CN116339267B - Automatic production line control system based on Internet of things - Google Patents
Automatic production line control system based on Internet of things Download PDFInfo
- Publication number
- CN116339267B CN116339267B CN202310594355.7A CN202310594355A CN116339267B CN 116339267 B CN116339267 B CN 116339267B CN 202310594355 A CN202310594355 A CN 202310594355A CN 116339267 B CN116339267 B CN 116339267B
- Authority
- CN
- China
- Prior art keywords
- production
- value
- day
- equipment
- characteristic data
- 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.)
- Active
Links
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 507
- 238000012549 training Methods 0.000 claims abstract description 38
- 238000010801 machine learning Methods 0.000 claims abstract description 33
- 238000012545 processing Methods 0.000 claims abstract description 21
- 239000002994 raw material Substances 0.000 claims description 26
- 238000000034 method Methods 0.000 claims description 20
- 238000013480 data collection Methods 0.000 claims description 16
- 239000013598 vector Substances 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 12
- 238000002372 labelling Methods 0.000 claims description 3
- 238000003062 neural network model Methods 0.000 claims description 3
- 238000005520 cutting process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000000227 grinding Methods 0.000 description 2
- 238000003754 machining Methods 0.000 description 2
- 239000000843 powder Substances 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012795 verification Methods 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] or computer integrated manufacturing [CIM]
- G05B19/41865—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] or computer integrated manufacturing [CIM] characterised by job scheduling, process planning, material flow
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G05B2219/32252—Scheduling production, machining, job shop
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Automation & Control Theory (AREA)
- Quality & Reliability (AREA)
- Manufacturing & Machinery (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an automatic production line control system based on the Internet of things, which relates to the technical field of automatic control of production lines, wherein historical training data of production equipment in each production link in the production line is collected in advance, a machine learning model for predicting the next-day fault probability of the equipment is trained based on the historical training data, equipment relation data of the production line to be monitored and production characteristic data of the previous working day are collected, the production line to be monitored uses the machine learning model and the production characteristic data of the previous working day to predict the fault probability of each production equipment on the current working day, and the maximum capacity of the production line to be monitored on the current working day of the production line to be monitored is predicted when the current working day of the production line to be monitored begins, so that the number of processing pieces on the current day is automatically planned; the efficiency of the plan making of staff is improved, and the production efficiency is further improved.
Description
Technical Field
The invention belongs to the automatic control technology of production lines, and particularly relates to an automatic production line control system based on the Internet of things.
Background
In modern manufacturing, automated production line control systems play a critical role in improving production efficiency and reducing production costs. However, many production line control systems still rely on manual entry of the day workload data, which has the following problems:
blind filling in of workload data: conventional production line control systems typically require a worker to manually fill out the workload data for the day, which can lead to inaccuracy and subjectivity in the data. Manual filling may be affected by personal views, errors and negligence, resulting in deviations in the data.
The capacity plan needs to be continuously modified: due to inaccuracy and subjectivity of the data, as well as variations in the production line, the production plan may need to be continually modified. This results in interruption of the production line and a decrease in production efficiency, while increasing the workload and pressure of the staff.
Inefficient planning process: conventional production planning processes require personnel to manually integrate and analyze various data, including workload, equipment status, and the like. This requires a lot of time and effort and is prone to errors.
Therefore, the invention provides an automatic production line control system based on the Internet of things.
Disclosure of Invention
The present invention aims to solve at least one of the technical problems existing in the prior art. Therefore, the invention provides an automatic production line control system based on the Internet of things, which solves the problem that the work load on the day is filled in the system for manual blind purpose, and the capacity plan is required to be continuously modified later, thereby realizing the effects of improving the efficiency of planning by staff and further improving the production efficiency.
To achieve the above object, an embodiment according to a first aspect of the present invention provides an automated production line control system based on the internet of things, including a historical data collection module, a failure probability prediction model training module, a production line data collection module, a failure probability prediction module, and a productivity prediction module; wherein, each module is connected by a wired and/or wireless network mode;
the historical data collection module is mainly used for collecting historical training data of production equipment of each production link in the production line in advance;
the production link comprises a raw material processing link, a component manufacturing link and a finished product production link;
the raw material treatment link refers to that raw materials are subjected to specific treatment steps, such as cutting, grinding and the like, production equipment required to be used in the raw material treatment link is a plurality of raw material treatment equipment, and different raw material treatment equipment is used for treating different raw materials or carrying out different processing on the raw materials;
the component manufacturing link refers to processing steps of processing raw materials into various types of components required by industrial finished products, production equipment required in the component manufacturing link is component processing equipment, and different component manufacturing equipment is used for producing different components;
the production link of the finished product refers to a processing step of assembling all types of components into an industrial finished product, and production equipment used in the production link of the finished product is finished product production equipment;
further, the historical training data comprises a plurality of groups of production characteristic data and fault label data corresponding to each production device in each production link;
for each production device of each production link, generating a group of production characteristic data and fault tag data corresponding to the characteristic data by each corresponding production device in each working day;
the production characteristic data comprise the use time length, the average value of the temperature on the day, the variance value of the temperature on the day, the average value of the production efficiency on the day, the variance value of the production efficiency on the day, the average value of vibration on the day, the variance value of vibration on the day, the average value of sound intensity on the day and the variance value of sound intensity on the day of each production device in each working day;
the production characteristic data may further include a difference between a current day temperature mean value and a previous day temperature mean value, a difference between a current day temperature variance value and a previous day temperature variance value, a difference between a current day production efficiency mean value and a previous day production efficiency mean value, a difference between a current day production efficiency variance value and a previous day production efficiency variance value, a difference between a current day vibration mean value and a previous day vibration mean value, a difference between a current day vibration variance value and a previous day vibration variance value, a difference between a current day sound intensity mean value and a previous day sound intensity mean value, and a difference between a current day sound intensity variance value and a previous day sound intensity variance value in each working day;
the using time length of each production device is obtained by timing the using time length of each production device;
the temperature, the vibration value and the sound intensity of each production device are respectively obtained in real time through a temperature sensor, a vibration sensor and a sound intensity sensor;
the production efficiency is obtained by calculating the number of products produced by each production device in real time; the number of the products corresponds to the number of the output products of the production equipment, and particularly, the output products and the statistical method of the number of the output products are determined according to the type of the production equipment; for example: for the grinder, the product quantity is the quantity of the powder which is output by the grinder and ground into the raw material;
the values of the fault tag data comprise 0 and 1, and when the fault tag data is 0, the fault tag data indicate that the production equipment corresponding to the corresponding production characteristic data does not have faults in the next day; if the fault label data is 1, the production equipment corresponding to the corresponding production characteristic data is indicated to be faulty in the next day;
the historical data collection module sends historical training data of production equipment of each production link in the production line to the fault probability prediction model training module;
the failure probability prediction model training module is mainly used for training a machine learning model for predicting the next-day failure probability of the equipment based on historical training data;
the machine learning model for predicting the next-day fault probability of the equipment is trained in the following way:
the number of the production link is marked as i, and the number of the production equipment type of the ith production link is marked as ij; marking a production characteristic data set corresponding to the ij type production equipment as Fij, and marking a fault tag data set corresponding to the production characteristic data Fij as Gij; marking the serial number of the production characteristic data in the production characteristic data set as fij;
for the firstThe production equipment combines each group of production characteristic data in the production characteristic data set Fij into a characteristic vector form;
all feature vector sets as machinesAn input to a machine learning model having as output a probability of failure predicted for each set of production characteristic data, for the thThe group production characteristic data takes fij group fault label data as a prediction target, and takes the sum of prediction error degrees of all the predicted fault probabilities as a training target; the calculation formula of the prediction error degree is as follows; />Wherein->For the prediction error degree, afij is the predicted failure probability corresponding to the fij-th group of production characteristic data,/I>Production of a failure tag data set corresponding to characteristic data for the fij th group +.>The fij-th fault tag data; training the machine learning model until the sum of the prediction error degrees reaches convergence, and stopping training; the machine learning model is any one of a deep neural network model and a deep belief network model; thereby generating a corresponding machine learning model for each production facility;
the failure probability prediction model training module sends the trained machine learning model to the failure probability prediction module;
the production line data collection module is mainly used for collecting equipment relation data of a production line to be monitored and production characteristic data of the previous working day;
the device relationship data comprises a set of dependency relationships for each production device;
specifically, for the ij-th production equipment, marking a dependency relationship set as Dij, wherein each element in the dependency relationship set Dij is the number and the production ratio of part of the production equipment types in the i-1 production links, and each production equipment corresponding to each number in the number of the part of the production equipment types meets the conditions: the product output by the equipment is required to be processed by using the ij production equipment;
the serial numbers of the production equipment types in the dependency relation set Dij are marked as Dij, and the production proportion of the production equipment types Dij refers to the output product output required to be provided by the production equipment types Dij in each unit output product of the production equipment types ij; labeling the yield as Bdij;
the method for collecting the production characteristic data of the previous working day of the production line to be monitored is as follows:
before each working day of a production line to be monitored starts, collecting production characteristic data of each production device in the previous working day;
the production line data collection module sends production characteristic data of the production line to be monitored on the previous working day to the fault probability prediction module;
the fault probability prediction module is mainly used for predicting the fault probability of each production equipment on the current working day by using a machine learning model and production characteristic data of the previous working day on a production line to be monitored;
the mode of predicting the fault probability of each production device on the current working day is as follows:
the method comprises the steps of marking the number of ij production equipment in a production line to be monitored as Nij, and marking the number of the production equipment in the ij production equipment as Nij;
for the nij production equipment, converting the production characteristic data of the previous working day into a characteristic vector form, and inputting the characteristic vector into a machine learning model corresponding to the nij production equipment to obtain a predicted value of the probability of failure of the nij production equipment on the current day, which is output by the machine learning model;
the failure probability prediction module sends a predicted value of the failure probability of each production device in the production line to be monitored to the productivity prediction module;
the productivity prediction module is mainly used for predicting the maximum productivity of the production line to be monitored on the current working day, so that the number of the machining pieces on the current working day is automatically planned;
the method for predicting the maximum capacity of the production line to be monitored on the same day is as follows:
marking a predicted value of the current day fault probability of the nij production equipment as Pnij
Acquiring a current day production efficiency average value from production characteristic data of the previous working day of the nij production equipment, and marking the current day production efficiency average value of the previous working day of the nij production equipment as Xnij;
presetting the production time length T of the current working day;
in the 1 st link, calculating a total expected production value E1j of the 1 st production equipment, wherein a calculation formula of the total expected production value E1j is as follows；
Calculating the total expected production value E2j of the 2 nd production equipment in the 2 nd link, wherein the total expected production value；
Further calculating the maximum expected yield Rd2j provided by the D2j production equipment in the dependency relation set D2j of the 2j production equipment, wherein the calculation formula of the maximum expected yield Rd2j is as followsThe method comprises the steps of carrying out a first treatment on the surface of the Wherein->For the total expected production value corresponding to the production facility type of the 1 st production link corresponding to the production facility d2j, i.e. +.>May correspond to a certain value of the values E1 j; obviously, during the production process, the total production value of the production equipment of the 2 j-th type is limited by the maximum expected production value Rd2 j;
traversing the minimum value of the maximum expected production value Rd2j corresponding to each production device in the dependency relation set D2j, marking the minimum value as M2j, and enabling the maximum expected production value R2j of the 2 j-th production device to be the minimum value of E2j and M2 j;
calculating the total expected production value E3j of the 3j production equipment in the 3 rd link；
Further calculating the maximum expected yield Rd3j provided by the D3j production equipment in the dependency relation set D3j of the 3j production equipment, wherein the calculation formula of the maximum expected yield Rd3j is as followsThe method comprises the steps of carrying out a first treatment on the surface of the Wherein->The maximum expected production value for the production facility type of the 2 nd production link for the production facility d3j, i.e. +.>May correspond to a certain value in R2 j; obviously, in real production, the total production value of the 3 j-th production device is limited by the maximum expected production value Rd3 j;
traversing the minimum value of the maximum expected production value Rd3j corresponding to each production device in the dependency relation set D3j, marking the minimum value as M3j, and enabling the maximum expected production value R3j of the 3 j-th production device to be the minimum value of E3j and M3 j;
the maximum expected production value R3j of the production equipment of type 3j is set to the maximum capacity of the production line to be monitored on the same day.
Compared with the prior art, the invention has the beneficial effects that:
according to the invention, historical production characteristic data of each production device on a production line are collected in advance, the machine learning model is used for training the historical production characteristic data, so that the prediction of the fault probability of each production device on the current working day based on the performance of the device on the previous working day is realized, in the actual production line to be monitored, when each working day starts, the production characteristic data of the previous working day is collected, the prediction of the fault probability of each production device on the current working day is carried out, and the maximum productivity which can be produced by the production line finally is calculated based on the production dependency relationship and the production quantity relationship of the devices among production links; therefore, production data is automatically provided for production line staff to serve as a reference, the problem that the follow-up requirement of continuously modifying the productivity plan is caused by filling the daily workload in the system for manual blind purpose is avoided, the efficiency of planning by the staff is improved, and the production efficiency is further improved.
Drawings
Fig. 1 is a module connection diagram of an automated production line control system based on the internet of things.
Detailed Description
The technical solutions of the present invention will be clearly and completely described in connection with the embodiments, and it is obvious that the described embodiments are only some embodiments of the present invention, 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, an automated production line control system based on the internet of things comprises a historical data collection module, a fault probability prediction model training module, a production line data collection module, a fault probability prediction module and a productivity prediction module; wherein, each module is connected by a wired and/or wireless network mode;
the historical data collection module is mainly used for collecting historical training data of production equipment of each production link in the production line in advance;
in a preferred embodiment, the production links include a raw material processing link, a component manufacturing link, and a finished product production link;
it is understood that the raw material processing step refers to that raw materials are subjected to specific processing steps, such as cutting, grinding, etc., and production equipment required to be used in the raw material processing step is a plurality of raw material processing equipment, and different raw material processing equipment processes different raw materials or carries out different processing on the raw materials;
the component manufacturing link refers to processing steps of processing raw materials into various types of components required by industrial finished products, production equipment required in the component manufacturing link is component processing equipment, and different component manufacturing equipment is used for producing different components;
the production link of the finished product refers to a processing step of assembling all types of components into an industrial finished product, and production equipment used in the production link of the finished product is finished product production equipment;
further, the historical training data comprises a plurality of groups of production characteristic data and fault label data corresponding to each production device in each production link;
for each production device of each production link, generating a group of production characteristic data and fault tag data corresponding to the characteristic data by each corresponding production device in each working day;
in a preferred embodiment, the production characteristic data includes a use period, a mean value of a day temperature, a variance value of a day temperature, a mean value of a day production efficiency, a variance value of a day production efficiency, a mean value of a day vibration, a variance value of a day vibration, a mean value of a day sound intensity, and a variance value of a day sound intensity of each production device in each working day;
it can be understood that the temperature, production efficiency, vibration value, and mean and variance of sound intensity of the production apparatus represent the production state and stability of the production state of the production apparatus on the same day, respectively;
in another preferred embodiment of the present invention, the production characteristic data may further include a difference value of a current day temperature mean value and a previous day temperature mean value, a difference value of a current day temperature variance value and a previous day temperature variance value, a difference value of a current day production efficiency mean value and a previous day production efficiency mean value, a difference value of a current day production efficiency variance value and a previous day production efficiency variance value, a difference value of a current day vibration mean value and a previous day vibration mean value, a difference value of a current day vibration variance value and a previous day vibration variance value, a difference value of a current day sound intensity mean value and a previous day sound intensity mean value, and a difference value of a current day sound intensity variance value and a previous day sound intensity variance value in each working day;
it can be understood that by comparing the temperature, production efficiency, vibration value, and mean value and variance value of sound intensity of the production apparatus with the differences of the same characteristics of the previous day, the abnormality of the production apparatus on the current day can be estimated;
further, the use time length of each production device is obtained by timing the use time length of each production device;
the temperature, the vibration value and the sound intensity of each production device are respectively obtained in real time through a temperature sensor, a vibration sensor and a sound intensity sensor;
the production efficiency is obtained by calculating the number of products produced by each production device in real time; the number of the products corresponds to the number of the output products of the production equipment, and particularly, the output products and the statistical method of the number of the output products are determined according to the type of the production equipment; for example: for the grinder, the product quantity is the quantity of the powder which is output by the grinder and ground into the raw material;
further, the values of the fault tag data comprise 0 and 1, and when the fault tag data are 0, the fault tag data indicate that the production equipment corresponding to the corresponding production characteristic data does not have faults in the next day; if the fault label data is 1, the production equipment corresponding to the corresponding production characteristic data is indicated to be faulty in the next day;
the historical data collection module sends historical training data of production equipment of each production link in the production line to the fault probability prediction model training module;
the failure probability prediction model training module is mainly used for training a machine learning model for predicting the next-day failure probability of the equipment based on historical training data;
in a preferred embodiment, the machine learning model that predicts the probability of device failure the next day is trained in the following manner:
marking the number of the production link as i, and producing the ith production linkThe serial number of the production equipment type of the link is marked as ij; marking a production characteristic data set corresponding to the ij type production equipment as Fij, and marking a fault tag data set corresponding to the production characteristic data Fij as Gij; marking the serial number of the production characteristic data in the production characteristic data set as fij; it will be appreciated that the failure tag data setMiddle->The failure tag data corresponds to->Individual production characteristic data;
for the firstThe production equipment combines each group of production characteristic data in the production characteristic data set Fij into a characteristic vector form;
the elements in the feature vector include a use duration, a current day temperature average value, a current day temperature variance value, a current day production efficiency average value, a current day production efficiency variance value, a current day vibration average value, a current day vibration variance value, a current day sound intensity average value, a current day sound intensity variance value,
or the difference value of the current day temperature mean value and the previous day temperature mean value, the difference value of the current day temperature variance value and the previous day temperature variance value, the difference value of the current day production efficiency mean value and the previous day production efficiency mean value, the difference value of the current day production efficiency variance value and the previous day production efficiency variance value, the difference value of the current day vibration mean value and the previous day vibration mean value, the difference value of the current day vibration variance value and the previous day vibration variance value, the difference value of the current day sound intensity mean value and the previous day sound intensity mean value and the difference value of the current day sound intensity variance value and the previous day sound intensity variance value;
the set of all feature vectors is taken as input to a machine learning model that takes as output the predicted probability of failure for each set of production feature data, for the thThe group production characteristic data takes fij group fault label data as a prediction target, and takes the sum of prediction error degrees of all the predicted fault probabilities as a training target; the calculation formula of the prediction error degree is as follows; />Wherein->For the prediction error degree, afij is the predicted failure probability corresponding to the fij-th group of production characteristic data,/I>Production of a failure tag data set corresponding to characteristic data for the fij th group +.>The fij-th fault tag data; training the machine learning model until the sum of the prediction error degrees reaches convergence, and stopping training; preferably, the machine learning model is any one of a deep neural network model or a deep belief network model; thereby generating a corresponding machine learning model for each production facility;
it should be noted that, other model parameters of the machine learning model, such as the depth of the network model, the number of neurons in each layer, the activation function used by the network model, the convergence condition, the verification set proportion of the training set test set, the loss function and the like are all realized through actual engineering, and are obtained after experimental tuning is continuously performed;
the failure probability prediction model training module sends the trained machine learning model to the failure probability prediction module;
the production line data collection module is mainly used for collecting equipment relation data of a production line to be monitored and production characteristic data of the previous working day;
in a preferred embodiment, the device relationship data comprises a set of dependencies for each production device;
specifically, for the ij-th production equipment, marking a dependency relationship set as Dij, wherein each element in the dependency relationship set Dij is the number and the production ratio of part of the production equipment types in the i-1 production links, and each production equipment corresponding to each number in the number of the part of the production equipment types meets the conditions: the product output by the equipment is required to be processed by using the ij production equipment; it can be appreciated that the capacity of each production facility in the set of dependencies Dij affects the capacity of the ij-th production facility;
it should be noted that, for the type of production equipment in the 1 st production link, the corresponding set of dependency relationships is empty; for the production equipment types of the 3 rd production link, the elements in the corresponding dependency relation set are all the production equipment types of the 2 nd production link;
the serial numbers of the production equipment types in the dependency relation set Dij are marked as Dij, and the production proportion of the production equipment types Dij refers to the output product output required to be provided by the production equipment types Dij in each unit output product of the production equipment types ij; labeling the yield as Bdij;
the method for collecting the production characteristic data of the previous working day of the production line to be monitored is as follows:
before each working day of a production line to be monitored starts, collecting production characteristic data of each production device in the previous working day;
it can be understood that the collection method still uses a temperature sensor, a vibration sensor and a sound intensity sensor to obtain the temperature, the vibration value and the sound intensity value of each production device in real time in the previous working day, and then obtains the production characteristic data based on the temperature, the vibration value and the sound intensity value;
the production line data collection module sends production characteristic data of the production line to be monitored on the previous working day to the fault probability prediction module;
the fault probability prediction module is mainly used for predicting the fault probability of each production equipment on the current working day by using a machine learning model and production characteristic data of the previous working day on a production line to be monitored;
in a preferred embodiment, the probability of failure for each production facility on the current workday is predicted by:
the method comprises the steps of marking the number of ij production equipment in a production line to be monitored as Nij, and marking the number of the production equipment in the ij production equipment as Nij;
for the nij production equipment, converting the production characteristic data of the previous working day into a characteristic vector form, and inputting the characteristic vector into a machine learning model corresponding to the nij production equipment to obtain a predicted value of the probability of failure of the nij production equipment on the current day, which is output by the machine learning model;
the failure probability prediction module sends a predicted value of the failure probability of each production device in the production line to be monitored to the productivity prediction module;
the productivity prediction module is mainly used for predicting the maximum productivity of the production line to be monitored on the current working day, so that the number of the machining pieces on the current working day is automatically planned;
in a preferred embodiment, the maximum capacity of the line to be monitored on the same day is predicted by:
marking a predicted value of the current day fault probability of the nij production equipment as Pnij
Acquiring a current day production efficiency average value from production characteristic data of the previous working day of the nij production equipment, and marking the current day production efficiency average value of the previous working day of the nij production equipment as Xnij;
presetting the production time length T of the current working day;
in the 1 st link, calculating a total expected production value E1j of the 1 st production equipment, wherein a calculation formula of the total expected production value E1j is as follows；
Calculating the total expected production value E2j of the 2 nd production equipment in the 2 nd link, wherein the total expected production value；
Further calculating the maximum expected yield Rd2j provided by the D2j production equipment in the dependency relation set D2j of the 2j production equipment, wherein the calculation formula of the maximum expected yield Rd2j is as followsThe method comprises the steps of carrying out a first treatment on the surface of the Wherein->For the total expected production value corresponding to the production facility type of the 1 st production link corresponding to the production facility d2j, i.e. +.>May correspond to a certain value of the values E1 j; obviously, during the production process, the total production value of the production equipment of the 2 j-th type is limited by the maximum expected production value Rd2 j;
traversing the minimum value of the maximum expected production value Rd2j corresponding to each production device in the dependency relation set D2j, marking the minimum value as M2j, and enabling the maximum expected production value R2j of the 2 j-th production device to be the minimum value of E2j and M2 j;
calculating the total expected production value E3j of the 3j production equipment in the 3 rd link；
Further calculating the maximum expected yield Rd3j provided by the D3j production equipment in the dependency relation set D3j of the 3j production equipment, wherein the calculation formula of the maximum expected yield Rd3j is as followsThe method comprises the steps of carrying out a first treatment on the surface of the Wherein->The maximum expected production value for the production facility type of the 2 nd production link for the production facility d3j, i.e. +.>May correspond to a certain value in R2 j; obviously, in real production, the total production value of the 3 j-th production device is limited by the maximum expected production value Rd3 j;
traversing the minimum value of the maximum expected production value Rd3j corresponding to each production device in the dependency relation set D3j, marking the minimum value as M3j, and enabling the maximum expected production value R3j of the 3 j-th production device to be the minimum value of E3j and M3 j;
the maximum expected production value R3j of the production equipment of type 3j is set to the maximum capacity of the production line to be monitored on the same day.
The above embodiments are only for illustrating the technical method of the present invention and not for limiting the same, and it should be understood by those skilled in the art that the technical method of the present invention may be modified or substituted without departing from the spirit and scope of the technical method of the present invention.
Claims (5)
1. The automatic production line control system based on the Internet of things is characterized by comprising a historical data collection module, a fault probability prediction model training module, a production line data collection module, a fault probability prediction module and a productivity prediction module; wherein, each module is connected by a wired and/or wireless network mode;
the historical data collection module is used for collecting historical training data of production equipment of each production link in the production line in advance and sending the historical training data of the production equipment of each production link in the production line to the fault probability prediction model training module;
the failure probability prediction model training module trains a machine learning model for predicting the next-day failure probability of the equipment based on the historical training data, and sends the trained machine learning model to the failure probability prediction module;
the production line data collection module is used for collecting equipment relation data of the production line to be monitored and production characteristic data of the previous working day, and sending the production characteristic data of the previous working day of the production line to be monitored to the fault probability prediction module; the device relationship data comprises a set of dependency relationships for each production device;
the production line to be monitored uses the machine learning model and the production characteristic data of the previous working day to predict the fault probability of each production device on the current working day, and sends the predicted value of the fault probability of each production device to the productivity prediction module in the production line to be monitored;
the productivity prediction module predicts the maximum productivity of the production line to be monitored on the current working day of the production line to be monitored;
for the ij-th production equipment, marking a dependency relation set as Dij, wherein each element in the dependency relation set Dij is the number and the production ratio of part of the production equipment types in the i-1 production link, and each production equipment corresponding to each number in the number of the part of the production equipment types meets the condition: the product output by the equipment is required to be processed by using the ij production equipment;
the serial numbers of the production equipment types in the dependency relation set Dij are marked as Dij, and the production proportion of the production equipment types Dij refers to the output product output required to be provided by the production equipment types Dij in each unit output product of the production equipment types ij; labeling the yield as Bdij;
the number of the production link is marked as i, and the number of the production equipment type of the ith production link is marked as ij; marking a production characteristic data set corresponding to the ij type production equipment as Fij, and marking a fault tag data set corresponding to the production characteristic data Fij as Gij; marking the serial number of the production characteristic data in the production characteristic data set as fij;
the mode of predicting the fault probability of each production device on the current working day is as follows:
the method comprises the steps of marking the number of ij production equipment in a production line to be monitored as Nij, and marking the number of the production equipment in the ij production equipment as Nij;
for the nij production equipment, converting the production characteristic data of the previous working day into a characteristic vector form, and inputting the characteristic vector into a machine learning model corresponding to the nij production equipment to obtain a predicted value of the probability of failure of the nij production equipment on the current day, which is output by the machine learning model;
the method for predicting the maximum capacity of the production line to be monitored on the same day is as follows:
marking a predicted value of the current day fault probability of the nij production equipment as Pnij
Acquiring a current day production efficiency average value from production characteristic data of the previous working day of the nij production equipment, and marking the current day production efficiency average value of the previous working day of the nij production equipment as Xnij;
presetting the production time length T of the current working day;
in the 1 st link, calculating a total expected production value E1j of the 1 st production equipment, wherein a calculation formula of the total expected production value E1j is as follows；
Calculating the total expected production value E2j of the 2 nd production equipment in the 2 nd link, wherein the total expected production value；
Calculating the maximum expected yield Rd2j provided by the D2j production equipment in the dependency relationship set D2j of the 2j production equipment, wherein the calculation formula of the maximum expected yield Rd2j is as followsThe method comprises the steps of carrying out a first treatment on the surface of the Wherein->For the total expected production value corresponding to the production facility type of the 1 st production link corresponding to the production facility d2j, i.e. +.>May correspond to a certain value of the values E1 j;
traversing the minimum value of the maximum expected production value Rd2j corresponding to each production device in the dependency relation set D2j, marking the minimum value as M2j, and enabling the maximum expected production value R2j of the 2 j-th production device to be the minimum value of E2j and M2 j;
calculating the total expected production value E3j of the 3j production equipment in the 3 rd link；
Calculating the maximum expected yield Rd3j provided by the D3j production equipment in the dependency relation set D3j of the 3j production equipment, wherein the calculation formula of the maximum expected yield Rd3j is as followsThe method comprises the steps of carrying out a first treatment on the surface of the Wherein->The maximum expected production value for the production facility type of the 2 nd production link for the production facility d3j, i.e. +.>May correspond to a certain value in R2 j;
traversing the minimum value of the maximum expected production value Rd3j corresponding to each production device in the dependency relation set D3j, marking the minimum value as M3j, and enabling the maximum expected production value R3j of the 3 j-th production device to be the minimum value of E3j and M3 j;
the maximum expected production value R3j of the 3 j-th production equipment is set as the maximum production capacity of the production line to be monitored on the same day.
2. The automated production line control system based on the internet of things according to claim 1, wherein the production links include a raw material processing link, a component manufacturing link, and a finished product production link;
the production equipment used in the raw material treatment link is a plurality of raw material treatment equipment, and different raw material treatment equipment is used for treating different raw materials or processing the raw materials differently;
the production equipment required to be used in the component manufacturing link is component processing equipment, and different components are produced by different component manufacturing equipment;
the production equipment used in the production link of the finished product is finished product production equipment.
3. The automated production line control system based on the internet of things according to claim 2, wherein the historical training data comprises a plurality of groups of production characteristic data and fault tag data corresponding to each production device in each production link;
for each production device of each production link, generating a group of production characteristic data and fault tag data corresponding to the characteristic data by each corresponding production device in each working day;
the production characteristic data comprise the use time length, the average value of the temperature on the day, the variance value of the temperature on the day, the average value of the production efficiency on the day, the variance value of the production efficiency on the day, the average value of vibration on the day, the variance value of vibration on the day, the average value of sound intensity on the day and the variance value of sound intensity on the day of each production device in each working day; or:
the production characteristic data comprise a difference value between a current day temperature mean value and a previous day temperature mean value, a difference value between a current day temperature variance value and a previous day temperature variance value, a difference value between a current day production efficiency mean value and a previous day production efficiency mean value, a difference value between a current day production efficiency variance value and a previous day production efficiency variance value, a difference value between a current day vibration mean value and a previous day vibration mean value, a difference value between a current day vibration variance value and a previous day vibration variance value, a difference value between a current day sound intensity mean value and a previous day sound intensity mean value and a difference value between a current day sound intensity variance value and a previous day sound intensity variance value in each working day;
the value of the fault tag data comprises 0 and 1, and the fault tag data is 0, which indicates that the production equipment corresponding to the corresponding production characteristic data does not have faults in the next day; and if the fault label data is 1, indicating that the production equipment corresponding to the corresponding production characteristic data is faulty in the next day.
4. The automated production line control system based on the internet of things of claim 3, wherein forFirst, theThe production equipment combines each group of production characteristic data in the production characteristic data set Fij into a characteristic vector form;
the set of all feature vectors is taken as input to a machine learning model that takes as output the predicted probability of failure for each set of production feature data, for the thThe group production characteristic data takes fij group fault label data as a prediction target, and takes the sum of prediction error degrees of all the predicted fault probabilities as a training target; the calculation formula of the prediction error degree is as follows; />Wherein->For the prediction error degree, afij is the predicted failure probability corresponding to the fij-th group of production characteristic data,/I>Production of a failure tag data set corresponding to characteristic data for the fij th group +.>The fij-th fault tag data; training the machine learning model until the sum of the prediction error degrees reaches convergence, and stopping training; the machine learning model is any one of a deep neural network model and a deep belief network model; thereby generating a corresponding machine learning model for each production facility.
5. The automated production line control system based on the internet of things according to claim 4, wherein the means for collecting production characteristic data of a previous workday of the production line to be monitored is:
before each working day of a production line to be monitored starts, production characteristic data of each production device in the previous working day are collected.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310594355.7A CN116339267B (en) | 2023-05-25 | 2023-05-25 | Automatic production line control system based on Internet of things |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310594355.7A CN116339267B (en) | 2023-05-25 | 2023-05-25 | Automatic production line control system based on Internet of things |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116339267A CN116339267A (en) | 2023-06-27 |
| CN116339267B true CN116339267B (en) | 2023-08-08 |
Family
ID=86884352
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310594355.7A Active CN116339267B (en) | 2023-05-25 | 2023-05-25 | Automatic production line control system based on Internet of things |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116339267B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117074845A (en) * | 2023-10-16 | 2023-11-17 | 广州市零脉信息科技有限公司 | Electronic component quality detection platform |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006178920A (en) * | 2004-11-29 | 2006-07-06 | Jfe Steel Kk | Method and device for simulating production distribution, and a production method |
| CN108375715A (en) * | 2018-03-08 | 2018-08-07 | 中国电力科学研究院有限公司 | A kind of distribution network line fault risk day prediction technique and system |
| CN109981328A (en) * | 2017-12-28 | 2019-07-05 | 中国移动通信集团陕西有限公司 | A kind of fault early warning method and device |
| CN110334728A (en) * | 2019-05-06 | 2019-10-15 | 中国联合网络通信集团有限公司 | A kind of fault early warning method and device towards industry internet |
| CN111461555A (en) * | 2020-04-02 | 2020-07-28 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Production line quality monitoring method, device and system |
| CN112198857A (en) * | 2020-12-08 | 2021-01-08 | 浙江中自庆安新能源技术有限公司 | Industrial equipment control optimization method and system based on monitoring data |
| CN113393211A (en) * | 2021-06-22 | 2021-09-14 | 柳州市太启机电工程有限公司 | Method and system for intelligently improving automatic production efficiency |
| CN114298423A (en) * | 2021-12-30 | 2022-04-08 | 北京深度奇点科技有限公司 | APS dynamic production plan scheduling algorithm |
| EP4033421A1 (en) * | 2021-01-26 | 2022-07-27 | Arisk S.r.l. | Method and system for predicting a failure of a monitored entity |
| CN115730815A (en) * | 2022-10-10 | 2023-03-03 | 江苏安胜达安全科技有限公司 | Industrial production safety risk early warning system and method based on Internet of things perception |
| CN115860714A (en) * | 2022-11-28 | 2023-03-28 | 珠海德瑞斯科技有限公司 | Power equipment safe operation management system and method based on industrial Internet |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11232368B2 (en) * | 2019-02-20 | 2022-01-25 | Accenture Global Solutions Limited | System for predicting equipment failure events and optimizing manufacturing operations |
-
2023
- 2023-05-25 CN CN202310594355.7A patent/CN116339267B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006178920A (en) * | 2004-11-29 | 2006-07-06 | Jfe Steel Kk | Method and device for simulating production distribution, and a production method |
| CN109981328A (en) * | 2017-12-28 | 2019-07-05 | 中国移动通信集团陕西有限公司 | A kind of fault early warning method and device |
| CN108375715A (en) * | 2018-03-08 | 2018-08-07 | 中国电力科学研究院有限公司 | A kind of distribution network line fault risk day prediction technique and system |
| CN110334728A (en) * | 2019-05-06 | 2019-10-15 | 中国联合网络通信集团有限公司 | A kind of fault early warning method and device towards industry internet |
| CN111461555A (en) * | 2020-04-02 | 2020-07-28 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Production line quality monitoring method, device and system |
| CN112198857A (en) * | 2020-12-08 | 2021-01-08 | 浙江中自庆安新能源技术有限公司 | Industrial equipment control optimization method and system based on monitoring data |
| EP4033421A1 (en) * | 2021-01-26 | 2022-07-27 | Arisk S.r.l. | Method and system for predicting a failure of a monitored entity |
| CN113393211A (en) * | 2021-06-22 | 2021-09-14 | 柳州市太启机电工程有限公司 | Method and system for intelligently improving automatic production efficiency |
| CN114298423A (en) * | 2021-12-30 | 2022-04-08 | 北京深度奇点科技有限公司 | APS dynamic production plan scheduling algorithm |
| CN115730815A (en) * | 2022-10-10 | 2023-03-03 | 江苏安胜达安全科技有限公司 | Industrial production safety risk early warning system and method based on Internet of things perception |
| CN115860714A (en) * | 2022-11-28 | 2023-03-28 | 珠海德瑞斯科技有限公司 | Power equipment safe operation management system and method based on industrial Internet |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116339267A (en) | 2023-06-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106598791B (en) | Industrial equipment fault preventive identification method based on machine learning | |
| CN116339267B (en) | Automatic production line control system based on Internet of things | |
| WO2016086360A1 (en) | Wind farm condition monitoring method and system | |
| CN107122832A (en) | A kind of manufacture system preventative maintenance method of Quality Control And Reliability analysis | |
| CN113552840B (en) | Machining control system | |
| CN110310134B (en) | Customized furniture management tracing method based on coding information association | |
| DE102016015332A1 (en) | A preventive maintenance management system and method for creating a maintenance plan of a machine and cell controller | |
| CN115526482A (en) | Workshop scheduling and dynamic scheduling platform based on digital twin | |
| Aleš et al. | Methodology of overall equipment effectiveness calculation in the context of industry 4.0 environment | |
| CN116316613B (en) | Power equipment operation monitoring method, system, electronic equipment and storage medium | |
| TWI663569B (en) | Quality prediction method for multi-workstation system and system thereof | |
| CN112801313A (en) | Fully mechanized mining face fault judgment method based on big data technology | |
| Wang et al. | Intelligent predictive maintenance (IPdM) for elevator service-through CPS, IOT&S and data mining | |
| DE102017001578A1 (en) | A cell controller that displays an abnormal condition of a manufacturing machine for each area or operation | |
| CN112418548A (en) | Intelligent workshop management method and system based on artificial intelligence technology | |
| CN116562599A (en) | Efficient intelligent factory scheduling system based on deep learning | |
| CN116467592A (en) | Production equipment fault intelligent monitoring method and system based on deep learning | |
| FOURLAS | Reliability and maintainability analysis of a robotic system for industrial applications: a case study | |
| CN113836820A (en) | Equipment health assessment and fault diagnosis algorithm based on autocorrelation model and multivariate monitoring method | |
| Tabit et al. | Machine Learning: Strategies for industrial defect detection | |
| EP3961333A1 (en) | System and method for determining a cause of an operating anomaly of a machine and computer program and electronically readable data carrier | |
| CN112748663A (en) | Wind power torque fault-tolerant control method based on data-driven output feedback | |
| CN116976862B (en) | Factory equipment informatization management system and method | |
| Roy et al. | Investigation of Production Line Defects Using Root Cause Analysis: A Case Study on An Automobile Industry In Bangladesh | |
| CN116551467B (en) | Cutter abrasion state identification method based on internet of things monitoring data |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |