CN116777010B - Model training method and task execution method and device - Google Patents
Model training method and task execution method and device Download PDFInfo
- Publication number
- CN116777010B CN116777010B CN202311080508.2A CN202311080508A CN116777010B CN 116777010 B CN116777010 B CN 116777010B CN 202311080508 A CN202311080508 A CN 202311080508A CN 116777010 B CN116777010 B CN 116777010B
- Authority
- CN
- China
- Prior art keywords
- field data
- physical field
- mixture state
- moment
- space
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 75
- 238000012549 training Methods 0.000 title claims abstract description 62
- 239000000203 mixture Substances 0.000 claims abstract description 190
- 239000000126 substance Substances 0.000 claims description 62
- 239000012530 fluid Substances 0.000 claims description 28
- 238000003860 storage Methods 0.000 claims description 17
- 238000004590 computer program Methods 0.000 claims description 15
- 239000007787 solid Substances 0.000 claims description 15
- 238000005457 optimization Methods 0.000 claims description 10
- 239000007790 solid phase Substances 0.000 claims description 9
- 230000008569 process Effects 0.000 abstract description 18
- 238000010586 diagram Methods 0.000 description 11
- 230000006870 function Effects 0.000 description 9
- 230000006872 improvement Effects 0.000 description 8
- 238000012545 processing Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 4
- 239000000428 dust Substances 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000007664 blowing Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000012071 phase Substances 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 230000005055 memory storage Effects 0.000 description 1
- 230000000704 physical effect Effects 0.000 description 1
- 229920001296 polysiloxane Polymers 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 239000010979 ruby Substances 0.000 description 1
- 229910001750 ruby Inorganic materials 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Landscapes
- Feedback Control In General (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The method and the device for model training are characterized in that the obtained physical field data of the mixture state in the designated space at the first moment can be input into a prediction model to train the prediction model, so that when the trained prediction model is applied to the actual task execution process, compared with the prior art, excessive time is not needed to be spent for deducing the physical field data of the mixture state in the designated space at the next moment at one step, the efficiency of predicting the physical field data is improved, and in addition, in the training stage, the accuracy of the physical field data predicted by the prediction model in the actual application can be ensured because the quality distribution of the mixture state before and after the first moment and the second moment meets the quality constraint.
Description
Technical Field
The present disclosure relates to the field of artificial intelligence and the field of fluid mechanics, and in particular, to a method for model training and a method and apparatus for task execution.
Background
Multiphase flow has wide application in industrial production and daily life, including industrial energy, chemical industry, aerospace, metallurgical and other fields, dust removing process, etc. The method realizes the efficient and accurate prediction of physical fields (volume fraction fields, flow fields, pressure fields, temperature fields and the like) in multiphase flow, and has important significance for the optimization design of related processes.
Numerical simulation means are generally used at present to obtain details of physical fields under a specific condition (such as a physical state formed by mixing a fluid substance and a solid substance), but the time is often long.
Disclosure of Invention
The present disclosure provides a method for model training and a method and apparatus for task execution, so as to partially solve the above-mentioned problems in the prior art.
The technical scheme adopted in the specification is as follows:
the present specification provides a method of model training, comprising:
acquiring data of each physical field of a mixture state in a designated space at a first moment, wherein the mixture state at least comprises a substance in a fluid state, the designated space comprises an inlet and an outlet, and the mixture state enters the designated space from the inlet and leaves the designated space from the outlet;
Inputting the physical field data into a prediction model to be trained, so that the prediction model predicts the physical field data of the mixture state in the specified space at a second moment as prediction data according to the physical field data, wherein the second moment is a moment after the first moment;
determining, as a first deviation, a deviation between an inflow amount of the substance in the fluid state in the specified space per unit space in the specified space and an increment of a mass of the substance contained in the per unit space in the unit time, and determining, as a predicted deviation, a deviation between the predicted data and the tag physical field data of the mixture state in the specified space at the second time, based on the predicted data and the physical field data of the mixture state at the first time;
and training the prediction model according to the first deviation and the prediction deviation.
Optionally, the mixture further comprises a substance in a solid phase state;
before training the predictive model based on the first deviation and the predicted deviation, the method further includes:
Determining a deviation of an increment of the substance in the solid state in the designated space in unit time and an inflow amount of the substance in the solid state in the designated space in unit time as a second deviation according to the prediction data and the physical field data of the mixture state at the first moment;
training the prediction model according to the first deviation and the prediction deviation, wherein the training specifically comprises the following steps:
training the predictive model with the optimization objective of minimizing the first deviation, minimizing the second deviation, and minimizing the predictive deviation.
Optionally, the prediction model includes a first network layer and a second network layer;
inputting the physical field data into a prediction model to be trained, so that the prediction model predicts the physical field data of the mixture state in the specified space at the second time according to the physical field data, and the prediction model specifically comprises the following steps:
inputting each physical field data of the mixture state at a first time into a first network layer in the prediction model, so that the first network layer predicts the volume fraction field data of the mixture state in the specified space at the second time according to each physical field data of the mixture state at the first time;
And inputting the volume fraction field data and the physical field data of the mixture state at the first moment into a second network layer in the prediction model, so that the second network layer predicts other physical field data of the mixture state in the appointed space except the volume fraction field data at the second moment according to the physical field data of the mixture state at the first moment and the volume fraction field data.
Optionally, acquiring each physical field data of the mixture state in the designated space at the first moment specifically includes:
acquiring data of each physical field of the mixture state in the designated space at the first moment under a set physical condition;
inputting the physical field data into a prediction model to be trained, so that the prediction model predicts the physical field data of the mixture state in the specified space at a second moment as prediction data according to the physical field data, wherein the second moment is a moment after the first moment and specifically comprises:
and inputting the physical field data of the mixture state at the first moment and the condition parameters corresponding to the set physical conditions into the prediction model, so that the prediction model predicts the physical field data of the mixture state in the specified space at the second moment under the set physical conditions according to the physical field data of the mixture state at the first moment, and the physical field data are used as prediction data.
The specification provides a task execution method, which comprises the following steps:
acquiring data of each physical field of a mixture state in a designated space at the current time, wherein the mixture state at least comprises a substance in a fluid state, the designated space comprises an inlet and an outlet, and the mixture state enters the designated space from the inlet and leaves the designated space from the outlet;
inputting the physical field data into a pre-trained prediction model, so that the prediction model predicts the physical field data of the mixture state in the appointed space at the next moment according to the physical field data, wherein the prediction model is obtained by training by the model training method;
and executing tasks according to the predicted physical field data of the mixture state in the designated space at the next moment.
The present specification provides an apparatus for model training, comprising:
the system comprises an acquisition module, a control module and a control module, wherein the acquisition module is used for acquiring each physical field data of a mixture state in a designated space at a first moment, the mixture state at least comprises a substance in a fluid state, the designated space comprises an inlet and an outlet, and the mixture state enters the designated space from the inlet and leaves the designated space from the outlet;
The prediction module is used for inputting the physical field data into a prediction model to be trained, so that the prediction model predicts the physical field data of the mixture state in the specified space at a second moment according to the physical field data, wherein the second moment is a moment after the first moment and is used as prediction data;
a determining module, configured to determine, as a first deviation, a deviation between an inflow amount of the substance in the fluid state in the specified space in each unit space in the specified space and an increment of a mass of the substance contained in each unit space in the unit space in accordance with the prediction data and the respective physical field data of the mixture state in the specified space at the first time, and determine, as a prediction deviation, a deviation between the prediction data and the respective tag physical field data of the mixture state in the specified space at the second time;
and the training module is used for training the prediction model according to the first deviation and the prediction deviation.
Optionally, the mixture further comprises a substance in a solid phase state;
the determining module is further configured to determine, as a second deviation, a deviation between an increment of the substance in the solid state in the specified space in a unit time and an inflow of the substance in the solid state in the specified space in a unit time, based on the prediction data and each physical field data of the mixture state at the first time, before training the prediction model based on the first deviation and the prediction deviation;
The training module is specifically configured to train the prediction model with the first deviation minimized, the second deviation minimized, and the prediction deviation minimized as optimization objectives.
Optionally, the prediction model includes a first network layer and a second network layer;
the prediction module is specifically configured to input each physical field data of the mixture state at a first time into a first network layer in the prediction model, so that the first network layer predicts volume fraction field data of the mixture state in the specified space at the second time according to each physical field data of the mixture state at the first time; and inputting the volume fraction field data and the physical field data of the mixture state at the first moment into a second network layer in the prediction model, so that the second network layer predicts other physical field data of the mixture state in the appointed space except the volume fraction field data at the second moment according to the physical field data of the mixture state at the first moment and the volume fraction field data.
Optionally, the acquiring module is specifically configured to acquire, under a set physical condition, each physical field data of the mixture state in the specified space at the first moment;
The prediction module is specifically configured to input each physical field data of the mixture state at a first time and a condition parameter corresponding to the set physical condition into the prediction model, so that the prediction model predicts each physical field data of the mixture state in the specified space at the set physical condition at the second time as prediction data according to each physical field data of the mixture state at the first time.
The present specification provides a task execution device including:
the system comprises an acquisition module, a control module and a control module, wherein the acquisition module is used for acquiring each physical field data of a mixture state in a designated space at the current moment, the mixture state at least comprises a substance in a fluid state, the designated space comprises an inlet and an outlet, and the mixture state enters the designated space from the inlet and leaves the designated space from the outlet;
the prediction module is used for inputting the physical field data into a pre-trained prediction model so that the prediction model predicts the physical field data of the mixture state in the appointed space at the next moment according to the physical field data, and the prediction model is trained by the model training method;
And the execution module is used for executing tasks according to the predicted physical field data of the mixture state in the specified space at the next moment.
The present specification provides a computer readable storage medium storing a computer program which when executed by a processor implements the method of model training or task execution method described above.
The present specification provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method of model training or the task execution method described above when executing the program.
The above-mentioned at least one technical scheme that this specification adopted can reach following beneficial effect:
as can be seen from the above method, in the present specification, since the obtained physical field data of the mixture state in the specified space at the first time may be input into the prediction model to train the prediction model, when the trained prediction model is applied to the actual task execution process, compared with the prior art, it does not need to take too much time to deduce the physical field data of the mixture state in the specified space at the next time at a step, so that the efficiency of predicting the physical field data is improved by several times, and in the training stage, the quality of the material in the fluid state in the mixture state before and after the first time and the second time accords with the quality constraint, so that the accuracy of the physical field data predicted by the prediction model in the actual application may be ensured.
Drawings
The accompanying drawings, which are included to provide a further understanding of the specification, illustrate and explain the exemplary embodiments of the present specification and their description, are not intended to limit the specification unduly. In the drawings:
FIG. 1 is a flow chart of a method of model training provided in the present specification;
FIG. 2 is a flow chart of a task execution method provided in the present specification;
FIG. 3 is a schematic diagram of a model training apparatus provided herein;
FIG. 4 is a schematic diagram of a task performing device provided in the present specification;
fig. 5 is a schematic structural view of an electronic device corresponding to fig. 1 or fig. 2 provided in the present specification.
Detailed Description
For the purposes of making the objects, technical solutions and advantages of the present specification more apparent, the technical solutions of the present specification will be clearly and completely described below with reference to specific embodiments of the present specification and corresponding drawings. It will be apparent that the described embodiments are only some, but not all, of the embodiments of the present specification. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are intended to be within the scope of the present disclosure.
The following describes in detail the technical solutions provided by the embodiments of the present specification with reference to the accompanying drawings.
Fig. 1 is a flow chart of a method for model training provided in the present specification, including the following steps:
s101: physical field data of a mixture state in a specified space at a first time is acquired.
The execution subject of the model training method provided in the present specification may be a terminal device such as a desktop computer or a notebook computer, or may be a server, and for convenience of explanation, the present specification describes the provided model training method using only the terminal device as the execution subject.
Since the scheme provided in the specification is to train a prediction model to realize rapid and accurate prediction of physical field data through the trained prediction model, a description for training the prediction model is acquired before training the prediction model is started.
Therefore, in the present specification, the terminal device may acquire each physical field data of the mixture state in the specified space at the first time. Here, the designated space may refer to a space having two ports, that is, the designated space includes an inlet and an outlet, and the mixture state may flow into and out of the designated space, so this process may be regarded as "blowing" into the designated space from the inlet in the designated space, and at the same time, a part of the mixture state may flow into the designated space, and a part of the mixture state may flow out of the designated space.
Since the scheme provided in the present specification is mainly intended to predict physical field data in the field of multiphase flow, at least a substance in a fluid state may be included in a mixture state, but a substance in a solid state may be included in a mixture state.
It should be further explicitly noted that the mixture state mentioned in the present specification is not a fixed atomic or molecular composition state, just like the above mentioned "blowing" process, so that the mixture state in the present specification is a flowing state, and the physical field data to be predicted in the present specification should be understood as physical field data of the part of the mixture state enclosed in the designated space.
The physical field data may be real data acquired by a measuring instrument or simulation data constructed by simulation. But different physical field data because each physical field data corresponds to a different physical dimension. Each of the physical field data mentioned in the present specification may include, for example, temperature field data, velocity field data, pressure field data, volume fraction field, and the like. The different physical field data actually reflects the physical properties exhibited by the mixture state at different locations in the designated space, for example, the physical data may be 128 x 128 data, so that when the physical field data is temperature field data, then the 128 x 128 temperature field data may be understood as a temperature value for characterizing each of 128 x 128 points uniformly distributed in the designated space, and similarly, for the pressure field data, the 128 x 128 pressure field data may be understood as a pressure value for characterizing each of 128 x 128 points uniformly distributed in the designated space.
It is noted that in order to be able to guarantee the subsequent training effect on the predictive model, it is necessary to guarantee that the mixture state in the specified space follows conservation of mass, i.e. how much mixture state flows in at the inlet in the specified space if no mixture state flows out from the outlet in the specified space, the mixture state of the corresponding mass should be increased in the specified space. That is, the amount of the mixture state flowing in or out of the specified space should correspond to an increase or decrease in the mass of the mixture state in the specified space.
In addition, in practical application, each piece of physical field data of the mixture state in the designated space at the first moment can be each piece of physical field data under a set physical condition, and the set physical condition can have various specific forms, such as a specific set air pressure, a specific set temperature, a specific set flow rate and the like. For example, when the set physical condition is that the set mixture state is at a certain temperature, then each physical field data in the specified space is acquired when the mixture state is at this temperature at the first time.
S102: and inputting the physical field data into a prediction model to be trained, so that the prediction model predicts the physical field data of the mixture state in the specified space at the second time according to the physical field data, and takes the predicted physical field data as prediction data.
After the terminal device obtains the physical field data, the physical field data can be input into a prediction model to be trained, and the prediction model predicts the physical field data of the mixture state in the designated space at the second time as prediction data. Wherein the second time is a time after the first time, for example, the second time may be a time next to the first time.
When predicting the data of each physical field at the second time, the volume fraction field data of the mixture state in the specified space at the second time can be predicted first, and then other physical field data of the mixture state in the specified space at the second time can be predicted according to the volume fraction field data. Where the volumetric fraction field data can be understood as representing the volume of the mixture state at each location in a given space.
Therefore, the prediction model in the present specification may include a first network layer and a second network layer, and the two network layers may be convolutional neural networks specifically.
On the basis, the specific process of inputting the physical field data into the prediction model to be trained by the terminal equipment can be that the physical field data of the mixture state at the first moment is input into a first network layer in the prediction model, and the first network layer predicts the volume fraction field data of the mixture state in the appointed space at the second moment according to the physical field data of the mixture state at the first moment. And then inputting the obtained volume fraction field data and the obtained physical field data of the mixture state at the first moment into a second network layer in the prediction model, and predicting other physical field data of the mixture state in the appointed space except the volume fraction field data at the second moment according to the physical field data of the mixture state at the first moment and the volume fraction field data by the second network layer. Other physical field data mentioned herein is understood to be physical field data other than volumetric fraction field data such as temperature, pressure, etc.
S103: according to the prediction data and the physical field data of the mixture state at the first moment, determining the deviation between the inflow amount of the substance in the fluid state in the designated space in each unit space in the designated space and the increment of the mass of the substance contained in each unit space in unit time as a first deviation, and determining the deviation between the prediction data and the physical field data of the labels of the mixture state in the designated space at the second moment as a prediction deviation.
In training the above-described predictive model, in addition to using a conventional label, it is necessary to introduce an optimization target of mass constraint, that is, since the above-described inflow and outflow of the mixture state in the specified space is required to satisfy mass conservation with an increase and decrease in the mass of the mixture state in the specified space, in practice, the inflow/outflow amount of the substance in the fluid state in the specified space in each cell space and the mass increase/decrease of the substance contained in each cell space should be the same.
Therefore, after obtaining the above-mentioned prediction data, it is necessary to determine, as the prediction bias, the bias between the prediction data and the physical field data of each tag at the second time point in the mixture state in the specified space. At the same time, the terminal device also needs to determine, as the first deviation, a deviation between an inflow amount per unit space in the designated space in the unit time of the substance in the fluid state in the designated space and an increment per unit time of the mass of the substance contained in each unit space, based on the prediction data and the respective physical field data of the mixture state at the first time input into the prediction model.
Wherein each of the tag physical field data mentioned herein may refer to each of the real physical field data specifying a substance in a fluid state in a space at the second time.
And the above-mentioned unit space can be understood as uniformly dividing a designated space into respective minute spaces. That is, the cell space in this application can be understood as a concept similar to "infinitesimal".
S104: and training the prediction model according to the first deviation and the prediction deviation.
The terminal device can determine a loss value corresponding to the first deviation through the first deviation, determine a corresponding loss value according to the predicted deviation, and train the prediction model based on the determined loss value.
Specifically, the determined loss value is minimized as an optimization target, and the prediction model is trained, so that it is also understood that the first deviation is minimized and the prediction deviation is minimized as an optimization target, and the prediction model is trained.
The determining the loss value corresponding to the first deviation may be implemented by the following formula:
in the middle ofRepresenting the volume of the substance in the fluid state in the specified space flowing into the unit space, +. >Represents the density of the substance in the fluid state, +.>Representing the rate of inflow of the substance in the fluid state in the specified space.
As can be seen from the above formula, it is necessary to ensure that the inflow amount of the substance in the fluid state into each cell space in the specified space is kept consistent with the mass increment of the substance contained in each cell space, that is, the first deviation should approach 0. It should be noted that the inflow referred to in this application may have a positive or negative score, i.e. it represents an inflow into the cell space when the inflow takes a positive value, and it represents an outflow from the cell space when the inflow takes a negative value. Accordingly, the quality increment mentioned in the present application is also divided into positive and negative, that is, when the quality increment takes a positive value, it means that the quality is increased, and when the quality increment takes a negative value, it means that the quality is decreased.
In addition, since the above-described inflow and outflow of the mixture state in the specified space needs to be kept in agreement with the mass increment of the substance contained in the specified space, in practice, the inflow and outflow of the substance in the solid state in the specified space in the mixture state also needs to be kept in agreement with the mass increment of the substance contained in the specified space.
On the basis, in the subsequent model training process, the deviation between the increment of the substance in the solid state in the designated space in unit time and the inflow of the substance in the solid state in the designated space in unit time can be determined as the second deviation according to the prediction data and the physical field data of the mixture state at the first moment. And further training the prediction model with the minimum of the first deviation, the minimum of the second deviation and the minimum of the prediction deviation as optimization targets.
Therefore, the second deviation is minimized in order to ensure that inflow and outflow of the substance in the solid phase state is required to be consistent with the mass increment of the substance contained in the prescribed space.
In the actual training process, the terminal device needs to determine a loss value corresponding to the second deviation, and further trains the prediction model through the loss value corresponding to the second deviation and other deviations (namely, the first deviation and the prediction deviation). The loss value corresponding to the second deviation may be specifically determined by the following formula:
in the middle ofIndicating the mass of the substance in the solid phase state at position i in the designated cell space at the second moment, +.>Representing the total mass of the substance in the solid phase state in the designated space at the first moment +. >Representing the change of the total mass of the substance in the solid phase state in the designated space between the first moment and the second moment, and is determined by the flow rate of the inlet and the outlet.
In addition, since the obtained physical field data are obtained under the set physical conditions, the terminal device may actually input the obtained physical field data and the condition parameters corresponding to the set physical conditions into the prediction model, and the prediction model may predict the physical field data of the mixture state in the specified space under the set physical conditions at the second time according to the physical field data of the mixture state at the first time.
That is, if the acquired physical field data is obtained under a given physical condition, the physical field data predicted by the prediction model will also be obtained under the given physical condition.
The condition parameter corresponding to the above-mentioned set physical condition may be determined by a specific physical condition, for example, if the set physical condition refers to a specific temperature defining a state of a mixture in a specified space, the condition parameter may refer to a specific temperature value defined in the specified space.
Further, in the case that the prediction model includes the first network layer and the second network layer, the terminal device may input the obtained physical field data of the mixture state at the first time and the condition parameters corresponding to the set physical conditions into the first network layer in the prediction model, so as to predict the volume fraction field data of the mixture state in the specified space at the second time.
And then, the volume fraction field data of the mixture state in the predicted specified space at the second moment, the obtained physical field data of the mixture state at the first moment and the condition parameters corresponding to the set physical conditions are input into a second network layer in the prediction model, and the second network layer predicts other physical field data except the volume fraction field data of the mixture state in the specified space at the second moment.
This process can be understood as that the second network layer will predict other physical field data at the second time instant according to each physical field data of the acquired mixture state at the first time instant, with reference to the condition parameters corresponding to the set physical conditions and the predicted volume fraction field data.
After the model training described above is completed, the model training may be applied to the execution of the actual tasks, as will be described in detail below.
Fig. 2 is a flow chart of a task execution method provided in the present specification, including the following steps:
s201: and acquiring data of each physical field of the mixture state in the designated space at the current time.
When executing tasks, it is necessary to acquire the data of each physical field of the mixture state in the designated space at the current time, where the specific task to be executed may depend on the actual requirement, for example, when executing the dust removing task, it is necessary to analyze the data of the physical fields of various object states in the designated space to determine what mode is finally adopted to remove dust from the designated space.
S202: and inputting the physical field data into a pre-trained prediction model, so that the prediction model predicts the physical field data of the mixture state in the specified space at the next moment according to the physical field data.
The terminal device may input the obtained physical field data into the trained prediction model, so as to obtain physical field data of the mixture state in the specified space at the next moment. Various specific implementations of this process and the content of the training phase are described in detail, and therefore are not described in detail herein.
S203: and executing tasks according to the predicted physical field data of the mixture state in the designated space at the next moment.
After predicting each physical field data of the mixture state in the designated space at the next time, a task may be performed.
Therefore, as can be seen from the above description, since the obtained physical field data of the mixture state in the designated space at the first time can be input into the prediction model to train the prediction model, when the trained prediction model is applied to the actual task execution process, compared with the prior art, excessive time is not required to be spent to deduce the physical field data of the mixture state in the designated space at the next time step by step, so that the efficiency of predicting the physical field data is improved by several times, and in the training stage, the quality of the materials in the fluid state in the mixture state before and after the first time and the second time accords with the quality constraint, and the accuracy of the predicted physical field data in the actual application of the prediction model can be ensured.
In addition, the discrete characteristics existing in the multiphase materials and the volume fraction field data of the mixture state are predicted in the prediction process, and then other physical field data are predicted, so that the relevance among the physical field data is fully considered, and the accuracy of model prediction data is ensured.
The foregoing is a method implemented by one or more embodiments of the present disclosure, and based on the same ideas, the present disclosure further provides a corresponding model training device and a task execution device, as shown in fig. 3 and fig. 4.
Fig. 3 is a schematic diagram of a model training apparatus provided in the present specification, including:
an acquisition module 301, configured to acquire physical field data of a mixture state in a designated space at a first moment, where the mixture state includes at least a substance in a fluid state, and the designated space includes an inlet and an outlet, and the mixture state enters the designated space from the inlet and exits the designated space from the outlet;
the prediction module 302 is configured to input the respective physical field data into a prediction model to be trained, so that the prediction model predicts, according to the respective physical field data, respective physical field data of a mixture state in the specified space at a second time, where the second time is a time after the first time, as prediction data;
A determining module 303, configured to determine, as a first deviation, a deviation between an inflow amount of the substance in the fluid state in the specified space in each unit space in the specified space and an increment of a mass of the substance contained in each unit space in the unit space in accordance with the prediction data and the respective physical field data of the mixture state in the specified space at the first time, and determine, as a prediction deviation, a deviation between the prediction data and the respective tag physical field data of the mixture state in the specified space at the second time;
the training module 304 is configured to train the prediction model according to the first deviation and the prediction deviation.
Optionally, the mixture further comprises a substance in a solid phase state;
the determining module 303 is further configured to determine, as a second deviation, a deviation between an increment of the substance in the solid state in the specified space in a unit time and an inflow of the substance in the solid state in the specified space in a unit time, based on the prediction data and each physical field data of the mixture state at the first time, before training the prediction model based on the first mass deviation and the prediction deviation;
The training module 304 is specifically configured to train the prediction model with the first deviation minimized, the second deviation minimized, and the prediction deviation minimized as optimization objectives.
Optionally, the prediction model includes a first network layer and a second network layer;
the prediction module 302 is specifically configured to input each physical field data of the mixture state at a first time into a first network layer in the prediction model, so that the first network layer predicts the volume fraction field data of the mixture state in the specified space at the second time according to each physical field data of the mixture state at the first time; and inputting the volume fraction field data and the physical field data of the mixture state at the first moment into a second network layer in the prediction model, so that the second network layer predicts other physical field data of the mixture state in the appointed space except the volume fraction field data at the second moment according to the physical field data of the mixture state at the first moment and the volume fraction field data.
Optionally, the acquiring module 301 is specifically configured to acquire, under a set physical condition, each physical field data of the mixture state in the specified space at the first time;
The prediction module 302 is specifically configured to input, into the prediction model, each physical field data of the mixture state at the first time and a condition parameter corresponding to the set physical condition, so that the prediction model predicts, as prediction data, each physical field data of the mixture state in the specified space at the second time under the set physical condition according to each physical field data of the mixture state at the first time.
Fig. 4 is a schematic diagram of a task execution device provided in the present specification, including:
an obtaining module 401, configured to obtain data of each physical field of a mixture state in a specified space at a current time, where the mixture state includes at least a substance in a fluid state, and the specified space includes an inlet and an outlet, and the mixture state enters the specified space from the inlet and leaves the specified space from the outlet;
a prediction module 402, configured to input the respective physical field data into a pre-trained prediction model, so that the prediction model predicts, according to the respective physical field data, the respective physical field data of the mixture state in the specified space at a next time, where the prediction model is obtained by training by the model training method;
And the execution module 403 is configured to execute a task according to each predicted physical field data of the mixture state in the specified space at the next moment.
The present specification also provides a computer readable storage medium storing a computer program operable to perform a method of model training as provided in fig. 1 or a method of task execution as provided in fig. 2, as described above.
The present specification also provides a schematic structural diagram of an electronic device corresponding to fig. 1 or fig. 2 shown in fig. 5. At the hardware level, as shown in fig. 5, the electronic device includes a processor, an internal bus, a network interface, a memory, and a nonvolatile storage, and may of course include hardware required by other services. The processor reads the corresponding computer program from the nonvolatile memory into the memory and then runs to implement the model training method described above in fig. 1 or the task execution method described in fig. 2.
Of course, other implementations, such as logic devices or combinations of hardware and software, are not excluded from the present description, that is, the execution subject of the following processing flows is not limited to each logic unit, but may be hardware or logic devices.
In the 90 s of the 20 th century, improvements to one technology could clearly be distinguished as improvements in hardware (e.g., improvements to circuit structures such as diodes, transistors, switches, etc.) or software (improvements to the process flow). However, with the development of technology, many improvements of the current method flows can be regarded as direct improvements of hardware circuit structures. Designers almost always obtain corresponding hardware circuit structures by programming improved method flows into hardware circuits. Therefore, an improvement of a method flow cannot be said to be realized by a hardware entity module. For example, a programmable logic device (Programmable Logic Device, PLD) (e.g., field programmable gate array (Field Programmable Gate Array, FPGA)) is an integrated circuit whose logic function is determined by the programming of the device by a user. A designer programs to "integrate" a digital system onto a PLD without requiring the chip manufacturer to design and fabricate application-specific integrated circuit chips. Moreover, nowadays, instead of manually manufacturing integrated circuit chips, such programming is mostly implemented by using "logic compiler" software, which is similar to the software compiler used in program development and writing, and the original code before the compiling is also written in a specific programming language, which is called hardware description language (Hardware Description Language, HDL), but not just one of the hdds, but a plurality of kinds, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), lava, lola, myHDL, PALASM, RHDL (Ruby Hardware Description Language), etc., VHDL (Very-High-Speed Integrated Circuit Hardware Description Language) and Verilog are currently most commonly used. It will also be apparent to those skilled in the art that a hardware circuit implementing the logic method flow can be readily obtained by merely slightly programming the method flow into an integrated circuit using several of the hardware description languages described above.
The controller may be implemented in any suitable manner, for example, the controller may take the form of, for example, a microprocessor or processor and a computer readable medium storing computer readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), programmable logic controllers, and embedded microcontrollers, examples of which include, but are not limited to, the following microcontrollers: ARC 625D, atmel AT91SAM, microchip PIC18F26K20, and Silicone Labs C8051F320, the memory controller may also be implemented as part of the control logic of the memory. Those skilled in the art will also appreciate that, in addition to implementing the controller in a pure computer readable program code, it is well possible to implement the same functionality by logically programming the method steps such that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers, etc. Such a controller may thus be regarded as a kind of hardware component, and means for performing various functions included therein may also be regarded as structures within the hardware component. Or even means for achieving the various functions may be regarded as either software modules implementing the methods or structures within hardware components.
The system, apparatus, module or unit set forth in the above embodiments may be implemented in particular by a computer chip or entity, or by a product having a certain function. One typical implementation is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.
For convenience of description, the above devices are described as being functionally divided into various units, respectively. Of course, the functions of each element may be implemented in one or more software and/or hardware elements when implemented in the present specification.
It will be appreciated by those skilled in the art that embodiments of the present description may be provided as a method, system, or computer program product. Accordingly, the present specification may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present description can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
The present description is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the specification. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In one typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media.
Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises the element.
It will be appreciated by those skilled in the art that embodiments of the present description may be provided as a method, system, or computer program product. Accordingly, the present specification may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present description can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
The description may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for system embodiments, since they are substantially similar to method embodiments, the description is relatively simple, as relevant to see a section of the description of method embodiments.
The foregoing is merely exemplary of the present disclosure and is not intended to limit the disclosure. Various modifications and alterations to this specification will become apparent to those skilled in the art. Any modifications, equivalent substitutions, improvements, or the like, which are within the spirit and principles of the present description, are intended to be included within the scope of the claims of the present description.
Claims (10)
1. A method of model training, comprising:
acquiring data of each physical field of a mixture state in a designated space at a first moment, wherein the mixture state at least comprises a substance in a fluid state, the designated space comprises an inlet and an outlet, and the mixture state enters the designated space from the inlet and leaves the designated space from the outlet;
inputting the physical field data into a prediction model to be trained, which comprises a first network layer and a second network layer, so that the prediction model predicts the physical field data of the mixture state in the appointed space at a second moment according to the physical field data, wherein the second moment is a moment after the first moment, and the physical field data of the mixture state at the first moment is input into the first network layer in the prediction model, so that the first network layer predicts the volume fraction field data of the mixture state in the appointed space at the second moment according to the physical field data of the mixture state at the first moment, and inputs the volume fraction field data and the physical field data of the mixture state at the first moment into the second network layer in the prediction model, so that the second network layer predicts the volume fraction field data of the mixture state in addition to the volume fraction field data of the mixture state in the appointed space according to the physical field data of the mixture state at the first moment and the volume fraction field data of the mixture state at the second moment;
Determining, as a first deviation, a deviation between an inflow amount of the substance in the fluid state in the specified space per unit space in the specified space and an increment of a mass of the substance contained in the per unit space in the unit time, and determining, as a predicted deviation, a deviation between the predicted data and the tag physical field data of the mixture state in the specified space at the second time, based on the predicted data and the physical field data of the mixture state at the first time;
and training the prediction model according to the first deviation and the prediction deviation.
2. The method of claim 1, further comprising a substance in a solid phase in the mixture;
before training the predictive model based on the first deviation and the predicted deviation, the method further includes:
determining a deviation of an increment of the substance in the solid state in the designated space in unit time and an inflow amount of the substance in the solid state in the designated space in unit time as a second deviation according to the prediction data and the physical field data of the mixture state at the first moment;
Training the prediction model according to the first deviation and the prediction deviation, wherein the training specifically comprises the following steps:
training the predictive model with the optimization objective of minimizing the first deviation, minimizing the second deviation, and minimizing the predictive deviation.
3. The method of claim 1, wherein acquiring each physical field data of the mixture state at the first time in the specified space comprises:
acquiring data of each physical field of the mixture state in the designated space at the first moment under a set physical condition;
inputting the physical field data into a prediction model to be trained, so that the prediction model predicts the physical field data of the mixture state in the specified space at a second moment as prediction data according to the physical field data, wherein the second moment is a moment after the first moment and specifically comprises:
and inputting the physical field data of the mixture state at the first moment and the condition parameters corresponding to the set physical conditions into the prediction model, so that the prediction model predicts the physical field data of the mixture state in the specified space at the second moment under the set physical conditions according to the physical field data of the mixture state at the first moment, and the physical field data are used as prediction data.
4. A method of performing a task, comprising:
acquiring data of each physical field of a mixture state in a designated space at the current time, wherein the mixture state at least comprises a substance in a fluid state, the designated space comprises an inlet and an outlet, and the mixture state enters the designated space from the inlet and leaves the designated space from the outlet;
inputting the physical field data into a pre-trained prediction model, so that the prediction model predicts the physical field data of the mixture state in the specified space at the next moment according to the physical field data, wherein the prediction model is obtained by training the method according to any one of claims 1-3;
and executing tasks according to the predicted physical field data of the mixture state in the designated space at the next moment.
5. An apparatus for model training, comprising:
the system comprises an acquisition module, a control module and a control module, wherein the acquisition module is used for acquiring each physical field data of a mixture state in a designated space at a first moment, the mixture state at least comprises a substance in a fluid state, the designated space comprises an inlet and an outlet, and the mixture state enters the designated space from the inlet and leaves the designated space from the outlet;
The prediction module is used for inputting the physical field data into a prediction model to be trained, which comprises a first network layer and a second network layer, so that the prediction model predicts the physical field data of the mixture state in the specified space at a second moment according to the physical field data, the second moment is the moment after the first moment, the physical field data of the mixture state at the first moment is input into the first network layer in the prediction model, the first network layer predicts the volume fraction field data of the mixture state in the specified space at the second moment according to the physical field data of the mixture state at the first moment, and inputs the volume fraction field data and the physical field data of the mixture state in the first moment into the second network layer in the prediction model, so that the second network layer predicts the volume fraction field data and the physical field data of the mixture state in the specified space according to the volume fraction field data of the mixture state in the first moment and the physical field data of the mixture state in addition to the specified volume fraction data of the mixture state in the specified space;
A determining module, configured to determine, as a first deviation, a deviation between an inflow amount of the substance in the fluid state in the specified space in each unit space in the specified space and an increment of a mass of the substance contained in each unit space in the unit space in accordance with the prediction data and the respective physical field data of the mixture state in the specified space at the first time, and determine, as a prediction deviation, a deviation between the prediction data and the respective tag physical field data of the mixture state in the specified space at the second time;
and the training module is used for training the prediction model according to the first deviation and the prediction deviation.
6. The device of claim 5, wherein the mixture further comprises a substance in a solid phase;
the determining module is further configured to determine, as a second deviation, a deviation between an increment of the substance in the solid state in the specified space in a unit time and an inflow of the substance in the solid state in the specified space in a unit time, based on the prediction data and each physical field data of the mixture state at the first time, before training the prediction model based on the first deviation and the prediction deviation;
The training module is specifically configured to train the prediction model with the first deviation minimized, the second deviation minimized, and the prediction deviation minimized as optimization objectives.
7. The apparatus of claim 5, wherein the acquisition module is specifically configured to acquire, under a set physical condition, each physical field data of the mixture state in the specified space at the first time;
the prediction module is specifically configured to input each physical field data of the mixture state at a first time and a condition parameter corresponding to the set physical condition into the prediction model, so that the prediction model predicts each physical field data of the mixture state in the specified space at the set physical condition at the second time as prediction data according to each physical field data of the mixture state at the first time.
8. A task execution device, characterized by comprising:
the system comprises an acquisition module, a control module and a control module, wherein the acquisition module is used for acquiring each physical field data of a mixture state in a designated space at the current moment, the mixture state at least comprises a substance in a fluid state, the designated space comprises an inlet and an outlet, and the mixture state enters the designated space from the inlet and leaves the designated space from the outlet;
The prediction module is used for inputting the physical field data into a pre-trained prediction model so that the prediction model predicts the physical field data of the mixture state in the specified space at the next moment according to the physical field data, and the prediction model is obtained by training the method according to any one of the claims 1-3;
and the execution module is used for executing tasks according to the predicted physical field data of the mixture state in the specified space at the next moment.
9. A computer readable storage medium, characterized in that the storage medium stores a computer program which, when executed by a processor, implements the method of any of the preceding claims 1-4.
10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method of any of the preceding claims 1-4 when executing the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311080508.2A CN116777010B (en) | 2023-08-25 | 2023-08-25 | Model training method and task execution method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311080508.2A CN116777010B (en) | 2023-08-25 | 2023-08-25 | Model training method and task execution method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116777010A CN116777010A (en) | 2023-09-19 |
| CN116777010B true CN116777010B (en) | 2023-12-19 |
Family
ID=88013796
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311080508.2A Active CN116777010B (en) | 2023-08-25 | 2023-08-25 | Model training method and task execution method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116777010B (en) |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| RU2012124698A (en) * | 2012-06-14 | 2013-12-20 | Федеральное государственное бюджетное образовательное учреждение высшего профессионального образования "Казанский национальный исследовательский технический университет им. А.Н. Туполева - КАИ" (КНИТУ-КАИ) | METHOD FOR MEASURING PHYSICAL FIELD PARAMETERS AND A DEVICE FOR ITS IMPLEMENTATION |
| CN112052617A (en) * | 2020-09-11 | 2020-12-08 | 西安交通大学 | Method and system for predicting branch blood vessel flow field for non-disease diagnosis |
| CN114282448A (en) * | 2021-11-16 | 2022-04-05 | 北京百度网讯科技有限公司 | Flow field information acquisition method, model training method and device and electronic equipment |
| CN114329319A (en) * | 2021-12-27 | 2022-04-12 | 北京航空航天大学 | Stream thermosetting coupling calculation method based on physical neural network |
| CN114970338A (en) * | 2022-05-19 | 2022-08-30 | 北京百度网讯科技有限公司 | Vortex-induced vibration model training method, vortex-induced vibration prediction method and device |
| CN115270593A (en) * | 2022-05-19 | 2022-11-01 | 东南大学 | Multi-physical-field model separation type solving method based on deep learning |
| CN115392077A (en) * | 2022-08-16 | 2022-11-25 | 中国人民解放军军事科学院国防科技创新研究院 | Satellite physical field level digital twin model construction method based on deep learning |
| CN115879335A (en) * | 2022-01-12 | 2023-03-31 | 西南交通大学 | Fluid multi-physical-field parameter prediction method based on graph-generated neural network |
| CN116127844A (en) * | 2023-02-08 | 2023-05-16 | 大连海事大学 | Flow field time interval deep learning prediction method considering flow control equation constraint |
| CN116341399A (en) * | 2021-12-14 | 2023-06-27 | 中核武汉核电运行技术股份有限公司 | Thermodynamic hydraulic heat exchange coefficient prediction method based on physical constraint neural network |
| CN116362126A (en) * | 2023-03-30 | 2023-06-30 | 中国长江电力股份有限公司 | Model fusion-based three-dimensional physical field real-time simulation method |
| CN116432538A (en) * | 2023-04-24 | 2023-07-14 | 广东工业大学 | Multi-physical-field simulation method and system based on residual dense network |
| CN116486211A (en) * | 2023-05-08 | 2023-07-25 | 中山大学 | Model training method, fractional flow reserve calculation method, device and equipment |
| CN116484744A (en) * | 2023-05-12 | 2023-07-25 | 北京百度网讯科技有限公司 | Object simulation method, model training method, device, equipment and storage medium |
-
2023
- 2023-08-25 CN CN202311080508.2A patent/CN116777010B/en active Active
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| RU2012124698A (en) * | 2012-06-14 | 2013-12-20 | Федеральное государственное бюджетное образовательное учреждение высшего профессионального образования "Казанский национальный исследовательский технический университет им. А.Н. Туполева - КАИ" (КНИТУ-КАИ) | METHOD FOR MEASURING PHYSICAL FIELD PARAMETERS AND A DEVICE FOR ITS IMPLEMENTATION |
| CN112052617A (en) * | 2020-09-11 | 2020-12-08 | 西安交通大学 | Method and system for predicting branch blood vessel flow field for non-disease diagnosis |
| CN114282448A (en) * | 2021-11-16 | 2022-04-05 | 北京百度网讯科技有限公司 | Flow field information acquisition method, model training method and device and electronic equipment |
| CN116341399A (en) * | 2021-12-14 | 2023-06-27 | 中核武汉核电运行技术股份有限公司 | Thermodynamic hydraulic heat exchange coefficient prediction method based on physical constraint neural network |
| CN114329319A (en) * | 2021-12-27 | 2022-04-12 | 北京航空航天大学 | Stream thermosetting coupling calculation method based on physical neural network |
| CN115879335A (en) * | 2022-01-12 | 2023-03-31 | 西南交通大学 | Fluid multi-physical-field parameter prediction method based on graph-generated neural network |
| CN115270593A (en) * | 2022-05-19 | 2022-11-01 | 东南大学 | Multi-physical-field model separation type solving method based on deep learning |
| CN114970338A (en) * | 2022-05-19 | 2022-08-30 | 北京百度网讯科技有限公司 | Vortex-induced vibration model training method, vortex-induced vibration prediction method and device |
| CN115392077A (en) * | 2022-08-16 | 2022-11-25 | 中国人民解放军军事科学院国防科技创新研究院 | Satellite physical field level digital twin model construction method based on deep learning |
| CN116127844A (en) * | 2023-02-08 | 2023-05-16 | 大连海事大学 | Flow field time interval deep learning prediction method considering flow control equation constraint |
| CN116362126A (en) * | 2023-03-30 | 2023-06-30 | 中国长江电力股份有限公司 | Model fusion-based three-dimensional physical field real-time simulation method |
| CN116432538A (en) * | 2023-04-24 | 2023-07-14 | 广东工业大学 | Multi-physical-field simulation method and system based on residual dense network |
| CN116486211A (en) * | 2023-05-08 | 2023-07-25 | 中山大学 | Model training method, fractional flow reserve calculation method, device and equipment |
| CN116484744A (en) * | 2023-05-12 | 2023-07-25 | 北京百度网讯科技有限公司 | Object simulation method, model training method, device, equipment and storage medium |
Non-Patent Citations (3)
| Title |
|---|
| Multi-scale numerical simulation of fluidized beds:Model applicability assessment;Shuai Wang et al.;《Particuology 80》;11-41 * |
| 基于B样条神经网络的熔铸装药温度场预测;陶磊 等;《兵工学报》;第44卷(第5期);1339-1349 * |
| 流化床内流动、混合与反应的多尺度模拟研究;胡陈枢;《中国博士学位论文全文数据库》;1-302 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116777010A (en) | 2023-09-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110245047B (en) | Time sequence abnormality detection method, device and equipment | |
| CN117390585B (en) | Time sequence data prediction method and model training method based on three-dimensional full-connection fusion | |
| CN110263050A (en) | Data processing method, device, equipment and storage medium | |
| CN116776135B (en) | Physical field data prediction method and device based on neural network model | |
| CN117370536B (en) | Task execution method and device, storage medium and electronic equipment | |
| CN116777010B (en) | Model training method and task execution method and device | |
| CN113902356B (en) | Regional flow data analysis method and device | |
| CN108921375A (en) | A kind of data processing method and device | |
| CN112787664B (en) | ASIC automatic frequency division method and device based on multiple clock sources | |
| CN110516814A (en) | A kind of business model parameter value determines method, apparatus, equipment and medium | |
| CN109325127A (en) | A kind of Risk Identification Method and device | |
| CN116760871B (en) | Intelligent table management system based on multi-protocol cooperation | |
| CN116661574B (en) | Method and device for radiating of computing equipment, storage medium and electronic equipment | |
| CN116384515B (en) | Model training method and device, storage medium and electronic equipment | |
| CN117393140B (en) | Intelligent finger ring control method and device based on historical data | |
| CN116880905B (en) | Data storage method and device, storage medium and electronic equipment | |
| CN115827880B (en) | Business execution method and device based on emotion classification | |
| CN117873789B (en) | Checkpoint writing method and device based on segmentation quantization | |
| CN116822138B (en) | Method and device for calculating conductivity of gunpowder gas and storage medium | |
| CN115983398A (en) | Business model adjusting method, device and equipment | |
| CN118095907A (en) | Energy consumption monitoring and predicting method and device, storage medium and electronic equipment | |
| CN117891600A (en) | Task scheduling method and device, storage medium and electronic equipment | |
| CN118131159A (en) | Asynchronous track association method, device and system based on pseudo nearest neighbor distance | |
| CN116151466A (en) | Training method, device, medium and equipment for enterprise risk early warning model | |
| CN117744731A (en) | Model training method, device, medium and equipment based on resistive random access memory |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |