A method for warning of deposits in a vacuum furnace based on multi-modal perception

Abstract

The present application relates to the field of vacuum furnace early warning technology, and provides a method for early warning of deposits in a vacuum furnace based on multimodal sensing. The method comprises: obtaining historical material records; extracting a first historical record; performing multi-dimensional feature collection to obtain first material feature information; traversing the deposition database to obtain target deposition data; obtaining a first key feature, selectively activating a multimodal sensing device to obtain first key sensing information; using the first key sensing information as input information for a furnace deposition discriminator to obtain output information; when the output information meets the discrimination constraint, issuing an early warning instruction, and performing an early warning of the presence of a first deposit. The present application solves the technical problem that traditional early warning technology cannot comprehensively monitor the formation of deposits in a vacuum furnace, resulting in inaccurate and untimely deposit early warnings, and achieves the effect of improving the accuracy and timeliness of deposit early warnings by identifying deposits through multimodal sensing.

Description

Technical Field

[0001] The present application relates to the field of industrial furnace early warning technology, specifically to the field of vacuum furnace early warning technology, and in particular to a method for early warning of deposits in a vacuum furnace based on multimodal sensing. Background Art

[0002] With the continuous advancement of industrial production and the rapid development of process technology, vacuum furnaces, as important heat treatment equipment, play a vital role in fields such as materials processing, metallurgy, and semiconductor manufacturing. However, with the continuous expansion of production scale and increasingly stringent process requirements, the problems faced by vacuum furnaces during operation are becoming increasingly complex, particularly the accumulation and adhesion of deposits within the furnace. The formation of deposits within the vacuum furnace is inevitable during operation. These deposits may originate from volatilization of raw materials, precipitation of reaction products, and contamination of the furnace environment. The accumulation of deposits not only affects the heating efficiency of the vacuum furnace but can also seriously affect product quality and even cause equipment failures and safety accidents. Traditional methods for warning deposits within vacuum furnaces cannot comprehensively monitor deposit formation within the vacuum furnace, resulting in inaccurate and untimely deposit warnings. Therefore, how to achieve real-time and accurate warning of deposits within vacuum furnaces has become a pressing issue in current industrial production. Summary of the Invention

[0003] This application provides a vacuum furnace deposit early warning method based on multimodal sensing, aiming to solve the technical problem that traditional early warning technology cannot comprehensively monitor the deposit formation in the vacuum furnace, resulting in inaccurate and untimely deposit early warning.

[0004] In view of the above problems, the present application provides a vacuum furnace deposit early warning method based on multimodal sensing.

[0005] The present application provides a method for warning of deposits in a vacuum furnace based on multimodal sensing, the method comprising: obtaining a historical material processing record, the historical material processing record being a material processing record of the vacuum furnace after a predetermined maintenance and inspection process; extracting a first historical record from the historical material processing record, the first historical record being a historical processing record of a first material; reading a predetermined feature dimension, and performing multi-dimensional feature collection on the first material based on the predetermined feature dimension to obtain first material feature information; traversing a deposition database based on the first material feature information to obtain target deposition data of a target material; obtaining a first key feature of the first deposition in the target deposition data, and selectively activating a multimodal sensing device based on the first key feature to obtain first key sensing information; using the first key sensing information as input information of a furnace deposition discriminator to obtain output information, the furnace deposition discriminator being an intelligent model trained based on the support vector machine principle; issuing a warning instruction when the output information meets the discrimination constraint, and providing a warning to the vacuum furnace based on the warning instruction for the presence of the first deposit.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] The multimodal sensing-based method for early warning of deposits in vacuum furnaces obtains material handling records (i.e., historical material handling records) from the vacuum furnace after scheduled maintenance and inspection. It then extracts a first historical record of a first material from this record and analyzes the potential deposits it may have caused. To comprehensively analyze the characteristics of the first material, multi-dimensional features are collected based on predetermined feature dimensions to obtain characteristic information of the first material. Subsequently, based on this characteristic information, a deposition database is traversed to retrieve target deposit data related to the first material. After obtaining the target deposit data, key features of the first deposit are extracted. Based on these key features, the corresponding multimodal sensing device is selected and activated to obtain first key sensing information. This sensing information is then input into a furnace deposition discriminator trained based on support vector machine principles to determine whether the first deposit exists in the furnace. If the output information of the furnace deposition discriminator meets the preset discrimination constraints, an early warning instruction is issued to alert the operator of the risk of the first deposit in the vacuum furnace. This process improves the accuracy and timeliness of deposit warnings, ensuring that operators can take appropriate measures to address the problem promptly.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0010] Figure 1 1 is a flow chart of a method for early warning of deposits in a vacuum furnace based on multimodal sensing in one embodiment;

[0011] Figure 2 The present invention is a schematic diagram of a process for obtaining target deposit data in a vacuum furnace deposit early warning method based on multimodal sensing in one embodiment. DETAILED DESCRIPTION

[0012] The embodiment of the present application provides a vacuum furnace deposit early warning method based on multimodal sensing to solve the technical problem that traditional early warning technology cannot comprehensively monitor the deposit formation in the vacuum furnace, resulting in inaccurate and untimely deposit early warning.

[0013] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only some of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0014] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products or devices.

[0015] Example

[0016] like Figure 1 As shown, the present application provides a vacuum furnace deposit early warning method based on multimodal sensing, the method comprising:

[0017] A historical material processing record is obtained, where the historical material processing record refers to a material processing record of the vacuum furnace after a predetermined maintenance process.

[0018] Vacuum furnaces operate in high-temperature, high-vacuum environments, placing extremely high demands on the purity of the furnace environment and the stability of equipment operation. However, in actual production processes, due to factors such as impure raw materials, the generation of reaction byproducts, and equipment aging, deposits often form inside vacuum furnaces. These deposits not only affect the normal operation of the equipment but can also seriously affect product quality.

[0019] In this embodiment of the present application, to effectively provide early warning of deposits within a vacuum furnace, the system terminal first obtains the material handling records of the vacuum furnace after scheduled maintenance inspections, i.e., the historical material handling records. This historical material handling record includes the handling records of different materials between each maintenance inspection. This scheduled maintenance inspection refers to regular maintenance inspections of the vacuum furnace, ensuring that no deposits are present within the furnace after the inspections are completed. By collecting and organizing the vacuum furnace handling data before maintenance inspections, a historical material handling record can be generated, which will serve as the data source for subsequent analysis of possible deposits.

[0020] A first historical record in the historical material processing record is extracted, where the first historical record refers to a historical processing record of a first material.

[0021] In one embodiment, after obtaining the historical material processing record, the system terminal parses the historical material processing record. This historical material processing record stores historical processing records corresponding to different materials. The system terminal extracts the historical processing record of the first material from these historical processing records and records it as the first historical record. This first material refers to the material that is processed by the vacuum furnace for the first time after the scheduled maintenance and inspection. By extracting this first historical record, the system terminal can understand the material characteristics of the first material and analyze the possible carbides by combining professional knowledge and historical experience. If carbides are likely to exist, the system terminal will perform a carbide test on the vacuum furnace to confirm whether carbide deposits are actually present. If the test result is that they do not exist, the system terminal will repeat this process, extract the second historical record, and make a carbide deposit judgment until a new scheduled maintenance and inspection is performed.

[0022] The predetermined characteristic dimension is read, and multi-dimensional characteristics of the first material are collected based on the predetermined characteristic dimension to obtain first material characteristic information.

[0023] In one embodiment, the system terminal reads pre-defined characteristic dimensions, which serve as a crucial basis for assessing material properties. Subsequently, based on these pre-defined characteristic dimensions, a detailed feature collection is performed on the first material to obtain specific information about the first material across each characteristic dimension. After collecting these multi-dimensional features, the system terminal obtains detailed characteristic information about the first material and organizes this information to generate first material characteristic information. This first material characteristic information is then used to subsequently traverse the deposition database to ensure accurate acquisition of sensory information.

[0024] Furthermore, this application provides predetermined feature dimensions, including:

[0025] The predetermined characteristic dimensions include furnace process dimensions, material property dimensions, and furnace atmosphere dimensions.

[0026] Preferably, the predetermined feature dimensions are a set of key indicators designed to comprehensively assess material handling within a vacuum furnace. These dimensions encompass the furnace process dimension, the material property dimension, and the furnace atmosphere dimension. The furnace process dimension focuses on the specific process parameters and procedures involved in material handling within the vacuum furnace, such as heating temperature, heating time, and cooling rate. These process parameters directly impact the reactions and changes in the material within the furnace and are therefore crucial for evaluating material handling performance. The material property dimension refers to the inherent properties of the material itself, such as chemical composition, physical state, and particle size distribution. These properties determine the material's behavior and degree of reactivity within the vacuum furnace and are crucial for analyzing material handling results. The furnace atmosphere dimension focuses on the gas environment and composition within the vacuum furnace, such as gas type, pressure, and purity. The furnace atmosphere significantly influences material reactions and deposit formation, and is therefore a crucial aspect of assessing material handling performance. By comprehensively collecting relevant information from these three dimensions, a comprehensive assessment of the material handling performance and the risk of deposit formation within the vacuum furnace can be made, providing robust data support for subsequent analysis and early warning.

[0027] The deposition database is traversed based on the first material characteristic information to obtain target deposition data of the target material.

[0028] In one embodiment, after obtaining the first material characteristic information, the system terminal traverses the deposition database. This deposition database stores the deposition data of all materials processed in the vacuum furnace except the first material. During the traversal process, the system terminal calls the predetermined feature dimension to perform multi-dimensional feature collection on the traversed first deposition data group to obtain the corresponding material characteristic information. Subsequently, the first material characteristic information and the material characteristic information corresponding to the first deposition data are compared and analyzed according to the predetermined feature comparison strategy. If the comparison result is similar, the first deposition data group is used as the target deposition data, and the material corresponding to the first deposition data group is used as the target material. This process is repeated until all the data in the deposition database is traversed, thereby gradually improving the target deposition data of the target material.

[0029] Further, if Figure 2 As shown, this application provides methods for obtaining target sediment data, including:

[0030] Extracting a first deposition data group from the deposition database, where the first deposition data group refers to deposition data of a second material processed by the vacuum furnace; performing multi-dimensional feature collection on the second material based on the predetermined feature dimension to obtain second material feature information.

[0031] Preferably, the system terminal extracts the deposition data sets traversed from the deposition database to obtain a first deposition data set. This first deposition data set is deposition data for the vacuum furnace when processing a second material, i.e., the second material. Subsequently, using the same method as previously described for obtaining the first material characteristic information, multi-dimensional features of the second material are collected using predetermined characteristic dimensions to obtain the second material characteristic information, which is used to calculate a material characteristic similarity index with the first material characteristic information.

[0032] A predetermined feature comparison strategy is read, and a comparative analysis is performed between the first material feature information and the second material feature information based on the predetermined feature comparison strategy to obtain a material feature similarity index; when the material feature similarity index reaches a similarity index limit, the second material is used as the target material, and the first deposition data group is used as the target deposition data.

[0033] Preferably, after obtaining the second material characteristic information, the system terminal retrieves a predetermined characteristic comparison strategy. This predetermined characteristic comparison strategy guides how to compare characteristic information of different materials and how to quantify the differences between these characteristics. Subsequently, based on the retrieved predetermined characteristic comparison strategy, the obtained first and second material characteristic information are comparatively analyzed. This comparative analysis involves information tagging, tag statistics, and ratio calculation. Based on this comparative analysis, the system terminal can then calculate a material characteristic similarity index. This index is a quantitative value that indicates the degree of similarity between the characteristics of the first and second materials and is the result of the ratio calculation. The calculated material characteristic similarity index is then compared with a preset similarity index limit. This similarity index limit is set based on historical experience and is used to determine whether the characteristics of the two materials are sufficiently similar. If the material characteristic similarity index reaches the similarity index limit, the system terminal determines that the characteristics of the second material are sufficiently similar to those of the first material, and therefore selects the second material as the target material, and uses the first deposition data set associated with the second material as the target deposition data. This is because if two materials are similar enough in characteristics, the resulting deposition data sets will also be similar enough, so the latest deposition data set is used as the target deposition data, and the latest material is used as the target material.

[0034] Furthermore, the present application provides a method for calculating the similarity index of material characteristics, including:

[0035] The predetermined feature comparison strategy includes a predetermined feature marking scheme; based on the predetermined feature marking scheme, the first material feature information and the second material feature information are marked in sequence to obtain a first material marking vector and a second material marking vector, respectively.

[0036] Optionally, during the comparative analysis of material feature information, the system terminal retrieves a predetermined feature comparison strategy, which includes a predetermined feature labeling scheme. This scheme specifies how to label various material features. The purpose of labeling is to convert each material feature into a specific, quantifiable vector. Based on this predetermined feature labeling scheme, the system terminal labels the feature information of the first and second materials. First, specific feature items in the furnace process dimension, material property dimension, and furnace atmosphere dimension are extracted from the feature information of the first and second materials, respectively, and the feature set to be labeled is determined. Each feature item is then quantified. For numerical features, such as temperature in the furnace process dimension and pressure in the furnace atmosphere dimension, their measured values are used directly. For non-numerical features, such as gas type in the furnace atmosphere dimension and physical state in the material property dimension, one-hot encoding is used to convert them into numerical form. For example, if there are three gas types: nitrogen, oxygen, and argon, nitrogen might be encoded as [1,0,0], oxygen as [0,1,0], and argon as [0,0,1]. Afterwards, all the labeled feature values of the first and second materials are grouped and arranged according to the dimensions to form two multidimensional vectors. These two multidimensional vectors are the first material label vector and the second material label vector, which represent the feature distribution of the first and second materials on the selected feature set.

[0037] A first information tag pair set is formed in which the tags of the first material tag vector and the second material tag vector are consistent, and a first number of the first information tag pair set is counted; a second information tag pair set is formed in which the tags of the first material tag vector and the second material tag vector are inconsistent, and a second number of the second information tag pair set is counted.

[0038] Optionally, after obtaining the first material marking vector and the second material marking vector, the system terminal compares each mark in the first material marking vector and the second material marking vector one by one. If the marks in the two vectors at a certain position are the same, that is, the codes represented are consistent, then the system terminal regards this pair of marks as an information marking pair with consistent marks and adds it to the first information marking pair set. Subsequently, the number of information marking pairs in this set is counted to obtain the first quantity of the first information marking pair set. This first quantity reflects the degree of consistency between the two materials in multiple features. Afterwards, the same method is used to obtain the positions with inconsistent codes, and these information marking pairs with inconsistent marks are added to the information marking pair set with inconsistent marks, thereby generating a second quantity. The second quantity reveals in which features the two materials differ.

[0039] According to the predetermined feature comparison strategy, the ratio of the first quantity to the sum of the first quantity and the second quantity is recorded as the material feature similarity index.

[0040] Optionally, in order to quantitatively evaluate the similarity in characteristics between the first material and the second material, the system terminal calculates a material characteristic similarity index based on the obtained first quantity and second quantity according to the calculation method recorded in the predetermined characteristic comparison strategy, that is, the ratio of the first quantity to the sum of the first quantity and the second quantity is calculated, and the result of this ratio calculation is recorded as the material characteristic similarity index.

[0041] A first key feature of a first sediment in the target sediment data is obtained, and a multimodal sensing device is selectively activated according to the first key feature to obtain first key sensing information.

[0042] In one embodiment, when processing target sediment data, the system terminal calculates the material quantity ratio by counting the similarities and differences between the first sediment feature of the first sediment and the remaining sediments. If this material quantity ratio meets the specified ratio threshold, the first sediment is added to the first key feature. If not, another sediment is extracted from the first sediment and the same process is repeated. This process continues until all sediment features of the first sediment have been analyzed for similarities and differences. Subsequently, to obtain detailed information related to this first key feature, the system terminal selectively activates the sensor based on this first key feature. This selective activation is based on the functional characteristics of each sensor in the multimodal sensing device. Each key feature corresponds to one or more sensors capable of monitoring it. For example, if the key feature is vacuum furnace humidity, the system terminal activates the humidity-related sensor; if the key feature is sediment particle size distribution, the vision-related sensor is activated. By selectively activating sensors corresponding to the first key feature, data closely related to that feature, namely the first key perception information, can be collected. This approach not only improves data collection efficiency but also ensures data accuracy and relevance, providing strong support for subsequent analysis.

[0043] Furthermore, the present application provides methods for obtaining the first key feature, including:

[0044] Arbitrarily obtain the first sediment feature of the first sediment; determine whether a third material in the sediment database has the first sediment feature; if so, add the third material to the first list; if not, add the third material to the second list.

[0045] Preferably, when processing sediment data, the system terminal arbitrarily obtains a feature of the first sediment as the first sediment feature. Subsequently, the system terminal checks whether the third material in the sediment database possesses this first sediment feature, that is, whether the sediment of the third material exhibits characteristics similar to those of the first sediment. For example, if the color feature of the first sediment is black carbon deposits, the system terminal will check whether the sediment of the third material also exhibits the key feature of black. Based on the inspection results, the corresponding list addition operation is performed. If the sediment of the third material possesses the feature of the first sediment, then this feature of the first sediment will be added to the first list, indicating that this feature of the first sediment has a certain similarity or correlation. If the sediment of the third material does not possess the feature of the first sediment, then it will be added to the second list, indicating that this feature of the first sediment is different and unique to the first sediment. In this way, the sediment features of the first material can be quickly and effectively classified, providing a data foundation for subsequent key feature judgment.

[0046] Obtaining a material quantity ratio between the first list and the second list; and adding the first sediment feature to the first key feature of the first sediment when the material quantity ratio is within a defined ratio threshold.

[0047] Preferably, after the first sediment's features are classified, the system terminal calculates a ratio between the quantity in the first list and the amount of data in the second list to generate a material quantity ratio. If the calculated material quantity ratio reaches a predetermined threshold, this indicates that the first sediment's features are unique. Therefore, the system terminal adds the first sediment's features to the first key features of the first sediment.

[0048] Furthermore, the present application provides a multimodal sensing device, comprising:

[0049] The multimodal sensing device includes at least a visual sensor, a humidity sensor, a quality sensor and a position sensor.

[0050] Preferably, the multimodal sensing device is a device that integrates multiple different types of sensors to capture and analyze multiple types of information about the deposits and the internal environment of the vacuum furnace. The multimodal sensing device includes at least a visual sensor, a humidity sensor, a quality sensor, and a position sensor. Among them, the visual sensor: can capture visual images of the interior of the vacuum furnace and the deposits, helping the system terminal to understand the state, color, shape and other information of the deposits. The humidity sensor is responsible for monitoring the humidity level in the vacuum furnace, which is crucial for analyzing the environmental conditions for the formation of deposits and possible chemical reaction processes. The mass sensor can measure the mass of the deposits in real time, which helps to monitor changes in the deposition process, such as deposition speed, deposition amount, etc. The position sensor is used to determine the specific location of the deposits in the vacuum furnace and provide the system terminal with detailed information on the distribution of the deposits. Through the comprehensive use of these sensors, the multimodal sensing device can provide rich data support for the system terminal, which helps to understand the various characteristics and changes of the deposits and the internal environment of the vacuum furnace.

[0051] The first key perception information is used as input information of an in-furnace deposition discriminator to obtain output information. The in-furnace deposition discriminator is an intelligent model trained based on the support vector machine principle.

[0052] In one embodiment, to assess the deposit condition within a vacuum furnace, the system terminal normalizes the first key sensory information. This involves extracting the maximum and minimum values of the corresponding features within the first key sensory information, calculating the difference between each eigenvalue and the corresponding minimum value, and then calculating the ratio of the calculated difference to the difference between the maximum and minimum values of the corresponding feature to obtain the normalized eigenvalue for each eigenvalue. This process is repeated to obtain the normalized first key sensory information. Subsequently, a weight is assigned to each feature based on expert recommendations, and the eigenvalues in the normalized first key sensory information are weighted with the corresponding weights to form weighted first key sensory information. This weighted first key sensory information is then input into a furnace deposition discriminator. The furnace deposition discriminator calculates the input weighted first key sensory information based on learned knowledge and generates output information, including a first deposition severity index. The furnace deposition discriminator is an intelligent model trained based on support vector machine principles. Specifically, the system terminal first extracts historical perception information from the multimodal sensing device and, based on the time stamps of this historical perception information, selects the corresponding historical deposition severity index from the experimental database. Subsequently, using the same method described above, the historical perception information is normalized and weighted. The weighted historical perception information and the corresponding historical deposition severity index are then partitioned into training and test sets. The weighted historical perception information in the training set is then used as input, and the corresponding historical deposition severity index is used as the target output. A support vector machine (SVM) algorithm is then used to train the weighted historical perception information and the historical deposition severity index. During training, the SVM algorithm learns how to distinguish between different data categories and finds the optimal hyperplane. The performance of the furnace deposition discriminator is then gradually optimized by adjusting SVM parameters, such as the penalty coefficient C and kernel function type. After training, the system terminal evaluates the adjusted discriminator using the test set, calculating metrics such as precision, recall, and F1 score. The calculated evaluation index is then compared with the preset index to determine whether the furnace deposition discriminator meets the requirements. If not, the penalty coefficient, kernel function and other hyperparameters are reselected and the training data is used for retraining. Otherwise, the current furnace deposition discriminator is output.

[0053] When the output information meets the discrimination constraint, a warning instruction is issued, and based on the warning instruction, a warning is issued to the vacuum furnace that the first deposit exists.

[0054] In one embodiment, after receiving the output information, the system terminal compares the first deposition severity index within the output information with the discrimination constraints. If the first deposition severity index meets the discrimination constraints, indicating that the vacuum furnace has severe deposition, the system terminal will issue an early warning instruction and trigger the corresponding early warning mechanism based on this early warning instruction. This may include displaying a warning message on the control interface, emitting an audible sound, and sending a notification to the operator. In this way, the operator can understand the deposition situation within the furnace and take quick action, such as suspending furnace operation and cleaning the deposits, to avoid damage or production interruption.

[0055] Furthermore, the present application provides for issuing the warning instruction, including:

[0056] When the first sediment severity index does not reach the severity index limit, a key monitoring instruction is issued, and real-time perception monitoring of the first sediment is performed based on the key monitoring instruction.

[0057] Preferably, when the first sediment severity index does not reach the preset severity index limit, the system terminal will determine that the sediment accumulation, while not yet at the level requiring emergency treatment, is sufficiently high to warrant attention. Therefore, a focused monitoring instruction will be issued. The purpose of this focused monitoring instruction is to enable the system terminal to perform more detailed and frequent perception and monitoring of the first sediment, so as to promptly detect sediment change trends and whether the severity index limit is likely to be reached at some point in the future.

[0058] In summary, the embodiments of the present application have at least the following technical effects:

[0059] The embodiment of the present application first obtains the historical material processing records of the vacuum furnace after the predetermined maintenance and inspection processing, and extracts the historical processing records of the first material. After reading the predetermined feature dimension, multi-dimensional feature collection is performed on the first material to obtain the first material feature information. Based on the first material feature information, the deposition database is traversed to obtain the target sediment data of the target material. The first key feature of the first sediment in the target sediment data is obtained, and the multimodal sensing device is selectively activated according to the first key feature to obtain the first key sensing information. The first key sensing information is used as the input information of the furnace deposition discriminator, and the output information is obtained by the intelligent model trained by the support vector machine principle. When the output information meets the discrimination constraint, an early warning instruction is issued, and the vacuum furnace is warned based on the early warning instruction. These technical effects jointly solve the technical problem that the traditional early warning technology cannot comprehensively monitor the deposition formation in the vacuum furnace, resulting in inaccurate and untimely sediment warnings, and realize the recognition of sediments through multimodal sensing, thereby improving the accuracy and timeliness of sediment warnings.

[0060] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0061] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

[0062] This specification and drawings are merely illustrative of the present application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, this application is intended to include such modifications and variations as fall within the scope of this application and its equivalents.

Claims

1. A method for early warning of deposits in a vacuum furnace based on multimodal sensing, characterized in that: include: Obtaining historical material processing records, wherein the historical material processing records refer to material processing records of the vacuum furnace after a scheduled maintenance and inspection process; Extracting a first historical record from the historical material processing record, where the first historical record refers to a historical processing record of the first material; Reading a predetermined characteristic dimension, and collecting multi-dimensional characteristics of the first material based on the predetermined characteristic dimension to obtain first material characteristic information; Traversing a deposition database based on the first material characteristic information to obtain target deposition data of the target material; Acquiring a first key feature of a first sediment in the target sediment data, and selectively activating a multimodal sensing device according to the first key feature to obtain first key sensing information; Using the first key perception information as input information of a furnace deposition discriminator to obtain output information, wherein the furnace deposition discriminator is an intelligent model trained based on the support vector machine principle; When the output information meets the discrimination constraint, issuing a warning instruction, and issuing a warning to the vacuum furnace based on the warning instruction that the first deposit exists; Among them, include: Extracting a first deposition data group from the deposition database, where the first deposition data group refers to deposition data of a second material processed by the vacuum furnace; Collect multi-dimensional features of the second material based on the predetermined feature dimension to obtain second material feature information; Reading a predetermined feature comparison strategy, and performing a comparative analysis of the first material feature information and the second material feature information based on the predetermined feature comparison strategy to obtain a material feature similarity index; When the material characteristic similarity index reaches a similarity index limit, the second material is used as the target material, and the first deposition data set is used as the target deposition data; Among them, include: The predetermined feature comparison strategy includes a predetermined feature labeling scheme; Marking the first material feature information and the second material feature information in sequence based on the predetermined feature marking scheme to obtain a first material marking vector and a second material marking vector respectively; forming a first information tag pair set having the same tags as the first material tag vector and the second material tag vector, and counting a first number of the first information tag pair sets; forming a second information tag pair set in which the tags of the first material tag vector and the second material tag vector are inconsistent, and counting a second number of the second information tag pair sets; According to the predetermined feature comparison strategy, recording the ratio of the first quantity to the sum of the first quantity and the second quantity as the material feature similarity index; Among them, include: arbitrarily obtaining a first sediment characteristic of the first sediment; determining whether a third material in the deposition database has the first deposition characteristic; If available, add the third material to the first list; if not available, add the third material to the second list; Obtaining a ratio of material quantities in the first list to those in the second list; When the material quantity ratio is at a defined ratio threshold, the first deposit characteristic is added to the first key characteristic of the first deposit.

2. The method for early warning of deposits in a vacuum furnace based on multimodal sensing according to claim 1, characterized in that: The predetermined characteristic dimensions include furnace process dimensions, material property dimensions, and furnace atmosphere dimensions.

3. The method for early warning of deposits in a vacuum furnace based on multimodal sensing according to claim 1, characterized in that: The multimodal sensing device includes at least a visual sensor, a humidity sensor, a quality sensor and a position sensor.

4. The method for early warning of deposits in a vacuum furnace based on multimodal sensing according to claim 1, characterized in that: Also includes: The first key sensing information after weighted normalization processing is used to obtain a first deposition severity index, where the first deposition severity index is used to characterize the deposition severity of the first deposit in the vacuum furnace; When the first deposition severity index reaches a severity index limit, the warning instruction is issued.

5. The method for early warning of deposits in a vacuum furnace based on multimodal sensing according to claim 4, characterized in that: When the first sediment severity index does not reach the severity index limit, a key monitoring instruction is issued, and real-time perception monitoring of the first sediment is performed based on the key monitoring instruction.

Images