阅读说明：本技术 一种基于香烟包装机综合检测平台的大数据分析方法 (Big data analysis method based on comprehensive detection platform of cigarette packaging machine ) 是由 张乐年 郑启旺 徐邓 于 2019-11-15 设计创作，主要内容包括：本发明涉及香烟检测技术领域,具体为一种基于香烟包装机综合检测平台的大数据分析方法,包括获取原始数据、建立数据库、清洗数据、分析数据和展示结果步骤。该基于香烟包装机综合检测平台的大数据分析方法,通过在香烟包装机综合检测平台上部署大数据分析软件,运行信息得以集中、充分、丰富的显示,检测数据的分析结果可以及时通知用户解决当前的故障,并通过具体的分析能定位到具体的故障点,大数据分析软件的运用能及时发现问题,快速定位故障,降低废品率,提高合格品率,从而不断的提高生产制造水平,促进生产效率的提升,降低废品消耗,不断提高精益化生产管理水平,向智能制造迈进。(The invention relates to the technical field of cigarette detection, in particular to a big data analysis method based on a comprehensive detection platform of a cigarette packaging machine. This big data analysis method based on cigarette packagine machine comprehensive testing platform, through dispose big data analysis software on cigarette packagine machine comprehensive testing platform, the operating information can be concentrated, abundant demonstration, the analysis result of detected data can in time inform the user to solve current trouble, and can fix a position to specific fault point through specific analysis, big data analysis software's application can in time discover the problem, fix a position the trouble fast, reduce the rejection rate, improve the qualification rate, thereby continuous improvement manufacturing level, promote the promotion of production efficiency, reduce waste product consumption, continuous improvement lean production management level, make a step forward to intelligent manufacturing.)

1. A big data analysis method based on a cigarette packaging machine comprehensive detection platform is characterized by comprising the following steps: the method comprises the following analysis steps:

s1, acquiring original data: acquiring a large amount of detection data through a plurality of detection devices of a comprehensive detection platform of the cigarette packaging machine;

s2, establishing a database: building a logistics private cloud big data center by using a cloud computing technology, and pooling resource data detected by cigarettes by using a virtualization technology to form a total database;

s3, cleaning data: cleaning each acquired detection data to remove the data lower than or exceeding a preset value;

s4, analysis data: analyzing the cleaned data, establishing different analysis models according to the types of the detected data, and processing different data by the different analysis models;

s5, displaying the result: and a foreground display module is adopted to realize data interaction between the total database and the data receiving terminal and display data.

2. The big data analysis method based on the comprehensive detection platform of the cigarette packing machine according to claim 1, characterized in that: the original data acquired in S1 includes system operating status, alarm information, display information, reject information, detection method type, detection number, defect proportion, detection image, and detection object data.

3. The big data analysis method based on the cigarette packing machine comprehensive detection platform as claimed in claim 2, characterized in that: the detection object data comprises the type, parameters, methods, positions, detection time, running conditions and detection proportion of the detection object.

4. The big data analysis method based on the comprehensive detection platform of the cigarette packing machine according to claim 1, characterized in that: and establishing a database in the S2, and constructing and modifying a total database by adopting SQLserver 2005 software.

5. The big data analysis method based on the comprehensive detection platform of the cigarette packing machine according to claim 1, characterized in that: the cleaning data in the S3 comprises a defect number module, a proportion data module, a time data module, a normal detection image module and a detection object data module;

the defect number module is used for removing data lower than a preset defect number;

the proportion data module is used for removing data lower than a preset detected proportion;

the time data module is used for removing data which is lower than preset detection time;

the normal detection image module is used for removing normal detection images exceeding a preset value;

the detection object data module is used for removing the detection object data which is not enabled.

6. The big data analysis method based on the comprehensive detection platform of the cigarette packing machine according to claim 1, characterized in that: the analysis data comprises a model building module, a mapping sample module, a production discrimination model module and an analysis comparison module.

7. The big data analysis method based on the comprehensive detection platform of the cigarette packing machine according to claim 1, characterized in that: the foreground display module comprises a user interaction module and a data display module;

the user interaction module is used for realizing data interaction between the total database and the data receiving terminal;

and the data display module is used for receiving the detected data by the data receiving terminal and displaying the detected data in a chart form.

8. The big data analysis method based on the cigarette packing machine comprehensive detection platform as claimed in claim 7, wherein: the chart in the data display module comprises a real-time curve, a histogram, a pie chart and a statistical table.

9. The big data analysis method based on the cigarette packing machine comprehensive detection platform as claimed in claim 7, wherein: the foreground display module also comprises an alarm module, and the alarm module is used for giving an alarm prompt when error data occurs.

10. The big data analysis method based on the cigarette packing machine comprehensive detection platform as claimed in claim 7, wherein: the foreground display module also comprises a data storage module, and the data storage module is used for storing the displayed data.

Technical Field

The invention relates to the technical field of cigarette detection, in particular to a big data analysis method based on a comprehensive detection platform of a cigarette packaging machine.

Background

Cigarette packaging is a very important process in cigarette production, and products with unqualified packaging quality flow into the market to bring very serious influence on the reputation of enterprises. At present, cigarette production enterprises add a plurality of detection devices on a packaging machine for improving the product quality, such as cigarette appearance detection, cigarette end missing mouth detection, aluminum foil paper paperboard detection, small package appearance detection, stay wire detection, large strip appearance detection, missing package detection and the like, wherein the detection devices can be integrated on a comprehensive detection platform of the cigarette packaging machine, the comprehensive detection platform of the cigarette packaging machine can obtain a large amount of detection data, but the data collection application and mining analysis are still in a primary stage, and how to deeply utilize the detection data resources by means of a data mining analysis technology is an important supporting means for improving the production management level and the information construction level of the enterprises and also for improving the production and manufacturing level.

Currently, the installed detection devices only realize the function of removing or alarming after detecting defects, do not analyze the reasons of abnormal products produced, and cannot guide users to improve the production level and reduce the proportion of defective products, thereby not really improving the production rate and the production and manufacturing level. In view of the above, we propose a big data analysis method based on a cigarette packing machine comprehensive detection platform.

Disclosure of Invention

The invention aims to provide a big data analysis method based on a comprehensive detection platform of a cigarette packaging machine, which aims to solve the problems that the reason of abnormal products cannot be analyzed and the productivity and the production and manufacturing level are not really improved in the background technology.

In order to achieve the purpose, the invention provides a big data analysis method based on a comprehensive detection platform of a cigarette packaging machine, which comprises the following analysis steps:

s1, acquiring original data: acquiring a large amount of detection data through a plurality of detection devices of a comprehensive detection platform of the cigarette packaging machine;

s2, establishing a database: building a logistics private cloud big data center by using a cloud computing technology, and pooling resource data detected by cigarettes by using a virtualization technology to form a total database;

s3, cleaning data: cleaning each acquired detection data to remove the data lower than or exceeding a preset value;

s4, analysis data: analyzing the cleaned data, establishing different analysis models according to the types of the detected data, and processing different data by the different analysis models;

s5, displaying the result: and a foreground display module is adopted to realize data interaction between the total database and the data receiving terminal and display data.

Preferably, the raw data acquired in S1 includes system operating status, alarm information, display information, rejection information, detection method type, detection number, defect proportion, detection image, and detection object data.

Preferably, the detection target data includes a type, a parameter, a method, a position, a detection time, an operation state, and a detection ratio of the detection target.

Preferably, the database established in S2 is built and modified by SQLserver 2005 software.

Preferably, the cleaning data in S3 includes a defect number module, a proportion data module, a time data module, a normal detection image module, and a detection object data module;

the defect number module is used for removing data lower than a preset defect number;

the proportion data module is used for removing data lower than a preset detected proportion;

the time data module is used for removing data which is lower than preset detection time;

the normal detection image module is used for removing normal detection images exceeding a preset value;

the detection object data module is used for removing the detection object data which is not enabled.

Preferably, the analysis data comprises a model building module, a sample mapping module, a production discrimination model module and an analysis comparison module.

Preferably, the foreground display module comprises a user interaction module and a data display module;

the user interaction module is used for realizing data interaction between the total database and the data receiving terminal;

and the data display module is used for receiving the detected data by the data receiving terminal and displaying the detected data in a chart form.

Preferably, the graph in the data display module comprises a real-time curve, a histogram, a pie chart and a statistical table.

Preferably, the foreground display module further comprises an alarm module, and the alarm module is used for giving an alarm prompt when error data occurs.

Preferably, the foreground display module further comprises a data storage module, and the data storage module is used for storing the displayed data.

Compared with the prior art, the invention has the beneficial effects that:

1. in the big data analysis method based on the cigarette packer comprehensive detection platform, one-line real-time detection data directly reflects the current production condition and the packaging quality, the big data analysis method is applied to the data to carry out centralized extraction and analysis, the analysis result can timely inform a user of solving the current fault, a specific fault point can be positioned through specific analysis, and the technical requirements and the working intensity of the user and a maintainer can be greatly reduced. The quality control mode in the cigarette packaging process is improved by reasonably planning, storing and utilizing big data and utilizing an information technology, so that on one hand, the human experience intervention can be reduced, and the workload is reduced; on the other hand, the accuracy and the stability of quality control can be improved to a certain extent by adopting a continuous self-learning prediction control model, the monitoring, the prediction and the control of the tobacco production process data are carried out by utilizing an informatization means, the method has important significance for improving the quality of the tobacco finished products, and the continuous improvement of the production and manufacturing level is finally promoted.

2. According to the big data analysis method based on the comprehensive detection platform of the cigarette packing machine, the big data analysis software is deployed on the comprehensive detection platform of the cigarette packing machine, the operation information can be displayed in a centralized, sufficient and abundant mode, the alarm information is notified and displayed in a centralized mode, the problem that notification and display are inconvenient due to the fact that a plurality of detectors are deployed in a scattered mode is avoided, the analysis result of the detection data can notify a user of the current fault in time, the specific fault point can be located through specific analysis, and the technical requirements and the working intensity of a user and a maintainer can be greatly reduced. The application of big data analysis software can find problems in time, locate faults quickly, reduce the rejection rate and improve the qualified product rate, thereby improving the production and manufacturing level continuously, promoting the improvement of the production efficiency, reducing the waste consumption, improving the lean production management level continuously and advancing to intelligent manufacturing.

Drawings

FIG. 1 is a schematic view of the overall structure of the present invention;

FIG. 2 is a database framework architecture diagram of the present invention;

FIG. 3 is a block diagram of a cleaning data module of the present invention;

FIG. 4 is a flow chart of a neural network algorithm of the present invention;

FIG. 5 is a diagram of an analytical data block according to the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.