CN116463687A

CN116463687A - Abnormal monitoring method and system based on out-of-groove information detection

Info

Publication number: CN116463687A
Application number: CN202210030216.7A
Authority: CN
Inventors: 林建平; 胡夏斌; 林建灶; 叶栋; 徐关峰
Original assignee: Hangzhou Sanal Environmental Technology Co ltd
Current assignee: Hangzhou Sanal Environmental Technology Co ltd
Priority date: 2022-01-12
Filing date: 2022-01-12
Publication date: 2023-07-21
Also published as: CN114318426A; CN114318426B

Abstract

The embodiment of the specification provides an anomaly monitoring method based on out-of-groove information detection, which comprises the following steps: obtaining detection information of a sample to be tested, wherein the sample to be tested comprises at least one of metal generated in electrolytic production and metal obtained after the electrolytic production is completed, and the detection information comprises at least one of image information, component information and temperature information; judging whether the electrolytic production is abnormal or not based on the detection information.

Description

Abnormal monitoring method and system based on out-of-groove information detection

Technical Field

The specification relates to the field of electrolytic production, in particular to an abnormality monitoring method and system based on out-of-tank information detection.

Background

The electrolytic discharging tank is the last link in the electrolytic production period, and the abnormal condition detection and monitoring of the electrolytic discharging tank is an important means for ensuring the normal operation of the electrolytic production. In the detection of the electrolytic cell, the detection of parameters such as a cathode plate of the cell, a residual anode, electrolyte in the electrolytic cell and the like is helpful for judging abnormal conditions generated in the electrolytic production link, so that the parameters and operation in the production can be timely adjusted and improved.

At present, the detection of the out-slot information mainly depends on manual work, and the detection mode is single, the efficiency is low and the subjectivity is often provided. In addition, acidic substances in the electrolysis shop have corrosiveness and toxicity, and long-time detection work can harm human health. Therefore, it is necessary to provide a more efficient method for detecting and monitoring the out-slot information to solve the above-mentioned problems.

Disclosure of Invention

One of the embodiments of the present disclosure provides an anomaly monitoring method based on out-of-groove information detection, the anomaly monitoring method including: obtaining detection information of a sample to be tested, wherein the sample to be tested comprises at least one of metal generated in electrolytic production and metal obtained after the electrolytic production is completed, and the detection information comprises at least one of image information, component information and temperature information; judging whether the electrolytic production is abnormal or not based on the detection information.

One of the embodiments of the present disclosure provides an anomaly monitoring system based on out-of-groove information detection, where the anomaly monitoring system includes an information acquisition module and a status judgment module; the information acquisition module is used for acquiring detection information of a sample to be tested, wherein the sample to be tested comprises at least one of metal generated in electrolytic production and metal obtained after the electrolytic production is completed, and the detection information comprises at least one of image information, component information and temperature information; the state judgment module is used for judging whether the electrolytic production is abnormal or not based on the detection information.

One of the embodiments of the present disclosure provides an anomaly monitoring device based on out-of-groove information detection, which includes a processor configured to execute the out-of-groove information detection-based anomaly monitoring method described in any one of the embodiments of the present disclosure.

One of the embodiments of the present disclosure provides a computer-readable storage medium storing computer instructions, where when the computer reads the computer instructions in the storage medium, the computer executes the anomaly monitoring method based on out-slot information detection according to any one of the embodiments of the present disclosure.

One of the embodiments of the present specification provides an exception handling method based on out-of-slot information detection, the exception handling method including: obtaining detection information of a sample to be tested, wherein the sample to be tested comprises at least one of metal generated in electrolytic production and metal obtained after the electrolytic production is completed, and the detection information comprises at least one of image information, component information and temperature information; determining production anomaly information based on the detection information; feedback information is determined based on the production anomaly information.

One embodiment of the present disclosure provides an exception handling system based on out-of-slot information detection, where the exception handling system includes an information acquisition module, a status determination module, and an exception handling module; the information acquisition module is used for acquiring detection information of a sample to be tested, wherein the sample to be tested comprises at least one of metal generated in electrolytic production and metal obtained after the electrolytic production is completed, and the detection information comprises at least one of image information, component information and temperature information; the state judging module is used for determining production abnormality information based on the detection information; the abnormality processing module is used for determining feedback information based on the production abnormality information.

One of the embodiments of the present disclosure provides an exception handling apparatus based on out-slot information detection, which includes a processor configured to execute the out-slot information detection-based exception handling method described in any one of the embodiments of the present disclosure.

One of the embodiments of the present disclosure provides a computer-readable storage medium storing computer instructions, where when the computer reads the computer instructions in the storage medium, the computer executes the anomaly handling method based on out-slot information detection according to any one of the embodiments of the present disclosure.

Drawings

The present specification will be further elucidated by way of example embodiments, which will be described in detail by means of the accompanying drawings. The embodiments are not limiting, in which like numerals represent like structures, wherein:

FIG. 1 is a schematic illustration of an application scenario of an anomaly monitoring system based on out-of-groove information detection, according to some embodiments of the present disclosure;

FIG. 2 is a schematic diagram of an anomaly monitoring system and/or anomaly handling system based on out-of-slot information detection, according to some embodiments of the present specification;

FIG. 3 is an exemplary flow chart of anomaly monitoring based on out-of-slot information detection, according to some embodiments of the present disclosure;

FIG. 4 is a schematic diagram of data acquisition of an exception monitoring system and/or an exception handling system, according to some embodiments of the present description;

FIG. 5 is a schematic diagram illustrating a data processing and anomaly determination flow according to some embodiments of the present disclosure;

FIG. 6 is a schematic illustration of a machine learning model shown in some embodiments of the present description determining production anomaly information.

Detailed Description

In order to more clearly illustrate the technical solutions of the embodiments of the present specification, the drawings that are required to be used in the description of the embodiments will be briefly described below. It is apparent that the drawings in the following description are only some examples or embodiments of the present specification, and it is possible for those of ordinary skill in the art to apply the present specification to other similar situations according to the drawings without inventive effort. Unless otherwise apparent from the context of the language or otherwise specified, like reference numerals in the figures refer to like structures or operations.

It will be appreciated that "system," "apparatus," "unit" and/or "module" as used herein is one method for distinguishing between different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

As used in this specification and the claims, the terms "a," "an," "the," and/or "the" are not specific to a singular, but may include a plurality, unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that the steps and elements are explicitly identified, and they do not constitute an exclusive list, as other steps or elements may be included in a method or apparatus.

A flowchart is used in this specification to describe the operations performed by the system according to embodiments of the present specification. It should be appreciated that the preceding or following operations are not necessarily performed in order precisely. Rather, the steps may be processed in reverse order or simultaneously. Also, other operations may be added to or removed from these processes.

Fig. 1 is a schematic view of an application scenario of an anomaly monitoring system based on out-of-groove information detection according to some embodiments of the present disclosure.

In some embodiments, the application scenario 100 may include a server 110, a network 120, one or more terminal devices 130, a storage device 140, and an acquisition detection apparatus 150. The server 110 may determine whether the sample to be tested is abnormal by the collection testing device 150 by practicing the methods and/or processes disclosed herein. In some embodiments, the sample to be tested may be an out-of-cell plate (e.g., anode plate or cathode plate) after completion of the electrolysis or an in-process plate (e.g., anode plate or cathode plate) being generated by the electrolysis. By detecting the metal on the polar plate, the system can determine abnormal information of the production link based on the detected abnormal condition, so that the abnormal information is fed back to the production link and correspondingly adjusted and improved, and the production efficiency of subsequent electrolysis is improved.

The server 110 may communicate with the terminal device 130, the storage device 140, and/or the acquisition detection apparatus 150 to provide various functions of the application scenario 100. In some embodiments, the server 110 may receive, from the terminal device 130 via, for example, the network 120, relevant detection information after detection of the sample to be tested by the collection detection means 150.

In some embodiments, the server 110 may be a single server or a group of servers. In some embodiments, server 110 may be connected locally to network 120 or remotely from network 120. In some embodiments, server 110 may be implemented on a cloud platform.

The network 120 may facilitate the exchange of information and/or data. In some embodiments, one or more components in the application scenario 100 (e.g., the server 110, the network 120, etc.) may send information and/or data to another component in the application scenario 100 via the network 120.

In some embodiments, the user may be the owner of the terminal device 130. The terminal device 130 may receive the user request and transmit information related to the request to the server 110 via the network 120. For example, the terminal device 130 may receive a request from a user to transmit the detection information of the sample to be tested, and transmit information related to the request to the server 110 via the network 120. Terminal device 130 may also receive information from server 110 via network 120. For example, the terminal device 130 may receive detection information of the sample to be tested related to the collection detection means 150 from the server 110. The determined one or more detection information may be displayed on the terminal device 130. For another example, the server 110 may transmit abnormality information generated based on the detection information to the terminal device 130. The feedback information may include electrolysis related information (e.g., in the case of impurities) of the sample to be tested, production improvement advice, and the like.

In some embodiments, the terminal device 130 may include a mobile device, a tablet computer, a laptop computer, an in-vehicle device, or the like, or any combination thereof. In some embodiments, the terminal device 130 may include a signal transmitter and a signal receiver configured to communicate with the acquisition detection apparatus 150 to obtain detection information of the sample to be tested.

In some embodiments, the storage device 140 may be connected to the network 120 to communicate with one or more components of the application scenario 100 (e.g., the server 110, the terminal device 130, the acquisition detection apparatus 150). In some embodiments, the storage device 140 may be part of the server 110.

The storage device 140 may store data and/or instructions. The data may include data related to the user, the terminal device 130, the acquisition detection means 150, etc. In some embodiments, the storage device 140 may store data acquired from the terminal device 130 and/or the acquisition detection apparatus 150. In some embodiments, storage device 140 may store data and/or instructions used by server 110 to perform or use to accomplish the exemplary methods described herein.

In some embodiments, storage device 140 may include mass storage, removable storage, volatile read-write memory, read-only memory (ROM), and the like, or any combination thereof. In some embodiments, storage device 140 may be implemented on a cloud platform.

The collection and detection device 150 may obtain detection information of the sample to be tested, and further description of the detection information may be referred to fig. 3 and related description thereof, which are not repeated herein. In some embodiments, the acquisition and detection device 150 may include an infrared detection module 150-1, an acquisition image module 150-2, a temperature detection module 150-3, an ultrasonic detection module 150-4, and a sampling detection module 150-5, and further description of the infrared detection module 150-1, the acquisition image module 150-2, the temperature detection module 150-3, the ultrasonic detection module 150-4, and the sampling detection module 150-5 may be found in FIG. 4 and related descriptions thereof, which are not repeated herein.

It should be noted that the application scenario 100 is provided for illustrative purposes only and is not intended to limit the scope of the present application. Many modifications and variations will be apparent to those of ordinary skill in the art in light of the present description. For example, the application scenario 100 may also include a database. However, such changes and modifications do not depart from the scope of the present application.

FIG. 2 is a schematic diagram of an anomaly monitoring system and/or anomaly handling system based on out-slot information detection, according to some embodiments of the present specification.

In some embodiments, the anomaly monitoring system 200 based on out-of-groove information detection may include an information acquisition module 210 and a status determination module 220. In some embodiments, exception monitoring system 200 may be subdivided into a plurality of specific sub-systems based on different functions, e.g., exception monitoring system 200 may include an exception handling system. In some embodiments, the exception handling system may include an information acquisition module 210, a status determination module 220, and an exception handling module 230.

The information acquisition module 210 may be configured to acquire detection information of a sample to be tested. Wherein the sample to be tested comprises at least one of metal generated in electrolytic production and metal obtained after electrolytic production. In some embodiments, the detection information includes at least one of image information, composition information, and temperature information.

In some embodiments, the information acquisition module 210 may determine the target object based on at least one of random sampling result determination, preliminary detection result determination, and other detection feedback determination; detection information of a target object can be acquired at a target time point, wherein the target time point comprises at least one time node, and the target object comprises at least one cathode plate. In some embodiments, the information obtaining module 210 may determine a key object based on a preliminary detection result of the target object, obtain a preliminary detection image of the key object, and send the preliminary detection image to a remote detection device, where the preliminary detection image meets a preset condition; acquiring an adjustment instruction and/or a demand instruction returned by the remote detection equipment; and acquiring detection information of the key object based on the adjustment instruction and/or the demand instruction.

In some embodiments, the status determination module 220 may be configured to determine whether the electrolytic production is abnormal based on the detection information. Further, the state determining module 220 may determine whether the detection information meets a preset condition; if the detection information meets the preset condition, judging that the electrolytic production is normal; and if the detection information does not meet the preset condition, judging that the electrolytic production is abnormal.

In another embodiment, the status determination module 220 may be configured to determine production anomaly information based on the detection information.

In some embodiments, the anomaly handling module 230 may be configured to determine feedback information based on the detection information in response to the electrolytic production anomaly. In another embodiment, the anomaly handling module 230 may be configured to determine feedback information based on the production anomaly information.

In some embodiments, the feedback information may include at least one of temperature adjustment information, electrolyte adjustment information, and abnormal plate position.

It should be noted that the above description of the anomaly monitoring system 200 and its modules is for convenience only and is not intended to limit the present disclosure to the scope of the illustrated embodiments. It will be appreciated by those skilled in the art that, given the principles of the system, various modules may be combined arbitrarily or a subsystem may be constructed in connection with other modules without departing from such principles. In some embodiments, the information acquisition module, the status determination module, and the exception handling module disclosed in fig. 2 may be different modules in one system, or may be one module to implement the functions of two or more modules described above. For example, each module may share one memory module, or each module may have a respective memory module. Such variations are within the scope of the present description.

FIG. 3 is an exemplary flow chart of anomaly monitoring based on out-of-slot information detection, according to some embodiments of the present description. As shown in fig. 2, the process 300 includes the following steps. In some embodiments, the process 300 may be performed by the status determination module 220.

In step 310, detection information of the sample to be tested is obtained. In some embodiments, step 310 may be performed by information acquisition module 210.

The sample to be tested may be an object or device associated with electrolytic production. Such as electrolyte, anode plate, cathode plate, etc. In some embodiments, the sample to be tested may also be an electrolytically produced substance, for example, at least one of a metal produced in electrolytic production and a metal obtained by completion of electrolytic production, and the like.

The detection information may be information related to the sample to be tested.

In some embodiments, the detection information may include appearance information, e.g., image information. Illustratively, a plate image, an electrolyte image, an image of the metal generated, and the like.

In some embodiments, the detection information may also include physical information such as electrolyte color, electrolyte temperature information, plate temperature information, quality information of the metal of the surface of the plate (e.g., surface flatness, integrity, texture and thickness uniformity, color, etc.).

In some embodiments, the electrolyte temperature information may include an electrolyte overall temperature, an electrolyte interlayer temperature difference, an electrolyte end-to-end temperature difference, and the like, and further description of the electrolyte overall temperature, the electrolyte interlayer temperature difference, and the electrolyte end-to-end temperature difference may be referred to in fig. 4 and the related description thereof, and will not be repeated here.

In some embodiments, the detection information may also be chemical information, e.g., compositional information. For example, the composition of the electrolyte, the composition of the metal generated in the electrolytic production, the composition of the metal obtained by the completion of the electrolytic production, and the like.

In some embodiments, the information acquisition module 210 may acquire the detection information through the acquisition detection device 150, and further description of the detection information may be referred to in fig. 4 and related description thereof, which are not repeated herein.

Step 320, judging whether the electrolytic production is abnormal or not based on the detection information. In some embodiments, step 310 may be performed by the status determination module 220.

The abnormal judgment of the electrolytic production can be realized based on manual or automatic identification.

In some embodiments, the operator may receive the detection information through the terminal device 130, determine whether the electrolytic production is abnormal, and then feed back the information of whether the electrolytic production is abnormal to the status determination module 220 through the terminal device 130.

In some embodiments, the status determination module 220 may determine whether the electrolytic production is abnormal based on the detection information through an abnormality determination model.

In some embodiments, the input of the anomaly determination model may be at least a portion of the detection information, and the output of the anomaly determination model may be whether the electrolytic production is abnormal.

In some embodiments, the anomaly determination model may be trained based on a plurality of labeled training samples, wherein the training samples are at least a portion of the historical detection information, and the labels of the training samples may be vectors that characterize normal electrolytic production or vectors that characterize abnormal electrolytic production.

In some embodiments, the training sample may be a combination of multiple types of information in the detection information, e.g., a combination of at least two of appearance information, physical information, and chemical information. For example, the training sample may include image information (e.g., plate image, electrolyte image, image of metal generated, etc.) in the historical detection information and quality information (e.g., surface flatness, integrity, texture, thickness uniformity of the plate, etc.) of the surface of the sample to be tested. For another example, the training sample may include quality information (e.g., surface flatness, integrity, texture, thickness uniformity of the electrode plate, etc.) of the surface of the sample to be tested in the historical detection information, and components of the electrolyte, components of the metal generated in the electrolytic production, components of the metal obtained by the electrolytic production, and the like.

In some embodiments, the state determination module 220 may perform multiple rounds of training on the abnormal determination model based on the multiple labeled training samples until the trained abnormal determination model meets a preset condition, where the preset condition may be that the loss function of the updated initial base model is less than a threshold, converges, or the number of training iterations reaches a threshold. In some embodiments, the anomaly determination model may include a convolutional neural network (Convolutional Neural Networks, CNN) model, a graph neural network (Graph Neural Networks, GNN) model, or the like.

In some embodiments, the state determining module 220 may determine whether the electrolytic production is abnormal based on the detection information through the preset condition, if the detection information satisfies the preset condition, determine that the electrolytic production is normal, and if the detection information does not satisfy the preset condition, determine that the electrolytic production is abnormal, and further description about determining whether the electrolytic production is abnormal through the preset condition may be referred to fig. 5 and related description thereof, which are not repeated herein.

In some embodiments, the process 300 may further include a step 330 of determining feedback information based on the detection information in response to the electrolytic production anomaly. In some embodiments, step 330 may be performed by exception handling module 230.

The feedback information may be information related to the adjustment of the electrolytic production. For example, the electrolytic production is suspended, parameters of the electrolytic production are adjusted (for example, the amount of purification of the electrolytic solution is increased, etc.), and the like.

In some embodiments, the feedback information may also include a later-produced adjustment scheme, which may be a scheme of electrolytic production performed at least one point in time in the future. In some embodiments, the adjustment scheme for subsequent production may include process criteria related to at least one feedstock used for electrolytic production. For example, the process criteria may include temperature range criteria for the electrolyte, electrolyte composition criteria, additive composition criteria, and the like.

The temperature adjustment information may be information related to adjusting the temperature of the electrolyte. The overall temperature of the electrolyte, the interlayer temperature difference and the end-to-end temperature difference can greatly influence the electrolytic production, for example, the overall temperature of the electrolyte is too low to cause crystallization near the polar plate, the overall temperature of the electrolyte can cause uneven crystallization on the surface of the polar plate, and therefore, the quality of metal produced by subsequent electrolysis can be influenced in both cases. Therefore, when the overall temperature of the electrolyte is abnormal, the interlayer temperature difference is abnormal, or the inter-terminal temperature difference is abnormal, the overall temperature of the electrolyte, the interlayer temperature difference, and the inter-terminal temperature difference can be increased or decreased by decreasing or increasing the flow rate of the electrolyte.

The electrolyte adjustment information may be information related to adjusting the electrolyte composition. The content of the additive in the electrolyte greatly influences the appearance of the metal generated by electrolysis, and thus, electrolyte adjustment information can be determined according to the appearance of the metal generated in the electrolysis production and the metal obtained by the completion of the electrolysis production. For example, the surface of the electrolytically produced metal is uneven and textured, and the amount of bone glue added needs to be increased. For example, the surface of the electrolytically generated metal has particles and needle-like projections, and it is necessary to reduce the amount of bone cement added.

The abnormal plate location may be a location of an electrolytically produced abnormal plate. In some embodiments, the anomaly handling module 230 may obtain the anomaly plate location via the acquisition detection device 150. For example, the anomaly handling module 230 may obtain an image of the environment in which the plate is located via the capture image module 150-2 and identify the position of the plate based on the image.

In some embodiments, the exception handling module 230 may determine the feedback information directly based on the detection information. In some embodiments, the anomaly processing module 230 may determine the feedback information based on at least a portion of the detection information when the at least a portion of the detection information exceeds a preset condition. For example, when the electrolyte temperature in the electrolyte information is detected to be outside of a preset range, the abnormality processing module 230 may determine to increase or decrease the flow rate of the electrolyte into and out of the electrolytic cell based on the electrolyte temperature to adjust the electrolyte temperature.

In some embodiments, the anomaly handling module 230 may determine production anomaly information based on the detection information and then determine feedback information based on the production anomaly information.

The production anomaly information may characterize information related to an electrolytic production anomaly. In some embodiments, the production anomaly information may be anomaly information during electrolytic production or anomaly information after electrolytic production is complete.

In some embodiments, the anomaly handling module 230 may determine production anomaly information based on the detection information of the sample to be tested acquired by the information acquisition module 210. In some embodiments, the production anomaly information may include at least one of: electrolyte anomaly information, plate anomaly information, and generated metal anomaly information.

In some embodiments, the electrolyte anomaly information may further include an excessive electrolyte temperature, an abnormal electrolyte composition, etc., the plate anomaly information may further include a non-uniform thickness, a surface unevenness, etc., and the generated metal anomaly information may include excessive metal impurities, etc. Generating the metal anomaly information may include anomalies in metal composition, texture anomalies, and the like.

In some embodiments, the anomaly processing module 230 may determine the anomaly information based on at least a portion of the detection information when the at least a portion of the detection information exceeds a preset condition. For example, when it is detected that the electrolyte temperature in the electrolyte information exceeds the preset range, the abnormality processing module 230 may determine production abnormality information based on the electrolyte temperature: the electrolyte temperature is too high or too low.

In some embodiments, the anomaly handling module 230 may determine production anomaly information based on the detection information via a machine learning model, and further description of the machine learning model may be found in FIG. 6 and its associated description, which is not repeated herein.

In some embodiments, the exception handling module 230 may determine the feedback information based on production exception information. In some embodiments, the exception handling module 230 may determine the corresponding feedback information based on the type of production exception information. For example, when the production anomaly information includes an excessive electrolyte temperature, the anomaly handling module 230 may determine feedback information: the flow rate of electrolyte into or out of the cell is increased.

In some embodiments, by acquiring detection information of a sample to be tested and judging whether the electrolytic production is abnormal based on the detection information of the sample to be tested, the electrolytic production can be adjusted in time when the electrolysis is abnormal, and the efficiency and quality of the electrolytic production are improved.

It should be noted that the above description of the process 300 is for purposes of example and illustration only and is not intended to limit the scope of applicability of the present disclosure. Various modifications and changes to flow 300 will be apparent to those skilled in the art in light of the present description. However, such modifications and variations are still within the scope of the present description.

FIG. 4 is a schematic diagram of data acquisition of an exception monitoring system and/or an exception handling system, according to some embodiments of the present description.

The anomaly monitoring system and/or the anomaly handling system may obtain detection information via the acquisition detection device 150.

The acquisition and detection device 150 is a device for detecting relevant data (e.g., detection information) of a sample to be tested. The related data may be appearance information, physical information and chemical information of the sample to be tested, and further description about the appearance information, the physical information and the chemical information may be referred to fig. 3 and related description thereof, which are not repeated herein.

In some embodiments, the acquisition and detection device 150 may acquire the detection information based on at least one of image, infrared, laser, sampling, and ultrasonic. For example, the acquisition detection device 150 may include an infrared acquisition detection device 150-1, an image acquisition device 150-2, a temperature acquisition device 150-3, an ultrasonic device 150-4, and a sampling detection module 150-5.

In some embodiments, temperature information of the sample to be tested may be obtained based on the acquisition detection device 150.

In some embodiments, the infrared acquisition and detection device 150-1 may be used to obtain temperature information of the sample to be tested. For example, the infrared acquisition and detection device 150-1 (e.g., a thermal infrared imager, etc.) can perform infrared thermal imaging on the sample to be tested, receive an infrared specific band signal of thermal radiation of the sample to be tested, convert the signal into an image and a graph that can be distinguished by human vision, and further calculate a temperature value of the sample to be tested.

In some embodiments, a temperature detection module 150-3 (e.g., a temperature sensor, etc.) may be used to obtain temperature information for the sample to be tested. For example, the temperature detection module 150-3 may include a plurality of temperature sensors mounted at different locations of the electrolytic cell to obtain temperatures of different locations of the electrolyte or the plate. For example, a temperature sensor mounted in the electrolyte may be used to obtain the electrolyte temperature, and for example, a temperature sensor mounted on the surface of the plate may be used to obtain the plate temperature.

In some embodiments, excessive electrolyte temperature differences (e.g., temperature differences greater than 1 ℃) at different locations in the cell, or too low electrolyte temperatures, can cause crystallization near the plates, thereby affecting the quality of the metal produced by subsequent electrolysis. The electrolyte temperature to be obtained may include at least two electrolyte temperatures measured at least two different depths of the electrolytic cell, wherein a difference between the two electrolyte temperatures measured at the two depths is a temperature difference between electrolyte layers.

In some embodiments, at least two temperature sensors may be disposed at different depths of the electrolytic cell to enable measurement of at least two electrolyte temperatures at different depths. For example, the structure in which the temperature sensors are arranged in the electrolyte may be a three-layer structure, the uppermost temperature sensor being located below the electrolyte level in the electrolytic cell, the middle temperature sensor being located at a middle height of the electrolytic cell, and the lowermost temperature sensor being located at the bottom of the electrolytic cell.

In some embodiments, the obtained electrolyte temperature may further include an electrolyte temperature at the water inlet end and the water outlet end of the electrolyte, wherein a difference between the electrolyte temperatures measured at the water inlet end and the water outlet end is a temperature difference between the electrolyte ends. Accordingly, the temperature detection module 150-3 may include two temperature sensors respectively installed at the water inlet end and the water outlet end of the electrolyte.

In some embodiments, image information of the sample to be tested may be acquired based on the acquisition detection device 150.

In some embodiments, the acquisition image module 150-2 (e.g., camera, image sensor, etc.) may be used to acquire image information of the sample to be tested. For example, the image module 150-2 may collect image information of a sample to be tested and perform image recognition to obtain information of color, texture, shape, spatial relationship, etc. of the cathode plate and/or the anode plate during or after the electrolytic production.

Image recognition may include processing, analysis, and understanding to identify targets and objects in various different modes. For example, image acquisition may be performed by the acquisition image module 150-2; preprocessing the acquired image, eliminating irrelevant information in the image, recovering useful real information, enhancing the detectability of related information and simplifying data to the maximum extent; and extracting the characteristics of the preprocessed image, wherein the color characteristics, texture characteristics, shape characteristics and spatial relation characteristics of the sample to be tested can be extracted based on algorithms such as a direction gradient Histogram (HOG) characteristic, a Local Binary Pattern (LBP) characteristic, a Haar-like characteristic and the like.

In some embodiments, quality information of the metal of the surface of the plate may be obtained based on the acquisition detection device 150.

In some embodiments, an ultrasonic detection module 150-4 (e.g., a pulse-reflection ultrasonic flaw detector) may be used to collect quality information (e.g., surface flatness, integrity, texture, thickness uniformity of the plate, etc.) of the surface of the sample to be tested.

In some embodiments, the infrared acquisition detection device 150-1 may determine quality information by infrared thermal imaging of the acquired sample to be tested and based on the thermally imaged image.

In some embodiments, a laser device (e.g., a laser thickness gauge) may be used to collect quality information of the sample to be tested. For example, the laser device may measure the thickness of the cathode plate and/or anode plate at different locations.

In some embodiments, the composition information of the sample to be tested may be acquired based on the acquisition detection device 150.

In some embodiments, the sampling detection module 150-5 can be used to randomly sample a sample to be tested and analyze the chemical composition of the sample to be tested. For example, the sample detection module 150-5 may randomly sample the anode plate, punch holes for the cathode plate and/or the anode plate, and analyze the sample for chemical components; as another example, the sample detection module 150-5 may perform chemical component analysis on particulate samples of the surfaces of the cathode and/or anode plates, for example, to analyze the purity of the metal product (e.g., copper) and impurity conditions, including the type and content of the primary impurities (e.g., silver and its content).

In some embodiments, the infrared acquisition detection device 150-1, the image acquisition device 150-2, the temperature acquisition device 150-3, the ultrasonic device 150-4, and the sample detection module 150-5 described above may also perform other functions. For example, the infrared acquisition and detection device 150-1 may also be used to obtain positional information of a sample to be tested. For example, the infrared acquisition and detection device 150-1 may acquire an infrared thermal image of the electrolytic cell through a thermal imager, process the infrared thermal image to obtain a pixel point of the polar plate with abnormal temperature, and finally determine a corresponding polar plate position according to the pixel point.

In some embodiments, the collection and detection device 150 may further include other devices for obtaining detection information, for example, the collection and detection device 150 may further include a smart socket, a smart meter, and other devices for collecting in-cell voltage, in-cell current, and the like in the electrolytic cell.

The multiple detection modes shown in some embodiments can be implemented as required, for example, the same sample to be detected (such as electrolyte, cathode plate and/or anode plate during production or after leaving the tank) can be detected sequentially by multiple detection modes to obtain more accurate results, or the initial detection can be performed in part of the detection modes, and the further detection targets and modes can be determined according to the initial detection results.

In some embodiments, the infrared acquisition and detection device 150-1, the image acquisition device 150-2, the temperature acquisition device 150-3, the ultrasonic device 150-4, and the sample detection module 150-5 may sequentially perform data acquisition on the sample to be tested.

For example, for a cathode plate in production to be detected or completed production, the surface quality of the cathode plate can be obtained by infrared thermal imaging of the cathode plate by the infrared acquisition and detection device 150-1, the image of the cathode plate can be obtained by the image acquisition device 150-2, the surface temperature of the cathode plate can be measured by the temperature acquisition device 150-3, the flatness of the surface of the cathode plate can be measured by the ultrasonic device 150-4, the cathode plate can be punched and sampled by the sampling detection module 150-5, the components of metal generated by electrolysis on the cathode plate can be obtained, and finally the thicknesses of different positions of the cathode plate can be measured by the laser device.

For another example, for the anode plate and electrolyte in the electrolysis to be detected or after the electrolysis is completed, the electrolyte is subjected to infrared thermal imaging outside the electrolytic tank through an infrared acquisition and detection device 150-1, the temperature in the tank is acquired, an image of the anode plate is acquired through an image acquisition device 150-2, the surface temperature of the anode plate is measured through a temperature acquisition device 150-3, the flatness of the surface of the anode plate is measured through ultrasonic waves of an ultrasonic device 150-4, the anode plate is punched and sampled through a sampling detection module 150-5, the components of the metal remained on the anode plate in the electrolysis are acquired, and finally the thicknesses of different positions of the anode plate are measured through a laser device.

In another embodiment, each sample to be tested may be tested in a partial test mode, and the targets and modes for further testing may be determined based on the test results.

For example, the physical appearance of the cathode plate and/or the anode plate may be detected in two ways (e.g., image acquisition detection, infrared thermal imaging), and if the detection results obtained in the two ways are consistent, no further detection is performed; if the detection results obtained by the two methods are inconsistent, a third method (such as component detection) is adopted for further detection.

For example, firstly randomly sampling, punching holes on the sampled cathode plate and/or anode plate, sampling and analyzing chemical components, and if the copper content and impurity content of all the sampled cathode plates reach the standard, indicating that the product qualification rate of the batch is higher, then only adopting other modes (such as component detection) to identify and judge the physical appearance of the anode plate; if the polar plate with copper content or impurity content not reaching the standard appears, two or more other modes (such as laser detection and ultrasonic detection) are adopted to carry out physical appearance identification on the cathode plate, so that the accuracy is improved.

In some embodiments, the collection testing device 150 may be moved and/or the sample to be tested may be moved to complete the test.

In some embodiments, the sample to be tested may be moved by one or more movement means to complete the test. The moving device can be a conveyor belt, a translational hoist, a pulley, etc. For example, one or more components of the acquisition detection device 150 may be disposed in a fixed location. One or more transfer devices may be disposed within the detection zone. The cathode plate after leaving the groove can be placed on a conveying device. The cathode plate moves to the position of the collecting and detecting device 150 under the driving of the conveying device, and the collecting and detecting device 150 can correspondingly detect the cathode plate. After the cathode plate is detected, the sorting device (such as a mechanical arm and the like) can sort the qualified products and the unqualified products to different conveying channels, or place the polar plates to be subjected to subsequent detection (such as chemical component analysis) in corresponding detection areas. In this case, the command control may be sent to trigger a certain detection mode or to switch off a certain detection mode according to the detected requirement.

In some embodiments, the acquisition detection device 150 may be driven by one or more mobile devices to perform movement completion detection. For example, the mobile device may drive the acquisition and detection device 150 mounted on the mobile device to move to an area in need of detection for detection. The installation position of the mobile device can be determined according to acquisition requirements. Such as ceilings, floors, etc. The mobile device can adaptively adjust parameters such as distance, angle and the like between the acquisition and detection device and the detection area and/or the sample to be tested, so as to obtain the optimal detection effect.

In some embodiments, the mobile device may be one or more mobile robots. For example, one or more mobile robots may be provided at the electrolytic production plant as desired. The collection and detection device 150 may be mounted on a mobile robot, and send instructions to the mobile robot (or control the mobile robot to move to a corresponding detection area according to a preset rule) for detection according to different detection requirements. For example, chemical analysis by random sampling reveals that the abnormality rate of an electrolytic cell or cells is high, and important detection is required. The anomaly monitoring system can mobilize the mobile robot to the corresponding key monitoring area, and the acquisition and detection device 150 can detect the area.

In some embodiments, the mobile device may be one or more drones. For example, the acquisition and detection device 150 may be equipped on a drone for movement detection. For another example, image acquisition may be performed by a camera and/or an image sensor built into the drone. The image of the sample to be detected, which is acquired by the unmanned aerial vehicle, can be subjected to image recognition, and the sample to be detected is judged. If the unmanned aerial vehicle can shoot the image of the cathode plate, the image is subjected to the processes of enhancing, denoising, image segmentation, edge denoising and the like according to the characteristics of the image, the image characteristics are extracted, and finally the image is judged and classified (qualified or unqualified).

In some embodiments, the drone may be autonomously positioned and attitude adjusted. For example, the unmanned aerial vehicle can adopt optical flow technology to realize indoor positioning, adopts an ultrasonic sensor to control indoor positioning height, adopts an Inertial Measurement Unit (IMU) to detect the attitude change of the aircraft and adjusts in real time. The unmanned aerial vehicle converts the acquired information such as pixel distribution, color, brightness and the like of the image information into digital signals through a built-in optical flow sensor, and transmits the digital signals to a built-in image processing system (which can be a processing system of an image recognition module) to perform various operations so as to extract the characteristics of a target, and further controls the unmanned aerial vehicle to act according to the judging result; judging the relative height by an ultrasonic sensor; through efficient vision processor calculation, the unmanned aerial vehicle realizes accurate indoor positioning hovering and stable flying.

In some embodiments, the drone may perform positioning and attitude adjustments on command. For example, the unmanned aerial vehicle may collect an image, transmit the image to the terminal device, and the processor or the user of the terminal device may analyze the image condition (such as resolution of the image, integrity of the image of the sample to be tested, angle and distance of shooting, etc.), send an adjustment instruction according to the analysis result, and adjust parameters (such as movement direction, distance, etc.) of the unmanned aerial vehicle.

In some embodiments, the terminal device may intelligently generate and send the adjustment instruction according to the acquired image condition. For example, by inputting image conditions into the machine learning model, adjustment instructions are regenerated and output. In some embodiments, the image condition may be fed back on the interface of the terminal device, and after analysis by the user, the user issues an adjustment instruction through the terminal device.

In some embodiments, the position of the sample to be tested may be adjusted by an adjustment device. The adjusting device can be a mechanical arm, a pulley and the like. For example, an adjusting device may be used to adjust the position and direction of the cathode plate during the detection process, so as to facilitate the detection by the collection and detection device 150. The abnormality monitoring system or the abnormality processing system can receive an instruction to sort the polar plates based on the detected polar plate abnormality.

In some embodiments, the collection detection device 150 may include auxiliary lights. The auxiliary light may be used to illuminate one or more components of the collection testing device 150 when a sample to be tested is being tested.

In some embodiments, the auxiliary light in the detection process can be adjusted according to lighting conditions in different time periods. For example, the light may be insufficient in the morning or evening, and the light may be automatically adjusted according to the current brightness and the brightness required for detection to perform intelligent light supplement.

In some embodiments, the auxiliary light may be adjusted according to the detected demand. In some embodiments, the brightness level of the auxiliary light can be controlled correspondingly according to different brightness levels of the light required by different detection modes. For example, the detection of image information requires stronger illumination brightness, so the light brightness level should be correspondingly increased; the brightness of the infrared detection is not particularly limited, so that the brightness level of the lamplight can be correspondingly high or low.

In some embodiments, detection information of a sample to be tested may be obtained, including: determining a target object based on at least one of random sampling result determination, preliminary detection result determination, and other detection feedback determination; and acquiring detection information of the target object at a target time point, wherein the target time point comprises at least one time node, and the target object comprises at least one cathode plate.

The random sampling result determination refers to determining whether a sample is abnormal or not based on one or more pieces of detection information obtained after at least one sample is randomly selected from a plurality of samples to be tested for detection. For example, at least one cathode plate is randomly extracted from the plurality of cathode plates, the extracted cathode plates are punched and sampled, chemical composition analysis is performed to obtain composition information, and whether the extracted cathode plates are abnormal or not is determined according to the composition information (such as copper content).

The preliminary detection result determination refers to determining whether a sample is abnormal or not based on detection information obtained after primary screening of one or more samples to be tested. For example, the physical appearance of each cathode plate after the discharge is subjected to naked eye identification or image acquisition to obtain detection information related to the physical appearance, and whether the cathode plate is abnormal or not is primarily screened according to the detection information related to the physical appearance.

In some embodiments, the failed sample may be screened based on the preliminary test results, and other tests may be performed on the failed sample. For example, defective cathode plates are screened out based on physical appearance, and the defective cathode plates are punched or copper packets on the surface of the cathode plates are collected for chemical analysis.

Other detection feedback determination refers to determining whether a sample is abnormal based on detection information of relevant data in the production process of the sample to be tested. For example, the electrolyte and/or the additive are subjected to component detection to obtain component information, and whether the electrolyte and/or the additive are abnormal or not is determined based on the component information.

In some embodiments, if the other detection results of the sample to be tested in a certain batch are abnormal, the product in the certain batch is subjected to important detection. For example, the electrolyte and/or additives in at least a portion of the cells in the electrolysis system 410 are subjected to composition detection. In some embodiments, the electrolysis system 410 may include a plurality of electrolysis cells, such as electrolysis cell 410-1, electrolysis cells 410-2, …, electrolysis cell 410-n. The electrolyte and/or additives in the electrolytic cell 410-1 may be subjected to composition detection, and if it is determined that the electrolyte and/or additives in the electrolytic cell 410-1 are abnormal in composition, the cathode plate produced in the electrolytic cell 410-1 is subjected to focus detection.

The target object refers to at least one sample which is screened and determined from a plurality of samples to be tested and is used for reflecting whether production is abnormal or not. For example, the target object comprises at least one cathode plate. In some embodiments, the at least one target object may be determined based on at least one of a random sampling result determination, a preliminary detection result determination, and other detection feedback determination. For example, an abnormal cathode plate determined by random sampling results, an abnormal cathode plate determined by preliminary detection results, and an abnormal cathode plate of a certain batch determined by other detection feedback can be used as target objects.

The target time point refers to a time at which the target object is detected.

In some embodiments, the target point in time may be a point in time. For example, the time the cathode plate is out of the tank. In other embodiments, the target time point may be a plurality of time points. For example, at several points in time during electrolysis, several points in time after the cell exit. In some embodiments, multiple time points may be set at certain time intervals within one electrolysis cycle. For example, 1 hour apart. In some embodiments, a plurality of points in time may be set for the problem-prone node based on the historical data. For example, because of a lower air temperature, anomalies are more likely to occur in the morning and evening, multiple points in time in the morning and/or evening may be determined as target points in time.

In some embodiments, the target point in time may be determined according to at least one of a predetermined rule, a user instruction, a previous detection result, and other monitoring results.

The predetermined rule refers to a predetermined regular point in time. In some embodiments, the predetermined rule may be a fixed detection time point. For example, 1 hour apart. In some embodiments, the pre-rules may be that the target points in time of the accent areas should be less spaced than the target points in time of the non-accent areas. For example, the emphasized areas are spaced 1 hour apart and the non-emphasized areas are spaced 2 hours apart.

The user instruction refers to information about a set time point issued by the user. In some embodiments, the user may set the detection time point by himself according to the production situation requirement. For example, electrolytic production is greater in noon, and can be detected at 1 hour intervals set in noon; the electrolytic production has smaller yield in the morning and evening, and can be detected at 2-hour intervals in the morning and evening.

In some embodiments, the detection results include one or more of random sampling detection, initial detection, and other detection results. It is possible to determine when abnormality is likely to occur based on the previous detection data, which is set as an important detection time point. For example, because of a lower air temperature, anomalies are more likely to occur in the morning and evening, multiple points in time in the morning and/or evening may be determined as target points in time.

The monitoring result comprises a result of monitoring related data in the production process in real time. For example, electrolyte monitoring results, additive monitoring results, circulation amount monitoring results, and the like. In some embodiments, the detection time point may be determined according to the condition of the monitoring result. For example, if the circulating amount of the electrolyte is monitored to be stable, the detection is performed at intervals of 2 hours; if the circulating amount of the electrolyte is detected to be unstable, the detection is carried out at 1 hour intervals.

In some embodiments, the detection information of the target object may be acquired based on the acquisition detection device, and the detection information may be acquired by at least one of the following ways: moving the target object to an acquisition and detection device for detection and acquisition; detecting and acquiring the target object through the acquisition and detection device; based on unmanned aerial vehicle acquisition equipped with acquisition detection means.

For more description of moving the target object to the collection and detection device for detection, reference may be made to the above description of moving the sample to be detected and the related description thereof, which is not repeated here.

For more description of detection acquisition by acquiring the detection device to the target object, reference may be made to the foregoing description of acquisition of detection information and related description thereof, which is not repeated here.

For further description of acquisition based on a drone equipped with acquisition detection means, reference may be made to the previous description of the drone and its related description, which is not repeated here.

In some embodiments, a key object may be determined based on the preliminary detection of the target object; acquiring a preliminary detection image of the key object and sending the preliminary detection image to remote detection equipment, wherein the preliminary detection image meets preset conditions; acquiring an adjustment instruction and/or a demand instruction returned by the remote detection equipment; and acquiring detection information of the key object based on the adjustment instruction and/or the demand instruction.

The key object refers to the production lot and/or region in which the target object needs to be further detected. In some embodiments, the key object may be obtained according to the preliminary detection result of the target object. For example, if the surface of the cathode plate determined by the preliminary detection result is uneven, the batch where the cathode plate is located is determined to be an important object. For another example, if it is determined by chemical composition analysis that the residual copper content of the anode plate exceeds a threshold, then the lot in which the anode plate is located is determined to be the subject of emphasis. For another example, if the temperature of the electrolytic cell is abnormal and the product quality is possibly affected by the preliminary detection of the infrared thermal image, the batch of products corresponding to the electrolytic cell are taken as key objects, and meanwhile, the condition of the electrolytic cell is monitored in a key way.

In some embodiments, the detection device detects key objects (such as the batch of products and other links where the electrolysis may be abnormal) to obtain a preliminary detection graph; transmitting the preliminary detection image to the remote detection equipment, wherein the preliminary detection image needs to meet preset conditions. In some embodiments, the preset conditions may include parameters such as resolution of the image and image integrity of the key object. In some embodiments, the processor or user of the remote detection device may analyze the image quality to determine whether a preset condition is met.

In some embodiments, the detection information of the key object may be obtained based on the adjustment instruction and/or the demand instruction. The adjustment instruction refers to information for adjusting parameters (e.g., angle, distance) of the detection device. In some embodiments, an adjustment instruction may be sent according to the condition of the preliminary detection image (such as resolution of the image, integrity of the image of the sample to be tested, shooting angle and distance, etc.), so as to adjust parameters of the detection device, so as to obtain detection information of the key object.

In some embodiments, the remote detection device may intelligently generate the adjustment instruction according to the condition of the preliminary detection image, and send the adjustment instruction to the detection device.

In some embodiments, the image condition may be fed back at the interface of the remote detection device, and after analysis by the user, the remote detection device sends an adjustment instruction to the detection device according to the information of the demand instruction (e.g., the angle is rotated by 30 °/45 ° counterclockwise/clockwise, etc., the distance is detected to be more or less 1cm, the lens is moved back and forth, etc.) input at the remote detection device.

FIG. 5 is a flow chart illustrating data processing and anomaly determination according to some embodiments of the present disclosure. As shown in fig. 5, the process 500 includes the following steps.

Step 510, obtaining a preset condition. In some embodiments, step 510 may be performed by a status determination module.

The preset conditions are one or more conditions which are preset and are required to be met for judging whether the sample to be tested is qualified.

In some embodiments, the setting of the preset conditions includes, but is not limited to, physical appearance, chemical composition of the reference cathode plate; temperature distribution of the electrolyte, electrolyte flow rate, current density, electrolyte composition, impurity content, and the like. For example, the preset conditions for judging that the cathode plate after the electrolysis bath is qualified are as follows: copper generated on the surface of the cathode plate is flat and has no texture; the surface of the copper is free of particles and needled bulges; the thickness of the cathode plate is uniform; the copper and silver content of the cathode plate is not less than 99.95 percent.

In some embodiments, the preset condition may be determined based on national standards and/or international standards. For example, when the cathode plate is sampled for chemical composition analysis, the copper content and the impurity content can be determined according to national standard GB/T467-2010 according to different production requirements: the copper content of the grade A copper (Cu-CATH-1) is not less than 99.9935 percent, and the silver content is less than 0.0025 percent; the content of copper and silver of the No. 1 standard copper (Cu-CATH-2) is not less than 99.95 percent, and the content of silver is not less than 0.015 percent and is more than 0.005 percent; the purity of copper of standard copper No. 2 (Cu-CATH-3) needs to reach 99.90%: the silver content should be less than 0.025%. The higher the purity of copper of the cathode plate, the lower the content of impurities such as silver, bismuth and the like.

In some embodiments, the preset conditions may be determined based on industry experience and/or historical experience data. For example, according to industry experience and/or historical experience, the qualified out-of-bath cathode plate generates copper on the surface of the cathode plate that is flat and free of lines; the surface of the copper is free of particles and needled bulges; the thickness of the cathode plate is uniform.

In some embodiments, a database may be built for out-of-groove results and production processes based on industry experience and/or historical data, and preset conditions may be determined based on information from the database. In some embodiments, the database may be updated in real-time or periodically based on the out-of-groove results and parameters of the production process.

Step 520, it is determined whether the electrolytic production is abnormal.

In some embodiments, the status determination module may determine whether the electrolytic production is abnormal based on the preset conditions obtained in step 510. For example, if the detection information meets the preset condition, judging that the electrolytic production is normal; and if the detection information does not meet the preset condition, judging that the electrolytic production is abnormal.

For example, the preset condition is that the temperature difference between the surface and the bottom of the electrolyte in the electrolytic cell is not more than 5 degrees celsius. The surface temperature of the electrolyte was detected to be 55 degrees celsius and the temperature of the bottom was detected to be 59 degrees celsius. The temperature difference is 4 ℃, then the preset condition is met, and the state judgment module can judge that the electrolytic production is normal. For another example, if the preset condition is that the surface of the cathode plate is flat and has no texture, and the surface roughness of the cathode plate of the outlet groove is detected, the preset condition is not satisfied, and the electrolytic production can be judged to be abnormal.

In some embodiments, when all preset conditions are met, it may be judged that the electrolysis production is normal; and judging that the electrolytic production is abnormal when at least one preset condition is not met. For example, the detection information of the cathode plate is obtained so that the surface is flat, no lines are formed, the color is normal, the thickness is uniform, but the impurity content of the cathode plate exceeds a threshold range (for example, 0.0025%), and the electrolytic production is judged to be abnormal.

In some embodiments, the machine learning model may first process the data acquired by the acquisition detection device 150 based on a certain data acquisition mode to determine the high risk target. The manner in which the acquisition and detection device 150 acquires data when determining the high risk target may be any one of image-based, infrared, laser, sampling, and ultrasonic. The high risk target may refer to a certain high risk area in the whole electrolytic cell, such as any position of the end part or the middle part of the electrolytic cell, and the high risk target may also be a specific area of the polar plate, such as one corner of the polar plate. The high risk target may be determined based on the preset condition, and may be considered as the high risk target for a part of the targets that do not satisfy the preset condition, for a description of the preset condition, see step 510.

The input of the machine learning model may be data acquired based on a certain data acquisition mode, and the output of the machine learning model may be a high risk target. The state judgment module may train the initial machine learning model through a plurality of training samples with labels, where one training sample corresponds to collected data of one historical time point, the collected data may include data collected by the collection detection device 150 at the historical time point based on a certain data obtaining manner, and the labels of the training samples may include high risk targets. The state judgment module can train the initial machine learning model for a plurality of times in a common mode (such as gradient descent and the like) until the trained initial machine learning model meets preset conditions, and the trained initial machine learning model is used as a machine learning model for predicting the high-risk target. The preset condition may be that the loss function of the updated initial basic model is smaller than a threshold, converges, or the number of training iterations reaches the threshold.

In some embodiments, the machine learning model may also be pre-trained by the server 110 or a third party and stored in the storage device 140, and the exception handling module 230 may invoke the machine learning model directly from the storage device 140. In some embodiments, the machine learning model may be a RNN (Recurrent Neural Network) model, an LSTM (Long Short-Term Memory) model, or the like.

After the high risk area is determined, further data acquisition may be performed on the high risk targets by the acquisition detection device 150. The manner of acquiring the data of the high-risk target by the acquisition and detection device 150 may be at least two manners of image-based, infrared, laser, sampling and ultrasonic. In some embodiments, the state determination module may determine at least two ways to acquire data for the high risk target according to a type of preset condition that is not satisfied by the high risk target. For example, if the high risk target is a cathode plate with particles or needled protrusions on the surface, it can be understood that when the content of bone glue in the electrolyte is too high or thiourea is too high, the surface of the cathode plate is uneven, so the method for determining the data acquired by the state judgment module to the high risk target may include: and (3) sampling electrolyte of the cathode plate and sampling metal on the cathode plate.

In some embodiments, after further data collection of the high risk target by the collection detection device 150, the further collected data may be sent to a remote monitoring center (e.g., server 110, terminal device 130, or other device) to facilitate the remote monitoring center to determine whether the electrolytic production is abnormal based on the further collected data and to determine a treatment regimen when the electrolytic production is abnormal.

In some embodiments, the status determination module may determine whether the electrolytic production is abnormal based on the further collected data. The state judgment module may judge whether the electrolytic production is abnormal based on the further collected data according to the preset condition, and when at least one mode of collected data does not satisfy the preset condition, the state judgment module may judge that the electrolytic production is abnormal, and the description of the preset condition is referred to as step 510.

In some embodiments, the machine learning model is used to process the data collected by the collection detection device 150 based on a certain data acquisition mode, determine the high-risk target, further determine the data acquisition mode, and further perform data collection on the high-risk target, so that invalid data can be effectively reduced, and the efficiency of the subsequent determination processing scheme can be improved.

As shown in fig. 6, the schematic diagram 600 includes one or more of the detection information 420, a machine learning model 610, and production anomaly information 620.

In some embodiments, the production anomaly information 620 may be determined based on processing at least a portion of the detection information by the machine learning model 610, and further description of the detection information, the production anomaly information 620, and their associated description may be found in fig. 3 and will not be repeated here.

In some embodiments, the input of the machine learning model 610 may be one or more pieces of detection information 420 of the sample under test, and the output of the machine learning model 610 may be anomaly information 620. For example, image information (e.g., appearance flatness, integrity, texture, color, thickness, etc.) and/or composition information (e.g., copper content, impurity content, etc.) of a number of cathode plates may be input, outputting abnormal cathode plate position information.

In some embodiments, multiple detection information may be combined as input to the machine learning model 610. For example, image information, component information, and temperature information of the cathode plate are combined as inputs.

In some embodiments, one or more of the detection information 420 at various points in time may be organized into a sequence and the sequence used as an input to the machine learning model 610. For example, one or more detected information acquired in the morning, noon, evening may be sequenced as input to the machine learning model 610.

In some embodiments, the machine learning model 610 may include various models and structures. In some embodiments, the machine learning model may include, but is not limited to, a neural network model, a support vector machine model, a k-nearest neighbor model, a decision tree model, or the like. The neural network model may include one or more combinations of Convolutional Neural Network (CNN), recurrent Neural Network (RNN), multi-layer neural network (MLP), antagonistic neural network (GAN), and the like, among others.

In some embodiments, the machine learning model 610 may be trained based on a number of training samples with identifications. Each training sample may include detection information and the tag may be a plurality of anomaly information. For example, detection information of a sample to be detected at a plurality of time points in a period of time (such as one day, one week, one month, etc.) may be collected as a sample, and a determination result (such as a result of manual determination) of the sample detection information may be obtained, where the determination result includes abnormality information corresponding to the plurality of sample detection information, respectively.

In some embodiments, individual ones of the machine learning models 610 may be co-trained. For example, convolutional Neural Networks (CNNs) may be trained in conjunction with Recurrent Neural Networks (RNNs). Inputting training sample data, such as sample image information of a plurality of time points, to the CNN to obtain a plurality of image feature vectors output by the CNN; and then, inputting the plurality of image feature vectors as training sample data into the RNN to obtain abnormal information corresponding to the plurality of image information output by the RNN, establishing a loss function based on the sample label and the output of the RNN, and iteratively updating parameters of the CNN and the RNN based on the loss function until a preset condition is met.

While the basic concepts have been described above, it will be apparent to those skilled in the art that the foregoing detailed disclosure is by way of example only and is not intended to be limiting. Although not explicitly described herein, various modifications, improvements, and adaptations to the present disclosure may occur to one skilled in the art. Such modifications, improvements, and modifications are intended to be suggested within this specification, and therefore, such modifications, improvements, and modifications are intended to be included within the spirit and scope of the exemplary embodiments of the present invention.

Meanwhile, the specification uses specific words to describe the embodiments of the specification. Reference to "one embodiment," "an embodiment," and/or "some embodiments" means that a particular feature, structure, or characteristic is associated with at least one embodiment of the present description. Thus, it should be emphasized and should be appreciated that two or more references to "an embodiment" or "one embodiment" or "an alternative embodiment" in various positions in this specification are not necessarily referring to the same embodiment. Furthermore, certain features, structures, or characteristics of one or more embodiments of the present description may be combined as suitable.

Furthermore, the order in which the elements and sequences are processed, the use of numerical letters, or other designations in the description are not intended to limit the order in which the processes and methods of the description are performed unless explicitly recited in the claims. While certain presently useful inventive embodiments have been discussed in the foregoing disclosure, by way of various examples, it is to be understood that such details are merely illustrative and that the appended claims are not limited to the disclosed embodiments, but, on the contrary, are intended to cover all modifications and equivalent arrangements included within the spirit and scope of the embodiments of the present disclosure. For example, while the system components described above may be implemented by hardware devices, they may also be implemented solely by software solutions, such as installing the described system on an existing server or mobile device.

Likewise, it should be noted that in order to simplify the presentation disclosed in this specification and thereby aid in understanding one or more inventive embodiments, various features are sometimes grouped together in a single embodiment, figure, or description thereof. This method of disclosure, however, is not intended to imply that more features than are presented in the claims are required for the present description. Indeed, less than all of the features of a single embodiment disclosed above.

In some embodiments, numbers describing the components, number of attributes are used, it being understood that such numbers being used in the description of embodiments are modified in some examples by the modifier "about," approximately, "or" substantially. Unless otherwise indicated, "about," "approximately," or "substantially" indicate that the number allows for a 20% variation. Accordingly, in some embodiments, numerical parameters set forth in the specification and claims are approximations that may vary depending upon the desired properties sought to be obtained by the individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and employ a method for preserving the general number of digits. Although the numerical ranges and parameters set forth herein are approximations that may be employed in some embodiments to confirm the breadth of the range, in particular embodiments, the setting of such numerical values is as precise as possible.

Each patent, patent application publication, and other material, such as articles, books, specifications, publications, documents, etc., referred to in this specification is incorporated herein by reference in its entirety. Except for application history documents that are inconsistent or conflicting with the content of this specification, documents that are currently or later attached to this specification in which the broadest scope of the claims to this specification is limited are also. It is noted that, if the description, definition, and/or use of a term in an attached material in this specification does not conform to or conflict with what is described in this specification, the description, definition, and/or use of the term in this specification controls.

Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments of this specification. Other variations are possible within the scope of this description. Thus, by way of example, and not limitation, alternative configurations of embodiments of the present specification may be considered as consistent with the teachings of the present specification. Accordingly, the embodiments of the present specification are not limited to only the embodiments explicitly described and depicted in the present specification.

Claims

1. An anomaly monitoring method based on out-of-groove information detection comprises the following steps:

obtaining detection information of a sample to be tested, wherein the sample to be tested comprises at least one of metal generated in electrolytic production and metal obtained after the electrolytic production is completed, and the detection information comprises at least one of image information, component information and temperature information;

judging whether the electrolytic production is abnormal or not based on the detection information.

2. The method of claim 1, the determining whether the electrolytic production is abnormal based on the detection information, comprising:

judging whether the detection information meets a preset condition or not;

if the detection information meets the preset condition, judging that the electrolytic production is normal;

and if the detection information does not meet the preset condition, judging that the electrolytic production is abnormal.

3. The method of claim 1, further comprising:

feedback information is determined based on the detection information in response to the electrolytic production anomaly.

4. The method of claim 3, wherein the feedback information includes at least one of temperature adjustment information, electrolyte adjustment information, and abnormal plate position.

5. An abnormality monitoring system based on out-of-groove information detection comprises an information acquisition module and a state judgment module;

the information acquisition module is used for acquiring detection information of a sample to be tested, wherein the sample to be tested comprises at least one of metal generated in electrolytic production and metal obtained after the electrolytic production is completed, and the detection information comprises at least one of image information, component information and temperature information;

the state judgment module is used for judging whether the electrolytic production is abnormal or not based on the detection information.

6. The system of claim 5, the status determination module further to:

judging whether the detection information meets a preset condition or not;

7. The system of claim 5, further comprising an anomaly handling module for determining feedback information based on the detection information in response to the electrolytic production anomaly.

8. The system of claim 7, wherein the feedback information includes at least one of temperature adjustment information, electrolyte adjustment information, and abnormal plate position.

9. An abnormality monitoring device based on out-of-groove information detection, comprising a processor configured to execute the out-of-groove information detection-based abnormality monitoring method according to any one of claims 1 to 4.

10. A computer-readable storage medium storing computer instructions that, when read by a computer in the storage medium, the computer performs the abnormality monitoring method based on out-of-slot information detection as claimed in any one of claims 1 to 4.