CN117150877A - Method for predicting optimal pressing process of press-loading mixed explosive based on Bagging algorithm - Google Patents
Method for predicting optimal pressing process of press-loading mixed explosive based on Bagging algorithm Download PDFInfo
- Publication number
- CN117150877A CN117150877A CN202210562572.3A CN202210562572A CN117150877A CN 117150877 A CN117150877 A CN 117150877A CN 202210562572 A CN202210562572 A CN 202210562572A CN 117150877 A CN117150877 A CN 117150877A
- Authority
- CN
- China
- Prior art keywords
- optimal
- press
- samples
- bagging
- mixed explosive
- 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
- 238000000034 method Methods 0.000 title claims abstract description 57
- 239000002360 explosive Substances 0.000 title claims abstract description 28
- 238000003825 pressing Methods 0.000 title claims abstract description 17
- 238000012549 training Methods 0.000 claims abstract description 22
- 238000005070 sampling Methods 0.000 claims abstract description 6
- 238000004220 aggregation Methods 0.000 claims abstract description 4
- 230000002776 aggregation Effects 0.000 claims abstract description 4
- 238000004321 preservation Methods 0.000 claims description 4
- 238000012935 Averaging Methods 0.000 claims description 2
- 238000010276 construction Methods 0.000 claims 1
- 238000013461 design Methods 0.000 abstract description 5
- 230000003247 decreasing effect Effects 0.000 abstract 1
- 238000012360 testing method Methods 0.000 description 10
- 239000003814 drug Substances 0.000 description 6
- 230000000694 effects Effects 0.000 description 3
- 230000004931 aggregating effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000001351 cycling effect Effects 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012856 packing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/20—Ensemble learning
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Pure & Applied Mathematics (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Artificial Intelligence (AREA)
- Computational Mathematics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Operations Research (AREA)
- Probability & Statistics with Applications (AREA)
- Geometry (AREA)
- Algebra (AREA)
- Computer Hardware Design (AREA)
- Databases & Information Systems (AREA)
- Computing Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to the field of explosive press-fitting, and discloses a Bagging algorithm-based optimal press-fitting process prediction method for press-fitting mixed explosive; randomly collecting samples with the same number as 21 samples of the training set, wherein the content of the samples is different, carrying out 1000 times of random sampling on the training set with 21 samples, and the 1000 sampling sets are different due to randomness; carrying out arithmetic average on regression results obtained by 1000 weak learners by using an aggregation strategy of a simple average method to obtain final model output; gradually decreasing from the theoretical density according to the output model, and searching the corresponding optimal process condition under the maximum density; the method for predicting the optimal pressing process of the press-fit mixed explosive based on the Bagging algorithm has the advantages that the optimal process condition is accurately predicted, and the difficult problem that the charging condition is difficult to determine is solved. Providing reference for selecting proper technological conditions and optimizing the technological design of the pressed mixed explosive.
Description
Technical Field
The invention relates to a Bagging algorithm-based method for predicting an optimal pressing process of a press-fit mixed explosive, and belongs to the field of explosive press-fit.
Background
In the process of producing the pressed mixed explosive, the higher the percentage of the pressed density to the theoretical density, the better the performance of the explosive. The density of the explosive is closely related to the process conditions, and the process conditions and the pressed density are changed. How to determine the optimal technological conditions, so that the pressed explosive density is infinitely close to the theoretical density, is a great difficulty in the design and research of the pressed mixed explosive.
In order to solve the problems, a Bagging algorithm-based method for predicting the optimal pressing process of the press-fit mixed explosive is provided, and the method has the characteristics of high accuracy, high safety and good economy, and has been applied to the design of CL-20-base press-fit mixed explosive.
Bagging algorithm (Bootstrap aggregating, guide aggregation algorithm) is also called Bagging algorithm, and is a group learning algorithm in the field of machine learning.
In the patent, technological parameters such as specific pressure, temperature and pressure maintaining time of the pressed medicine are taken as sample characteristics, the medicine charge density is taken as a sample label, and supervised regression learning training is carried out through a Bagging algorithm, so that a medicine charge density prediction model is obtained. And (3) using a trained model, starting from the theoretical density, gradually reducing the density according to a designated step length, and obtaining the optimal process condition.
Disclosure of Invention
The invention aims to solve the problem that the technological parameters of the pressed mixed explosive are difficult to determine, and provides a Bagging algorithm-based method for predicting the optimal pressed explosive process of the pressed mixed explosive.
The aim of the invention is achieved by the following technical scheme.
A Bagging algorithm-based optimal pressing process prediction method for press-loading mixed explosive comprises the following specific steps:
step one, relevant data are sorted and collected, and the data are divided into training sets x_train and y_train and test sets x_test and y_test. 80% of which are training sets and 20% of which are test sets.
Step two, calling a Baggingregress module in a sklearn library of Python. Setting parameters: base_counter=decision tree, n_counter=1000. The Bagging algorithm flow is as follows:
1) 21 samples were randomly collected from the training set of step one. As a sampling set D, 1000 sampling sets were acquired in total.
2) The 1000 weak learners were trained with sample set D.
3) And (3) carrying out arithmetic average on regression results obtained by the 1000 weak learners to obtain a value which is output as a final model.
And thirdly, evaluating the Bagging training model by using the test set in the first step, and carrying out error analysis by comparing the true value with the predicted value. And further drawing a learning curve, and evaluating the fitting generalization effect of the model.
Writing a process-density program, and setting the specific drug pressure x1 to be 2000-3600kg/cm respectively 2 300 steps, the range of the pressing temperature x2 is 20-90 ℃, 5 steps, the range of the pressure maintaining time x3 is 20-75min, 5 steps, and the maximum charge density value y and the error limit a=1.03×10 are input -5 Sequentially cycling through x1, x2 and x3, taking one group as input each time, putting into the prediction model trained in the third step for training, and charging densityThe predicted value is compared with the set value, the method adopts the mean square error, if the error is larger than the set error limit a, the method does not meet the requirements, otherwise, the method meets the requirements and outputs.
And fifthly, selecting the optimal process conditions from the output results, and verifying through experiments.
Advantageous effects
1. The method for predicting the optimal pressing process of the press-fit mixed explosive based on the Bagging algorithm can determine the optimal process conditions, and is extremely high in accuracy.
2. According to the method for predicting the optimal pressing process of the pressed mixed explosive based on the Bagging algorithm, the Bagging algorithm generally collects samples with the same number as the number m of samples of the training set randomly. The number of samples of the obtained sampling set and the training set is the same, but the content of the samples is different. If we do T random samples of m sample training sets, then the T sample sets are different due to randomness. The Bagging aggregation strategy is also simpler, and for regression problems, a simple average method is generally used to carry out arithmetic average on regression results obtained by the T weak learners to obtain final model output. The method provides a new thought for predicting the charge density, and has guiding significance for the production and process design of the press-fit mixed explosive.
3. According to the method for predicting the optimal pressing process of the press-fit mixed explosive based on the Bagging algorithm, disclosed by the invention, the corresponding process conditions can be predicted by continuously changing the designated density, so that the aim of predicting the optimal process is fulfilled, and the problem that the process conditions are difficult to determine is solved.
Drawings
FIG. 1 is a Bagging model learning curve graph in the present invention;
FIG. 2 is a graph of evaluation scores for the number of Bagging training devices of 10, 100, 500, 1000, 2000 and 5000 in the invention;
FIG. 3 is a graph comparing actual and predicted values of packing density in accordance with the present invention;
FIG. 4 is a schematic diagram of a Bagging algorithm;
Detailed Description
The invention is further described below with reference to examples and figures.
Example 1 training a predictive model
Firstly, a CL-20 base press-fit mixed explosive is subjected to arrangement of technological parameters and a charging density data set, wherein the data set comprises 3 characteristics of a pressing specific pressure, a pressing temperature, a pressure maintaining time and the like. The label is the numerical size of the charge density. After data loading, the data sets are divided into two groups, wherein 80% of the data sets are training sets, 20% of the data sets are test sets, and a Bagging regressor is started for training. Next, a learning curve of the Bagging regression is drawn, and as the number of samples increases, the test curve gradually decreases, the training curve gradually increases, and the training curve gradually approach each other, as shown by the Bagging regression learning curve (fig. 1), so that the Bagging regression effect is better.
The numbers of the sub-trainers are respectively set to 10, 100, 500, 1000, 2000 and 5000, and the models are evaluated to obtain model scores (figure 2) under different numbers of the sub-trainers, and the results are shown in the following table.
Table one: model scoring for different numbers of sub-trainers
As can be seen from the results in table one, for the data set in this example, the model score increased and then tended to be flat as the number of sub-trainers increased. When the number of the sub-trainers is increased, the model training cost is increased, so that the number of the sub-trainers is set to be 1000, and the bagging algorithm principle is shown in figure 4.
Example 2 prediction Using trained models
After training, the model was predicted using the test set data (fig. 3), and the results are shown in table two.
And (II) table: comparing the true value to the predicted value of the test set
As can be seen from the results shown in Table II, the model is predicted by using the test set data, and the method provided by the invention has a smaller error range for the charge density predicted value and the actual measured value which are basically the same.
Example 3
To determine the optimum process conditions, a density of 2.05 g.cm was taken -3 The reverse prediction method of the invention is adopted to predict the technological parameters, and no proper technological conditions corresponding to the technological parameters are available. Gradually reducing input density with 0.005 as step length to obtain density of 2.04g cm -3 The results shown in Table III are obtained.
Table three: taking density of 2.04 g.cm -3 Predicting the technological parameters
| Results | 1 | 2 |
| Temperature of pressing medicine | 85 | 85 |
| Specific pressure for pressing medicine | 3200 | 3500 |
| Holding time | 65 | 60 |
As can be seen from the results shown in Table three, the density was specifiedIs 2.04 g.cm -3 When the method is used, 2 technological condition parameters are reversely predicted, and after averaging, the optimal temperature is 85 ℃, and the optimal specific pressure is 3350kg/cm 2 The optimal heat preservation time is 62.5min, and the optimal heat preservation time is equal to the real experimental value of 85 ℃ and 3300kg/cm 2 The prediction is very accurate after 60 minutes and the similar.
The method for predicting the optimal pressing process of the press-fit mixed explosive based on the Bagging algorithm has the advantages that the optimal process condition is accurately predicted, and the difficult problem that the charging condition is difficult to determine is solved. Providing reference for selecting proper technological conditions and optimizing the technological design of the pressed mixed explosive.
Claims (2)
1. A Bagging algorithm-based optimal pressing process prediction method for a press-fit mixed explosive is characterized by comprising the following steps of: the Bagging algorithm construction thought is as follows:
the Bagging algorithm will randomly collect as many samples as there are training set samples 21. The number of samples of the obtained sampling set and the training set is the same, but the content of the samples is different. If we do 1000 random samples of the 21 sample training set, the 1000 sample sets are different due to randomness. The Bagging aggregation strategy is also simpler, and for regression problems, a simple average method is generally used for carrying out arithmetic average on regression results obtained by 1000 weak learners to obtain final model output.
2. A Bagging algorithm-based optimal pressing process prediction method for a press-fit mixed explosive is characterized by comprising the following steps of: the optimal process determination method comprises the following steps:
taking density of 2.05g cm -3 The reverse prediction method of the invention is adopted to predict the technological parameters, and no proper technological conditions corresponding to the technological parameters are available. Gradually reducing input density with 0.005 as step length to obtain density of 2.04g cm -3 2 reverse predicted technological condition parameters are adopted, and after averaging, the optimal temperature is 85 ℃, and the optimal specific pressure is 3350kg/cm 2 The optimal heat preservation time is 62.5min, and the optimal heat preservation time is equal to the real experimental value of 85 ℃ and 3300kg/cm 2 The time of 60 minutes is very similar, and the prediction is very accurate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210562572.3A CN117150877A (en) | 2022-05-23 | 2022-05-23 | Method for predicting optimal pressing process of press-loading mixed explosive based on Bagging algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210562572.3A CN117150877A (en) | 2022-05-23 | 2022-05-23 | Method for predicting optimal pressing process of press-loading mixed explosive based on Bagging algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117150877A true CN117150877A (en) | 2023-12-01 |
Family
ID=88885396
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210562572.3A Pending CN117150877A (en) | 2022-05-23 | 2022-05-23 | Method for predicting optimal pressing process of press-loading mixed explosive based on Bagging algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117150877A (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105117525A (en) * | 2015-07-31 | 2015-12-02 | 天津工业大学 | Bagging extreme learning machine integrated modeling method |
| CN113515891A (en) * | 2021-06-04 | 2021-10-19 | 浙江永联民爆器材有限公司 | Method for predicting and optimizing quality of emulsion explosive |
-
2022
- 2022-05-23 CN CN202210562572.3A patent/CN117150877A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105117525A (en) * | 2015-07-31 | 2015-12-02 | 天津工业大学 | Bagging extreme learning machine integrated modeling method |
| CN113515891A (en) * | 2021-06-04 | 2021-10-19 | 浙江永联民爆器材有限公司 | Method for predicting and optimizing quality of emulsion explosive |
Non-Patent Citations (3)
| Title |
|---|
| 付忠良;: "关于AdaBoost有效性的分析", 计算机研究与发展, no. 10, 15 October 2008 (2008-10-15), pages 119 - 127 * |
| 唐子惠: "医学人工智能导论", vol. 1, 30 April 2020, 上海科学技术出版社, pages: 173 - 174 * |
| 曹兴: "粉体炸药压制成型工艺仿真与质量预测", 中国优秀硕士学位论文全文数据库 工程科技II辑, no. 9, 15 September 2021 (2021-09-15), pages 81 - 97 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104849050B (en) | A kind of Fault Diagnosis of Roller Bearings based on compound multiple dimensioned arrangement entropy | |
| CN105319071B (en) | Diesel Engine Fuel System Fault Diagnosis method based on least square method supporting vector machine | |
| CN108664010A (en) | Generating set fault data prediction technique, device and computer equipment | |
| CN106770939A (en) | A kind of Diagnosis Method of Transformer Faults based on support vector description and the nearly barycenter neighbours of K | |
| CN111340248A (en) | Transformer fault diagnosis method and system based on intelligent integration algorithm | |
| CN101504443A (en) | Prediction method for discharge capacity of lithium ion battery | |
| CN111368904A (en) | Electrical equipment identification method based on electric power fingerprint | |
| CN112270129B (en) | Plant growth prediction method based on big data analysis | |
| CN115166563A (en) | Power battery aging state evaluation and decommissioning screening method and system | |
| CN114595956B (en) | Eucalyptus soil fertility analysis method based on gray scale correlation method fuzzy clustering algorithm | |
| CN113821976A (en) | Lithium battery fault diagnosis modeling method based on integrated algorithm | |
| CN111208436A (en) | Energy storage battery overcharge diagnosis method based on IMF energy moment and genetic algorithm optimization SVM | |
| CN117556369A (en) | Power theft detection method and system for dynamically generated residual error graph convolution neural network | |
| CN111626559A (en) | Main factor analysis method-based medium-voltage distribution network line loss key characteristic index extraction method and system | |
| CN112287980A (en) | Power battery screening method based on typical feature vector | |
| CN117473424A (en) | Transformer fault diagnosis method, system, equipment and medium based on random forest | |
| CN117150877A (en) | Method for predicting optimal pressing process of press-loading mixed explosive based on Bagging algorithm | |
| CN118114149A (en) | Detection method and system for liquid crystal display panel | |
| CN113722898A (en) | Durability evaluation method and device for fuel cell stack | |
| CN117760732A (en) | Multi-domain feature selection bearing fault diagnosis method | |
| CN110276478B (en) | Short-term wind power prediction method based on segmented ant colony algorithm optimization SVM | |
| CN116822355A (en) | Method for real-time on-line monitoring and controlling fermentation quality of Daqu | |
| Bergmeir et al. | Using balanced random forests on load spectrum data for classifying component failures of a hybrid electric vehicle fleet | |
| CN117706377B (en) | Battery inconsistency identification method and device based on self-adaptive clustering | |
| CN117150876A (en) | Method for controlling charge density of pressed mixed explosive based on ridge regression algorithm |
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 |