CN109739902A

CN109739902A - A kind of data analysing method, equipment and computer readable storage medium

Info

Publication number: CN109739902A
Application number: CN201811644367.1A
Authority: CN
Inventors: 马辉; 夏蕴; 杨汇成
Original assignee: Lenovo Beijing Ltd
Current assignee: Lenovo Beijing Ltd
Priority date: 2018-12-29
Filing date: 2018-12-29
Publication date: 2019-05-10

Abstract

The embodiment of the present invention discloses a kind of data analysing method, it include: the first information and corresponding first parameter of the object to be trained for obtaining the object to be trained for being determined as substandard products, the first information, which is used to characterize, causes the object to be trained to be the factor information of substandard products, and first parameter, which is used to characterize, produces the historical production data that the first production equipment of the object to be trained generates in the first predetermined amount of time；Substandard products sample set is generated based on the first information and first parameter；When monitoring object to be detected is substandard products, corresponding second parameter of the object to be detected is obtained, second parameter produces the first real-time production data that the second production equipment of the object to be detected generates in the second predetermined amount of time for characterizing；Based on second parameter and the substandard products sample set, obtaining causes the object to be detected to be the target factor information of substandard products.The embodiment of the present invention further simultaneously discloses a kind of data analysis equipment and computer readable storage medium.

Description

Data analysis method and device and computer readable storage medium

Technical Field

The invention relates to an optical fiber manufacturing technology in the field of industrial automation control, in particular to a data analysis method, data analysis equipment and a computer readable storage medium.

Background

Optical fiber is a short term optical fiber, a fiber made of glass or plastic, and has wide application as a basic material for current communication transmission. At present, the manufacturing of optical fiber is mainly completed through a drawing tower, and the production process from the production of a raw material preform to the drawing and coiling of the optical fiber mainly comprises the production process flows of rod feeding, heating, drawing, cooling, diameter measuring, coating, solidifying, secondary cooling, wire winding and the like. In the optical fiber manufacturing process, the problems of discontinuous and uneven drawing in the optical fiber drawing process are caused by various reasons such as deviation and production environment factors caused by the setting of the control system parameters of the drawing tower, and finally the occurrence of fiber breakage is caused.

In the actual production process, in order to avoid the problems, the experienced engineering personnel are mainly used for judging the reason of fiber breakage according to the known fiber breakage data and the production experience, and then the reason is fed back to the production process to correct the parameters of the control system, so that the fiber breakage rate is reduced. The method for manually judging the reason of the broken fiber has the problems that the historical data of the broken fiber is not fully utilized, and the analysis on the complex cause of the broken fiber is incomplete.

Disclosure of Invention

In order to solve the above technical problems, embodiments of the present invention are expected to provide a data analysis method, a data analysis device, and a computer-readable storage medium, so as to solve the problem of incomplete analysis of a fiber breakage complex cause in optical fiber production, and achieve quick and accurate acquisition of a fiber breakage cause without manual judgment when a fiber breakage occurs, thereby saving labor cost and time.

In order to achieve the above purpose, the technical solution of the embodiment of the present invention is realized as follows:

the embodiment of the invention provides a data analysis method, which comprises the following steps:

acquiring first information of an object to be trained determined to be a defective object and a first parameter corresponding to the object to be trained, wherein the first information is used for representing factor information causing the object to be trained to be a defective object, and the first parameter is used for representing historical production data generated by first production equipment for producing the object to be trained in a first preset time period;

generating a set of substandard product samples based on the first information and the first parameter;

when the detected object is a defective product, acquiring a second parameter corresponding to the detected object, wherein the second parameter is used for representing first real-time production data generated by second production equipment for producing the detected object in a second preset time period;

and acquiring target factor information which causes the object to be detected to be a defective product based on the second parameter and the defective product sample set.

In the foregoing scheme, the acquiring first information of the object to be trained determined to be an inferior product and a first parameter corresponding to the object to be trained includes:

acquiring first information of the object to be trained;

acquiring historical control data of the first production equipment, historical environment data corresponding to the object to be trained and historical raw material information corresponding to the object to be trained within a first preset time period;

correspondingly, the generating a set of substandard product samples based on the first information and the first parameter includes:

and inputting the first information, the historical control data, the historical environment data and the historical raw material information into a preset model, and generating a defective sample set through the preset model.

In the above scheme, when it is monitored that the object to be detected is a defective product, acquiring a second parameter corresponding to the object to be detected includes:

when the detected object is a defective product, acquiring first real-time control data of the second production equipment, first real-time environment data corresponding to the detected object and first real-time raw material information corresponding to the detected object within a second preset time period; wherein the second parameters comprise the first real-time control data, the first real-time environmental data, and the first real-time raw material information;

correspondingly, the obtaining the reason that the object to be detected is a defective product based on the second parameter and the defective product sample set includes:

acquiring matched factor information from the defective sample set based on the first real-time control data, the first real-time environment data and the first real-time raw material information to obtain the target factor information; and the defective product sample set comprises control data of the production equipment, environmental data of the product and the corresponding relation between the raw material information of the product and the factor information causing the product to be defective in a preset time period.

In the above scheme, the method further comprises:

if the defective sample set does not have matched target factor information, generating prompt information for prompting the input of the reasons of the defective samples;

and acquiring input information aiming at the prompt information, and updating the first information based on the input information.

In the above scheme, the method further comprises:

generating an early warning threshold based on the first parameter;

when the object to be detected is monitored not to be a defective product, acquiring a third parameter corresponding to the object to be detected, wherein the third parameter is used for representing second real-time production data currently generated by the second production equipment;

generating a predicted value based on the third parameter;

and if the predicted value is greater than or equal to the early warning threshold value, generating warning information for prompting that the object to be detected has a defective product risk.

In the foregoing solution, the generating an early warning threshold based on the first parameter includes:

acquiring attribute parameters of the object to be trained in a first preset time period from the first parameters;

generating a time sequence model for outputting the change of the production efficiency along with the time based on the attribute parameters and the first parameters;

and processing the time sequence model to generate the production efficiency corresponding to the object to be trained, and generating the early warning threshold value based on the production efficiency.

In the foregoing scheme, the obtaining of the third parameter corresponding to the object to be detected includes:

acquiring second real-time control data of third production equipment at a real-time point, second real-time environment data corresponding to the object to be detected and real-time raw material information corresponding to the object to be detected; the third parameters comprise second real-time control data, second real-time environment data and second real-time raw material information;

correspondingly, the generating a predicted value based on the third parameter includes:

and inputting the third parameter into the time sequence model, obtaining real-time production efficiency through the time sequence model, and generating the predicted value based on the real-time production efficiency.

An embodiment of the present invention provides a data analysis device, where the device includes: a processor, a memory, and a communication bus; wherein,

the communication bus is used for realizing communication connection between the processor and the memory;

the processor is configured to execute a data analysis program stored in the memory to implement the steps of:

In the foregoing solution, the processor is further configured to:

acquiring first information of the object to be trained;

Embodiments of the present invention provide a computer readable storage medium storing one or more programs, which are executable by one or more processors to implement the steps of the data analysis method according to any one of the first aspect.

According to the data analysis method, the data analysis equipment and the computer-readable storage medium, first information of an object to be trained determined to be a defective item and a first parameter corresponding to the object to be trained are obtained; generating a set of substandard product samples based on the first information and the first parameter; when the detected object is a defective product, acquiring a second parameter corresponding to the detected object; acquiring target factor information which causes the object to be detected to be a defective product based on the second parameter and the defective product sample set; therefore, the fiber breakage reason of the current fiber breakage is obtained by establishing a fiber breakage sample set corresponding to the fiber breakage reason and the fiber breakage production data based on the historical data of the fiber breakage, and comparing the real-time fiber breakage production data with the historical fiber breakage production data in the fiber breakage sample set when the fiber breakage occurs in real-time production. Because the reason for fiber breakage does not need to be judged manually, the labor cost and time in the optical fiber production process are saved.

Drawings

Fig. 1 is a schematic flow chart of a data analysis method according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of a neural network model according to an embodiment of the present invention;

FIG. 3 is a schematic flow chart of another data analysis method according to an embodiment of the present invention;

FIG. 4 is a schematic flow chart of another data analysis method according to an embodiment of the present invention;

FIG. 5 is a schematic flow chart of a data analysis method according to another embodiment of the present invention;

FIG. 6 is a schematic flow chart of another data analysis method according to another embodiment of the present invention;

FIG. 7 is a flow chart illustrating a further data analysis method according to another embodiment of the present invention;

FIG. 8 is a flow chart illustrating a data analysis method according to another embodiment of the present invention;

fig. 9 is a schematic diagram of a hardware structure of a data analysis device according to an embodiment of the present invention.

Detailed Description

The technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

An embodiment of the present invention provides a data analysis method, which is applied to data analysis equipment, and as shown in fig. 1, the data analysis method of the present embodiment includes the following steps:

step 101: the method comprises the steps of obtaining first information of an object to be trained determined to be a defective object and first parameters corresponding to the object to be trained, wherein the first information is used for representing factor information causing the object to be trained to be the defective object, and the first parameters are used for representing historical production data generated by first production equipment for producing the object to be trained in a first preset time period.

It is understood that, in the embodiment of the present invention, the object to be trained refers to an optical fiber, and if a fiber breakage occurs during the production process, the length of the optical fiber is determined as a defective fiber because the length of the optical fiber cannot reach the production requirement.

It is understood that in step 101, the first information refers to the reason for fiber breakage, and the information may be the reason for fiber breakage obtained through manual analysis and is stored in a database of the production equipment or other storage equipment in advance. Here, several factors that are common in optical fiber production leading to fiber breakage are listed, including but not limited to: 1) poor preform: fiber breakage occurs in drawing due to microcracks and foreign particles on the surface of the preform; 2) and (3) the environment of the wire drawing furnace: because the internal temperature of the wire drawing furnace is above 2000 ℃, and the heating parts in the wire drawing furnace are all made of graphite materials, particles can be generated due to long-time high temperature, and the prefabricated rod can be sublimated to form impurity particles during melting, so that fiber breakage can be caused by the impurity particles; 3) and (3) wire drawing channel environment: in the process of drawing the optical fiber, the optical fiber often passes through air at a high speed of 1500 m/min or even 2000 m/min, the cleanliness of the air in the channel greatly affects the strength of the optical fiber, and suspended particles in the air can rub the surface of the optical fiber, so that the strength of the optical fiber is reduced, even the surface of the optical fiber is cracked, and finally the fiber breakage is caused; 4) the quality of the optical fiber coating is as follows: the particles of the optical fiber coating can also cause damage to the surface of an optical fiber, so that the strength of the optical fiber is reduced, generally, the particles with the diameter of 1-3 pm have no influence on the strength of the optical fiber, but the particles with the diameters of 6-10 pm and 15-20 pm can reduce the strength of the optical fiber, so that the use of the coating which does not reach the standard can also cause fiber breakage; 5) the process conditions are as follows: the temperature control, the wire drawing tension control, the coating thickness selection and the process concentricity of the wire drawing furnace are all related to the strength of the optical fiber, and when the process conditions of the wire drawing furnace are changed, the fiber breakage can be probably caused; 6) personnel control: control errors of production operators on production equipment can directly cause fiber breakage. In view of the above situation, after the fiber breakage is detected and analyzed, the production personnel can obtain and record the reason for the fiber breakage.

It is understood that in step 101, the first parameter refers to production data of the optical fiber production equipment within a first predetermined time before fiber breakage. Because the fiber drawing speed in the drawing tower is very high, the first preset time is preferably a timing unit of seconds, such as 1 second or 5 seconds, and the analysis and comparison of the fiber breakage data by production personnel are facilitated. The production data refers to various data related to the wire drawing production, and can be directly acquired by the production equipment, including but not limited to control data of the wire drawing equipment and environmental data of the production environment. Taking a wire drawing furnace as an example, the control data can be control parameters such as heating temperature, wire drawing speed, cooling temperature and the like, the environmental data can be environmental parameters such as temperature, humidity, air pressure value and the like of a production environment, and in addition, the environmental data can also comprise preform batch and other raw material information. For other specific production data related to the production of optical fibers, further description is omitted here.

Optionally, the first predetermined time may be other time duration, such as related to the internal sampling period of the optical fiber production equipment, and may be adjusted and modified in the production practice, which is not described herein again.

Step 102: a set of substandard sample is generated based on the first information and the first parameter.

It can be understood that by acquiring the data of the cause of fiber breakage and the production data corresponding to the fiber breakage, a fiber breakage sample set corresponding to the production data of the cause of fiber breakage and the fiber breakage can be established. The sample set is stored in an analysis database, and model iteration training is performed by using samples in the fiber breaking sample set, so that a classification model about fiber breaking can be obtained.

Specifically, in the embodiment of the present invention, a Neural Network model (ANN) is used to perform model iterative training on a broken fiber sample, and referring to a schematic diagram of the Neural Network model shown in fig. 2, it can be understood that in the Neural Network model, x1, x2, and x3 refer to production data corresponding to a first parameter in the embodiment of the present invention, and are located in an input layer of the Neural Network model; wherein w1, w2 and w3 refer to equipment control data in the production data corresponding to x1, x2 and x3, and r1, r2 and r3 refer to production environment data in the production data corresponding to x1, x2 and x 3; y1 and y2 are fiber breakage factors corresponding to the first information in the embodiment of the invention, the fiber breakage factors are used as feature labels, and a function s of a second layer can be obtained through x data and y data and is used for representing different fiber breakage classifications; finally, the model function z(s) of the output layer of the neural network model can be obtained by carrying out iteration and classification algorithm calculation on all s.

It can be understood that by further refining various parameters in the production data, the hierarchy of the neural network can be further increased under the condition that the condition allows, so as to realize more accurate classification of the fiber breakage condition, and further description is omitted here.

Optionally, the Neural Network model in the embodiment of the present invention may select any one of a BP Neural Network (BPNN), a Hopfield Network (HN), an ART Network (ART n), a Kohonen Network (KN), and the like, and details thereof are not repeated herein.

Step 103: when the detected object is a defective product, acquiring a second parameter corresponding to the detected object, wherein the second parameter is used for representing first real-time production data generated by second production equipment for producing the detected object in a second preset time period.

Step 104: and acquiring target factor information which causes the object to be detected to be a defective product based on the second parameter and the defective product sample set.

It will be appreciated that historical data for fiber breakage is obtained in steps 101 and 102, and real-time data for fiber production is obtained in step 103. The second parameter is the production data obtained from the optical fiber production equipment in the second predetermined time period when the fiber breakage occurs in the real-time production. The specific content and acquisition method of the production data are the same as those in step 101.

Optionally, the second predetermined time here may be the same time length as the first predetermined time in step 101, or may be another time length, and is not described here again.

It can be understood that after the production data within the preset time is obtained, the data can be brought into the neural network model, and the classification result of the fiber breakage reason is obtained through the neural network model algorithm, so that the manual analysis process of the fiber breakage reason is not needed, and the time and the labor are saved.

Through the technical scheme provided by the embodiment corresponding to fig. 1, it can be seen that the embodiment of the present invention obtains the cause of fiber breakage and the production data in the predetermined time period before the occurrence of fiber breakage, substitutes the cause of fiber breakage into the neural network model for training to obtain the network model classified by the cause of fiber breakage, and substitutes the production data in the predetermined time period into the network model again when a new fiber breakage occurs, so as to quickly and accurately obtain the cause of fiber breakage, thereby avoiding the process of manually performing fiber breakage analysis, and saving labor and time. And foreseeably, each broken fiber occurrence can be changed into historical data to be added into the classification model of the broken fiber cause, so that broken fiber samples can be increased more and more, and the analysis of the broken fiber cause is more and more accurate.

Based on the foregoing embodiments, an embodiment of the present invention provides a data analysis method, as shown in fig. 3, in an implementation manner of the method embodiment, the method includes the following steps:

step 201: acquiring first information of an object to be trained;

step 202: acquiring historical control data of first production equipment, historical environment data corresponding to an object to be trained and historical raw material information corresponding to the object to be trained in a first preset time period;

step 203: and inputting the first information, the historical control data, the historical environment data and the historical raw material information into a preset model, and generating a defective sample set through the preset model.

It is understood that the object to be trained in the embodiment of the present invention refers to an optical fiber, and if a fiber breakage occurs during the production process, the length of the optical fiber is determined as a defective fiber because the length of the optical fiber cannot reach the production requirement.

It is understood that the first information in the embodiment of the present invention refers to a cause of fiber breakage, and the information may be a reason for fiber breakage obtained through manual analysis and stored in a predetermined storage device.

It is understood that the first parameter in the embodiment of the present invention refers to the production data of the optical fiber production equipment within the first predetermined time before fiber breakage, and includes but is not limited to: historical control data of the optical fiber production equipment, historical environmental data when fiber breakage occurs and historical raw material information when the fiber breakage occurs. Since the speed of drawing the optical fiber is very fast, the predetermined time is preferably a time duration in seconds, for example, 1 second or 5 seconds, which can improve the comparison efficiency of the production data.

It will be appreciated that production data is often stored in a storage medium of the optical fiber production facility and can be read directly. The optical fiber production equipment mainly refers to a drawing tower, and also refers to other production equipment for optical fiber production. In production data, control data includes, but is not limited to: furnace temperature, concentricity, drawing speed, cooling temperature, and the like, and environmental data including, but not limited to: temperature, humidity, and air pressure values in the production environment, and the raw material information includes but is not limited to: preform lot and coating lot information. It should be noted here that all process data related to the fiber drawing production should be considered and incorporated into the production data in the practice of the present invention, and will not be described in detail here.

It can be understood that the preset model in the embodiment of the present invention is a neural network model, based on obtaining production data related to fiber breakage in optical fiber production and known fiber breakage reason data, the production data can be used as output layer data, the fiber breakage reason data is used as a feature value, the feature value is brought into the preset neural network model for training, and a calculation function for classifying fiber breakage causes is obtained through iterative operation. Through continuous training and classification of the model, a broken fiber sample set can be determined based on known broken fiber historical data.

Optionally, the Neural Network model in the embodiment of the present invention may select any one of a BP Neural Network (BPNN), a Hopfield Network (HN), an ART Network (ART n, Adaptive response Network) and a Kohonen Network (KN, Kohonen Network), and the like, for the Neural Network model, the more the types of the input layer data are, the more the hierarchy of the model is, the more accurate the model function obtained by the output layer is, and details of a specific classification and calculation process of the Neural Network model are omitted here.

Step 204: when the detected object is a defective product, acquiring a second parameter corresponding to the detected object, wherein the second parameter is used for representing first real-time production data generated by second production equipment for producing the detected object in a second preset time period.

Step 205: and acquiring target factor information which causes the object to be detected to be a defective product based on the second parameter and the defective product sample set.

It is understood that, in the embodiment of the present invention, the method and content of step 204 and step 205 are the same as those of step 103 and step 104 in the foregoing embodiment, and are not described herein again.

Through the technical scheme provided by the embodiment corresponding to fig. 3, it can be seen that the fiber breakage cause data in the historical data, the historical control data, the historical environment data and the historical raw material information of the broken fibers within the preset time are acquired, and the data are input into the preset neural network model for training, so as to finally obtain the fiber breakage sample set classified by the fiber breakage cause, wherein the fiber breakage sample set comprises the corresponding relation among the control data of the production equipment, the environment data of the product, the raw material information of the product and the factor information causing the product to be defective within the preset time period.

Based on the foregoing embodiments, an embodiment of the present invention provides a data analysis method, as shown in fig. 4, in an implementation manner of the method embodiment, the method includes the following steps:

step 301: the method comprises the steps of obtaining first information of an object to be trained determined to be a defective object and first parameters corresponding to the object to be trained, wherein the first information is used for representing factor information causing the object to be trained to be the defective object, and the first parameters are used for representing historical production data generated by first production equipment for producing the object to be trained in a first preset time period.

Step 302: generating a set of substandard product samples based on the first information and the first parameter;

step 303: when the detected object is a defective product, acquiring first real-time control data of second production equipment in a second preset time period, first real-time environment data corresponding to the detected object and first real-time raw material information corresponding to the detected object; the second parameters comprise first real-time control data, first real-time environment data and first real-time raw material information;

step 304: acquiring matched factor information from the defective sample set based on the first real-time control data, the first real-time environment data and the first real-time raw material information to obtain target factor information; the defective sample set comprises control data of production equipment in a preset time period, environmental data of products and corresponding relation between raw material information of the products and factor information causing the products to be defective.

It is understood that, in the embodiment of the present invention, the method and content of step 301 and step 302 are the same as those of step 101 and step 102 in the foregoing embodiment, and are not described here again.

It is understood that the second parameter in the embodiment of the present invention is real-time production data within a second preset time period when fiber breakage occurs in production, where the second preset time period may be the same as the first preset time period, so as to unify validity of production data in comparison, where the real-time production data includes, but is not limited to, first real-time control data, second real-time control data, first real-time environment data, second real-time environment data, first real-time raw material information, and second real-time raw material information. The specific content of the first real-time control data, the second real-time control data, the first real-time environment data, the second real-time environment data, the first real-time raw material information, and the second real-time raw material information is the same as the data type and content of the historical production data in the foregoing embodiment, so as to facilitate the comparison process of the sample set, and the details are not repeated here.

By substituting the acquired first real-time control data, second real-time control data, first real-time environment data, second real-time environment data, first real-time raw material information and second real-time raw material information into the fiber breakage sample set established through the historical data, the fiber breakage factor information of the fiber breakage at this time can be obtained by comparing the same data types based on the corresponding relation between the production data stored in the fiber breakage sample set and the fiber breakage reason.

Through the technical scheme provided by the embodiment corresponding to fig. 4, it can be seen that when fiber breakage occurs in actual production, the real-time data of the production data is acquired and compared with historical data of the production data in a known fiber breakage sample set, so that factor information of the real-time fiber breakage can be acquired from the fiber breakage sample set. Therefore, extra labor is not required to be paid by production personnel in the production process for fiber breakage analysis, and time and production cost are greatly saved.

Based on the foregoing embodiments, embodiments of the present invention provide a data analysis method, which is expected to solve the problem that corresponding fiber breakage factor information cannot be obtained from a fiber breakage sample set, and as shown in fig. 5, in an implementation manner of the method embodiment, the method includes the following steps:

step 401: if the defective sample set does not have matched target factor information, generating prompt information for prompting the input of the cause of the defective;

step 402: input information for the prompt information is acquired, and the first information is updated based on the input information.

It can be understood that due to the limitation of the historical data, the fiber breakage factor information in the fiber breakage sample set obtained based on the fiber breakage historical data is limited. Therefore, fiber breaking factors which do not appear in the fiber breaking sample set may appear in the actual production process, so that the fiber breaking factor information corresponding to real-time fiber breaking cannot be obtained from the fiber breaking sample set generated based on historical data when fiber breaking occurs.

Based on the assumption, the embodiment of the invention provides a solution, when the fiber breaking sample set cannot be matched with the factor information of the fiber breaking at the time, the reason of the fiber breaking is identified in a manual mode, and the obtained fiber breaking factor information is input into the historical database, so that the real-time updating of the fiber breaking sample set is realized.

Specifically, the fiber breakage can be identified manually by using the prior art, for example, by observing the crack growth direction trajectory and the mirror position of the cross section through a high-power microscope, and the method is not described herein again.

Through the technical scheme provided by the embodiment corresponding to fig. 5, it can be seen that when the fiber breaking factor information cannot be obtained from the fiber breaking sample set, the fiber breaking factor information of the current fiber breaking is manually input, so that the fiber breaking sample set is continuously expanded and updated, and the problem that the sample set cannot be analyzed in the future when the same fiber breaking condition occurs is avoided.

Based on the foregoing embodiment, an embodiment of the present invention provides a data analysis method, as shown in fig. 6, in an implementation manner of the embodiment of the method, the data analysis method further includes the following steps:

step 501: generating an early warning threshold value based on the first parameter;

step 502: when the detected object is not a defective product, acquiring a third parameter corresponding to the detected object, wherein the third parameter is used for representing second real-time production data currently generated by second production equipment;

step 503: generating a predicted value based on the third parameter;

step 504: and if the predicted value is greater than or equal to the early warning threshold value, generating warning information for prompting that the object to be detected has a defective product risk.

It can be understood that the embodiment of the invention is expected to generate the predicted fiber breakage value and the early warning fiber breakage value by further mining the historical data of fiber breakage, and when the predicted value based on the real-time production data of the optical fiber reaches the early warning value, the embodiment gives an alarm to production personnel so as to prevent the occurrence of fiber breakage.

It is understood that the first parameter is production data within a predetermined time of occurrence of a fiber break in the historical data, including but not limited to historical control data, historical environmental data, and historical material information for the production equipment. Based on the historical data, common data closely related to fiber breakage can be analyzed and obtained, for example, through analysis of the temperature data of the drawing furnace, fiber breakage can be found when the temperature fluctuation value of the drawing furnace exceeds 20 ℃; through the analysis of the drawing speed data, the fiber breakage and the like can be found when the drawing speed exceeds 2000 m/min. Based on the common data, the production data containing the common data can be set as an early warning threshold, for example, a warning is set when the temperature fluctuation value in the drawing furnace exceeds 15 ℃, or a warning is set when the purity of the optical fiber raw material preform does not reach the average value requirement.

It can be understood that the third parameter in the embodiment of the present invention refers to real-time production data when no fiber breakage occurs, and based on the data, a predicted value can be generated for a preset alarm threshold. For example, if the temperature fluctuation value of the drawing furnace exceeds 15 ℃ and is set as an alarm threshold, a predicted value can be obtained according to the change data of the temperature of the drawing furnace in real-time production, and when the predicted value is greater than 15 ℃, a fiber breakage alarm is triggered.

It can be understood that, because the production data in the embodiment of the present invention is production data including multiple data types, a person skilled in the art may set multiple alarm thresholds according to each type of production data, or may perform comprehensive consideration according to multiple types of production data, for example, obtain one alarm threshold by using a regression algorithm or other operations to reduce the number of production pauses caused by alarms, thereby improving production efficiency.

Optionally, the reason for the alarm can be displayed during the alarm, so that the production personnel can be conveniently instructed to perform corresponding correction operation.

Through the technical scheme provided by the embodiment corresponding to fig. 6, it can be seen that an alarm threshold is set by further mining the history data of fiber breakage, and an alarm is issued to a producer when the real-time data reaches the alarm threshold, so that the producer can adjust the production process in time after receiving the alarm information, thereby fundamentally avoiding the occurrence of fiber breakage and realizing the saving of production cost.

Based on the foregoing embodiment, an embodiment of the present invention provides a data analysis method, as shown in fig. 7, in an implementation manner of the method embodiment, the generating an early warning threshold value based on a first parameter in step 501 may also be implemented by the following manner:

step 601: acquiring attribute parameters of an object to be trained in a first preset time period from the first parameters;

step 602: generating a time sequence model for outputting the change of the production efficiency along with the time based on the attribute parameters and the first parameters;

step 603: and processing the time sequence model to generate the production efficiency corresponding to the object to be trained, and generating an early warning threshold value based on the production efficiency.

It can be understood that the embodiment of the present invention expects to set the warning threshold for fiber breakage by establishing historical production data of fiber breakage and change information of fiber production efficiency at a corresponding time point.

Specifically, in the embodiment of the present invention, the production data in a predetermined time before the fiber breakage occurs in the historical data is obtained, for example, a curve of a temperature of the drawing furnace changing with time, a curve of a drawing speed changing with time, a curve of a humidity of a production environment changing with time, and the like in 5 seconds before the fiber breakage occurs.

It is expected that the fiber breakage is finally caused by the change of production data related to the fiber breakage cause, for example, a group of generated data related to the fiber breakage cause being the too fast drawing speed is the drawing speed within 5 seconds before the fiber breakage occurs, and by analyzing the time series data of the drawing speed, the trend that the drawing speed is continuously increased within 5 seconds is found, and the maximum value is reached immediately before the fiber breakage. The time sequence data can be combined with the production efficiency to generate a time sequence curve of the time sequence data of the fiber breaking historical data and the production efficiency. Training the historical data of each group of broken fibers and the production efficiency data to finally obtain a time sequence model for judging the broken fibers.

It can be understood that an early warning threshold value can be set based on the change condition of the production efficiency in the obtained time sequence model, and when the real-time production efficiency changes (specifically, when a descending trend occurs) and reaches the early warning threshold value, an alarm message is sent to a producer to prompt the producer to adjust and correct the production equipment so as to prevent fiber breakage.

Through the technical scheme provided by the embodiment corresponding to fig. 7, it can be seen that a time sequence model of which the production efficiency changes along with time can be finally obtained by further mining the history data of broken fibers and correlating the time sequence data with the production efficiency, an alarm threshold value can be set through the time sequence model, and an alarm is initiated to a producer when the real-time data reaches the alarm threshold value, so that the producer can adjust the production process in time after receiving the alarm information, thereby avoiding the occurrence of broken fibers and realizing the saving of the production cost.

Based on the foregoing embodiment, an embodiment of the present invention provides a data analysis method, as shown in fig. 8, in an implementation manner of the method embodiment, the obtaining of the third parameter corresponding to the object to be detected in step 502 may also be implemented in the following manner:

step 701: acquiring second real-time control data of third production equipment at a real-time point, second real-time environment data corresponding to the object to be detected and real-time raw material information corresponding to the object to be detected; the third parameters comprise second real-time control data, second real-time environment data and second real-time raw material information;

step 702: and inputting the third parameter into the time sequence model, obtaining the real-time production efficiency through the time sequence model, and generating a predicted value based on the real-time production efficiency.

It is understood that the embodiment of the present invention produces predicted values based on real-time data of the optical fiber.

In particular, real-time data in the practice of the present invention includes, but is not limited to: real-time control data, real-time environmental data, and real-time raw material information. By inputting these real-time data into the time-series model of the production efficiency with time, which is established in the foregoing embodiment, the curve information of the real-time production efficiency value with time can be obtained.

Optionally, the real-time production efficiency value obtained by the model may be directly used as the real-time predicted fiber breakage value, or the real-time predicted fiber breakage value may be obtained by subjecting the real-time production efficiency value to other mathematical changes.

By comparing the predicted value corresponding to the real-time production data with a preset alarm threshold value, when the predicted value reaches the alarm threshold value, alarm information is sent to production personnel to prompt the production personnel to adjust and correct production equipment so as to prevent fiber breakage.

Through the technical scheme provided by the embodiment corresponding to fig. 8, it can be seen that real-time production efficiency can be obtained by bringing real-time production data of the optical fiber into a preset production efficiency time sequence model, a fiber breakage predicted value is obtained based on the real-time production efficiency, and an alarm can be triggered when the predicted value reaches a threshold value through comparison between the fiber breakage predicted value and an alarm threshold value, so that a producer can adjust a production process in time after receiving alarm information, thereby avoiding fiber breakage and realizing saving production cost.

Referring to fig. 9, which shows a data analysis device provided by an embodiment of the present invention, the data analysis device 8 may include: a memory 82 and a processor 83; the various components are coupled together by a communication bus 81. It will be appreciated that the communication bus 81 is used to enable communications among the components. The communication bus 91 includes a power bus, a control bus, and a status signal bus in addition to a data bus. But for clarity of illustration the various buses are labeled in figure 9 as communication bus 81.

A memory 82 for storing a data analysis method program executable on the processor 83;

a processor 83, configured to execute the following steps when executing the data analysis method program:

acquiring first information of the to-be-trained object determined to be a defective item and a first parameter corresponding to the to-be-trained object, wherein the first information is used for representing factor information causing the to-be-trained object to be a defective item, and the first parameter is used for representing historical production data generated by first production equipment for producing the to-be-trained object in a first preset time period;

In other embodiments of the present invention, the processor 83 is configured to execute the following steps when executing the data analysis method program:

acquiring first information of an object to be trained;

acquiring historical control data of first production equipment, historical environment data corresponding to an object to be trained and historical raw material information corresponding to the object to be trained in a first preset time period;

correspondingly, a set of substandard product samples is generated based on the first information and the first parameter, including:

When the detected object is a defective product, acquiring first real-time control data of second production equipment in a second preset time period, first real-time environment data corresponding to the detected object and first real-time raw material information corresponding to the detected object; the second parameters comprise first real-time control data, first real-time environment data and first real-time raw material information;

correspondingly, based on the second parameter and the defective sample set, obtaining the reason that the object to be detected is defective includes:

acquiring matched factor information from the defective sample set based on the first real-time control data, the first real-time environment data and the first real-time raw material information to obtain target factor information; the defective sample set comprises control data of production equipment in a preset time period, environmental data of products and corresponding relation between raw material information of the products and factor information causing the products to be defective.

If the defective sample set does not have matched target factor information, generating prompt information for prompting the input of the cause of the defective;

input information for the prompt information is acquired, and the first information is updated based on the input information.

Generating an early warning threshold value based on the first parameter;

when the detected object is not a defective product, acquiring a third parameter corresponding to the detected object, wherein the third parameter is used for representing second real-time production data currently generated by second production equipment;

generating a predicted value based on the third parameter;

Acquiring attribute parameters of an object to be trained in a first preset time period from the first parameters;

and processing the time sequence model to generate the production efficiency corresponding to the object to be trained, and generating an early warning threshold value based on the production efficiency.

accordingly, generating a predicted value based on the third parameter includes:

and inputting the third parameter into the time sequence model, obtaining the real-time production efficiency through the time sequence model, and generating a predicted value based on the real-time production efficiency.

It will be appreciated that the memory 82 in embodiments of the invention may be either volatile memory or nonvolatile memory, or may include both volatile and nonvolatile memory. The non-volatile Memory may be a Read-Only Memory (ROM), a Programmable ROM (PROM), an Erasable PROM (EPROM), an Electrically Erasable PROM (EEPROM), or a flash Memory. Volatile Memory can be Random Access Memory (RAM), which acts as external cache Memory. By way of example, but not limitation, many forms of RAM are available, such as Static random access memory (Static RAM, SRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic random access memory (Synchronous DRAM, SDRAM), Double Data rate Synchronous Dynamic random access memory (ddr SDRAM ), Enhanced Synchronous SDRAM (ESDRAM), Synchlink DRAM (SLDRAM), and direct memory bus RAM (DRRAM). The memory 82 of the systems and methods described herein is intended to comprise, without being limited to, these and any other suitable types of memory.

And processor 83 may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware or instructions in the form of software in the processor 83. The Processor 83 may be a general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf programmable Gate Array (FPGA) or other programmable logic device, discrete Gate or transistor logic device, discrete hardware component. The various methods, steps and logic blocks disclosed in the embodiments of the present invention may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the embodiments of the present invention may be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software module may be located in ram, flash memory, rom, prom, or eprom, registers, etc. storage media as is well known in the art. The storage medium is located in the memory 82, and the processor 83 reads the information in the memory 82 and performs the steps of the above method in combination with the hardware thereof.

Based on the foregoing embodiments, an embodiment of the present invention provides a computer-readable medium, in which a data analysis program is stored, and the data analysis program, when executed by at least one processor, implements the steps of the data analysis method in any of the above embodiments.

It is understood that the method steps in the above embodiments may be stored in a computer-readable storage medium, and based on such understanding, part of the technical solutions of the embodiments of the present invention that essentially or contributes to the prior art, or all or part of the technical solutions may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions to enable a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor (processor) to execute all or part of the steps of the method of the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.

It is to be understood that the embodiments described herein may be implemented in hardware, software, firmware, middleware, microcode, or any combination thereof. For a hardware implementation, the Processing units may be implemented within one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), general purpose processors, controllers, micro-controllers, microprocessors, other electronic units configured to perform the functions of the present application, or a combination thereof.

For a software implementation, the techniques herein may be implemented with modules (e.g., procedures, functions, and so on) that perform the functions herein. The software codes may be stored in a memory and executed by a processor. The memory may be implemented within the processor or external to the processor.

Specifically, when the processor 83 in the user terminal is further configured to run the computer program, the method steps in the foregoing embodiments are executed, which is not described herein again.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, and the like) having computer-usable program code embodied therein.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention.

Claims

1. A method of data analysis, the method comprising:

2. The method according to claim 1, wherein the acquiring first information of the object to be trained determined to be the inferior product and the first parameter corresponding to the object to be trained comprises:

acquiring first information of the object to be trained;

3. The method according to claim 1, wherein when it is monitored that the object to be detected is a defective product, acquiring a second parameter corresponding to the object to be detected comprises:

4. The method of claim 3, further comprising:

5. The method according to any one of claims 1-4, further comprising:

generating an early warning threshold based on the first parameter;

generating a predicted value based on the third parameter;

6. The method of claim 5, the generating an early warning threshold based on the first parameter, comprising:

7. The method according to claim 6, wherein the obtaining of the third parameter corresponding to the object to be detected includes:

8. A data analysis apparatus, characterized in that the apparatus comprises: a processor, a memory, and a communication bus; wherein,

9. The device of claim 8, wherein the processor is further configured to:

acquiring first information of the object to be trained;

10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores one or more programs which are executable by one or more processors to implement the steps of the data analysis method according to any one of claims 1 to 7.