areas correspond one to one with the first electrolyte data; When the first task type is a mechanism prediction type, the number of the first electrode + electrolyte systems in the first electrode + electrolyte system set of the first task file is counted to obtain the corresponding total number of first systems; and a sub-data area of ​​the total number of the first systems is created in the first data area and recorded as the corresponding first system data area; and the three status bits of the first task configuration of the first task file are identified, if the molecular reaction mechanism prediction status bit is in an activated state, a corresponding molecular reaction mechanism prediction data area is created in each of the first system data areas, if the interface reaction mechanism prediction status bit is in an activated state, a corresponding interface reaction mechanism prediction data area is created in each of the first system data areas, and if the SEI / CEI film formation mechanism prediction status bit is in an activated state, a corresponding SEI / CEI film formation mechanism prediction data area is created in each of the first system data areas; the first system data area corresponds one-to-one to the first electrode + electrolyte system; When the first task type is a performance prediction type, counting the number of the second electrolyte data in the second electrolyte data set of the first task file to obtain a corresponding total number of second electrolytes; and creating a sub-data area for the total number of the second electrolytes in the first data area and recording it as the corresponding second electrolyte data area; The second electrolyte data corresponds one-to-one to the second electrolyte data area.

4. The workflow task processing system for assisting high-voltage electrolyte experiments according to claim 3 is characterized in that: The workflow task scheduling module is specifically configured to identify the first task type of the first task file when the task execution module is called to perform task execution processing on the first task file; When the first task type is a high-throughput screening type, calling the task execution module and the first data area to perform high-throughput screening task execution processing according to the first task file; and confirming the completion of the current high-throughput screening task execution processing when the current high-throughput screening task execution processing is completed; When the first task type is a mechanism prediction type, calling the task execution module and the first data area to perform mechanism prediction task execution processing according to the first task file; and confirming the completion of the current task execution processing when the current mechanism prediction task execution processing ends; When the first task type is a performance prediction type, the task execution module and the first data area are called to perform performance prediction task execution processing according to the first task file; and the task execution processing is confirmed to be completed when the performance prediction task execution processing ends.

5. The workflow task processing system for assisting high-voltage electrolyte experiments according to claim 4 is characterized in that: The workflow task scheduling module is specifically configured to extract each first electrolyte data of the first electrolyte data set of the first task file as a corresponding first prediction input data when the task execution module and the first data area are called to perform high-throughput screening task execution processing according to the first task file, and set a corresponding first prediction unit type as a basic physical property prediction type, and form a corresponding first task execution file by the first prediction unit type corresponding to each first electrolyte data and the first prediction input data; and send each first task execution file obtained this time to the first scheduling unit of the task execution module; and receiving the first execution result files corresponding to the first task execution files returned by the first scheduling unit; and whenever a first execution result file is received, taking the first physical property prediction data area of ​​the first electrolyte data area corresponding to the first electrolyte data corresponding to the current first execution result file in the first data area as the corresponding current data area, and extracting the corresponding viscosity prediction data, dielectric constant prediction data, melting point temperature prediction data, and boiling point temperature prediction data from the current first execution result file and storing them in the current data area; After confirming that the first-level physical property prediction data areas of all the first electrolyte data areas in the first data area have been stored in corresponding execution result files, all prediction data of each of the first-level physical property prediction data areas are verified based on the first-level physical property indicator set configured in the first task of the first task file to obtain corresponding first verification results; and the first electrolyte data corresponding to the first-level physical property prediction data area for which the first verification result indicates that the first electrolyte data meets the standards is recorded as corresponding first-level qualified electrolyte data; and forming a corresponding first-level standard electrolyte data set from all the obtained first-level standard electrolyte data; The first verification result includes meeting the standard and failing to meet the standard; and extracting each first-level qualified electrolyte data from the first-level qualified electrolyte data set as a corresponding first prediction input data, setting a corresponding first prediction unit type as a transport property prediction type, and forming a corresponding first task execution file from the first prediction unit type and the first prediction input data corresponding to each first-level qualified electrolyte data; and sending each first task execution file obtained this time to the first scheduling unit of the task execution module; and receiving the first execution result files corresponding to the first task execution files returned by the first scheduling unit; and whenever a first execution result file is received, taking the secondary transport property prediction data area of ​​the first electrolyte data area corresponding to the first electrolyte data corresponding to the current first execution result file in the first data area as the corresponding current data area, and extracting the corresponding diffusion coefficient prediction data, conductivity prediction data, and migration number prediction data from the current first execution result file and storing them in the current data area; After confirming that the secondary transport property prediction data areas of the first electrolyte data area corresponding to all the first-level qualified electrolyte data in the first data area are stored in the corresponding execution result file, all prediction data of each of the second-level transport property prediction data areas are verified based on the second-level transport property indicator set configured in the first task of the first task file to obtain a corresponding second verification result; and the first-level qualified electrolyte data corresponding to the second-level qualified electrolyte data area for which the second verification result indicates that the electrolyte data meets the standard is recorded as the corresponding second-level qualified electrolyte data; and forming a corresponding secondary standard electrolyte data set from all the obtained secondary standard electrolyte data; The second verification result includes meeting the standard and failing to meet the standard; and extracting each of the secondary qualified electrolyte data from the secondary qualified electrolyte data set as a corresponding first prediction input data, setting a corresponding first prediction unit type as an electrochemical property prediction type, and forming a corresponding first task execution file from the first prediction unit type and the first prediction input data corresponding to each of the secondary qualified electrolyte data; and sending each of the first task execution files obtained this time to the first scheduling unit of the task execution module; and receiving the first execution result files corresponding to the first task execution files returned by the first scheduling unit; and whenever a first execution result file is received, taking the three-level electrochemical property prediction data area of ​​the first electrolyte data area corresponding to the first electrolyte data corresponding to the current first execution result file in the first data area as the corresponding current data area, and extracting the corresponding oxidation potential prediction data, reduction potential prediction data, and electrochemical window prediction data from the current first execution result file and storing them in the current data area; After confirming that the tertiary electrochemical property prediction data area of ​​the first electrolyte data area corresponding to all the secondary qualified electrolyte data in the first data area is stored in the corresponding execution result file, all the predicted data of each of the tertiary electrochemical property prediction data areas are verified based on the tertiary electrochemical property indicator set of the first task configuration of the first task file to obtain a corresponding third verification result; and the secondary qualified electrolyte data corresponding to the tertiary electrochemical property prediction data area whose third verification result is qualified is recorded as the corresponding tertiary qualified electrolyte data; and a corresponding tertiary qualified electrolyte data set is formed by all the obtained tertiary qualified electrolyte data; The third verification result includes meeting the standard and failing to meet the standard; and storing the obtained three-level standard electrolyte data set into the high-throughput screening task output data area of ​​the first data area; When the data is successfully stored, it is confirmed that the execution of this high-throughput screening task is completed.

6. The workflow task processing system for assisting high-voltage electrolyte experiments according to claim 5 is characterized in that: The workflow task scheduling module is specifically used to extract the viscosity prediction data, dielectric constant prediction data, melting point temperature prediction data and boiling point temperature prediction data of the current first-level physical property prediction data area as the corresponding current viscosity, current dielectric constant, current melting point temperature and current boiling point temperature when the first-level physical property indicator set configured by the first task file is verified for all prediction data of each first-level physical property prediction data area to obtain a corresponding first verification result; and when the current viscosity matches the viscosity indicator of the first-level physical property indicator set according to the preset viscosity matching rule, and the current dielectric constant matches the dielectric constant indicator of the first-level physical property indicator set according to the preset dielectric constant matching rule, and the When the current melting point temperature matches the melting point temperature index of the primary physical property index set according to the preset melting point temperature matching rule, and the current boiling point temperature matches the boiling point temperature index of the primary physical property index set according to the preset boiling point temperature matching rule, the first verification result is set to be up to standard; and when the current viscosity does not match the viscosity index of the primary physical property index set according to the viscosity matching rule, or the current dielectric constant does not match the dielectric constant index according to the dielectric constant matching rule, or the current melting point temperature does not match the melting point temperature index according to the melting point temperature matching rule, or the current boiling point temperature does not match the boiling point temperature index according to the boiling point temperature matching rule, the first verification result is set to be unsatisfactory; The workflow task scheduling module is specifically configured to extract the diffusion coefficient prediction data, conductivity prediction data and migration number prediction data of the current secondary transport property prediction data area as the corresponding current diffusion coefficient, current conductivity and current migration number when the secondary transport property indicator set configured by the first task file based on the first task file verifies all the prediction data of each secondary transport property prediction data area to obtain a corresponding second verification result; and extract the diffusion coefficient prediction data, conductivity prediction data and migration number prediction data of the current secondary transport property prediction data area as the corresponding current diffusion coefficient, current conductivity and current migration number; and when the current diffusion coefficient matches the diffusion coefficient indicator of the secondary transport property indicator set according to a preset diffusion coefficient matching rule, and the When the current conductivity matches the conductivity index of the secondary transport property index set according to the preset conductivity matching rule, and the current mobility number matches the mobility number index of the secondary transport property index set according to the preset mobility number matching rule, the second verification result is set as meeting the standard; and when the current diffusion coefficient does not match the diffusion coefficient index according to the diffusion coefficient matching rule, or the current conductivity does not match the conductivity index according to the conductivity matching rule, or the current mobility number does not match the mobility number index according to the mobility number matching rule, the second verification result is set as failing to meet the standard; The workflow task scheduling module is specifically used to, when the three-level electrochemical property indicator set configured by the first task based on the first task file verifies all the prediction data of each three-level electrochemical property prediction data area to obtain a corresponding third verification result, extract the oxidation potential prediction data, reduction potential prediction data and electrochemical window prediction data of the current three-level electrochemical property prediction data area as the corresponding current oxidation potential, current reduction potential and current electrochemical window; and when the current oxidation potential is matched with the oxidation potential indicator of the three-level electrochemical property indicator set according to a preset oxidation potential matching rule, and the current reduction potential is matched with the three-level electrochemical property indicator set according to a preset reduction potential matching rule. When the reduction potential index of the chemical property index set matches, and the current electrochemical window matches the electrochemical window index of the three-level electrochemical property index set according to the preset electrochemical window matching rules, the third verification result is set to be up to standard; and when the current oxidation potential does not match the oxidation potential index of the three-level electrochemical property index set according to the oxidation potential matching rules, or the current reduction potential does not match the reduction potential index of the three-level electrochemical property index set according to the reduction potential matching rules, or the current electrochemical window does not match the electrochemical window index of the three-level electrochemical property index set according to the electrochemical window matching rules, the third verification result is set to be unsatisfactory.

7. The workflow task processing system for assisting high-voltage electrolyte experiments according to claim 4 is characterized in that: The workflow task scheduling module is specifically configured to identify the molecular reaction mechanism prediction status bit, the interface reaction mechanism prediction status bit, and the SEI / CEI film formation mechanism prediction status bit of the first task configuration of the first task file when the task execution module is called and the first data area performs the mechanism prediction task execution processing according to the first task file; When the molecular reaction mechanism prediction state bit is in an activated state, one or more types of molecular structures in each of the first electrode + electrolyte systems of the first electrode + electrolyte system set of the first task file are extracted to form a corresponding first electrolyte molecule set, and each of the first electrolyte molecule sets is used as a corresponding first prediction input data, and a corresponding first prediction unit type is set as a molecular reaction mechanism prediction type, and a corresponding first task execution file is formed by the first prediction unit type and the first prediction input data corresponding to each of the first electrode + electrolyte systems; and each of the first task execution files obtained this time is sent to the first scheduling unit of the task execution module; and the first execution result file corresponding to each of the first task execution files this time returned by the first scheduling unit is received; and each time a first execution result file is received, taking the molecular reaction mechanism prediction data area of ​​the first system data area corresponding to the first electrode+electrolyte system corresponding to the current first execution result file in the first data area as the corresponding current data area, and storing the current first execution result file in the current data area; When the interface reaction mechanism prediction state bit is in an activated state, each of the first electrode + electrolyte systems of the first electrode + electrolyte system set of the first task file is extracted as a corresponding first prediction input data, and a corresponding first prediction unit type is set as an interface reaction mechanism prediction type, and the first prediction unit type corresponding to each of the first electrode + electrolyte systems and the first prediction input data form a corresponding first task execution file; and each of the first task execution files obtained this time is sent to the first scheduling unit of the task execution module; and the first execution result file corresponding to each of the first task execution files this time returned by the first scheduling unit is received; and each time a first execution result file is received, the interface reaction mechanism prediction data area of ​​the first system data area corresponding to the first electrode+electrolyte system corresponding to the current first execution result file in the first data area is used as the corresponding current data area, and the current first execution result file is stored in the current data area; When the SEI / CEI film formation mechanism prediction status bit is in an activated state, each of the first electrode + electrolyte systems of the first electrode + electrolyte system set of the first task file is extracted as a corresponding first prediction input data, and a corresponding first prediction unit type is set as the SEI / CEI film formation mechanism prediction type, and the first prediction unit type corresponding to each of the first electrode + electrolyte systems and the first prediction input data form a corresponding first task execution file; and each of the first task execution files obtained this time is sent to the first scheduling unit of the task execution module; and the first execution result file corresponding to each of the first task execution files this time returned by the first scheduling unit is received; and each time a first execution result file is received, the SEI / CEI film formation mechanism prediction data area of ​​the first system data area corresponding to the first electrode+electrolyte system corresponding to the current first execution result file in the first data area is used as the corresponding current data area, and the current first execution result file is stored in the current data area; After all execution result files of all the first electrode+electrolyte systems are stored, it is confirmed that the execution of this mechanism prediction task is completed.

8. The workflow task processing system for assisting high-voltage electrolyte experiments according to claim 4 is characterized in that: The workflow task scheduling module is specifically configured to extract each second electrolyte data of the second electrolyte data set of the first task file as a corresponding first prediction input data when the task execution module and the first data area are called to perform the performance prediction task execution processing according to the first task file, and set a corresponding first prediction unit type as a battery performance prediction type, and form a corresponding first task execution file by the first prediction unit type corresponding to each first electrolyte data and the first prediction input data; and send each first task execution file obtained this time to the first scheduling unit of the task execution module; and receiving the first execution result files corresponding to the first task execution files returned by the first scheduling unit; and whenever a first execution result file is received, taking the second electrolyte data area in the first data area corresponding to the second electrolyte data corresponding to the current first execution result file as the corresponding current data area, and storing the current first execution result file in the current data area; After the corresponding execution result files are stored in all the second electrolyte data areas in the first data area, it is confirmed that the execution processing of this performance prediction task is completed.

9. The workflow task processing system for assisting high-voltage electrolyte experiments according to claim 3 is characterized in that: The workflow task scheduling module is specifically configured to identify the first task type of the first task file when performing task report preparation processing according to the first task file and the first data area to generate a corresponding first report file and send it back to the high-voltage electrolyte experiment party; When the first task type is a high-throughput screening type, the stored data in the high-throughput screening task output data area of ​​the first data area is extracted as the corresponding first screening electrolyte set; and the first report file corresponding to the first screening electrolyte set is composed and sent back to the high-voltage electrolyte experiment party; When the first task type is a mechanism prediction type, the stored data of each first system data area in the first data area is extracted as a corresponding first system mechanism prediction sub-report, and all the obtained first system mechanism prediction sub-reports are combined to form a corresponding first report file and sent back to the high-voltage electrolyte experiment party; When the first task type is a performance prediction type, the stored data of each of the second electrolyte data areas in the first data area are extracted as the corresponding first electrolyte performance prediction sub-report, and the corresponding first report file is composed of all the first electrolyte performance prediction sub-reports obtained and sent back to the high-voltage electrolyte experiment party.

Images