CN117055502A - Intelligent control system based on Internet of things and big data analysis - Google Patents
Intelligent control system based on Internet of things and big data analysis Download PDFInfo
- Publication number
- CN117055502A CN117055502A CN202311180588.9A CN202311180588A CN117055502A CN 117055502 A CN117055502 A CN 117055502A CN 202311180588 A CN202311180588 A CN 202311180588A CN 117055502 A CN117055502 A CN 117055502A
- Authority
- CN
- China
- Prior art keywords
- unit
- module
- model
- parameters
- scheduling
- 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.)
- Withdrawn
Links
- 238000007405 data analysis Methods 0.000 title claims abstract description 35
- 238000004519 manufacturing process Methods 0.000 claims abstract description 54
- 238000005457 optimization Methods 0.000 claims abstract description 42
- 238000004088 simulation Methods 0.000 claims abstract description 20
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 18
- 238000012417 linear regression Methods 0.000 claims abstract description 6
- 238000004891 communication Methods 0.000 claims abstract description 5
- 230000008859 change Effects 0.000 claims description 36
- 230000002159 abnormal effect Effects 0.000 claims description 30
- 238000004458 analytical method Methods 0.000 claims description 10
- 238000011156 evaluation Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 8
- 230000006855 networking Effects 0.000 claims description 7
- 238000000265 homogenisation Methods 0.000 claims description 6
- 238000012544 monitoring process Methods 0.000 claims description 6
- 238000013112 stability test Methods 0.000 claims description 6
- 230000000007 visual effect Effects 0.000 abstract description 2
- 238000007726 management method Methods 0.000 description 30
- 238000000034 method Methods 0.000 description 13
- 230000008569 process Effects 0.000 description 8
- 230000004044 response Effects 0.000 description 6
- 230000006872 improvement Effects 0.000 description 4
- 230000007774 longterm Effects 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 238000009499 grossing Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000002790 cross-validation Methods 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000002068 genetic effect Effects 0.000 description 2
- 238000003058 natural language processing Methods 0.000 description 2
- 238000007637 random forest analysis Methods 0.000 description 2
- 238000002922 simulated annealing Methods 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000003339 best practice Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 238000012043 cost effectiveness analysis Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000011031 large-scale manufacturing process Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000003449 preventive effect Effects 0.000 description 1
- 238000003908 quality control method Methods 0.000 description 1
- 230000001932 seasonal effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000012731 temporal analysis Methods 0.000 description 1
- 238000000700 time series analysis Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 238000011179 visual inspection Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS], computer integrated manufacturing [CIM]
- G05B19/41875—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS], computer integrated manufacturing [CIM] characterised by quality surveillance of production
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/32—Operator till task planning
- G05B2219/32368—Quality control
Abstract
The invention relates to the technical field of production equipment control, in particular to an intelligent control system based on the Internet of things and big data analysis, which comprises a digital twin management unit, a data acquisition unit, a prediction scheduling unit and a twin model optimization unit. The digital twin management unit is used as a system core and provides a visual interface to monitor the production flow in real time. The data acquisition unit is in communication connection with the sensor, collects real-time and historical data and fault information, and provides a basis for big data analysis. The predictive scheduling unit performs big data analysis on the data to predict potential faults and generate production scheduling instructions. The twin model optimizing unit is responsible for generating and updating a digital twin model, and performs model optimization by using a multiple linear regression and optimization algorithm. Through the comprehensive system, the invention realizes the high simulation and real-time optimization of the production environment, remarkably improves the production efficiency and the system stability, and reduces the human error and the failure rate.
Description
Technical Field
The invention relates to the technical field of production equipment control, in particular to an intelligent control system based on the Internet of things and big data analysis.
Background
The digital twin technology is tightly combined with the Internet of things in industrial production application, so that seamless connection between the reality and the virtual model is realized. Under this architecture, sensors and devices of the internet of things are responsible for collecting data in the factory production lines, devices and flows in real time, which are used to build and continuously update the digital twin model. The method can realize the high simulation of the production environment and monitor and predict in real time, thereby achieving the purposes of optimizing the production efficiency and reducing the failure rate.
However, in real-world applications, since existing systems often lack sufficient intelligence, potential failure points cannot be automatically identified, which leads to excessive reliance on human intervention, making the production line not fully automated, consuming a lot of human resources and increasing the risk of human error. Meanwhile, even if a visualized digital twin model exists, due to the complexity of a large-scale production line and the overlarge data volume, the production line is difficult to adjust and optimize manually in the model.
Disclosure of Invention
In order to solve the problems, the invention provides an intelligent control system based on the Internet of things and big data analysis.
In order to achieve the above purpose, the invention adopts the following technical scheme:
intelligent control system based on thing networking and big data analysis contains: the system comprises a digital twin management unit, a data acquisition unit, a prediction scheduling unit and a twin model optimizing unit;
the digital twin management unit is respectively connected with the twin model optimizing unit, the data acquisition unit and the prediction scheduling unit, the data acquisition unit is respectively connected with the prediction scheduling unit and the twin model optimizing unit, and the twin model optimizing unit is connected with the prediction scheduling unit;
the digital twin management unit is used for visually managing and controlling the system;
the data acquisition unit is used for communicating with the networking sensor and collecting real-time parameters, historical parameters, fault data and big data analysis results;
the prediction scheduling unit is used for carrying out deep analysis on the collected real-time parameters, historical parameters and fault data, capturing data rules and correlations, realizing the fault prediction of the equipment, carrying out data processing on a historical fault processing mode, generating a production scheduling adjustment instruction and sending the production scheduling adjustment instruction to the digital twin management unit;
the twin model optimizing unit is used for manually generating and updating a digital twin model of the factory, evaluating the performance of each part in the model through the parameter change relation, and optimizing the model.
Further, the data acquisition unit comprises an acquisition module and a data archiving module;
the acquisition module is in communication connection with various sensors and cameras arranged on the production line to collect real-time parameters;
the data archiving module is used for storing real-time parameters, historical parameters and fault data and synchronizing the real-time parameters to the digital twin management unit and the prediction scheduling unit;
the data archiving module is also used for storing log data manually scheduled by staff when the production line faults occur.
Further, the predictive scheduling unit comprises a prediction module;
the prediction module is used for executing the following steps:
identifying, extracting and marking common abnormal parameters in the historical data when the same fault occurs as high-risk parameters, and classifying the high-risk parameters into an abnormal parameter group of the fault;
respectively extracting parameters on a time line from the high risk parameters on the abnormal parameter group before the fault occurs to obtain an abnormal change curve of the high risk parameters;
carrying out big data homogenization on the abnormal change curve;
and monitoring curve change of the high-risk parameters in real time, comparing the curve change with the homogenized abnormal change curve, and generating an early warning report when an abnormal trend is detected.
Further, the prediction scheduling unit also comprises a scheduling module;
the scheduling module is used for executing the following steps:
big data analysis is carried out on log data manually scheduled by staff when faults occur to obtain a scheduling scheme;
matching the early warning report received by the prediction module with a scheduling scheme;
and sending the matched scheduling scheme to a digital twin management unit.
Further, the twin model optimization unit includes a manual update module for receiving model updates or revisions uploaded by an engineer or operator.
Further, the manual update module is further used for recording the update history and change log of each model.
Further, the twin model optimizing unit further comprises a parameter evaluation module;
the parameter evaluation module evaluates historical parameters or parameter change relations by using a multiple linear regression algorithm, and evaluates the performance utilization rate of each part in the production line by using a preset parameter safety range as a standard and using real-time parameters.
Further, the twin model optimizing unit further comprises an optimizing and identifying module;
the optimization recognition module is used for carrying out simulation optimization on the production line through the performance utilization rate, and a simulation optimization scheme comprises the addition of branch lines and the addition of equipment in the production line.
Further, the twin model optimization unit further comprises a virtual environment module;
the virtual environment module is used for generating a virtual production environment model to apply a simulation optimization scheme, performing scene simulation through a parameter change relation and performing stability test.
Further, the twin model optimization unit further comprises a model updating module;
the model updating module is used for sending the simulation optimization scheme which is passed by the stability test to the digital twin management unit.
The invention has the beneficial effects that: according to the invention, through big data analysis in the prediction scheduling unit, real-time and historical data are mined, and data rules and correlations are captured, so that automatic fault prediction is realized, faults are solved, and dependence on manual fault detection is reduced. The prediction is based on real-time parameters, historical fault data and manual scheduling logs are combined, and the early warning accuracy and the feasibility of a fault solution are further improved. The twin model optimization unit performs performance analysis on the model by utilizing a multiple linear regression algorithm to provide continuous optimization, and adapts to the complexity of the production line. Through the cooperation of multiple modules, the production efficiency is improved, the failure rate and the risk of human errors are greatly reduced, and the high automation is realized to reduce the labor cost.
Drawings
Fig. 1 is a schematic structural diagram of an intelligent control system based on internet of things and big data analysis in the invention.
FIG. 2 is a schematic diagram of a predictive scheduling unit.
FIG. 3 is a schematic diagram of a twin model optimization unit.
Detailed Description
Referring to fig. 1-3, the present invention relates to an intelligent control system based on internet of things and big data analysis.
Example 1
Intelligent control system based on thing networking and big data analysis contains: the system comprises a digital twin management unit, a data acquisition unit, a prediction scheduling unit and a twin model optimizing unit;
the digital twin management unit is respectively connected with the twin model optimizing unit, the data acquisition unit and the prediction scheduling unit, the data acquisition unit is respectively connected with the prediction scheduling unit and the twin model optimizing unit, and the twin model optimizing unit is connected with the prediction scheduling unit;
the digital twin management unit is used for visually managing and controlling the system;
the data acquisition unit is used for communicating with the networking sensor and collecting real-time parameters, historical parameters, fault data and big data analysis results;
the prediction scheduling unit is used for carrying out deep analysis on the collected real-time parameters, historical parameters and fault data, capturing data rules and correlations, realizing the fault prediction of the equipment, carrying out data processing on a historical fault processing mode, generating a production scheduling adjustment instruction and sending the production scheduling adjustment instruction to the digital twin management unit;
the twin model optimizing unit is used for manually generating and updating a digital twin model of the factory, evaluating the performance of each part in the model through the parameter change relation, and optimizing the model.
In this embodiment, the digital twin management unit is the core of the whole intelligent control system, and is directly connected with the data acquisition unit, the predictive scheduling unit and the twin model optimizing unit in data and control. The unit provides a comprehensive and intuitive view for plant management personnel to monitor and control the operating conditions of the entire plant in real time. The digital twinning management unit also provides a user-friendly Graphical User Interface (GUI) containing various visualization tools such as instrument boards, charts and real-time data streams. These tools enable the manager to have a clear view of various aspects of the production line including, but not limited to, equipment status, production efficiency, energy consumption, and fault records. Through the connection with other units, the digital twin management unit can display the operation parameters of each device and production line in real time. Meanwhile, the digital twin management unit allows a manager to remotely control equipment, including starting, stopping, speed adjustment and the like, so that fine control on the production flow is realized. When the prediction scheduling unit identifies potential faults or risks, the digital twin management unit immediately displays early warning information and provides corresponding processing suggestions or automatically executes control commands such as emergency stop and the like according to a preset response strategy. In addition to real-time data and control functions, the digital twin management unit is also tightly integrated with the twin model optimization unit. When the model optimizing unit updates or optimizes the digital twin model, the changes are reflected on the visual interface of the digital twin management unit in real time so that the manager can know and verify the effect of the optimizing measures.
The data acquisition unit is a component in the intelligent control system responsible for communication with various networking sensors and devices in the factory. Once the connection is established, the data acquisition unit continuously collects real-time data from these devices. These data are stored in a structured manner and transferred in real time to the digital twin management unit and predictive scheduling unit to support real-time monitoring and fault prediction. In addition to real-time data, the unit is also responsible for collecting and storing historical data, including operational records, production data, and fault records for the equipment. These data are used for long term trend analysis, fault diagnosis and production optimization.
The predictive scheduling unit uses machine learning and time series analysis data rules and correlations to deep mine real-time and historical data from the data acquisition unit. The process can accurately predict possible faults and performance degradation of equipment, generate early warning information and immediately transmit the early warning information to the digital twin management unit to take preventive measures. Meanwhile, the unit is also responsible for generating an optimal production scheduling scheme based on data driving so as to improve the production efficiency and the resource utilization rate. The highly integrated and intelligent operation not only improves the response capability of the system to the emergency, but also optimizes the whole production flow.
The twin model optimizing unit not only allows a worker to adjust the model, but also can automatically carry out fine adjustment and optimization on the model, the continuous optimizing process not only improves the accuracy of the digital twin model, but also adjusts the performance utilization rate of each part in the production line so as to avoid the too low utilization rate of a certain part and realize reasonable scheduling of the performance.
Through the four highly interconnected units, the intelligent control system can effectively solve the common problems in the traditional factory, such as lack of automatic fault prediction, excessive dependence of manual intervention, difficult management of complex production environments and the like. Through real-time data and a large amount of historical data analysis in the system, the intelligent and more automatic operation of the factory is realized, so that the production efficiency is improved, the failure rate is reduced, and the waste of human resources is reduced.
Example 2
The intelligent control system based on the internet of things and big data analysis according to embodiment 1, wherein the data acquisition unit comprises an acquisition module and a data archiving module;
the acquisition module is in communication connection with various sensors and cameras arranged on the production line to collect real-time parameters;
the data archiving module is used for storing real-time parameters, historical parameters and fault data and synchronizing the real-time parameters to the digital twin management unit and the prediction scheduling unit;
the data archiving module is also used for storing log data manually scheduled by staff when the production line faults occur.
In this embodiment, the acquisition module is responsible for real-time data collection by communicating with various sensors and cameras on the factory production line. The sensors may be temperature sensors, pressure sensors, speed sensors, rotational speed sensors, etc. for monitoring various production parameters and environmental variables, and cameras for visual inspection and quality control. The data archiving module employs a database management system to structurally store large-scale, high-speed data streams collected from various sensors and devices. These data are not only synchronized in real time to the digital twin management unit and predictive scheduling unit to support data analysis and real time decisions, but also stored for long term for trend analysis and fault diagnosis. The data archiving module also records fault data and manual operation logs, the fault data including, but not limited to, the specific time at which the fault occurred, the unique identifier and detailed description of the fault, the severity level of the fault, key parameters and environmental variables of the fault, and the specific impact of the fault on production or service. The manual operation log comprises, but is not limited to, the specific time of operation execution, the staff or system account of the operation, the executed operation or change, the operation target and the equipment or system state after the operation execution, ensures the integrity, the safety and the usability of the data, and provides a basis for the data driving decision and the optimization of the whole system.
Example 3
The intelligent control system based on the internet of things and big data analysis according to embodiment 2, wherein the predictive scheduling unit includes a predictive module;
the prediction module is used for executing the following steps:
identifying, extracting and marking common abnormal parameters in the historical data when the same fault occurs as high-risk parameters, and classifying the high-risk parameters into an abnormal parameter group of the fault;
respectively extracting parameters on a time line from the high risk parameters on the abnormal parameter group before the fault occurs to obtain an abnormal change curve of the high risk parameters;
carrying out big data homogenization on the abnormal change curve;
and monitoring curve change of the high-risk parameters in real time, comparing the curve change with the homogenized abnormal change curve, and generating an early warning report when an abnormal trend is detected.
The prediction scheduling unit also comprises a scheduling module;
the scheduling module is used for executing the following steps:
big data analysis is carried out on log data manually scheduled by staff when faults occur to obtain a scheduling scheme;
matching the early warning report received by the prediction module with a scheduling scheme;
and sending the matched scheduling scheme to a digital twin management unit.
In this embodiment, the prediction module first groups the historical fault data using a density-based spatial clustering application with a clustering algorithm such as noise (DBSCAN) or K-means (K-means). The identification of common features or parameters occurring in similar fault scenarios is achieved. The module then uses a Support Vector Machine (SVM) or Random Forest (Random Forest) class classification algorithm to further screen and classify the features generated by these clusters. The purpose of this step is to determine which parameters have significantly changed before and when the fault occurred and thus can be marked as high risk parameters. Once these high risk parameters are identified, they are categorized into a particular database table or data structure, which is classified as an "abnormal parameter class group". This class set contains not only the names and types of parameters, but also statistical measures, such as mean, variance, skewness, etc., about the parameters in normal operation and fault conditions. The comprehensive and deep analysis process ensures accurate identification of high risk parameters, and provides a solid data base for subsequent predictive analysis and fault early warning. The advantage of this approach is that it can accommodate a wide variety of different types and sizes of data, thereby providing highly accurate and reliable fault predictions.
After the high risk parameters are successfully identified and categorized, the prediction module enters a timeline parameter extraction phase, where the module first uses a data window technique to segment the historical data, typically creating a time window in terms of minute, hour, or day time intervals. For each time window, the prediction module employs an autoregressive moving average model (ARIMA) to analyze seasonal, trend, and noise contributions of the parameters. In this way, the prediction module can generate an abnormal change curve for each high risk parameter that reveals the pattern of dynamic change of the parameter before the failure threshold is reached. Meanwhile, the prediction module also uses a long-short-time memory network (LSTM), namely a deep learning algorithm specially designed for processing time series data. LSTM captures long-term dependencies and non-linearities, providing more accurate warning of impending failures. The time line parameter extraction process not only generates an abnormal change curve with high descriptive and predictive properties, but also provides an accurate reference for real-time early warning generation, and greatly enhances the capability of the system in early recognition and fault prevention.
After the timeline parameter extraction generates a highly descriptive outlier variation, the prediction module further performs big data homogenization to enhance the accuracy and reliability of the model. The prediction module processes the abnormal change curve using a smoothing algorithm such as Moving Average (MA), exponential smoothing (EMA), or gaussian smoothing. These algorithms effectively filter out short-term random fluctuations, highlighting long-term trends and patterns in the data. The prediction module also adopts statistical methods such as Tukey fences or mahalanobis distances to identify and process outliers. These values are typically caused by sensor errors, data transmission problems, or other abnormal factors, and may negatively impact the accuracy of the model. The prediction module then performs data normalization, i.e., scaling all parameter variables to within the same numerical range. This is typically achieved by subtracting the mean and dividing by the standard deviation, helping to improve the comparability of the model between different scales and units of data. After the steps, the module generates an abnormal change curve after homogenization as a baseline model. The baseline model not only has high statistical robustness, but also can reflect the real behavior mode of the high risk parameters before the fault occurs more accurately.
After the big data homogenization and the baseline model generation are completed, the prediction module enters a real-time early warning generation stage. The prediction module first establishes a real-time data stream monitor for continuously receiving real-time data of high risk parameters from the data acquisition unit. In order to compare the real-time data to the homogenized outlier, the prediction module uses an advanced time series similarity algorithm, dynamic Time Warping (DTW). DTW can effectively align similar patterns in two time series even though they differ in time scale. An adaptive threshold detection mechanism is then implemented, which is set based on past data and statistical methods, such as confidence intervals. Only when the real-time data exceeds these thresholds will the system consider an abnormal trend. Upon detection of an abnormal trend, the module immediately initiates an early warning report generation flow, including a snapshot of the abnormal parameters, the likely type and severity of the fault.
The scheduling module performs big data analysis on log data manually scheduled by a worker when a fault stored in the data archiving module occurs. This analysis uses Natural Language Processing (NLP) techniques and association rule learning algorithms, which in this embodiment employ the Apriori algorithm or the FP-growth algorithm to extract the effective fault response patterns and best practices. Based on the analysis results, the module generates a set of predefined scheduling schemes. Each of these schemes involves a series of response measures to a specific type or level of failure, such as equipment restart, line speed adjustment, or emergency shutdown. When the prediction module generates an early warning report and sends the early warning report to the scheduling module, the later can immediately carry out the matching of the report and the scheduling scheme. This process can be implemented using a machine-learned classifier, such as a decision tree, to ensure that the selected scheduling scheme is highly matched to the type and severity of potential faults identified in the early warning report. Once the matching is successful, the scheduling scheme is immediately sent to the digital twin management unit for real-time regulation and control, so that timeliness of fault response is ensured, the requirement of human intervention is greatly reduced, and the response speed and accuracy of the whole system in the face of faults are improved.
Example 4
The intelligent control system based on internet of things and big data analysis according to embodiment 3, wherein the twin model optimization unit includes a manual update module for receiving model updates or revisions uploaded by an engineer or operator. The manual updating module is also used for recording the updating history and change log of each model.
In this embodiment, the manual update module first provides a web-based interface that allows an engineer or operator to directly upload or revise the model. The manual update module also records the update history and change log of each model. Each time a new model is uploaded or an old model is revised, the module automatically generates a detailed change log including, but not limited to, update time, operator identity, change content, and difference comparisons with previous versions. The above arrangement provides support for the version rollback functionality, allowing for restoration to any previous model version if necessary, enabling flexibility and security of model updates, ensuring that the digital twin model maintains its accuracy and validity all the time even in highly dynamic and uncertain production environments.
Example 5
The intelligent control system based on the internet of things and big data analysis according to embodiment 4, wherein the twin model optimization unit further comprises a parameter evaluation module;
the parameter evaluation module evaluates historical parameters or parameter change relations by using a multiple linear regression algorithm, and evaluates the performance utilization rate of each part in the production line by using a preset parameter safety range as a standard and using real-time parameters.
The twin model optimizing unit further comprises an optimizing and identifying module;
the optimization recognition module is used for carrying out simulation optimization on the production line through the performance utilization rate, and a simulation optimization scheme comprises the addition of branch lines and the addition of equipment in the production line.
The twin model optimizing unit further comprises a virtual environment module;
the virtual environment module is used for generating a virtual production environment model to apply a simulation optimization scheme, performing scene simulation through a parameter change relation and performing stability test.
The twin model optimizing unit further comprises a model updating module;
the model updating module is used for sending the simulation optimization scheme which is passed by the stability test to the digital twin management unit.
In this embodiment, the parameter evaluation module uses a multiple linear regression algorithm to evaluate the relevance and impact between parameters. Model training typically involves gradient descent algorithms and cross-validation techniques to ensure robustness and accuracy of the model. Feature selection is then performed, using correlation analysis and Lasso/Ridge regularization to screen for important features. Next, the dataset is divided into a training set and a test set, a gradient descent algorithm is applied to minimize the loss function, and k-fold cross-validation is employed to ensure robustness of the model. The performance of the model is assessed by the R-squared value, as well as the mean square error or mean absolute error. Finally, the module is also provided with a preset parameter safety range, and the trained model is used for carrying out performance evaluation on the real-time data and comparing the real-time data with the preset safety range. The series of processes not only accurately evaluate the relevance and influence among the parameters, but also provide real-time equipment and process performance monitoring, thereby providing basis for further optimization.
The optimization recognition module uses, for example, genetic algorithms or simulated annealing to find the best solution to improve line efficiency. The module receives as input performance utilization data from the parameter assessment module and integrates it with other factors such as cost, time and resource constraints. In genetic algorithms, a number of possible improvements are encoded and the optimal solution is found by selection, crossover and mutation operations. And the simulated annealing starts from an initial solution, and searches for a globally optimal solution through random interference and condition judgment. The module then generates a specific optimization scheme, which may include adding new line branches, introducing new equipment, or adjusting existing workflows. These schemes will be pushed to execution after cost-effectiveness analysis and simulation verification, ensuring that recommended improvements are scientific and viable.
The virtual environment module is a simulation platform, and combines a data twinning technology to simulate the actual production environment in the virtual space. The module receives the improvement from the optimization recognition module and performs simulation testing in this virtual environment. Specifically, the virtual environment module first creates or updates a digital copy of an existing production flow using data twinning techniques. The digital copy contains not only the physical device and layout, but also historical and real-time data, as well as associated performance metrics. In a virtual environment, the system may perform a series of stability and efficiency tests, such as by varying input parameters or operating conditions to evaluate the toughness and performance of the production line. Thus, the management team can safely evaluate the feasibility and benefits of each optimization scheme without affecting the actual production. And finally, the model updating module is responsible for pushing the optimized scheme verified by the virtual environment module to the digital twin management unit.
The above embodiments are merely illustrative of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention, and various modifications and improvements made by those skilled in the art to the technical solution of the present invention should fall within the scope of protection defined by the claims of the present invention without departing from the spirit of the design of the present invention.
Claims (10)
1. Intelligent control system based on thing networking and big data analysis, its characterized in that includes: the system comprises a digital twin management unit, a data acquisition unit, a prediction scheduling unit and a twin model optimizing unit;
the digital twin management unit is respectively connected with the twin model optimizing unit, the data acquisition unit and the prediction scheduling unit, the data acquisition unit is respectively connected with the prediction scheduling unit and the twin model optimizing unit, and the twin model optimizing unit is connected with the prediction scheduling unit;
the digital twin management unit is used for visually managing and controlling the system;
the data acquisition unit is used for communicating with the networking sensor and collecting real-time parameters, historical parameters, fault data and big data analysis results;
the prediction scheduling unit is used for carrying out deep analysis on the collected real-time parameters, historical parameters and fault data, capturing data rules and correlations, realizing the fault prediction of the equipment, carrying out data processing on a historical fault processing mode, generating a production scheduling adjustment instruction and sending the production scheduling adjustment instruction to the digital twin management unit;
the twin model optimizing unit is used for manually generating and updating a digital twin model of the factory, evaluating the performance of each part in the model through the parameter change relation, and optimizing the model.
2. The intelligent control system based on the internet of things and big data analysis according to claim 1, wherein the data acquisition unit comprises an acquisition module and a data archiving module;
the acquisition module is in communication connection with various sensors and cameras arranged on the production line to collect real-time parameters;
the data archiving module is used for storing real-time parameters, historical parameters and fault data and synchronizing the real-time parameters to the digital twin management unit and the prediction scheduling unit;
the data archiving module is also used for storing log data manually scheduled by staff when the production line faults occur.
3. The intelligent control system based on the internet of things and big data analysis according to claim 1, wherein the predictive scheduling unit comprises a predictive module;
the prediction module is used for executing the following steps:
identifying, extracting and marking common abnormal parameters in the historical data when the same fault occurs as high-risk parameters, and classifying the high-risk parameters into an abnormal parameter group of the fault;
respectively extracting parameters on a time line from the high risk parameters on the abnormal parameter group before the fault occurs to obtain an abnormal change curve of the high risk parameters;
carrying out big data homogenization on the abnormal change curve;
and monitoring curve change of the high-risk parameters in real time, comparing the curve change with the homogenized abnormal change curve, and generating an early warning report when an abnormal trend is detected.
4. The intelligent control system based on internet of things and big data analysis according to claim 3, wherein the predictive scheduling unit further comprises a scheduling module;
the scheduling module is used for executing the following steps:
big data analysis is carried out on log data manually scheduled by staff when faults occur to obtain a scheduling scheme;
matching the early warning report received by the prediction module with a scheduling scheme;
and sending the matched scheduling scheme to a digital twin management unit.
5. The intelligent control system based on internet of things and big data analysis according to claim 1, wherein the twin model optimization unit comprises a manual update module for receiving model updates or revisions uploaded by engineers or operators.
6. The intelligent control system based on internet of things and big data analysis of claim 5, wherein the manual update module is further configured to record an update history and a change log for each model.
7. The intelligent control system based on internet of things and big data analysis according to claim 1, wherein the twin model optimization unit further comprises a parameter evaluation module;
the parameter evaluation module evaluates historical parameters or parameter change relations by using a multiple linear regression algorithm, and evaluates the performance utilization rate of each part in the production line by using a preset parameter safety range as a standard and using real-time parameters.
8. The intelligent control system based on internet of things and big data analysis according to claim 7, wherein the twin model optimization unit further comprises an optimization recognition module;
the optimization recognition module is used for carrying out simulation optimization on the production line through the performance utilization rate, and a simulation optimization scheme comprises the addition of branch lines and the addition of equipment in the production line.
9. The intelligent control system based on internet of things and big data analysis of claim 8, wherein the twin model optimization unit further comprises a virtual environment module;
the virtual environment module is used for generating a virtual production environment model to apply a simulation optimization scheme, performing scene simulation through a parameter change relation and performing stability test.
10. The intelligent control system based on internet of things and big data analysis according to claim 9, wherein the twin model optimization unit further comprises a model update module;
the model updating module is used for sending the simulation optimization scheme which is passed by the stability test to the digital twin management unit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311180588.9A CN117055502A (en) | 2023-09-13 | 2023-09-13 | Intelligent control system based on Internet of things and big data analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311180588.9A CN117055502A (en) | 2023-09-13 | 2023-09-13 | Intelligent control system based on Internet of things and big data analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117055502A true CN117055502A (en) | 2023-11-14 |
Family
ID=88653676
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311180588.9A Withdrawn CN117055502A (en) | 2023-09-13 | 2023-09-13 | Intelligent control system based on Internet of things and big data analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117055502A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117393076A (en) * | 2023-12-13 | 2024-01-12 | 山东三岳化工有限公司 | Intelligent monitoring method and system for heat-resistant epoxy resin production process |
| CN117389236A (en) * | 2023-12-11 | 2024-01-12 | 山东三岳化工有限公司 | Propylene oxide production process optimization method and system |
| CN117425165A (en) * | 2023-12-18 | 2024-01-19 | 江苏泽宇智能电力股份有限公司 | System for managing novel power communication board card by using intelligent terminal |
-
2023
- 2023-09-13 CN CN202311180588.9A patent/CN117055502A/en not_active Withdrawn
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117389236A (en) * | 2023-12-11 | 2024-01-12 | 山东三岳化工有限公司 | Propylene oxide production process optimization method and system |
| CN117389236B (en) * | 2023-12-11 | 2024-02-13 | 山东三岳化工有限公司 | Propylene oxide production process optimization method and system |
| CN117393076A (en) * | 2023-12-13 | 2024-01-12 | 山东三岳化工有限公司 | Intelligent monitoring method and system for heat-resistant epoxy resin production process |
| CN117393076B (en) * | 2023-12-13 | 2024-02-09 | 山东三岳化工有限公司 | Intelligent monitoring method and system for heat-resistant epoxy resin production process |
| CN117425165A (en) * | 2023-12-18 | 2024-01-19 | 江苏泽宇智能电力股份有限公司 | System for managing novel power communication board card by using intelligent terminal |
| CN117425165B (en) * | 2023-12-18 | 2024-04-09 | 江苏泽宇智能电力股份有限公司 | System for managing novel power communication board card by using intelligent terminal |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10902368B2 (en) | Intelligent decision synchronization in real time for both discrete and continuous process industries | |
| CN112052979B (en) | Equipment spare part demand prediction system based on fault prediction and health management | |
| CN117055502A (en) | Intelligent control system based on Internet of things and big data analysis | |
| Alves et al. | Deployment of a smart and predictive maintenance system in an industrial case study | |
| US7827122B1 (en) | Data mining of unfiltered controller data | |
| KR101825881B1 (en) | Method of managing a manufacturing process and system using the same | |
| US10444746B2 (en) | Method for managing subsystems of a process plant using a distributed control system | |
| US10521457B2 (en) | Subject matter expert knowledge mapping using dynamic clustering | |
| US20190171168A1 (en) | Computer-implemented method and system for automatically monitoring and determining the status of entire process sections in a process unit | |
| CN104412190A (en) | Systems and methods for health assessment of a human-machine interface (HMI) device | |
| CN116681250A (en) | Building engineering progress supervisory systems based on artificial intelligence | |
| CN117390529A (en) | Multi-factor traceable data center information management method | |
| CN113468022B (en) | Automatic operation and maintenance method for centralized monitoring of products | |
| CN117348503B (en) | Propylene oxide production data monitoring system and method | |
| KR20220089853A (en) | Method for Failure prediction and prognostics and health management of renewable energy generation facilities using machine learning technology | |
| CN117041312A (en) | Enterprise-level information technology monitoring system based on Internet of things | |
| US20230205161A1 (en) | Method and apparatus for monitoring industrial devices | |
| WO2021111936A1 (en) | Prediction system, information processing device, and information processing program | |
| CN114265891A (en) | Intelligent workshop system and method based on multi-source data fusion and storage medium | |
| CN114168409A (en) | Service system running state monitoring and early warning method and system | |
| CN113835387A (en) | Operation and maintenance management method, system and medium | |
| CN117234785B (en) | Centralized control platform error analysis system based on artificial intelligence self-query | |
| Elbazi et al. | Digital Twin-Enabled Monitoring of Mining Haul Trucks with Expert System Integration: A Case Study in an Experimental Open-Pit Mine | |
| CN117118862A (en) | Large data center monitoring data visualization system and method | |
| Nashivochnikov et al. | The System for Operational Monitoring and Analytics of Industry Cyber-Physical Systems Security in Fuel and Energy |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20231114 |