Production data analysis systemTechnical Field
The invention relates to the technical field of data analysis, in particular to a production data analysis system.
Background
The data generated during the production process is often very huge and contains a large number of variables and parameters, the processing and analysis of such data require corresponding hardware and software resources, and the capability of processing complex data structures and algorithms is required, the processing of large amounts of data is important, the analysis of data and the extraction of useful information are important, but the processing of large amounts of data is huge, the selection of proper data analysis methods and models is a challenge, different data types and problems require different analysis techniques and model selections, and meanwhile, the interpretation and the interpretability of analysis results are considered so as to convert the analysis results into an actual action plan.
In modern industrial production, a large amount of data is generated and collected, and the data can include sensor data, equipment state information, operation records of staff and the like in the production process, and the data is used for analysis to help enterprises to achieve the aims of optimizing, quality control, fault prediction and the like in the production process, so that the production data analysis system is necessary to be used in the existing data analysis technical field.
Disclosure of Invention
(one) solving the technical problems
Aiming at the problems of huge data volume and low information analysis efficiency in the prior art, the invention provides a production data analysis system which has the advantages of efficiently analyzing production data and improving production efficiency so as to solve the problems presented by the background technology.
(II) technical scheme
In order to achieve the above purpose, the present invention provides the following technical solutions:
a production data analysis system comprises a control host,
the control host computer is fixedly equipped with operation platform, and fixed mounting has telescopic tube on the operation platform, and telescopic tube activity is equipped with flexible loop bar, and fixed mounting has the display screen on the flexible loop bar, and fixed mounting has the equipment sensor on the operation platform, and fixed mounting has the control sensor on the operation platform, and fixed mounting has the remote control pole on the operation platform.
Preferably, the device sensors are connected with the sensors of different devices and integrate data, and the collected data may include parameters such as temperature, pressure, humidity, speed, quality and the like.
Preferably, the control host stores the collected data in a reliable and safe database, so that subsequent data processing and analysis are facilitated, and the control host has high availability, expandability and data security.
Preferably, the control host needs to clean and preprocess the data, ensure the quality and consistency of the data, and reduce the error of subsequent analysis.
Preferably, the control host can perform various data analysis tasks by analyzing production data, which can identify potential problems, improve production processes, optimize resource utilization, and the like.
Preferably, the results of the manipulation host data analysis are typically presented in a display screen visual manner to help users better understand and interpret the data and to help them make decisions or take action.
Preferably, the control host monitors the production data in real time through the monitoring sensor, can quickly find potential problems or abnormal conditions, and timely takes measures to adjust or correct.
Preferably, the control host can gradually increase the efficiency, quality and reliability of the production process by continuously collecting and analyzing data, finding problems and taking improvement measures.
A workflow of a production data analysis system comprising the steps of:
s1, controlling host equipment data to be collected in a mode of equipment sensors, monitoring equipment or system logs and the like, ensuring accuracy and timeliness of data collection to obtain real-time equipment state and performance data, wherein the data can come from various sensors, equipment or monitoring systems on a production line, and the collected data comprise parameters such as temperature, pressure, humidity, speed, quality and the like;
s2, after data is collected, a production data analysis system in a control host machine can establish a data warehouse and is specially used for storing and managing a large amount of structured data, a relational database (such as MySQL and Oracle) is adopted to support complex query and analysis operation, meanwhile, the archiving and history recording functions of the data are provided, apache Kafka is adopted to store real-time streaming data, a large amount of real-time data can be efficiently received and processed, streaming calculation and real-time analysis are supported, and then the NoSQL database is used for storing and managing the data, so that the system provides persistent storage and efficient query support for structured and semi-structured data, and a distributed file system is used for storing and managing large-scale original data, the system has high expandability and fault tolerance, can support redundant backup and parallel processing of the data, and provides functions of access authority management and data backup while having high reliability, reliability and flexibility.
S3, controlling a host to clean and preprocess the stored data, guaranteeing the effectiveness of the data, identifying and processing the missing value of the data, optionally deleting the record containing the missing value, filling by using a mean value or a median, or interpolating by using a regression model and the like, detecting and processing the abnormal value, judging according to data distribution or business rules, selecting to delete, correct or replace the abnormal value, detecting and deleting repeated data record when the repeated value appears, so as to avoid the influence on statistical analysis and the model, converting the data into a correct format, such as a date and time format, a numerical format and the like, selecting the characteristic with obvious influence or relevance on a target variable, selecting by using a statistical method such as a correlation coefficient, a variance analysis and the like, standardizing or normalizing the characteristic of different scales in statistical analysis and model training, converting a classification variable into a binary form so as to facilitate the use of a machine learning algorithm, smoothing the time sequence data, e.g. a method such as sliding average, index and the like, polymerizing the data of granularity, polymerizing the data into data of minute, increasing the data size, reducing the data size of the data size by reducing the data size by the data size of the data size analysis, greatly reducing the data size by the data size analysis;
s4, processing the completed data, controlling a host to analyze the data through data mining, statistical analysis and artificial intelligent algorithm, having the effects of extracting insight, finding patterns and problems, supporting improvement, optimization and decision making of the production process, summarizing and describing the data through methods such as summary metering, frequency distribution, box diagram and the like, knowing the distribution, central trend, change degree and the like of the data, helping to obtain overall knowledge and initial insight of the production process, converting the data into information in a graphic form through visual tools such as charts, graphs and maps and the like, helping to identify patterns, trends and anomalies, evaluating the association degree between the variables through indexes such as correlation coefficients, covariance and the like among the calculated variables, helping to determine key influences of the variables on the production process, guiding further data analysis and modeling, using methods such as time sequence analysis, historical analysis, machine learning and the like, establishing a model based on the data, predicting future and performance, helping to identify potential problems or challenges in advance, and comparing the potential problems with the corresponding devices or the failure models through the methods such as early-stage analysis, optimizing the data, optimizing the factors and the failure detection rules, optimizing the performance, optimizing the factors and the failure detection rules can be realized through the error rate, the error rate and the failure rate detection, the error rate can be optimized through the analysis and the threshold value, the error rate is optimized, the threshold value is optimized, and the error rate is calculated, patterns, rules, and associations hidden in the data are discovered using data mining techniques and machine learning algorithms. This can be used for tasks such as prediction, classification, clustering, and aggregation;
s5, controlling the host to analyze the data to be presented on a display screen, displaying the data analysis result by using various chart types and graphs, visually displaying key indexes and performance parameters by creating an instrument panel and a real-time monitoring interface, enabling the instrument panel to provide real-time data updating and visualization, helping a user to track the state and important indexes of the production process in real time, using technologies such as maps, geographic coordinates and thermodynamic diagrams, presenting data related to geographic positions in a visual manner, helping to find a spatial distribution mode, optimizing resource allocation and solving the problem related to geographic, providing an interactive visual interface, allowing the user to explore and analyze the data in different visual angles and dimensions, customizing and exploring interesting data by selecting, filtering, zooming and the like, generating a report by using the analysis result, including elements such as text description, charts, graphs, data and the like, integrating and conclusion tables for data analysis, enabling the report to be provided in a visual or electronic form, enabling the presentation of data to change trend and evolution in time by technologies such as dynamic animation, sliding blocks, time axes and the like, enabling the user to better understand the time and change trend and evolution, enabling the user to respond to the visual information to be better designed to be in a visual response to the visual interface and visual information, and the visual information is not suitable for the user to be in the visual information, and the visual information is suitable for the user to be provided by the visual interface, and the user has the visual information, and the visual interface, and the user has better visual information;
s6, the monitoring sensor uses real-time data to conduct model training and continuous learning so as to improve accuracy and timeliness of fault detection and prediction, the model can be corrected and updated in real time through real-time monitoring and feedback, new abnormal modes and changes can be found in time, a real-time data analysis result is applied to an automatic control system to realize real-time feedback and adjustment, for example, the control system based on real-time monitoring can automatically adjust control parameters such as temperature, pressure or flow so as to maintain stability and consistency of a production process, and in order to conduct fault detection and analysis, the real-time data is recorded, stored and played back so as to review and re-analyze events and conditions in the production process at any time;
s7, the control host determines key performance indexes associated with the production process, such as production capacity, rejection rate, energy efficiency and the like, the indexes can quantify production performance, provide references for evaluation and tracking improvement, establish a proper data collection mechanism, collect data related to the key performance indexes, including real-time data, historical data, monitoring data and the like, identify pain points, bottlenecks and problems in the production process through data analysis, find opportunities for improvement, conduct root cause analysis on the production problems, use fish bone map tools and technical assistance to determine root causes of the problems, facilitate targeted improvement measures rather than just coping with surface symptoms of the problems, define definite improvement targets based on the root cause analysis and the problem identification, and formulate an improvement plan, set proper targets to enable targeted and quantifiable improvement actions, and plan ensures execution and tracking of the improvement, executes the improvement plan, implements designed improvement measures, continuously monitors the effect of the improvement, pays attention to changes and trends of the key performance indexes, evaluates the effectiveness and effectiveness of the improvement measures, and applies the improvement and teaching to continuous experience and training items to the improvement and the experience and training items to be continuously used in the optimization through the study and the evaluation.
(III) beneficial effects
Compared with the prior art, the invention has the following beneficial effects:
1. according to the production data analysis system, the production data is analyzed through intelligent big data analysis, so that the problems of huge data volume and low information analysis efficiency in the prior art are solved.
2. According to the production data analysis system, data are collected and processed in different modes, so that the collected data are more in line with equipment requirements, and effective collection of the data is realized.
Drawings
FIG. 1 is a schematic flow diagram of a production data analysis system according to the present invention;
FIG. 2 is a schematic diagram of a data collection flow of a production data analysis system according to the present invention;
FIG. 3 is a schematic side view of the operational body of the production data analysis system of the present invention;
FIG. 4 is a schematic top plan view of the operational body of the production data analysis system according to the present invention.
In the figure:
10. controlling a host; 11. an operating platform; 12. a driving motor; 13. driving a screw rod;
14. a movable block; 15. operating a keyboard; 16. an operation key; 17. a telescoping sleeve;
18. a telescopic loop bar; 19. a display screen; 20. a device sensor; 21. monitoring a sensor;
22. a remote control lever.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1
As shown in fig. 1 to 4, which are schematic structural views of a production data analysis system according to a preferred embodiment of the present invention, the production data analysis system of this embodiment includes a control host 10, an operation platform 11 fixedly mounted on the control host 10, a driving motor 12 fixedly mounted in the operation platform 11, a driving screw 13 fixedly mounted at an output end of the driving motor 12, a movable block 14 movably mounted on the driving screw 13, a movable block 15 movably mounted between the driving screw 13 and the movable block 14, an operation keyboard 15 fixedly mounted on the movable block 14, an operation key 16 fixedly mounted on the operation keyboard 15, a telescopic sleeve 17 fixedly mounted on the operation platform 11, a telescopic sleeve 18 movably mounted on the telescopic sleeve 18, a display 19 fixedly mounted on the telescopic sleeve 18, an apparatus sensor 20 fixedly mounted on the operation platform 11, a monitoring sensor 21 fixedly mounted on the operation platform 11, and a remote control lever 22 fixedly mounted on the operation platform 11.
A workflow of a production data analysis system comprising the steps of:
s1, device data of the control host 10 can be collected through a device sensor 20, monitoring equipment or a system log and the like, so that accuracy and timeliness of data collection are ensured, real-time device state and performance data can be obtained, the data can be from various sensors, monitoring systems on the device or a production line, and the collected data comprise parameters such as temperature, pressure, humidity, speed, quality and the like.
S2, after data collection is completed, a production data analysis system in the control host 10 is specially used for storing and managing a large amount of structured data, a relational database (such as MySQL and Oracle) is adopted to support complex query and analysis operation, meanwhile, the archiving and history recording functions of the data are provided, apache Kafka is adopted to store real-time streaming data, a large amount of real-time data can be efficiently received and processed, streaming calculation and real-time analysis are supported, and then the NoSQL database is used for storing and managing the data, so that the system provides persistent storage and efficient query support for structured and semi-structured data, and a distributed file system is used for storing and managing large-scale original data, the system has high expandability and fault tolerance, can support redundant backup and parallel processing of the data, and the functions of access authority management and data backup are provided while the functions of high reliability, reliability and flexibility are provided.
S3, controlling the host 10 to clean and preprocess the stored data, guaranteeing the effectiveness of the data, identifying and processing the missing value of the data, optionally deleting the record containing the missing value, filling by using a mean value or a median, or interpolating by using a regression model and the like, detecting and processing the abnormal value, judging according to the data distribution or business rules, selecting to delete, correct or replace the abnormal value, detecting and deleting repeated data record when the repeated value appears, avoiding the influence on statistical analysis and the model, converting the data into a correct format, such as a date and time format, a numerical format and the like, selecting the characteristics with obvious influence or relevance on a target variable, selecting by using a statistical method such as a correlation coefficient, a variance analysis and the like, standardizing or normalizing the characteristics of different scales in statistical analysis and model training, converting a classification variable into a binary form, facilitating the use of a machine learning algorithm, smoothing the time sequence data, such as a sliding average, an index and the like, polymerizing the data with granularity, increasing the data aggregation level to the data with small data size or the data size of the data size, reducing the data sampling rate greatly, and reducing the data size and analyzing the data size greatly.
S4, processing the completed data, the control host 10 analyzes the data through data mining, statistical analysis and artificial intelligence algorithms, has the functions of extracting insight, finding patterns and problems, supports improvement, optimization and decision making of the production process, summarizes and describes the data through methods of summary metering, frequency distribution, box diagram and the like, so as to know the distribution, central trend, change degree and the like of the data, is helpful for obtaining the overall knowledge and initial insight of the production process, converts the data into information in the form of graphs through visualization tools such as charts, graphs and maps and the like, so as to help identify patterns, trends and anomalies, enables users to more intuitively understand the data and find problems, evaluates the association degree between variables through indexes such as correlation coefficients, covariance and the like among the calculated variables, this can help determine which variables have a critical impact on the production process and direct further data analysis and modeling, using methods such as time series analysis, regression analysis, machine learning, etc., build models based on historical data, predict future trends and performance, help identify potential problems or challenges in advance, and take corresponding measures, analyze data of devices or systems to identify potential failure modes and abnormal behavior, can help achieve early failure diagnosis and prediction, reduce downtime and costs, optimize the production process by modeling and simulating the data, try different strategies and parameters, can be achieved by optimizing algorithms, scheme comparison and sensitivity analysis, etc., identify abnormal events that do not match predefined rules or thresholds using rule engines and abnormality detection algorithms, thereby timely discovering potential problems or risks, patterns, rules, and associations hidden in the data are discovered using data mining techniques and machine learning algorithms. This can be used for tasks such as prediction, classification, clustering, and aggregation.
S5, the data analyzed by the control host 10 are presented on the display screen 19, the data analysis results are displayed by various chart types and graphs, key indexes and performance parameters are visually displayed by creating an instrument panel and a real-time monitoring interface, the instrument panel can provide real-time data updating and visualization to help a user to track the state and important indexes of the production process in real time, the data related to the geographic position are presented in a visual mode by using technologies such as a map, geographic coordinates and a thermodynamic diagram, the data related to the geographic position are presented in a visual mode, the discovery of a spatial distribution mode, the optimization of resource allocation and the solution of the geographic related problem are facilitated, an interactive visual interface is provided, the user is allowed to search and analyze the data with different visual angles and dimensions, the user can customize and search the data interested by selecting, filtering, zooming and other operations, the analysis results are generated into a report, the report comprises elements such as word description, charts, graphs, data tables and the like, the report can be integrated and communicated with the main discovery and conclusion of the data analysis, the report can be provided in a printable or electronic mode, the data trend and evolution of the data can be changed in time through technologies such as dynamic animation, a sliding block, a time axis and the like, the data trend and the time-dependent trend can be better can be displayed, the user can respond to the visual information and the data can be better understood by the user to the user has the visual information and the visual results and the visual information.
S6, the monitoring sensor 21 uses real-time data to conduct model training and continuous learning so as to improve accuracy and timeliness of fault detection and prediction, the model can be corrected and updated in real time through real-time monitoring and feedback, new abnormal modes and changes can be found timely, a real-time data analysis result is applied to an automatic control system to realize real-time feedback and adjustment, for example, the control system based on real-time monitoring can automatically adjust control parameters such as temperature, pressure or flow so as to maintain stability and consistency of a production process, and for fault detection and analysis, the real-time data is recorded, stored and played back so as to review and re-analyze events and conditions in the production process at any time.
S7, the control host 10 determines key performance indexes associated with the production process, such as production capacity, rejection rate, energy efficiency and the like, the indexes can quantify production performance and provide references for evaluation and tracking improvement, an appropriate data collection mechanism is established, data related to the key performance indexes, including real-time data, historical data, monitoring data and the like are collected, pain points, bottlenecks and problems in the production process are identified through data analysis, opportunities for improvement are found, root cause analysis is conducted on the production problems, root cause of the problems is determined through using fish bone map tools and technology assistance, targeted improvement measures are facilitated, rather than just surface symptoms of the problems are dealt with, clear improvement targets are defined based on the root cause analysis and the problem identification, an improvement plan is formulated, appropriate targets are set to enable the improvement work to be targeted and quantifiable, the plan ensures execution and tracking of the improvement action, the designed improvement measures are implemented, the effect of the improvement is continuously monitored, the change and trend of the key performance indexes are focused, the effectiveness and the effect of the improvement measures are improved, and the improvement and the teaching are applied to the continuous experience and the optimization are carried out through the study and the optimization process.
Example 2
As shown in fig. 1 to 4, which are schematic structural views of a production data analysis system according to another preferred embodiment of the present invention, when the production data analysis system analyzes data, the data collection is particularly critical, the collected data is large in quantity and needs to be processed in advance, so that improvement is made on the basis of example 1, a driving motor 12 is fixedly assembled in the operation platform 11, a driving screw 13 is fixedly assembled at the output end of the driving motor 12, a movable block 14 is movably assembled on the driving screw 13, an operation keyboard 15 is fixedly assembled on the movable block 14, an operation key 16 is fixedly assembled on the operation keyboard 15, and data can be manually input to the system through the operation key 16.
A workflow of a production data analysis system comprising the steps of:
determining a data source: the data sources that need to be collected during the production process are determined, which may include sensors, equipment monitoring systems, manual inputs, etc., and which types of data, such as temperature, pressure, speed, quality metrics, etc., need to be collected as needed.
Selecting a data collection method: according to the characteristics and access modes of the data sources, a proper data collection method is selected, including real-time data collection, periodic data collection, event trigger-based collection and the like, and different technologies such as sensors, data interfaces, database connection and the like can be used for realizing data collection.
Setting up a data acquisition point: data acquisition points are set in the production process, so that accurate and timely data acquisition can be ensured, including sensor installation, monitoring equipment installation or specific data acquisition time marking at key production nodes.
Data transmission and storage: the collected data is transmitted to a data storage system, and the method can be realized through network transmission, data interface synchronization, batch uploading and the like, so that the safety and stability of the data transmission are ensured, and an appropriate data storage scheme such as a database, a data warehouse, cloud storage and the like is selected.
Data preprocessing: before storage, necessary preprocessing is performed on the data, including invalid data removal, missing data processing, abnormal value correction, data cleaning and conversion, and some simple data statistics and aggregation operations can be performed in the preprocessing process.
Checking data quality: and (3) performing data quality inspection to ensure that the quality of the acquired data meets the requirements, wherein the data quality inspection can comprise verification of the accuracy, the integrity and the consistency of the data, and eliminating the influence of the data quality problem on subsequent analysis.
Data annotation and tag: annotating and tagging the collected data as needed may facilitate subsequent data analysis and querying, such as adding timestamps, event markers, product lot information, etc. to the data points.
Data backup and protection: data is backed up periodically and necessary security measures are taken to protect the confidentiality and integrity of the data, including data backup policies, disaster recovery plans, data access rights management, etc.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.