CN117725541B

CN117725541B - Intelligent monitoring and fault diagnosis system for running state of annealing furnace

Info

Publication number: CN117725541B
Application number: CN202410174327.4A
Authority: CN
Inventors: 陈佳宇; 周韦军; 周文华
Original assignee: Shanghai Qianghua Industrial Co ltd
Current assignee: Shanghai Qianghua Industrial Co ltd
Priority date: 2024-02-07
Filing date: 2024-02-07
Publication date: 2024-04-16
Anticipated expiration: 2044-02-07
Also published as: CN117725541A

Abstract

The invention discloses an intelligent monitoring and fault diagnosis system for the running state of an annealing furnace, in particular to the technical field of intelligent monitoring of the annealing furnace, which comprises a data acquisition module, a data processing module and a data processing module, wherein the data acquisition module is used for acquiring a temperature monitoring coefficient time sequence set of an nth annealing furnace, inputting a pre-constructed coefficient prediction model based on a historical temperature monitoring coefficient in the temperature monitoring coefficient time sequence set, and predicting a temperature monitoring coefficient at a future T moment; the annealing furnace abnormality judging module is used for presetting a temperature coefficient gradient threshold value, and comparing a temperature monitoring coefficient at a future T moment with the preset temperature coefficient gradient threshold value to obtain m abnormal annealing furnaces; the abnormal analysis module is used for acquiring abnormal characteristic data of the mth abnormal annealing furnace and determining abnormal type data of the abnormal annealing furnace according to the abnormal characteristic data; the invention is beneficial to comprehensively monitoring the whole running state of the annealing furnace, timely responding and determining the abnormal reason of the fault when the fault occurs, and reduces the production time cost.

Description

Intelligent monitoring and fault diagnosis system for running state of annealing furnace

Technical Field

The invention relates to the technical field of intelligent monitoring of annealing furnaces, in particular to an intelligent monitoring and fault diagnosis system for the running state of an annealing furnace.

Background

Annealing furnace refers to an apparatus for performing a heat treatment process, one of the most common processes being annealing, which is a process of changing material properties by heating and cooling in the field of material preparation. The basic principle of the annealing furnace is to heat the material to a certain temperature, and then control the cooling rate to cool the material slowly, so as to change the structure, hardness, strength and other physical properties of the material, thereby obtaining the final product. The annealing furnace is commonly used for controlling parameters such as temperature, heat preservation time, cooling rate and the like to change the internal thermal stress of the material, particularly when quartz materials are produced, the monitoring of the running condition of the annealing furnace is particularly important, and the quality of an annealing result determines whether a final product meets quality standards.

At present, the running state and fault detection of the annealing furnace are usually determined by experience of operators, and high-quality products produced by the annealing furnace are difficult to ensure effectively; the existing fault diagnosis method of the annealing furnace is realized by analyzing the fault of the heating tube and grounding, for example, chinese patent publication No. CN112505584A discloses a positioning system of the fault point of the earth leakage of the heating tube and the heating tube of the annealing furnace, and although the fault detection of the annealing furnace can be realized by the method, the inventor researches and application discovers that the method and the prior art have at least the following part of defects:

(1) Considering the problem of annealing furnace faults singly from the angle of a heating pipe, detecting the faults of the annealing furnace incompletely;

(2) During mass production, the abnormal reason of the abnormal annealing furnace cannot be determined in time, so that the rapid response of the fault alarm information of the annealing furnace is difficult to ensure, and the production time cost is increased.

Therefore, the invention provides an intelligent monitoring and fault diagnosis system for the running state of the annealing furnace.

Disclosure of Invention

In order to overcome the defects in the prior art, the invention provides an intelligent monitoring and fault diagnosis system for the running state of an annealing furnace, so as to solve the problems in the background art.

In order to achieve the above purpose, the present invention provides the following technical solutions: an intelligent monitoring and fault diagnosis system for the running state of an annealing furnace comprises:

the data acquisition module is used for acquiring a temperature monitoring coefficient time sequence set of the nth annealing furnace, inputting a pre-constructed coefficient prediction model based on a historical temperature monitoring coefficient in the temperature monitoring coefficient time sequence set, and predicting a temperature monitoring coefficient at a future T moment;

the annealing furnace abnormality judging module is used for presetting a temperature coefficient gradient threshold value and comparing a temperature monitoring coefficient at a future T moment with the preset temperature coefficient gradient threshold value so as to obtain m abnormal annealing furnaces;

The abnormal analysis module is used for acquiring abnormal characteristic data of the mth abnormal annealing furnace, determining abnormal type data of the abnormal annealing furnace according to the abnormal characteristic data, and determining abnormal analysis data corresponding to the abnormal type data based on a preset relation between the abnormal type data and the abnormal analysis data, wherein the abnormal analysis data comprises a plurality of abnormal reasons and standard fault current fluctuation intervals associated with the abnormal reasons; the anomaly type data includes an anomaly component and an anomaly component location; the abnormal characteristic data comprise an abnormal temperature spectrogram and an abnormal atmosphere pressure spectrogram;

the abnormal cause analysis module is used for acquiring a real-time current data set of the abnormal component in a set production period according to the position of the abnormal component and determining the abnormal cause of the abnormal annealing furnace based on the real-time current data; the real-time current data set includes a plurality of real-time current data.

Further, the method for obtaining the temperature monitoring coefficient time series set comprises the following steps:

extracting temperature monitoring coefficients in the history monitoring process of an nth annealing furnace from an annealing furnace database, marking the temperature monitoring coefficients as history temperature monitoring coefficients, constructing a temperature monitoring coefficient time sequence set by the extracted history temperature monitoring coefficients, wherein the temperature monitoring coefficient time sequence set comprises i history temperature monitoring coefficients, the time intervals obtained by the i history temperature monitoring coefficients are equal, and the i history temperature monitoring coefficients correspond to the production period of one annealing furnace; the production period is hours, days or weeks;

The pre-construction logic of the coefficient prediction model is as follows: presetting a time step J, a sliding step H and a sliding window length C according to the actual experience of annealing furnace staff; converting historical temperature monitoring coefficients in the temperature monitoring coefficient time sequence set into a plurality of training samples by using a sliding window method, taking the training samples as input of a coefficient prediction model, taking the temperature monitoring coefficients after the prediction time step J as output, taking the subsequent temperature monitoring coefficients of each training sample as a prediction target, taking the prediction accuracy rate as a training target, and training the coefficient prediction model; generating a coefficient prediction model for predicting a temperature monitoring coefficient at a future time T according to a historical temperature monitoring coefficient in the temperature monitoring coefficient time sequence set; wherein the coefficient prediction model is an RNN neural network model.

Further, the method for obtaining the historical temperature monitoring coefficient comprises the following steps:

acquiring temperature characteristic data of an nth annealing furnace; the temperature characteristic data comprise a heating rate, a product quality coefficient and a cooling rate;

extracting the heat preservation coefficient of each annealing furnace based on the preset relation between the annealing furnace and the heat preservation coefficient;

carrying out dimensionless calculation based on the temperature characteristic data and the heat preservation coefficient to obtain a historical temperature monitoring coefficient of the nth annealing furnace; the calculation formula is as follows:

；

Wherein: Representing historical temperature monitoring coefficients of an nth annealing furnace,/> representing product quality coefficients of the nth annealing furnace,/> representing heat preservation coefficients of the nth annealing furnace,/> representing heating rate of the nth annealing furnace,/> representing cooling rate of the nth annealing furnace,/> representing natural constant,/> representing product quality coefficient weight factors of the nth annealing furnace; A weight factor representing the heat retention coefficient of the nth annealing furnace; the/> represents a weight factor of the temperature rise rate of the nth annealing furnace; and/> is the cooling rate weight factor of the nth annealing furnace, and ln is a logarithmic function.

Further, the logic for generating the product quality coefficient of the nth annealing furnace is as follows:

Acquiring an annealed image of each product through a camera device; extracting standard annealing images corresponding to the products, which are pre-stored in an annealing furnace monitoring database;

taking the annealed image of the product as a first annealed image;

dividing the first annealing image and the standard annealing image into G areas, wherein G is a positive integer greater than zero;

comparing pixel points of the same position areas of the first annealing image and the standard annealing image one by one, and recording defect areas of the first annealing image and the standard annealing image;

Counting the number of defective areas with differences of each product to obtain the total number of defective areas; marking the total number of defect areas as , marking the production cycle of an nth annealing furnace as/> , and acquiring the total number G of annealing image areas of each product;

Carrying out formula calculation on the total number of defect areas, the production cycle/> of the nth annealing furnace and the total number G of annealing image areas of each product to obtain the product quality coefficient/> of the nth annealing furnace; the formula is as follows:

；

Wherein denotes a production cycle weight factor of the nth annealing furnace.

Further, acquiring heat preservation characteristic data of an nth annealing furnace, wherein the heat preservation characteristic data comprise furnace tightness , furnace atmosphere pressure/> and heat preservation material thickness/> ; the calculation formula of the heat preservation coefficient of the nth annealing furnace is as follows:

；

Wherein: The furnace tightness of the nth annealing furnace is represented by/> , the furnace atmosphere pressure of the nth annealing furnace is represented by/> , the furnace tightness weight factor of the nth annealing furnace is represented by/> , the furnace atmosphere pressure weight factor of the nth annealing furnace is represented by , and the furnace atmosphere pressure weight factor of the nth annealing furnace is represented by/> .

Further, the method for obtaining m abnormal annealing furnaces comprises the following steps:

Extracting preset temperature coefficient gradient thresholds and/> of the annealing furnace, wherein/> ＞, and comparing a temperature monitoring coefficient/> at a future time T with the preset temperature coefficient gradient threshold;

If ＞, judging the corresponding annealing furnace as an abnormal annealing furnace;

if , judging the corresponding annealing furnace to be a normal annealing furnace;

if , judging the corresponding annealing furnace as an abnormal annealing furnace;

counting all abnormal annealing furnaces to obtain m abnormal annealing furnaces.

Further, obtaining abnormal characteristic data of the abnormal annealing furnace, including:

101: acquiring an annealing operation assembly of the abnormal annealing furnace and acquiring temperature data of the annealing operation assembly in a set production period; constructing a temperature trend graph by taking time in the temperature data as a horizontal axis and taking a temperature value in the temperature data as a vertical axis;

102: dividing the temperature trend graph in equal parts according to K temperature intervals to obtain K actual temperature line graphs, wherein the obtained K actual temperature line graphs form an actual temperature line graph set; k is an integer greater than zero;

103: extracting a d-th actual temperature line graph in the actual temperature line graph set, wherein d epsilon K, and the initial value of d is 1;

104: acquiring a temperature interval of the annealing operation assembly, extracting a standard temperature line graph related to the temperature interval, calculating the similarity between an actual temperature line graph and the standard temperature line graph, and jumping to the step 105 if the similarity between the actual temperature line graph and the standard temperature line graph is greater than or equal to a first preset similarity threshold; if the similarity between the actual temperature line graph and the standard temperature line graph is smaller than the first preset similarity threshold, marking the actual temperature line graph as an abnormal temperature line graph, and jumping to the step 105;

105: let d=d+1 and jump back to step 103;

106: repeating the steps 103-105 until d=K, ending the cycle to obtain A abnormal temperature line graphs, wherein A epsilon K;

107: extracting the similarity corresponding to each abnormal temperature line graph, and carrying out Fourier transformation on the abnormal temperature line graph with the minimum similarity to obtain an abnormal temperature spectrogram;

Further, obtaining abnormal characteristic data of the abnormal annealing furnace, further comprising:

201: acquiring an annealing operation assembly of the abnormal annealing furnace and acquiring atmosphere pressure data of the annealing operation assembly in a set production period; taking the time in the atmosphere pressure data as a horizontal axis and the atmosphere pressure value in the atmosphere pressure data as a vertical axis, constructing an atmosphere pressure trend graph;

202: dividing the atmosphere pressure trend graph in equal parts according to W atmosphere pressure intervals to obtain W actual atmosphere pressure line graphs, wherein the W actual atmosphere pressure line graphs form an actual atmosphere pressure line graph set, and W is an integer greater than zero;

203: extracting an f-th actual atmosphere pressure line graph in the actual atmosphere pressure line graph set, wherein f epsilon W and the initial value of f are 1;

204: acquiring a corresponding atmosphere pressure interval of the annealing operation assembly, extracting a standard atmosphere pressure line graph associated with the corresponding atmosphere pressure interval, calculating the similarity between an actual atmosphere pressure line graph and the standard atmosphere pressure line graph, and jumping to the step 205 if the similarity between the actual atmosphere pressure line graph and the standard atmosphere pressure line graph is greater than or equal to a second preset similarity threshold; if the similarity between the actual atmosphere pressure line graph and the standard atmosphere pressure line graph is smaller than a second preset similarity threshold, marking the actual atmosphere pressure line graph as an abnormal atmosphere pressure line graph, and jumping to the step 205;

205: let f=f+1 and jump back to step 203;

206: repeating the steps 203-205 until f=w, ending the cycle to obtain B abnormal atmosphere pressure line graphs, B e W;

207: and extracting the similarity corresponding to each abnormal atmosphere pressure line graph, and carrying out Fourier transformation on the abnormal atmosphere pressure line graph with the minimum similarity to obtain an abnormal atmosphere pressure spectrogram.

Further, the method for determining the abnormal type data of the abnormal annealing furnace according to the abnormal characteristic data comprises the following steps:

acquiring an abnormal temperature spectrogram and an abnormal atmosphere pressure spectrogram of the abnormal annealing furnace;

Inputting the abnormal temperature spectrogram and the abnormal atmosphere pressure spectrogram into a second machine learning model to determine an abnormal component of the abnormal annealing furnace, and determining the position of the abnormal component according to the abnormal component.

Further, determining an abnormality cause of the abnormal annealing furnace, comprising:

301: extracting a standard current interval according to a preset relation between the abnormal component and the standard current interval; the standard current interval is maximum standard current and minimum standard current/> ;

302: comparing pieces of real-time current data in the real-time current data set with a standard current interval, obtaining real-time current data larger than the maximum standard current/> in the real-time current data set, and obtaining real-time current data smaller than the minimum standard current/> in the real-time current data set;

303: taking real-time current data larger than the maximum standard current as first real-time current data and taking real-time current data smaller than the minimum standard current/> as second real-time current data;

304: respectively counting the number of the first real-time current data and the second real-time current data to obtain the total number of the first real-time current data and the total number of the second real-time current data;

305: comparing the total number of the first real-time current data with the total number of the second real-time current data, if the total number of the first real-time current data is greater than or equal to the total number of the second real-time current data, acquiring the variance and the average value of the first real-time current data, and jumping to step 306; if the total number of the first real-time current data is smaller than the total number of the second real-time current data, obtaining the variance and the average value of the second real-time current data, and jumping to step 307;

306: taking the variance of the first real-time current data as a first variance, taking the average value of the first real-time current data as a first average value, comparing the first variance with the standard variances of the standard fault current fluctuation intervals associated with a plurality of abnormal reasons, and simultaneously comparing the first average value with the standard average value of the standard fault current fluctuation intervals associated with a plurality of abnormal reasons, wherein if the first variance and the first average value fall into one standard fault current fluctuation interval; taking the corresponding abnormal reason corresponding to the standard fault current fluctuation interval as the abnormal reason of the abnormal annealing furnace;

307: taking the variance of the second real-time current data as a second variance, taking the average value of the second real-time current data as a second average value, comparing the second variance with the standard variances of the standard fault current fluctuation intervals associated with a plurality of abnormal reasons, and simultaneously comparing the second average value with the standard average value of the standard fault current fluctuation intervals associated with a plurality of abnormal reasons, wherein if the second variance and the second average value both fall into one standard fault current fluctuation interval; taking the corresponding abnormal reason corresponding to the standard fault current fluctuation interval as the abnormal reason of the abnormal annealing furnace;

308: if the first variance and the first average value do not fall into a standard fault current fluctuation interval, a first early warning instruction is generated; if the second variance and the second average value do not fall into a standard fault current fluctuation interval, a second early warning instruction is generated; and when the first early warning instruction or the second early warning instruction is generated, notifying an annealing furnace manager to check the fault reason.

The invention has the technical effects and advantages that:

according to the invention, the temperature monitoring coefficient of each annealing furnace is obtained, the abnormal annealing furnace in the preset running time is determined based on the temperature monitoring coefficient, then the abnormal characteristic data of the abnormal annealing furnace is obtained, the abnormal type data of the abnormal annealing furnace is determined according to the abnormal characteristic data, the abnormal analysis data corresponding to the abnormal type data is determined based on the preset relation between the abnormal type data and the abnormal analysis data, the real-time current data set of the abnormal assembly in the set production period is acquired according to the position of the abnormal assembly, the abnormal reason of the abnormal annealing furnace is determined based on the real-time current data, the whole running state of the annealing furnace is monitored comprehensively, the abnormal reason of the fault is responded and determined in time when the fault occurs, and the production time cost is reduced.

Drawings

FIG. 1 is a schematic diagram of a system of example 1;

FIG. 2 is a comparison of the first annealing image and the standard annealing image of example 1;

FIG. 3 is a temperature line graph of example 1;

FIG. 4 is an atmospheric pressure line graph of example 1;

FIG. 5 is a schematic diagram of an electronic device according to embodiment 2;

Fig. 6 is a schematic diagram of a computer-readable storage medium according to embodiment 3.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

Furthermore, the drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus a repetitive description thereof will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. The functional entities may be implemented in software or in one or more hardware modules or integrated circuits or in different networks and/or processor methods and/or microcontroller methods.

It will be understood that, although the terms "first," "second," etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another element. For example, a first element could be termed a second element, and a similar second element could be termed a first element, without departing from the scope of example embodiments. The term "and/or" as used herein includes any and all combinations of one or more of the associated listed items.

Example 1

Referring to fig. 1, the disclosure of the present embodiment provides an intelligent monitoring and fault diagnosis system for an annealing furnace running state, including:

It should be noted that n annealing furnaces are provided in the target factory, n is an integer greater than zero, annealing operation components are disposed in each annealing furnace, and each annealing operation component executes an automatic product annealing process in a respective set production period; the annealing furnace is internally provided with a plurality of sensors, and the annealing furnace is provided with a monitoring device, wherein the sensors comprise but are not limited to an infrared temperature sensor, a pressure sensor, a humidity sensor, a current sensor and the like;

extracting temperature monitoring coefficients in the history monitoring process of an nth annealing furnace from an annealing furnace database, marking the temperature monitoring coefficients as history temperature monitoring coefficients, constructing a temperature monitoring coefficient time sequence set by the extracted history temperature monitoring coefficients, wherein the temperature monitoring coefficient time sequence set comprises i history temperature monitoring coefficients, the time intervals obtained by the i history temperature monitoring coefficients are equal, and the i history temperature monitoring coefficients correspond to the production period of one annealing furnace; the production cycle may be hours, days or weeks.

Further, the pre-construction logic of the coefficient prediction model is as follows: presetting a time step J, a sliding step H and a sliding window length C according to the actual experience of annealing furnace staff; converting historical temperature monitoring coefficients in the temperature monitoring coefficient time sequence set into a plurality of training samples by using a sliding window method, taking the training samples as input of a coefficient prediction model, taking the temperature monitoring coefficients after the prediction time step J as output, taking the subsequent temperature monitoring coefficients of each training sample as a prediction target, taking the prediction accuracy rate as a training target, and training the coefficient prediction model; generating a coefficient prediction model for predicting a temperature monitoring coefficient at a future time T according to a historical temperature monitoring coefficient in the temperature monitoring coefficient time sequence set; wherein the coefficient prediction model is an RNN neural network model; predicting temperature monitoring coefficients of n annealing furnaces at future T moments by using a temperature monitoring coefficient time sequence set, so that the temperature monitoring coefficients based on the future T moments are realized, the condition of the annealing furnaces to be predicted is predicted, and the temperature monitoring coefficients are predicted from multiple aspects;

The method includes the steps that a temperature monitoring coefficient time sequence set is divided into Z sliding windows with equal sizes, the temperature monitoring coefficient in each window is used as a sample, the temperature monitoring coefficient at the future T moment of the window is used as a digital label, the one sample corresponds to one digital label, one sample and the corresponding digital label form coefficient training data, and a plurality of groups of coefficient training data form a coefficient training set;

By way of further illustration, assuming that the temperature monitoring coefficient time series set a contains 10 sets of historical temperature monitoring coefficients, , is the/> set of historical temperature monitoring coefficients, constructing a plurality of training samples using a sliding window, setting the prediction time step J to 1, the length H of the sliding window to 5, and the sliding step C to 1, generating each training sample to contain consecutive 5 historical temperature monitoring coefficients, taking the next temperature monitoring coefficient of the consecutive 5 historical temperature monitoring coefficients as the prediction target; for example:

As coefficient training data, the prediction target corresponding to the coefficient training data is/> ;

Similarly, a coefficient prediction model is used for training a temperature monitoring coefficient for predicting future time.

In the implementation process, the method for acquiring the historical temperature monitoring coefficient comprises the following steps:

；

Wherein: Representing historical temperature monitoring coefficients of an nth annealing furnace,/> representing product quality coefficients of the nth annealing furnace,/> representing heat preservation coefficients of the nth annealing furnace,/> representing heating rate of the nth annealing furnace,/> representing cooling rate of the nth annealing furnace,/> representing natural constant,/> representing product quality coefficient weight factors of the nth annealing furnace; A weight factor representing the heat retention coefficient of the nth annealing furnace; the/> represents a weight factor of the temperature rise rate of the nth annealing furnace; the/> is the cooling rate weight factor of the nth annealing furnace, and ln is a logarithmic function;

It should be noted that, in the production period, the larger the value of the heating rate or the cooling rate is, the smaller the value of the product quality coefficient or the heat preservation coefficient is, the smaller the temperature monitoring coefficient is, the more unstable the temperature of the annealing furnace is, and the greater the possibility of abnormality of the annealing furnace is; conversely, the smaller the value of the temperature rising rate or the temperature reducing rate is, the larger the value of the product quality coefficient or the heat preservation coefficient is, the larger the temperature monitoring coefficient is, the more stable the temperature of the annealing furnace is, and the less the possibility of abnormality of the annealing furnace is;

Referring to fig. 2, the logic for generating the product quality coefficient of the nth annealing furnace is as follows:

taking the annealed image of the product as a first annealed image;

dividing the first annealing image and the standard annealing image into G areas, wherein G is an integer greater than zero;

it should be noted that: the dividing mode of dividing the area in the first annealing image and the standard annealing image is identical to the size of the area, then the areas at the same position in the first annealing image and the standard annealing image are compared one by one, and if the pixel points with differences between the areas at the same position exceed the preset percentage, the difference between the areas at the same position is judged;

；

Wherein represents a production cycle weight factor of the nth annealing furnace;

it should be appreciated that: the smaller the product quality coefficient of the nth annealing furnace is, the worse the quality of the annealed product of the annealing furnace is; conversely, the larger the product quality coefficient of the nth annealing furnace is, the better the quality of the annealed product of the annealing furnace is;

The method for acquiring the heat preservation coefficient of the nth annealing furnace comprises the following steps:

Acquiring heat preservation characteristic data of an nth annealing furnace, wherein the heat preservation characteristic data comprise furnace tightness , furnace atmosphere pressure/> and heat preservation material thickness/> ; the calculation formula of the heat preservation coefficient of the nth annealing furnace is as follows:

；

Wherein: Represents the degree of tightness in the furnace of the nth annealing furnace,/> represents the furnace atmosphere pressure of the nth annealing furnace,/> represents the thickness of the insulating material of the nth annealing furnace,/> represents the weight factor of tightness in the furnace of the nth annealing furnace, represents the weight factor of the furnace atmosphere pressure of the nth annealing furnace, and/> represents the weight factor of thickness of the insulating material of the nth annealing furnace;

It should be noted that: the larger the heat preservation coefficient of the nth annealing furnace is, the better the heat preservation effect of the annealing furnace is; the smaller the heat preservation coefficient of the nth annealing furnace is, the worse the heat preservation effect of the annealing furnace is; the furnace tightness of the nth annealing furnace, the furnace atmosphere pressure/> of the nth annealing furnace and the insulation material thickness/> of the nth annealing furnace are respectively obtained by a gas detector, a pressure sensor and ultrasonic measurement in advance;

In the implementation process, the method for obtaining m abnormal annealing furnaces comprises the following steps:

It should be noted that: the calculation process of the preset temperature coefficient gradient threshold is the same as the generation process of the temperature monitoring coefficient, the details are referred to above, and the preset temperature coefficient gradient threshold and/> are determined by a manager, so that redundant description is omitted;

The abnormal analysis module is used for acquiring abnormal characteristic data of the mth abnormal annealing furnace, determining abnormal type data of the abnormal annealing furnace according to the abnormal characteristic data, and determining abnormal analysis data corresponding to the abnormal type data based on a preset relation between the abnormal type data and the abnormal analysis data, wherein the abnormal analysis data comprises a plurality of abnormal reasons and standard fault current fluctuation intervals associated with the abnormal reasons; the anomaly type data includes an anomaly component and an anomaly component location;

Specifically, the abnormal characteristic data comprises an abnormal temperature spectrogram and an abnormal atmosphere pressure spectrogram;

In one embodiment, obtaining abnormal characteristic data of an abnormal annealing furnace comprises:

101: acquiring an annealing operation assembly of the abnormal annealing furnace and acquiring temperature data of the annealing operation assembly in a set production period; constructing a temperature trend graph by taking time in temperature data as a horizontal axis and taking a temperature value in the temperature data as a vertical axis, and referring to fig. 3;

102: dividing the temperature trend graph in equal parts according to K temperature intervals to obtain K actual temperature line graphs, wherein the obtained K actual temperature line graphs form an actual temperature line graph set; k is an integer greater than zero; the temperature interval is preset by a manager based on the preset production period (heating stage, heat preservation stage and cooling stage);

It should be noted that: a plurality of temperature intervals are prestored in an annealing furnace monitoring database, and each temperature interval is associated with a standard temperature line diagram which reflects the normal temperature of the annealing furnace under the condition of no fault;

105: let d=d+1 and jump back to step 103;

It should be appreciated that: the Fourier transform is specifically one of fast Fourier transform or short-time Fourier transform;

in one specific embodiment, obtaining abnormal characteristic data of the abnormal annealing furnace further comprises:

201: acquiring an annealing operation assembly of the abnormal annealing furnace and acquiring atmosphere pressure data of the annealing operation assembly in a set production period; constructing an atmosphere pressure trend chart by taking time in the atmosphere pressure data as a horizontal axis and taking an atmosphere pressure value in the atmosphere pressure data as a vertical axis, and referring to fig. 4;

It should be noted that: each current interval is also associated with a standard atmosphere pressure line graph reflecting the normal atmosphere pressure characterization of the annealing furnace under the fault-free condition;

205: let f=f+1 and jump back to step 203;

207: extracting the similarity corresponding to each abnormal atmosphere pressure line graph, and carrying out Fourier transformation on the abnormal atmosphere pressure line graph with the minimum similarity to obtain an abnormal atmosphere pressure spectrogram;

in the implementation process, the method for determining the abnormal type data of the abnormal annealing furnace according to the abnormal characteristic data comprises the following steps:

Inputting the abnormal temperature spectrogram and the abnormal atmosphere pressure spectrogram into a second machine learning model to determine components of the abnormal annealing furnace, and determining the positions of the abnormal components according to the abnormal components.

Specifically, the generating logic of the second machine learning model is: acquiring second historical characteristic data which is pre-stored in an annealing furnace monitoring database and is used for training a second machine learning model, wherein the second historical characteristic data comprises an abnormal temperature spectrogram, an abnormal atmosphere pressure spectrogram and an abnormal component; dividing second historical characteristic data for training a second machine learning model into a type training set and a type testing set, constructing the second machine learning model, taking an abnormal temperature spectrogram and an abnormal atmosphere pressure spectrogram in the type training set as input of the second machine learning model, taking an abnormal component in the type training set as output of the second machine learning model, and training the second machine learning model to obtain an initial second machine learning model; performing model test on the initial second machine learning model by using the type test set, and screening the corresponding initial second machine learning model with the accuracy greater than or equal to the preset test accuracy as a second machine learning model;

Wherein the second machine learning model includes, but is not limited to, a long-term memory network model, a convolutional neural network model, a deep neural network model, or the like;

It should be noted that, in the second historical characteristic data, a group of abnormal temperature spectrogram and abnormal atmosphere pressure spectrogram correspond to an abnormal component in the annealing furnace, because the influence on the temperature and the atmosphere pressure is different when the component in the annealing furnace is abnormal, each component corresponds to a group of abnormal temperature spectrogram and abnormal atmosphere pressure spectrogram when being abnormal, and the abnormal component is determined through the abnormal temperature spectrogram and the abnormal atmosphere pressure spectrogram, compared with the unilateral consideration of burner faults only in the prior art, the method is more accurate;

The abnormal temperature spectrogram and the abnormal atmosphere pressure spectrogram are associated with the abnormal components, a technician can acquire the real-time abnormal temperature spectrogram and the abnormal atmosphere pressure spectrogram when the annealing furnace fails, then overhaul the annealing furnace to determine the specific abnormal components, and multiple times of collection are carried out to acquire abnormal component associated data corresponding to multiple groups of abnormal temperature spectrograms and abnormal atmosphere pressure spectrograms;

for example, when an annealing operation component of an annealing furnace is abnormal, a sudden fluctuation or abnormal peak occurs in a corresponding temperature spectrogram; an abnormal peak or change corresponding to the abnormal temperature spectrum is observed in the atmosphere pressure spectrum.

Note that, the abnormal component includes r abnormal component names, for example: annealing operation components (such as resistors, capacitors, coils and the like), electrical components or heat preservation components and the like, determining the positions of the abnormal components according to the names of the abnormal components, and particularly obtaining the installation positions of the corresponding abnormal components according to an annealing furnace design drawing.

The abnormal cause analysis module is used for acquiring a real-time current data set of the abnormal component in a set production period according to the position of the abnormal component and determining the abnormal cause of the abnormal annealing furnace based on the real-time current data; the real-time current data set comprises a plurality of real-time current data;

In the implementation process, determining the abnormality reason of the abnormal annealing furnace comprises the following steps:

301: extracting a standard current interval according to a preset relation between the abnormal component and the standard current interval; the standard current interval is maximum standard current and minimum standard current/> ; /(I)

It should be noted that: the annealing furnace monitoring database is pre-stored with a plurality of standard current intervals under different conditions (such as production period and equipment power), and each standard current interval reflects the standard current fluctuation range of the annealing furnace under the condition of no abnormality;

306: taking the variance of the first real-time current data as a first variance, taking the average value of the first real-time current data as a first average value, comparing the first variance with the standard variances of the standard fault current fluctuation intervals associated with a plurality of abnormal reasons, and simultaneously comparing the first average value with the standard average value of the standard fault current fluctuation intervals associated with a plurality of abnormal reasons, wherein if the first variance and the first average value both fall into the first standard fault current fluctuation interval; taking the corresponding abnormal reason corresponding to the standard fault current fluctuation interval as the abnormal reason of the abnormal annealing furnace;

307: taking the variance of the second real-time current data as a second variance, taking the average value of the second real-time current data as a second average value, comparing the second variance with the standard variances of the standard fault current fluctuation intervals associated with a plurality of abnormal reasons, and simultaneously comparing the second average value with the standard average value of the standard fault current fluctuation intervals associated with a plurality of abnormal reasons, wherein if the second variance and the second average value both fall into the second standard fault current fluctuation interval; taking the corresponding abnormal reason corresponding to the standard fault current fluctuation interval as the abnormal reason of the abnormal annealing furnace;

308: if the first variance and the first average value do not fall into the corresponding standard fault current fluctuation interval, a first early warning instruction is generated; if the second variance and the second average value do not fall into a standard fault current fluctuation interval, a second early warning instruction is generated;

further, when the first early warning instruction or the second early warning instruction is generated, an annealing furnace manager is informed to check the fault reason.

It should be noted that, there are a plurality of annealing operation components (such as resistors, capacitors, coils, etc.) sensitive to temperature in the annealing furnace, the operation states of the annealing operation components are closely related to the current characteristics, and when the annealing operation components fail, the abnormal changes of the voltage are easier to directly reflect the abnormal reasons of the annealing operation components than the abnormal changes of the current; the invention utilizes the corresponding standard fault current fluctuation interval in which the variance of the real-time current data falls, and the variance represents the change of the current fluctuation more accurately due to the discrete degree of the variance on the current data, so that the abnormal cause can be rapidly and accurately analyzed even in transient abnormality, the rapid analysis and timely feedback of the abnormal cause in the operation process of the annealing furnace are facilitated, the timely adjustment of management personnel is facilitated, and the normal operation of the annealing furnace is ensured.

The formulas related in the above are all formulas with dimensions removed and numerical values calculated, and are a formula which is obtained by acquiring a large amount of data and performing software simulation to obtain the closest actual situation, and weight factors in the formulas and various preset thresholds in the analysis process are set by a person skilled in the art according to the actual situation or are obtained by simulating a large amount of data; the size of the weight factor is a specific numerical value obtained by quantizing each parameter, so that the subsequent comparison is convenient, and the size of the weight factor depends on the number of sample data and the corresponding processing coefficient is preliminarily set for each group of sample data by a person skilled in the art; as long as the proportional relation between the parameter and the quantized value is not affected.

Example 2

Referring to fig. 5, the present embodiment provides an electronic device, including: a processor and a memory, wherein the memory stores a computer program for the processor to call;

the processor executes the intelligent monitoring and fault diagnosis system of the annealing furnace running state of the embodiment 1 by calling the computer program stored in the memory.

Example 3

Referring to fig. 6, the present embodiment provides a computer readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the intelligent monitoring and fault diagnosis system for the operating state of the annealing furnace of embodiment 1.

The above embodiments may be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented in software, the above-described embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product comprises one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the processes or functions described in accordance with embodiments of the present application are produced in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, from one website site, computer, server, or data center over a wired network. The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains one or more sets of available media. The usable medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium may be a solid state disk.

Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, and are not repeated herein.

In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the units is merely one, and there may be additional divisions in actual implementation, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other forms.

The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution in the form of a software product stored in a storage medium, comprising instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a read-only memory (ROM), a random access memory (random access memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.

The foregoing is merely illustrative of the present application, and the present application is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Finally: the foregoing description of the preferred embodiments of the invention is not intended to limit the invention to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and principles of the invention are intended to be included within the scope of the invention.

Claims

1. The intelligent monitoring and fault diagnosis system for the running state of the annealing furnace is characterized by comprising the following components:

the abnormal cause analysis module is used for collecting a real-time current data set of the abnormal component in a set production period according to the position of the abnormal component and determining the abnormal cause of the abnormal annealing furnace based on the real-time current data; the real-time current data set comprises a plurality of real-time current data;

determining the abnormality reason of the abnormal annealing furnace, comprising the following steps:

2. The intelligent monitoring and fault diagnosis system for the operating state of an annealing furnace according to claim 1, wherein the method for obtaining the time series set of temperature monitoring coefficients comprises the following steps:

Extracting temperature monitoring coefficients in the history monitoring process of an nth annealing furnace from an annealing furnace database, marking the temperature monitoring coefficients as history temperature monitoring coefficients, constructing a temperature monitoring coefficient time sequence set by the extracted history temperature monitoring coefficients, wherein the temperature monitoring coefficient time sequence set comprises i history temperature monitoring coefficients, the time intervals obtained by the i history temperature monitoring coefficients are equal, and the i history temperature monitoring coefficients correspond to the production period of one annealing furnace; the production cycle is hours, days or weeks.

3. The intelligent monitoring and fault diagnosis system for the operation state of the annealing furnace according to claim 2, wherein the pre-construction logic of the coefficient prediction model is as follows: presetting a time step J, a sliding step H and a sliding window length C; converting historical temperature monitoring coefficients in the temperature monitoring coefficient time sequence set into a plurality of training samples by using a sliding window method, taking the training samples as input of a coefficient prediction model, taking the temperature monitoring coefficients after the prediction time step J as output, taking the subsequent temperature monitoring coefficients of each training sample as a prediction target, taking the prediction accuracy rate as a training target, and training the coefficient prediction model; generating a coefficient prediction model for predicting a temperature monitoring coefficient at a future time T according to a historical temperature monitoring coefficient in the temperature monitoring coefficient time sequence set; wherein the coefficient prediction model is an RNN neural network model.

4. The intelligent annealing furnace operating state monitoring and fault diagnosis system according to claim 3, wherein the method for obtaining the historical temperature monitoring coefficient comprises the following steps:

；

Wherein: Representing historical temperature monitoring coefficients of an nth annealing furnace,/> representing product quality coefficients of the nth annealing furnace,/> representing heat preservation coefficients of the nth annealing furnace,/> representing heating rate of the nth annealing furnace,/> representing cooling rate of the nth annealing furnace,/> representing natural constant,/> representing product quality coefficient weight factors of the nth annealing furnace; the/> represents the weight factor of the heat preservation coefficient of the nth annealing furnace; the/> represents a weight factor of the temperature rise rate of the nth annealing furnace; And the cooling rate weight factor of the nth annealing furnace is represented, and ln is a logarithmic function.

5. The intelligent monitoring and fault diagnosis system for the operating state of an annealing furnace according to claim 4, wherein the logic for generating the product quality coefficient of the nth annealing furnace is as follows:

taking the annealed image of the product as a first annealed image;

；

Wherein denotes a production cycle weight factor of the nth annealing furnace.

6. The intelligent monitoring and fault diagnosis system for the operating state of the annealing furnace according to claim 5, wherein the heat preservation characteristic data of the nth annealing furnace are obtained, wherein the heat preservation characteristic data comprise the tightness of the furnace, the pressure/> of the atmosphere in the furnace and the thickness/> of the heat preservation material; the calculation formula of the heat preservation coefficient of the nth annealing furnace is as follows:

；

Wherein: The furnace tightness of the nth annealing furnace is expressed, the furnace atmosphere pressure of the nth annealing furnace is expressed by/> , the thickness of the insulating material of the nth annealing furnace is expressed by ,/> , the furnace tightness weight factor of the nth annealing furnace is expressed by/> , the furnace atmosphere pressure weight factor of the nth annealing furnace is expressed by/> , and the thickness weight factor of the insulating material of the nth annealing furnace is expressed by the weight factor of the insulating material of the nth annealing furnace.

7. The intelligent monitoring and fault diagnosis system for the operation state of an annealing furnace according to claim 6, wherein the method for obtaining m abnormal annealing furnaces comprises the following steps:

8. The intelligent monitoring and fault diagnosis system for the operation state of an annealing furnace according to claim 7, wherein obtaining abnormal characteristic data of an abnormal annealing furnace comprises:

105: let d=d+1 and jump back to step 103;

107: and extracting the similarity corresponding to each abnormal temperature line graph, and carrying out Fourier transformation on the abnormal temperature line graph with the minimum similarity to obtain an abnormal temperature spectrogram.

9. The intelligent monitoring and fault diagnosis system for the operation state of an annealing furnace according to claim 8, wherein the abnormal characteristic data of the abnormal annealing furnace is obtained, further comprising:

205: let f=f+1 and jump back to step 203;

10. The intelligent monitoring and fault diagnosis system for the operation state of an annealing furnace according to claim 9, wherein the method for determining the abnormal type data of the abnormal annealing furnace according to the abnormal characteristic data comprises the following steps: