Aluminum casting aluminum leakage real-time monitoring method, system, electronic equipment and storage medium

Abstract

The present invention provides a method, system, electronic device and storage medium for real-time monitoring of aluminum leakage in aluminum casting, and relates to the technical field of metal casting process monitoring. The present invention can obtain the current sound data of the casting process; based on the current sound data and the preset environmental model, the preset basic sound model is corrected to obtain the target sound model; wherein the environmental model is obtained by training based on a pre-built environmental sound database, and the basic sound model is obtained by training based on a pre-built aluminum leakage sound database; based on the target sound model, the current sound data is subjected to aluminum leakage sound recognition to obtain an aluminum leakage recognition result. In this way, real-time monitoring of aluminum leakage in the aluminum casting process is realized based on sound event recognition, and the influence of environmental sound is considered when identifying aluminum leakage sound, thereby improving the accuracy and real-time performance of aluminum leakage detection. In addition, the use of additional environmental models enhances the accuracy of aluminum leakage discrimination and reduces the dependence on the number of aluminum leakage samples.

Description

Technical Field

[0001] The present invention relates to the technical field of metal casting process monitoring, and in particular to a method, system, electronic equipment and storage medium for real-time monitoring of aluminum leakage in aluminum casting. Background Art

[0002] During the casting process of aluminum alloy wheels or other castings, aluminum leakage often occurs due to mold aging, high pouring temperature, poor sealing, uneven force on the tail of the casting, etc. This not only affects product quality, but also causes waste of resources and brings safety hazards. Traditional aluminum leakage detection usually relies on manual observation or indirect physical detection methods, which are inefficient and have lags. Therefore, it is of great value to develop a technology that can accurately determine aluminum leakage in real time. Summary of the invention

[0003] The object of the present invention is to provide a method, system, electronic device and storage medium for real-time monitoring of aluminum leakage in aluminum smelting and casting, so as to improve the accuracy and real-time performance of aluminum leakage detection.

[0004] In a first aspect, an embodiment of the present invention provides a method for real-time monitoring of aluminum leakage during aluminum casting, comprising:

[0005] Get current sound data of the casting process;

[0006] Based on the current sound data and the preset environment model, the preset basic sound model is modified to obtain a target sound model; wherein the environment model is trained based on a pre-built environment sound database, and the basic sound model is trained based on a pre-built aluminum leakage sound database;

[0007] Based on the target sound model, aluminum leakage sound recognition is performed on the current sound data to obtain an aluminum leakage recognition result.

[0008] Furthermore, the obtaining of current sound data of the casting process includes:

[0009] The sound signals in the aluminum smelting and casting process are collected by a microphone array to obtain current sound data; wherein, the microphone array includes a plurality of microphones arranged at key positions of the smelting and casting equipment and at the edge guardrail of the casting deep well.

[0010] Furthermore, the step of modifying the preset basic sound model based on the current sound data and the preset environment model to obtain the target sound model includes:

[0011] Preprocessing and feature extraction are performed on the current sound data to obtain current sound features;

[0012] Inputting the current sound feature into the environment model to obtain a matching environment type;

[0013] The correction parameters corresponding to the matching environment type are loaded into the basic sound model to obtain a target sound model.

[0014] Further, the preprocessing and feature extraction of the current sound data to obtain the current sound features includes:

[0015] Preprocessing the current sound data to obtain preprocessed data; wherein the preprocessing includes one or more of filtering, noise reduction and enhancement;

[0016] Extract key features from the preprocessed data to obtain current sound features; wherein the key features include one or more of frequency features, amplitude features, and time features related to the aluminum leakage event.

[0017] Furthermore, after performing aluminum leakage sound recognition on the current sound data based on the target sound model and obtaining an aluminum leakage recognition result, the method further includes:

[0018] When the aluminum leakage identification result is that an aluminum leakage sound event is identified, an aluminum leakage alarm is issued and an emergency procedure is initiated.

[0019] Furthermore, the method further comprises:

[0020] Acquire the environmental sound database, wherein the environmental sound database includes a plurality of environmental sound data collected under different production conditions;

[0021] The environmental model is obtained by training based on the environmental sound database.

[0022] Furthermore, the method further comprises:

[0023] Acquire the aluminum leakage sound database, wherein the aluminum leakage sound database includes aluminum leakage sound data of multiple aluminum leakage types under different environment types;

[0024] The basic sound model is obtained by training based on the aluminum leakage sound database.

[0025] In a second aspect, an embodiment of the present invention further provides a real-time monitoring system for aluminum leakage during aluminum casting, comprising:

[0026] A data acquisition module, used to acquire current sound data of the casting process;

[0027] A model correction module, used to correct the preset basic sound model based on the current sound data and the preset environment model to obtain a target sound model; wherein the environment model is obtained by training based on a pre-built environment sound database, and the basic sound model is obtained by training based on a pre-built aluminum leakage sound database;

[0028] The aluminum leakage identification module is used to perform aluminum leakage sound identification on the current sound data based on the target sound model to obtain an aluminum leakage identification result.

[0029] In a third aspect, an embodiment of the present invention further provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and when the processor executes the computer program, the real-time monitoring method for aluminum leakage in aluminum casting described in the first aspect is implemented.

[0030] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for real-time monitoring of aluminum leakage in aluminum casting described in the first aspect is executed.

[0031] The real-time monitoring method, system, electronic device and storage medium for aluminum leakage in aluminum casting provided by the embodiments of the present invention can obtain the current sound data of the casting process; based on the current sound data and the preset environmental model, the preset basic sound model is corrected to obtain the target sound model; wherein the environmental model is obtained by training based on a pre-built environmental sound database, and the basic sound model is obtained by training based on a pre-built aluminum leakage sound database; based on the target sound model, the current sound data is subjected to aluminum leakage sound recognition to obtain an aluminum leakage recognition result. In this way, real-time monitoring of aluminum leakage in the aluminum casting process is achieved based on sound event recognition, and the influence of environmental sound is taken into account when identifying aluminum leakage sound, thereby improving the accuracy and real-time performance of aluminum leakage detection. In addition, the use of additional environmental models enhances the accuracy of aluminum leakage discrimination and reduces the dependence on the number of aluminum leakage samples. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0033] Figure 1 A schematic flow chart of a method for real-time monitoring of aluminum leakage in aluminum casting provided by an embodiment of the present invention;

[0034] Figure 2 A schematic diagram of a process for building an environment model provided by an embodiment of the present invention;

[0035] Figure 3 A schematic diagram of a process for constructing a basic sound model provided by an embodiment of the present invention;

[0036] Figure 4 A schematic diagram of a process for identifying aluminum leakage provided by an embodiment of the present invention;

[0037] Figure 5 A schematic structural diagram of a real-time monitoring system for aluminum leakage during aluminum casting provided by an embodiment of the present invention;

[0038] Figure 6 A schematic diagram of the module composition of a real-time monitoring system for aluminum leakage in aluminum casting provided by an embodiment of the present invention;

[0039] Figure 7 A schematic diagram of the module composition of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0040] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] As a non-contact monitoring method, sound recognition technology is used in the industrial field for fault diagnosis and abnormality detection, such as judging potential faults by analyzing operating noise in mechanical equipment maintenance. For aluminum leakage detection in the aluminum casting process, it is possible to achieve it through sound signal analysis, because aluminum leakage may produce sounds of specific frequencies or patterns, such as the sputtering sound and flow sound of aluminum liquid leakage, or other sounds caused by pressure changes caused by leakage.

[0042] In practical applications, the sound recognition system can combine modern signal processing technology and machine learning algorithms to conduct real-time analysis of the audio data generated by the melting and casting equipment when it is working, identify the special sound characteristics associated with aluminum leakage events, and thus achieve early warning.

[0043] At present, industrial application sound recognition monitoring technologies mainly include:

[0044] 1. Sound event detection

[0045] This technology is mainly used to identify and locate specific sound events in audio clips, such as knocks, collisions, horns, etc. The system analyzes the spectrum, energy distribution and other relevant features of the audio signal, combined with machine learning or deep learning models to determine whether a specific sound event has occurred.

[0046] 2. Voiceprint recognition

[0047] Voiceprint recognition in the industrial field mainly utilizes the characteristics of sound signals, including frequency components, duration, and environmental (temperature, humidity, wind speed, pressure) influences.

[0048] 3. Noise suppression and separation

[0049] In complex environments, effective noise suppression and reverberation elimination methods, as well as sound source separation technology, are used to ensure that the target sound can be accurately identified even in the presence of large background noise.

[0050] 4. Far-field sound recognition

[0051] Designed specifically for long-distance or noisy environments, it denoises, enhances, and extracts features from captured acoustic signals in order to identify the unique soundprint characteristics of aluminum leakage.

[0052] When using the sound recognition system for aluminum leakage detection, there are the following challenges:

[0053] 1) Noise interference: Aluminum casting sites are usually accompanied by various high-intensity mechanical noises, environmental noises, and noises generated by other production processes, which may lead to misjudgment of the sound recognition system.

[0054] 2) Complexity of signal features: The sound signal features generated when aluminum leaks may not be unique enough to be distinguished from the sounds of other normal production operations, requiring very sophisticated feature extraction and classification algorithms.

[0055] 3) Environmental adaptability: The harsh environment of the melting and casting workshop, such as high temperature and dust, places high demands on the durability and stability of the sound sensor, which may affect the quality of sound signal collection.

[0056] 4) Real-time requirements: In order to ensure safe production, a rapid response is required once an aluminum leakage incident occurs, which requires the sound recognition system to have high real-time processing capabilities and accuracy.

[0057] There are two drawbacks in using the sound recognition system for aluminum leakage detection:

[0058] 1) Training data limitation: Building a reliable aluminum leakage sound recognition model requires a large number of actual aluminum leakage case sound samples, but it is not easy to obtain such sample data under realistic conditions, and there may be sample bias problems.

[0059] 2) Model generalization ability: Even if the model performs well on the existing data set, it is necessary to verify its generalization ability under new production conditions or different types of aluminum leakage. Different factories have different environmental sounds during production in different seasons, which also leads to the poor performance of ordinary aluminum leakage sound recognition models.

[0060] Based on this, the embodiments of the present invention provide a real-time monitoring method, system, electronic device and storage medium for aluminum leakage in aluminum smelting and casting, which can improve the accuracy and real-time performance of aluminum leakage detection, while solving the defects of training data limitation and model generalization ability.

[0061] To facilitate understanding of this embodiment, a real-time monitoring method for aluminum leakage in aluminum casting disclosed in an embodiment of the present invention is first introduced in detail.

[0062] The embodiment of the present invention provides a method for real-time monitoring of aluminum leakage in aluminum casting, which can be executed by an electronic device with data processing capability. Figure 1 The flowchart of a method for real-time monitoring of aluminum leakage in aluminum casting is shown, and the method mainly includes the following steps S110 to S130:

[0063] Step S110, obtaining current sound data of the casting process.

[0064] The casting equipment may be, but is not limited to, casting equipment for aluminum alloys, etc., and the casting equipment may include a melting and casting equipment and a casting well, etc. The current sound data of the casting process may be acquired in real time by a microphone, and the microphones may be arranged in an array at multiple positions of the casting equipment, such as key positions of the melting and casting equipment and edge guardrails of the casting well, etc. Based on this, in some possible embodiments, step S110 may be implemented as follows: the sound signal in the aluminum melting and casting process is acquired by a microphone array, and the current sound data is acquired; wherein the microphone array includes multiple microphones arranged at key positions of the melting and casting equipment and edge guardrails of the casting well.

[0065] Considering that the microphone collects analog sound signals, in order to facilitate subsequent data processing, the analog sound signals can be converted into digital signals through an analog to digital converter (ADC) to obtain current sound data.

[0066] Step S120, based on the current sound data and the preset environmental model, the preset basic sound model is modified to obtain a target sound model; wherein the environmental model is trained based on a pre-built environmental sound database, and the basic sound model is trained based on a pre-built aluminum leakage sound database.

[0067] The influence of noise can be reduced by preprocessing the current sound data, and features related to aluminum leakage can be obtained by feature extraction; and considering that the sound of aluminum leakage is different in different environments, the environment type where the current sound data is located can be identified through the environment model, and then the presence of aluminum leakage sound can be accurately identified based on the sound model under the environment type (i.e., the target sound model). Based on this, in some possible embodiments, step S120 can be implemented in the following manner: preprocessing and feature extraction of the current sound data to obtain the current sound feature; inputting the current sound feature into the environment model to obtain the matching environment type; loading the correction parameters corresponding to the matching environment type into the basic sound model to obtain the target sound model.

[0068] In one possible implementation, the current sound features can be obtained through the following process: preprocessing the current sound data to obtain preprocessed data; wherein the preprocessing includes one or more of filtering, noise reduction and enhancement; extracting key features of the preprocessed data to obtain the current sound features; wherein the key features include one or more of frequency features, amplitude features and time features related to the aluminum leakage event.

[0069] Since the environmental sounds are separated from the aluminum leakage sounds, when building the aluminum leakage sound database, it is only necessary to collect different types of aluminum leakage sounds, which alleviates the training data limitation problem of the basic sound model (i.e. reduces the dependence on the number of aluminum leakage samples) and improves the generalization ability of the model.

[0070] Step S130, based on the target sound model, the current sound data is subjected to aluminum leakage sound recognition to obtain an aluminum leakage recognition result.

[0071] The current sound feature corresponding to the current sound data can be input into the target sound model, and the target sound model can output an aluminum leakage recognition result of whether the aluminum leakage sound event is recognized.

[0072] Furthermore, in order to ensure safe production, the above method also includes: when the aluminum leakage identification result is an aluminum leakage sound event, an aluminum leakage alarm is issued and an emergency procedure is initiated. When an aluminum leakage sound event is identified, an alarm can be issued through an alarm, and corresponding measures can be taken in conjunction with production equipment (such as smelting and casting equipment, etc.), such as closing valves, initiating emergency procedures, etc.

[0073] The real-time monitoring method for aluminum leakage in aluminum casting provided by the embodiment of the present invention can obtain the current sound data of the casting process; based on the current sound data and the preset environmental model, the preset basic sound model is corrected to obtain the target sound model; wherein the environmental model is obtained by training based on a pre-built environmental sound database, and the basic sound model is obtained by training based on a pre-built aluminum leakage sound database; based on the target sound model, the current sound data is subjected to aluminum leakage sound recognition to obtain an aluminum leakage recognition result. In this way, real-time monitoring of aluminum leakage in the aluminum casting process is achieved based on sound event recognition, and the influence of environmental sound is taken into account when identifying aluminum leakage sound, thereby improving the accuracy and real-time performance of aluminum leakage detection. In addition, the use of additional environmental models enhances the accuracy of aluminum leakage discrimination and reduces the dependence on the number of aluminum leakage samples.

[0074] The embodiment of the present invention also provides a process for constructing an environment model, see Figure 2 A schematic diagram of a process for building an environment model is shown, and the environment model is obtained through the following steps:

[0075] Step S210, obtaining an environmental sound database, where the environmental sound database includes multiple environmental sound data collected under different production conditions.

[0076] Step S220: training an environment model based on the environment sound database.

[0077] The input of the environmental model is sound data, and the output is the environment type that matches the input sound data. When training the environmental model, the environment can be divided into multiple types according to the production conditions, and multiple environmental sound data are obtained under each type. An environmental sound database is constructed based on multiple environmental sound data under different production conditions, so as to train the environmental model based on the environmental sound database; during training, each environmental sound data is used as the input of the environmental model, and the environmental type corresponding to the environmental sound data is the label. Among them, the production conditions include production address (such as different factories) and season. It should be noted that the environmental sound data here can be environmental sound data in the presence of aluminum leakage, or it can be environmental sound data in the absence of aluminum leakage.

[0078] The present invention also provides a basic sound model construction process, see Figure 3 A schematic diagram of a process for constructing a basic sound model is shown, and the basic sound model is obtained through the following steps:

[0079] Step S310, obtaining an aluminum leakage sound database, wherein the aluminum leakage sound database includes aluminum leakage sound data of various aluminum leakage types under different environment types.

[0080] Step S320: Based on the aluminum leakage sound database, a basic sound model is trained.

[0081] The input of the basic sound model is the aluminum leakage sound data and its environment type, and the output of the basic sound model is the aluminum leakage recognition result. Among them, the environment type can be used to modify the parameters of the basic sound model, so that the environmental sound in the aluminum leakage sound data can be separated to achieve accurate recognition of the aluminum leakage sound.

[0082] For ease of understanding, the above-mentioned aluminum casting aluminum leakage real-time monitoring method is exemplarily introduced in detail below.

[0083] The embodiment of the present invention adopts aluminum leakage recognition technology based on sound event recognition. The aluminum leakage recognition technology is an application example of cross-domain fusion innovation, integrating sensor technology, signal and information processing technology, artificial intelligence algorithm and other technologies. The aluminum leakage recognition technology is applied in aluminum melting and casting, using sound signal processing and pattern recognition technology to realize the method of real-time detection of aluminum leakage anomalies, realize online monitoring and real-time alarm, and improve the production safety and efficiency of the aluminum processing industry.

[0084] In the embodiment of the present invention, the real-time monitoring method for aluminum leakage in aluminum casting can be implemented by the following process:

[0085] 1) Modeling of environmental sound characteristics.

[0086] That is to build the above environment model, which can be referred to Figure 2 The process shown.

[0087] 2) Modeling of aluminum leakage sound characteristics.

[0088] That is to build the above basic sound model, which can be referred to Figure 3 The process shown.

[0089] 3) Aluminum leakage identification process.

[0090] See also Figure 4 The schematic diagram of a process of aluminum leakage identification is shown in FIG. The aluminum leakage identification process is as follows:

[0091] Step S410, obtaining the casting well sound signal (ie, current sound data).

[0092] The sound signals of the aluminum casting process are collected in real time by using microphones at the production site. These sound signals contain various processing sounds, including normal processing sounds and possible aluminum leakage sounds. Then preprocessing and feature extraction are performed: the preprocessing step can include noise reduction, filtering and enhancement processing to improve the quality of the sound signal; feature extraction is to extract key features related to aluminum leakage events from the sound signal, such as the frequency, amplitude and duration of the sound (these features will be used in the subsequent recognition algorithm) to obtain the current environmental sound features (i.e. the current sound features).

[0093] Step S420, loading current ambient sound features.

[0094] The current environmental sound features to be identified are input into the environmental model. The environmental model will determine the currently applicable environmental sound (i.e., obtain the matching environment type) based on the learned features, and provide model correction parameters for the aluminum leakage sound recognition algorithm model (i.e., the basic sound model) for the environmental sound.

[0095] Step S430, determine whether there is aluminum leakage feature, if yes, execute step S440; if no, end.

[0096] The modified aluminum leakage sound recognition algorithm model is run to determine the aluminum leakage characteristics. If the model determines that the sound signal is an aluminum leakage sound event, a corresponding alarm signal is sent through step S440.

[0097] Step S440: generating an alarm.

[0098] Alarm handling: When a sound signal that meets the characteristics of aluminum leakage is identified, an alarm will be sounded immediately, and corresponding measures can be taken in conjunction with production equipment to close valves and initiate emergency procedures.

[0099] The embodiment of the present invention uses acoustic signals as a new means to detect aluminum leakage during the aluminum melting and casting process, and utilizes an additional environmental sound model to enhance the accuracy of aluminum leakage identification and reduce the dependence on the number of aluminum leakage samples.

[0100] Corresponding to the above-mentioned aluminum casting aluminum leakage real-time monitoring method, the embodiment of the present invention also provides an aluminum casting aluminum leakage real-time monitoring system, see Figure 5 The structure diagram of a real-time monitoring system for aluminum leakage in aluminum casting is shown, and the system includes:

[0101] The data acquisition module 501 is used to acquire the current sound data of the casting process;

[0102] A model correction module 502 is used to correct the preset basic sound model based on the current sound data and the preset environment model to obtain a target sound model; wherein the environment model is obtained by training based on a pre-built environment sound database, and the basic sound model is obtained by training based on a pre-built aluminum leakage sound database;

[0103] The aluminum leakage identification module 503 is used to perform aluminum leakage sound identification on the current sound data based on the target sound model to obtain an aluminum leakage identification result.

[0104] The real-time monitoring system for aluminum leakage in aluminum casting provided by the embodiment of the present invention can obtain the current sound data of the casting process; based on the current sound data and the preset environmental model, the preset basic sound model is corrected to obtain the target sound model; wherein the environmental model is obtained by training based on a pre-built environmental sound database, and the basic sound model is obtained by training based on a pre-built aluminum leakage sound database; based on the target sound model, the current sound data is subjected to aluminum leakage sound recognition to obtain an aluminum leakage recognition result. In this way, real-time monitoring of aluminum leakage in the aluminum casting process is achieved based on sound event recognition, and the influence of environmental sound is taken into account when identifying aluminum leakage sound, thereby improving the accuracy and real-time performance of aluminum leakage detection. In addition, the use of additional environmental models enhances the accuracy of aluminum leakage discrimination and reduces the dependence on the number of aluminum leakage samples.

[0105] Furthermore, the data acquisition module 501 is also used to collect sound signals during the aluminum smelting process through a microphone array to obtain current sound data; wherein the microphone array includes multiple microphones arranged at key positions of the smelting equipment and at the edge guardrail of the casting deep well.

[0106] Furthermore, the above-mentioned model correction module 502 is also used to: preprocess and extract features of the current sound data to obtain current sound features; input the current sound features into the environment model to obtain a matching environment type; load the correction parameters corresponding to the matching environment type into the basic sound model to obtain a target sound model.

[0107] Furthermore, the above-mentioned model correction module 502 is also used to: pre-process the current sound data to obtain pre-processed data; wherein the pre-processing includes one or more of filtering, noise reduction and enhancement; extract key features of the pre-processed data to obtain current sound features; wherein the key features include one or more of frequency features, amplitude features and time features related to aluminum leakage events.

[0108] Furthermore, the above-mentioned aluminum casting aluminum leakage real-time monitoring system also includes:

[0109] The alarm control module is used to issue an aluminum leakage alarm and start an emergency procedure when the aluminum leakage identification result is an aluminum leakage sound event.

[0110] Furthermore, the above-mentioned aluminum casting aluminum leakage real-time monitoring system also includes:

[0111] The first training module is used to obtain the environmental sound database, which includes multiple environmental sound data collected under different production conditions; based on the environmental sound database, the environmental model is trained.

[0112] Furthermore, the above-mentioned aluminum casting aluminum leakage real-time monitoring system also includes:

[0113] The second training module is used to obtain the aluminum leakage sound database, which includes aluminum leakage sound data of multiple aluminum leakage types under different environment types; based on the aluminum leakage sound database, the basic sound model is trained.

[0114] For ease of understanding, the following is an introduction to the module composition of the above-mentioned aluminum casting aluminum leakage real-time monitoring system. Figure 6 The module composition diagram of a real-time monitoring system for aluminum leakage in aluminum casting is shown, and the system includes the following modules:

[0115] 1) Sound data acquisition module 601: includes multiple microphones (such as microphone 1, microphone 2, microphone 3 and microphone 4). The microphones are arranged at key positions of the melting and casting equipment and the edge guardrail of the casting deep well to continuously collect environmental sound data. The collected analog sound signals are converted into digital signals through an analog-to-digital converter (ADC).

[0116] 2) Sound data processing module 602: filtering, noise reduction and enhancement processing are performed on the collected sound signals to extract specific frequency, intensity and time series features related to aluminum leakage.

[0117] 3) Environmental sound matching module 603: uses the trained environmental model (established environmental sound feature database) to identify the sound environment that best matches the current environment, and automatically corrects the aluminum leakage sound model.

[0118] 4) Identification and output module 604: uses the modified sound model to identify the normal operation sound and the aluminum leakage sound in the current sound. When the aluminum leakage sound is identified, an aluminum leakage alarm signal is sent to the alarm and control module 605.

[0119] 5) Alarm and control module 605: When a sound signal that meets the characteristics of aluminum leakage is identified, an alarm is immediately issued, and corresponding measures can be taken in conjunction with production equipment to close valves and start emergency procedures.

[0120] This embodiment designs an integrated sound event monitoring and alarm system, which can immediately trigger an alarm after detecting the acoustic characteristics of aluminum leakage in real time, and guide on-site operators to respond quickly.

[0121] The real-time monitoring system for aluminum smelting and aluminum leakage provided in this embodiment has the same implementation principle and technical effects as those of the aforementioned real-time monitoring method for aluminum smelting and aluminum leakage embodiment. For the sake of brief description, for matters not mentioned in the embodiment of the real-time monitoring system for aluminum smelting and aluminum leakage, reference may be made to the corresponding contents in the aforementioned embodiment of the real-time monitoring method for aluminum smelting and aluminum leakage.

[0122] like Figure 7As shown, an electronic device 700 provided by an embodiment of the present invention includes: a processor 701, a memory 702 and a bus, the memory 702 stores a computer program that can be run on the processor 701, when the electronic device 700 is running, the processor 701 and the memory 702 communicate through the bus, and the processor 701 executes the computer program to realize the above-mentioned real-time monitoring method for aluminum leakage in aluminum casting.

[0123] Specifically, the memory 702 and the processor 701 can be general-purpose memories and processors, which are not specifically limited here.

[0124] The embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the real-time monitoring method for aluminum leakage in aluminum casting described in the above method embodiment is executed. The computer-readable storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a RAM, a magnetic disk or an optical disk, and other media that can store program codes.

[0125] The term "and / or" herein is only a description of the association relationship of the associated objects, indicating that there may be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the term "at least one" herein represents any combination of at least two of any one or more of a plurality of. For example, including at least one of A, B, and C can represent including any one or more elements selected from the set consisting of A, B, and C.

[0126] In all examples shown and described herein, any specific values ​​should be interpreted as merely exemplary and not as limiting, and thus other examples of the exemplary embodiments may have different values.

[0127] The flow charts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the system, method and computer program product according to multiple embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0128] In the several embodiments provided in the present application, it should be understood that the disclosed systems and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0129] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple 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.

[0130] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0131] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A real-time monitoring method for aluminum leakage in aluminum casting, characterized in that: include: Get current sound data of the casting process; Based on the current sound data and the preset environmental model, the preset basic sound model is modified to obtain a target sound model; wherein the environmental model is trained based on a pre-built environmental sound database, the environmental sound database includes multiple environmental sound data collected under different production conditions, and the basic sound model is trained based on a pre-built aluminum leakage sound database, the aluminum leakage sound database includes aluminum leakage sound data of multiple aluminum leakage types under different environmental types; Running the target sound model, performing aluminum leakage sound recognition on the current sound data, and obtaining an aluminum leakage recognition result; The method of modifying the preset basic sound model based on the current sound data and the preset environment model to obtain the target sound model includes: Preprocessing and feature extraction are performed on the current sound data to obtain current sound features; The current sound feature is input into the environment model, the environment model discriminates the currently applicable environment sound according to the learned feature, obtains the matching environment type, and provides model correction parameters for the basic sound model according to the matching environment type; The correction parameters are loaded into the basic sound model to obtain a target sound model.

2. The method according to claim 1, characterized in that The step of obtaining current sound data of the casting process comprises: The sound signals in the aluminum smelting and casting process are collected by a microphone array to obtain current sound data; wherein, the microphone array includes a plurality of microphones arranged at key positions of the smelting and casting equipment and at the edge guardrail of the casting deep well.

3. The method according to claim 1, characterized in that: The preprocessing and feature extraction of the current sound data to obtain the current sound features includes: Preprocessing the current sound data to obtain preprocessed data; wherein the preprocessing includes one or more of filtering, noise reduction and enhancement; Extract key features from the preprocessed data to obtain current sound features; wherein the key features include one or more of frequency features, amplitude features, and time features related to the aluminum leakage event.

4. The method according to claim 1, characterized in that: After performing aluminum leakage sound recognition on the current sound data based on the target sound model and obtaining an aluminum leakage recognition result, the method further includes: When the aluminum leakage identification result is that an aluminum leakage sound event is identified, an aluminum leakage alarm is issued and an emergency procedure is initiated.

5. The method according to claim 1, characterized in that The method further comprises: Acquire the environmental sound database, wherein the environmental sound database includes a plurality of environmental sound data collected under different production conditions; The environmental model is obtained by training based on the environmental sound database.

6. The method according to claim 1, characterized in that The method further comprises: Acquire the aluminum leakage sound database, wherein the aluminum leakage sound database includes aluminum leakage sound data of multiple aluminum leakage types under different environment types; The basic sound model is obtained by training based on the aluminum leakage sound database.

7. A real-time monitoring system for aluminum casting aluminum leakage, characterized in that: include: A data acquisition module, used to acquire current sound data of the casting process; A model correction module is used to correct the preset basic sound model based on the current sound data and the preset environmental model to obtain a target sound model; wherein the environmental model is trained based on a pre-built environmental sound database, the environmental sound database includes multiple environmental sound data collected under different production conditions, and the basic sound model is trained based on a pre-built aluminum leakage sound database, the aluminum leakage sound database includes aluminum leakage sound data of multiple aluminum leakage types under different environmental types; An aluminum leakage identification module is used to run the target sound model, perform aluminum leakage sound identification on the current sound data, and obtain an aluminum leakage identification result; The model correction module is specifically used to: preprocess and extract features of the current sound data to obtain current sound features; input the current sound features into the environmental model, and the environmental model determines the currently applicable ambient sound based on the learned features to obtain a matching environment type, and provides model correction parameters for the basic sound model according to the matching environment type; load the correction parameters into the basic sound model to obtain a target sound model.

8. An electronic device comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, characterized in that: When the processor executes the computer program, the real-time monitoring method for aluminum leakage in aluminum casting according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for real-time monitoring of aluminum leakage in aluminum casting according to any one of claims 1 to 6 is executed.

Images