CN115796011A - Hydrogen storage bed heat transfer performance optimization method based on neural network and genetic algorithm - Google Patents
Hydrogen storage bed heat transfer performance optimization method based on neural network and genetic algorithm Download PDFInfo
- Publication number
- CN115796011A CN115796011A CN202211436719.0A CN202211436719A CN115796011A CN 115796011 A CN115796011 A CN 115796011A CN 202211436719 A CN202211436719 A CN 202211436719A CN 115796011 A CN115796011 A CN 115796011A
- Authority
- CN
- China
- Prior art keywords
- neural network
- hydrogen storage
- storage bed
- genetic algorithm
- heat transfer
- 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.)
- Pending
Links
- UFHFLCQGNIYNRP-UHFFFAOYSA-N Hydrogen Chemical compound [H][H] UFHFLCQGNIYNRP-UHFFFAOYSA-N 0.000 title claims abstract description 64
- 229910052739 hydrogen Inorganic materials 0.000 title claims abstract description 64
- 239000001257 hydrogen Substances 0.000 title claims abstract description 64
- 238000000034 method Methods 0.000 title claims abstract description 56
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 47
- 238000012546 transfer Methods 0.000 title claims abstract description 24
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 23
- 230000002068 genetic effect Effects 0.000 title claims abstract description 23
- 238000005457 optimization Methods 0.000 title claims abstract description 17
- 238000012360 testing method Methods 0.000 claims abstract description 30
- 230000008569 process Effects 0.000 claims abstract description 16
- 238000012549 training Methods 0.000 claims abstract description 14
- 238000004364 calculation method Methods 0.000 claims abstract description 12
- 230000006870 function Effects 0.000 claims description 34
- 238000001816 cooling Methods 0.000 claims description 30
- 210000002569 neuron Anatomy 0.000 claims description 24
- 238000003062 neural network model Methods 0.000 claims description 18
- 238000013461 design Methods 0.000 claims description 15
- 238000004088 simulation Methods 0.000 claims description 10
- 238000013507 mapping Methods 0.000 claims description 7
- 230000004913 activation Effects 0.000 claims description 3
- 239000002826 coolant Substances 0.000 claims description 3
- 210000004205 output neuron Anatomy 0.000 claims description 3
- 238000002945 steepest descent method Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 2
- 238000010187 selection method Methods 0.000 claims description 2
- 238000010521 absorption reaction Methods 0.000 abstract description 2
- 238000011156 evaluation Methods 0.000 abstract 1
- 238000002474 experimental method Methods 0.000 abstract 1
- 238000005070 sampling Methods 0.000 abstract 1
- 230000000694 effects Effects 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000035772 mutation Effects 0.000 description 2
- 238000003795 desorption Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 229910052987 metal hydride Inorganic materials 0.000 description 1
- 150000004681 metal hydrides Chemical class 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a hydrogen storage bed heat transfer performance optimization method based on a neural network and a genetic algorithm, which adopts an orthogonal method to carry out numerical experiment sampling, utilizes COMSOL software to simulate a hydrogen absorption process model, and identifies the change of the temperature and the time of a hydrogen storage bed under different conditions; establishing a hybrid method based on the combination of a BP neural network and a genetic algorithm, firstly establishing a hydrogen storage bed temperature prediction model, training and testing the hydrogen storage bed temperature prediction model based on the BP neural network, and adjusting the weight and the threshold of the hydrogen storage bed temperature prediction model according to a test error; and then constructing a multi-objective genetic algorithm optimization model, taking the hydrogen storage bed prediction model as a fitness function, optimizing the input data through the genetic algorithm model, and finding out the optimal input parameter combination to obtain the optimal heat transfer performance. The invention can save the test cost, has simple calculation and provides guidance for the prediction, evaluation, model selection and operation of the hydrogen storage performance of the similar hydrogen storage bed.
Description
Technical Field
The invention relates to the technical field of design and operation of hydrogen storage beds, in particular to a method for optimizing the heat transfer performance of a hydrogen storage bed based on a neural network and a genetic algorithm.
Background
Many researchers have investigated the effect of only one parameter on hydrogen storage performance while the other parameters remain unchanged. The optimization method has the advantages of large workload and complex calculation process, and ignores the influence of interaction among various parameters on a calculation result. It reflects only a single influence of each parameter on the hydrogen storage device performance, and thus offers limited recommendations for hydrogen storage device design, and few researchers have systematically studied the influence of parameter and configuration design on the hydrogen storage performance of metal hydrides to obtain the best combination of different parameters. Therefore, in order to obtain better hydrogen storage performance, it is necessary to systematically optimize the performance affecting parameters.
The neural network can process the complex nonlinear problems of complex environmental information, unknown knowledge and undefined rules, the neural network is a genetic algorithm of a fitness function, has the capability of parallelization and global search, can avoid the phenomenon that the traditional single-point search method is easy to fall into local optimal distribution when searching a multi-modal parameter space, and has strong expandability.
Disclosure of Invention
The invention aims to provide a hydrogen storage bed heat transfer performance optimization method based on a neural network and a genetic algorithm, and aims to provide an optimal design parameter of a thin-wall double-layer annular hydrogen storage bed, which achieves the optimal heat transfer effect within a certain parameter range. The technical scheme adopted by the application is as follows:
a hydrogen storage bed heat transfer performance optimization method based on a neural network and a genetic algorithm comprises the following steps:
step 1: numerical modeling;
performing numerical simulation calculation on the hydrogen storage bed by using comsol software and combining a control equation according to the parameters, and verifying the numerical simulation calculation;
specifically, a geometric model is established according to solidworks, and comsol software is used for solving by combining a control equation.
Step 2: obtaining a sample;
the quality of the data determines the performance of the neural network; considering a large amount of data, constructing a training sample of the neural network by using an orthogonal method; geometric design parameters and operation parameters influencing the heat transfer performance of the hydrogen storage bed are taken as optimization parameters, and the parameters comprise the number (N) of cooling pipes, the distance (Pt, mm) between hydrogen storage layers, the diameter (D, mm) of the cooling pipes, the wall thickness (b, mm) of the cooling pipes, the flow rate (v, m/s) of cooling media, the thickness (t, mm) of the hydrogen storage layers and the Al doping rate (D,%); selecting a proper test scheme design by utilizing Minitab software according to an orthogonal method, generating an orthogonal test table with 36 test schemes, modifying relevant parameters according to the numerical model in the step 1 according to the orthogonal test table, carrying out numerical calculation to obtain the hot spot temperature and the cooling time of the hydrogen storage bed as output quantities, and determining a group of mapping sets which are input and corresponding to a group of outputs as test samples;
and step 3: establishing a BP neural network;
3.1 dividing the sample set in the step 2 into a training set and a testing set; establishing a neural network structure by the mapping relation of input and output of a training set, wherein the neural network structure comprises an input layer, a hidden layer and an output layer, the number of neurons of the input layer is determined by the number of input parameters, the number of neurons of the output layer is determined by the number of output quantities, and the number of layers of the hidden layer and the number of neurons are determined by a trial-and-error method; the BP neural network is a multi-layer feedforward network; the learning process of the method uses a steepest descent method, the weight and the threshold value of the network are continuously adjusted through back propagation, the sum of square errors of the network is minimized, and the process is continuously iterated to achieve convergence and complete the training process of the neural network; wherein the calculated output of each neuron is an activation function value having a parameter of the sum of the inputs multiplied by a weight, expressed as follows:
where y is an output variable, x i Is an input variable, w ij Is the connection weight, v, of input layer neuron i and intermediate layer neuron j j Is the connection weight, θ, of the intermediate layer neuron j and the output neuron j And gamma is the middle and output layer nerves, respectivelyA threshold for the element;
for a neural network with k as the total number of outputs, the Mean Square Error (MSE) function should be minimized, and the MSE function is calculated as follows:
where p is the number of output nodes, k is the number of training samples, S k For desired output value of network, O k Actual output values for the network;
3.2 after the neural network model is trained, testing the neural network model according to samples in the test set, and finally determining that the error between the output value and the target value meets the regulation, completing construction of the BP neural network model for predicting the hot spot temperature and the cooling time of the hydrogen storage bed;
and 4, step 4: based on the constructed BP neural network model, optimizing a multi-target solution of the established neural network model by using a parallel selection genetic algorithm;
4.1 using the constructed BP neural network model output as a fitness function, determining the temperature of the hot spot of the hydrogen storage bed and the cooling time as optimization targets, minimizing the target function by 2 targets, using the value range of the input parameters as constraint conditions, and expressing the optimization model as follows:
where V-min represents the minimization of the vector, i.e., each sub-targeting function fi (x) in the vector target is minimized to the maximum extent;
4.2 determining an initial population according to the parameter range, wherein each individual in the population represents a scheme configuration, and calling a neural network to calculate a fitness value;
4.3 adopting a parallel selection method, firstly dividing all individuals in the initial group into equal subgroups according to the number of the sub-objective functions; each subgroup corresponds to one sub-target function, and each sub-target function independently performs selection operation in the corresponding subgroup; then, selecting individuals with strong adaptability to form a new subgroup, and combining all the new subgroups into a complete subgroup; performing cross and variation operations on the group to generate a next generation complete population, and continuously performing operations of 'segmentation-collocation-selection-combination'; carrying out evolutionary search based on a fitness function process to finally obtain a pareto optimal solution;
compared with the prior art, the invention has the following advantages:
the invention provides a hydrogen storage bed heat transfer performance optimization method based on a neural network and a genetic algorithm, which can be used for quickly and effectively evaluating the heat transfer performance of a hydrogen storage bed by combining a numerical simulation method, an orthogonal method, a neural network model and a genetic algorithm to obtain the optimal configuration of the geometric parameters and the operating parameters of the hydrogen storage bed. Therefore, the invention can predict and optimize the heat transfer performance of the hydrogen storage bed, regulate and control the factors influencing the heat transfer performance of the hydrogen storage bed and improve the hydrogen absorption and desorption efficiency of the hydrogen storage system.
Drawings
FIG. 1 is an inventive flow chart;
FIG. 2 is a cross-sectional view of a double-walled hydrogen storage bed of an example;
FIG. 3 is a schematic view of a circular cooling tube embedded in a tank wall of an embodiment;
FIG. 4 is a diagram of a neural network architecture of an embodiment;
FIG. 5 is an embodiment of hot spot temperature optimization and performance tracking;
FIG. 6 is a cooling time optimum and performance trace for an embodiment;
FIG. 7 shows the variation of parameters in the convergence iteration process of the embodiment.
Detailed Description
The technical scheme of the invention is further explained in detail by combining the attached drawings and the implementation case.
The invention relates to a method for optimizing the heat transfer performance of a hydrogen storage bed based on a neural network and a genetic algorithm, which comprises the steps of firstly establishing a neural network model taking the number of cooling pipe sections, the distance between hydrogen storage layers, the diameter of a cooling pipe, the allowance of the cooling pipe, the flow rate of a cooling medium, the thickness of the hydrogen storage layer and the Al doping rate as input parameters, outputting hot spot temperature and cooling time of the hydrogen storage bed as output parameters, then selecting a proper test scheme design according to an orthogonal method by utilizing Minitab software, generating 36 orthogonal test tables of the test schemes, obtaining the output value of each test scheme by a comsol calculation numerical model, obtaining a data set of input and output mapping, training and testing the neural network model according to the data set, and obtaining the neural network model with superior prediction capability; all individuals in the initial group are divided into equal sub-groups according to the number of sub-objective functions. Each subgroup corresponds to one sub-target function, each sub-target function independently performs selection operation in the corresponding subgroup, trained BP neural network simulation results are used as fitness functions, then individuals with high adaptability are selected to form a new subgroup, and all the new subgroups are combined into a complete group. The group is interleaved and mutated to generate the next full population, and "split-concatenate-select-merge" operations are performed continuously. In the process, the optimal solution and the population number are searched from the ObjV matrix by using a minimum function until convergence, and finally the pareto optimal solution and the corresponding parameter scheme are obtained.
A method for optimizing the heat transfer performance of a hydrogen storage bed based on a neural network and a genetic algorithm is shown in figure 1 and comprises the following specific steps:
the cross section of the double-layer thin-wall hydrogen storage bed in the step 1 is shown in figure 2, the selected geometric parameters are the number of cooling pipe sections, the distance between hydrogen storage layers, the diameter of the cooling pipe, the distance from the wall of the cooling pipe to the edge of the bed wall and the thickness of the hydrogen storage layer, and other influencing parameters are the flow rate of a cooling medium and the Al doping rate of the hydrogen storage layer, wherein the circular cooling pipe embedded in the wall of the tank is arranged as shown in figure 3, and the average bed area of each cooling pipe can be obtained by the following formula under the assumption that the number of the cooling pipes in the inner ring of the hydrogen storage bed is n 1:
the area of the cooling pipe corresponding to the outer ring of the hydrogen storage bed is kept equal to that of the inner ring, and the number of the cooling pipes on the outer ring is as follows:
thus, the total number of cooling tubes in the arrangement is N = N1+ N2.
And (3) determining the horizontal number of each factor according to the seven factors selected in the step (1) as the design factors of the orthogonal design test, inputting the horizontal number into Minitab software, and designing 36 test schemes according to the orthogonal method.
The value ranges and levels of each factor are shown in table 1;
TABLE 1 influence factors and levels of orthogonal array design parameters based on the orthogonal method
And performing numerical simulation calculation on the hydrogen storage bed by using comsol software and combining a control equation according to the parameters to obtain the hot spot temperature and the cooling time of the optimized target hydrogen storage bed.
Taking the test data with the mapping relation as a sample set, wherein the specific test scheme and the numerical simulation result are shown in table 2;
table 2 set of 36 experimental protocols and results of numerical simulation design
And 2, establishing a neural network structure by the one-to-one mapping relation of the input set and the output set of the sample set, wherein the neural network structure comprises an input layer, a hidden layer and an output layer, the number of neurons in the input layer is 7, the number of neurons in the output layer is 2, the number of proper hidden layers and the number of neurons are determined by a trial and error method, and the regression coefficients R of neural network models of different hidden layers and neurons are calculated for comparison. According to empirical formula
Where m is the number of hidden layer nodes, n is the number of input layer nodes, l is the number of output layer nodes and is a constant between 1 and 10, the computation range of the hidden layer nodes is between 3 and 13. Finally determining the neural network result of 7-5-2-2, see figure 4;
the neural network learning process uses a steepest descent method, the weight and the threshold value of the network are continuously adjusted through back propagation, the sum of square errors of the network is minimized, and the process is continuously iterated to achieve convergence and complete the training process of the neural network.
The calculation output of each neuron in the neural network is an activation function value taking the sum of the inputs multiplied by the weight as a parameter, and is specifically represented as follows:
where y is an output variable, x i Is an input variable, w ij Is the connection weight, v, of input layer neuron i and intermediate layer neuron j j Is the connection weight, θ, of the intermediate layer neuron j and the output neuron j And γ are the thresholds for the middle and output layer neurons, respectively. For a neural network with k as the total number of outputs, the Mean Square Error (MSE) function should be minimized, and the MSE function is calculated as follows:
where p is the number of output nodes, k is the number of training samples, S k For desired output value of network, O k The actual output value of the network.
all individuals in the initial group are first divided into equal sub-groups according to the number of sub-objective functions. Each subgroup corresponds to one sub-target function, and each sub-target function independently performs selection operation in the corresponding subgroup;
using the simulation result of the trained BP neural network as a fitness function, adopting a roulette method to select, wherein the probability that the ith individual is selected is P i Then, there are:
wherein n is the number of population individuals, F i Is the i individual fitness value.
Then, the adaptable individuals are selected to form a new subgroup, and all the new subgroups are combined into a complete group. Performing crossing and mutation operations on the group to generate a next generation complete population, wherein the crossing rate is 0.7, the mutation rate is 1 percent,
the operation of "split-concatenate-select-merge" is performed.
The output result is two-dimensional, and the corresponding dimension is the corresponding sub-targeting function. And searching the optimal solution and the population number from the ObjV matrix by using a minimum function, and repeating iteration until convergence is stable to finally obtain the pareto optimal solution. The hot spot temperature and cooling time optima and performance tracking are shown in fig. 5 and 6, respectively, and the variation of each parameter during the convergence iteration is shown in fig. 7.
Claims (9)
1. A hydrogen storage bed heat transfer performance optimization method based on a neural network and a genetic algorithm is characterized by comprising the following steps:
step 1: numerical modeling;
performing numerical simulation calculation on the hydrogen storage bed according to the parameters and the control equation, and verifying the numerical simulation calculation;
step 2: obtaining a sample;
selecting a proper test scheme design by taking geometric design parameters and operation parameters influencing the heat transfer performance of the hydrogen storage bed as optimization parameters, generating an orthogonal test table, modifying relevant parameters according to the numerical model in the step 1 according to the orthogonal test table to carry out numerical calculation, obtaining the hot spot temperature and the cooling time of the hydrogen storage bed as output quantities, and determining a group of mapping sets input to and corresponding to a group of outputs as test samples;
and step 3: establishing a BP neural network;
3.1 dividing the sample set in the step 2 into a training set and a testing set; establishing a neural network structure by the mapping relation of input and output of a training set, wherein the neural network structure comprises an input layer, a hidden layer and an output layer, the number of neurons of the input layer is determined by the number of input parameters, the number of neurons of the output layer is determined by the number of output quantities, and the number of the proper hidden layer and the number of the neurons are determined by a trial-and-error method; the BP neural network is a multi-layer feedforward network; the learning process of the method uses a steepest descent method, the weight and the threshold value of the network are continuously adjusted through back propagation, the sum of square errors of the network is minimized, and the process is continuously iterated to achieve convergence and complete the training process of the neural network;
3.2 after the neural network model is trained, testing the neural network model according to samples in the test set, and finally determining that the error between the output value and the target value meets the regulation, completing construction of the BP neural network model for predicting the hot spot temperature and the cooling time of the hydrogen storage bed;
and 4, step 4: and based on the constructed BP neural network model, optimizing the multi-target solution of the established neural network model by using a parallel selection genetic algorithm.
2. The method for optimizing the heat transfer performance of a hydrogen storage bed based on a neural network and a genetic algorithm as claimed in claim 1, wherein: numerical modeling is carried out in the step 1, a geometric model is built according to solidworks, and comsol software is used for solving by combining a control equation.
3. The method for optimizing the heat transfer performance of a hydrogen storage bed based on a neural network and a genetic algorithm as claimed in claim 2, wherein: in the step 2, geometric design parameters and operation parameters influencing the heat transfer performance of the hydrogen storage bed are taken as optimization parameters, including the number of cooling pipes and N; hydrogen storage interlamellar spacing, pt, mm; diameter of cooling tube, d, mm; wall thickness of cooling tube, b, mm; cooling medium flow rate, v, m/s; thickness of hydrogen storage layer, t, mm; al doping rate, D,%.
4. The method for optimizing the heat transfer performance of a hydrogen storage bed based on a neural network and a genetic algorithm as claimed in claim 3, wherein: and 2, constructing a training sample of the neural network by using an orthogonal method, selecting a proper test scheme design by using Minitab software according to the orthogonal method, and generating an orthogonal test table with 36 test schemes.
5. The method for optimizing the heat transfer performance of a hydrogen storage bed based on a neural network and a genetic algorithm as claimed in claim 1, wherein: the calculated output of each neuron in step 3 is an activation function value with the parameter of the sum of the inputs multiplied by the weight, and is expressed as follows:
where y is an output variable, x i Is an input variable, w ij Is the connection weight, v, of input layer neuron i and intermediate layer neuron j j Is the connection weight, θ, of the intermediate layer neuron j and the output neuron j And γ are thresholds for middle and output layer neurons, respectively.
6. The method for optimizing the heat transfer performance of a hydrogen storage bed based on a neural network and a genetic algorithm as claimed in claim 5, wherein: in step 3, for the neural network taking k as the total output number, the mean square error MSE function should be minimized, and the mean square error is calculated as follows:
where p is the number of output nodes, k is the number of training samples, S k For desired output value of network, O k The actual output value of the network.
7. The method for optimizing the heat transfer performance of a hydrogen storage bed based on a neural network and a genetic algorithm as claimed in claim 1, wherein: step 4 comprises the following substeps:
4.1, outputting the constructed BP neural network model as a fitness function, determining the temperature of a hot spot of a hydrogen storage bed and the cooling time as optimization targets, minimizing 2 targets as an objective function, and taking the value range of input parameters as constraint conditions;
4.2 determining an initial population according to the parameter range, wherein each individual in the population represents a scheme configuration, and calling a neural network to calculate a fitness value;
4.3, finally obtaining the pareto optimal solution by adopting a parallel selection method.
8. The method of claim 7, wherein the method comprises the following steps: the optimization model in substep 4.1 is represented as:
where V-min represents the minimization of the vector, i.e., each sub-targeting function fi (x) in the vector target is minimized to the maximum extent.
9. The method of claim 7, wherein the method comprises the following steps: the method of substep 4.3 is to divide all individuals in the initial group into equal subgroups according to the number of sub-objective functions; each subgroup corresponds to one sub-target function, and each sub-target function independently performs selection operation in the corresponding subgroup; then, selecting individuals with strong adaptability to form a new subgroup, and combining all the new subgroups into a complete group; performing cross and variation operations on the group to generate a next generation complete population, and continuously performing operations of 'segmentation-collocation-selection-combination'; and (4) carrying out evolutionary search based on a fitness function process to finally obtain a pareto optimal solution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211436719.0A CN115796011A (en) | 2022-11-16 | 2022-11-16 | Hydrogen storage bed heat transfer performance optimization method based on neural network and genetic algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211436719.0A CN115796011A (en) | 2022-11-16 | 2022-11-16 | Hydrogen storage bed heat transfer performance optimization method based on neural network and genetic algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115796011A true CN115796011A (en) | 2023-03-14 |
Family
ID=85438279
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211436719.0A Pending CN115796011A (en) | 2022-11-16 | 2022-11-16 | Hydrogen storage bed heat transfer performance optimization method based on neural network and genetic algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115796011A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116595885A (en) * | 2023-05-29 | 2023-08-15 | 上海交通大学 | Lightweight shielding deep coupling optimization method for small reactor |
| CN117519054A (en) * | 2023-12-11 | 2024-02-06 | 广州智业节能科技有限公司 | High-efficient cold station control system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102072272A (en) * | 2011-01-14 | 2011-05-25 | 四川大学 | Corner-free rubber shock absorber for precision instrument |
| CN102507221A (en) * | 2011-10-12 | 2012-06-20 | 北京交通大学 | Method for forecasting comfortableness of train body |
| CN104597842A (en) * | 2015-02-02 | 2015-05-06 | 武汉理工大学 | BP neutral network heavy machine tool thermal error modeling method optimized through genetic algorithm |
| CN105160201A (en) * | 2015-09-30 | 2015-12-16 | 成都民航空管科技发展有限公司 | Genetic algorithm back propagation (GABP) neural network based controller workload prediction method and system |
| CN109034388A (en) * | 2018-07-27 | 2018-12-18 | 湖北中烟工业有限责任公司 | A kind of prediction model of cigarette material and mainstream smoke constituents based on Genetic Algorithm Optimized Neural Network |
| CN114004341A (en) * | 2021-11-15 | 2022-02-01 | 威海长和光导科技有限公司 | Optical fiber preform preparation process optimization method based on genetic algorithm and BP neural network |
| US20220129757A1 (en) * | 2020-10-26 | 2022-04-28 | Jiangxi University Of Science And Technology | Prediction method, device and system for rock mass instability stages |
-
2022
- 2022-11-16 CN CN202211436719.0A patent/CN115796011A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102072272A (en) * | 2011-01-14 | 2011-05-25 | 四川大学 | Corner-free rubber shock absorber for precision instrument |
| CN102507221A (en) * | 2011-10-12 | 2012-06-20 | 北京交通大学 | Method for forecasting comfortableness of train body |
| CN104597842A (en) * | 2015-02-02 | 2015-05-06 | 武汉理工大学 | BP neutral network heavy machine tool thermal error modeling method optimized through genetic algorithm |
| CN105160201A (en) * | 2015-09-30 | 2015-12-16 | 成都民航空管科技发展有限公司 | Genetic algorithm back propagation (GABP) neural network based controller workload prediction method and system |
| CN109034388A (en) * | 2018-07-27 | 2018-12-18 | 湖北中烟工业有限责任公司 | A kind of prediction model of cigarette material and mainstream smoke constituents based on Genetic Algorithm Optimized Neural Network |
| US20220129757A1 (en) * | 2020-10-26 | 2022-04-28 | Jiangxi University Of Science And Technology | Prediction method, device and system for rock mass instability stages |
| CN114004341A (en) * | 2021-11-15 | 2022-02-01 | 威海长和光导科技有限公司 | Optical fiber preform preparation process optimization method based on genetic algorithm and BP neural network |
Non-Patent Citations (3)
| Title |
|---|
| PING ZHAO, XIANGGUO ZENG, WEI LI, HAN ZHAO, AND FANG WANG: "System optimization of heat transfer performance of hydrogen storage bed based on backpropagation neural network-genetic algorithm", 《ENERGY SOURCES, PART A: RECOVERY, UTILIZATION, AND ENVIRONMENTAL EFFECTS》, pages 1 - 20 * |
| 杨增宇;陈华燕;赵黎;曾祥国;: "Ti-6Al-4V合金动态本构模型参数敏感性及其优化", 钛工业进展, no. 05 * |
| 盛鹰;曾祥国;韩悌信;陈军;: "钛合金动态本构模型参数敏感度分析及识别方法", 四川大学学报(工程科学版), no. 2 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116595885A (en) * | 2023-05-29 | 2023-08-15 | 上海交通大学 | Lightweight shielding deep coupling optimization method for small reactor |
| CN116595885B (en) * | 2023-05-29 | 2024-03-12 | 上海交通大学 | Lightweight shielding deep coupling optimization method for small reactor |
| CN117519054A (en) * | 2023-12-11 | 2024-02-06 | 广州智业节能科技有限公司 | High-efficient cold station control system |
| CN117519054B (en) * | 2023-12-11 | 2024-06-11 | 广州智业节能科技有限公司 | High-efficient cold station control system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115796011A (en) | Hydrogen storage bed heat transfer performance optimization method based on neural network and genetic algorithm | |
| Gosselin et al. | Review of utilization of genetic algorithms in heat transfer problems | |
| CN109932903A (en) | The air-blower control Multipurpose Optimal Method of more parent optimization networks and genetic algorithm | |
| JP4157477B2 (en) | Improving the performance of an artificial neural network model in the presence of mechanical noise and measurement errors | |
| Jadav et al. | Optimizing weights of artificial neural networks using genetic algorithms | |
| CN103605909B (en) | It is a kind of based on gray theory and the water quality prediction method of support vector machine | |
| CN111639838B (en) | Water quality monitoring point layout optimization method suitable for water supply pipe network | |
| Wang et al. | An interval uncertain optimization method using back-propagation neural network differentiation | |
| CN111881505A (en) | Multi-objective optimization transformation decision method for existing building based on GA-RBF algorithm | |
| Chen et al. | Geometry optimization and pressure analysis of a proton exchange membrane fuel cell stack | |
| Mishra et al. | Short term load forecasting using neural network trained with genetic algorithm & particle swarm optimization | |
| CN107947738A (en) | A kind of Forecasting Methodology of solar energy unmanned plane cell plate voltage | |
| CN105426959A (en) | Aluminium electrolysis energy conservation and emission reduction method based on BP neural network and self-adaptive MBFO algorithm | |
| CN117150896B (en) | Supercritical fluid heat transfer coefficient prediction method based on interpretable machine learning | |
| CN113255211A (en) | BP neural network and multi-objective optimization based organic Rankine cycle working medium screening method | |
| CN116415728B (en) | Carbon emission prediction method, system, electronic equipment and storage medium | |
| CN116595845A (en) | Structure optimization method of spiral fin tube type heat exchanger | |
| CN110033118A (en) | Elastomeric network modeling and the blower multiobjective optimization control method based on genetic algorithm | |
| CN115081293A (en) | Multi-objective optimization method of bubble column carbon capture process based on proxy model | |
| CN115600492A (en) | Laser cutting process design method and system | |
| CN114519299A (en) | Multi-objective optimization method for double-wall structure of integrated support plate flame stabilizer | |
| CN113935556A (en) | Temperature sensor optimal arrangement method based on DNA genetic algorithm | |
| CN105426960A (en) | BP neural network and MBFO algorithm-based aluminum electrolysis energy conservation and emission reduction control method | |
| Wang et al. | Neural network prediction of gasoline and diesel production at catalytic cracking main fractionator | |
| CN117313554B (en) | Multi-section combined multi-objective optimization method, system, equipment and medium for coking production |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20230314 |
|
| WD01 | Invention patent application deemed withdrawn after publication |