Method for analyzing production fluctuation of hard metal wear block

Abstract

The application is suitable for the technical field of alloy material production, and particularly relates to a hard alloy wear-resistant block production fluctuation analysis method. The hard alloy wear-resistant block production fluctuation analysis method comprises the following steps: obtaining hard alloy wear-resistant block production information; wherein the hard alloy wear-resistant block production information comprises a product quality set of the hard alloy wear-resistant block and a production parameter set corresponding to each production link; generating a fluctuation analysis task based on the product quality set; wherein the fluctuation analysis task is used to indicate the production link that needs to be analyzed at present; performing analysis based on the fluctuation analysis task and the corresponding production parameter set to obtain a fluctuation parameter set; wherein the fluctuation parameter set is used to reflect the fluctuation characteristics of the production parameter that causes the hard alloy wear-resistant block production to fluctuate; generating an analysis result based on the fluctuation parameter set; wherein the analysis result is used to reflect the root cause of the fluctuation of the production parameter. The method provides a clear improvement direction for production optimization and helps to quickly develop targeted measures.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of alloy material production, and particularly relates to a hard alloy wear-resistant block production fluctuation analysis method. BACKGROUND

[0002] As a key vulnerable component of heavy machinery such as engineering machinery and mining equipment, the quality stability of the hard alloy wear-resistant block directly affects the running efficiency and service life of the whole machine. In the production process of the hard alloy wear-resistant block, multiple process links such as mixing, forming, sintering and fine grinding are involved, and the fluctuation of the production parameters (such as mixing ratio, sintering temperature, holding time, etc.) of each link may cause the product quality to deviate, which specifically manifests in problems such as unqualified hardness, insufficient wear resistance, and size precision out of tolerance.

[0003] At present, the analysis method for the production fluctuation of the hard alloy wear-resistant block has formed certain technical accumulation. The data-driven method commonly used in the industry is to collect the production parameters and product quality detection data of each link, and use statistical analysis tools (such as correlation analysis and trend chart drawing), which separately analyzes the fluctuation of each production link. This method not only has a complex process and needs to process a large amount of data, but also may cause deviation in the finally located root cause. SUMMARY

[0004] The embodiments of the application provide a hard alloy wear-resistant block production fluctuation analysis method, which can solve the problems of low efficiency and inaccurate fluctuation tracing of the analysis method for the production fluctuation of the hard alloy wear-resistant block.

[0005] In a first aspect, the embodiments of the application provide a hard alloy wear-resistant block production fluctuation analysis method, comprising:

[0006] obtaining hard alloy wear-resistant block production information; wherein the hard alloy wear-resistant block production information comprises a product quality set of the hard alloy wear-resistant block and a production parameter set corresponding to each production link;

[0007] generating a fluctuation analysis task based on the product quality set; wherein the fluctuation analysis task is used to indicate the production link that needs to be analyzed at present;

[0008] analyzing the fluctuation analysis task and the corresponding production parameter set to obtain a fluctuation parameter set; wherein the fluctuation parameter set is used to reflect the fluctuation characteristics of the production parameter that causes the production fluctuation of the hard alloy wear-resistant block;

[0009] generating an analysis result based on the fluctuation parameter set; wherein the analysis result is used to reflect the root cause of the production parameter fluctuation.

[0010] The technical scheme described above in the embodiments of the application has at least the following technical effects:

[0011] The hard alloy wear-resistant block production fluctuation analysis method provided by the application obtains hard alloy wear-resistant block production information including a product quality set of hard alloy wear-resistant blocks and a production parameter set corresponding to each production link; then generates a fluctuation analysis task for indicating a production link that needs to be analyzed based on the product quality set; analyzes the fluctuation analysis task and the corresponding production parameter set to obtain a fluctuation parameter set for reflecting production parameters that cause fluctuations in the production of hard alloy wear-resistant blocks; and finally generates an analysis result for reflecting the source of the fluctuations in the production parameters based on the fluctuation parameter set. The method obtains the fluctuation characteristics of the production parameters by analyzing the obtained fluctuation parameter set, realizes deep mining from "parameter value fluctuations" to "fluctuation rules and influences", provides more abundant basis for tracing, forms a closed-loop analysis link through the correlation of "quality data-production links-parameter characteristics-fluctuation sources", and effectively solves the fluctuation tracing problem caused by the coupling of multiple links. In the production of hard alloy wear-resistant blocks, a certain quality index may be affected by multiple links such as mixing and sintering. The method quickly locates the fluctuation parameters by correlating the product quality set and the production links, and finally traces the specific source by analyzing the corresponding parameters of the production links. Compared with the one-sided judgment of simply equating "fluctuation parameters" to "fluctuation sources" by analyzing the "parameter fluctuations" of each production link, the method can quickly and effectively find out the problem of production fluctuations, and provides a clear improvement direction for production optimization, which helps to quickly develop targeted measures, thereby reducing quality fluctuations and improving product consistency. It is especially suitable for product production scenarios such as hard alloy wear-resistant blocks which have strict requirements on material performance.

[0012] In a possible implementation form of the first aspect, the generating the fluctuation analysis task based on the product quality set comprises:

[0013] determining a key quality index based on the product quality set; wherein the key quality index is used to indicate a main quality index of the quality data fluctuation exceeding the standard;

[0014] establishing a quality-link mapping rule, and obtaining a candidate production link corresponding to the key quality index based on the quality-link mapping rule; wherein the candidate production link is used to reflect a set of multiple production links;

[0015] generating the fluctuation analysis task based on the candidate production link.

[0016] In a possible implementation form of the first aspect, the determining the key quality index based on the product quality set comprises:

[0017] The product quality set is divided into a basic index layer and a performance index layer; wherein, the basic index layer is used for reflecting inherent attribute data of the hard alloy wear-resistant block, and the performance index layer is used for reflecting dynamic attribute data of the hard alloy wear-resistant block;

[0018] The standard deviations of the basic index layer and the performance index layer are calculated respectively to obtain a first fluctuation coefficient and a second fluctuation coefficient; wherein, the first fluctuation coefficient is used for reflecting a data fluctuation degree corresponding to the basic index layer, and the second fluctuation coefficient is used for reflecting a data fluctuation degree corresponding to the performance index layer;

[0019] The key quality index is determined based on the first fluctuation coefficient and the second fluctuation coefficient.

[0020] In a possible implementation manner of the first aspect, the determining of the key quality index based on the first fluctuation coefficient and the second fluctuation coefficient comprises:

[0021] The first fluctuation coefficient is compared with a first threshold value, and the second fluctuation coefficient is compared with a second threshold value to obtain a comparison result; wherein, the first threshold value is a critical value of data fluctuation of the basic index layer, and the second threshold value is a critical value of data fluctuation of the performance index layer;

[0022] The key quality index is determined based on the comparison result.

[0023] In a possible implementation manner of the first aspect, the determining of the key quality index based on the comparison result comprises:

[0024] If the comparison result is that the first fluctuation coefficient is greater than the first threshold value and the second fluctuation coefficient is greater than the second threshold value, an associated quality index of the basic index layer and the performance index layer is generated, and the associated quality index is determined as the key quality index;

[0025] If the comparison result is that only the first fluctuation coefficient is greater than the first threshold value, a quality index corresponding to the basic index layer is determined as the key quality index;

[0026] If the comparison result is that only the second fluctuation coefficient is greater than the second threshold value, a quality index corresponding to the performance index layer is determined as the key quality index.

[0027] In a possible implementation manner of the first aspect, the establishing of the quality-link mapping rule comprises:

[0028] A link-parameter correspondence table is established; wherein, the link-parameter correspondence table is used for reflecting a production process corresponding to each production procedure;

[0029] generate an association matrix based on historical production fluctuation data; wherein the association matrix is used to reflect the co-occurrence frequency of parameter fluctuation and quality index exceeding;

[0030] establish a quality-link mapping rule based on the association matrix and the link-parameter correspondence table.

[0031] In a possible implementation manner of the first aspect, the obtaining a fluctuation analysis task based on the candidate production link comprises:

[0032] generate a plurality of fluctuation transmission chains based on the process coupling relationship of the candidate production link; wherein the fluctuation transmission chain is used to reflect the conduction path of the parameter fluctuation of the previous link causing the parameter abnormality of the subsequent link through the process coupling relationship;

[0033] calculate the fluctuation contribution degree of each fluctuation transmission chain; wherein the fluctuation contribution degree is used to reflect the cumulative influence degree of the co-occurrence probability of each production link in the fluctuation transmission chain;

[0034] obtain a fluctuation analysis task based on the fluctuation contribution degree.

[0035] In a possible implementation manner of the first aspect, the obtaining a fluctuation analysis task based on the fluctuation contribution degree comprises:

[0036] sort the plurality of fluctuation transmission chains based on the fluctuation contribution degree to obtain a transmission chain sequence;

[0037] determine a core fluctuation chain based on the transmission chain sequence;

[0038] obtain a fluctuation analysis task based on the core fluctuation chain.

[0039] In a possible implementation manner of the first aspect, the obtaining a fluctuation analysis task based on the core fluctuation chain comprises:

[0040] calculate the association degree between each candidate production link and the key quality index based on the order position of each candidate production link in the core fluctuation chain;

[0041] determine the candidate production link with the largest association degree as the fluctuation analysis task.

[0042] In a possible implementation manner of the first aspect, the obtaining a fluctuation parameter set based on the fluctuation analysis task and the corresponding production parameter set comprises:

[0043] perform matching based on the fluctuation analysis task and the corresponding production parameter set to obtain a parameter analysis target; wherein the parameter analysis target is used to indicate the production parameter in the production parameter set corresponding to the fluctuation analysis task;

[0044] extract time sequence fluctuation data of the target in a preset time window based on the parameters; wherein, the time sequence fluctuation data comprises a parameter collection time point and a corresponding parameter value;

[0045] perform feature extraction based on the time sequence fluctuation data to obtain a fluctuation parameter set; wherein, the fluctuation parameter set comprises time domain features and fluctuation pattern features.

[0046] In a second aspect, an embodiment of the present application provides a cemented carbide wear-resistant block production fluctuation analysis system, comprising:

[0047] an acquisition module configured to acquire cemented carbide wear-resistant block production information; wherein, the cemented carbide wear-resistant block production information comprises a product quality set of cemented carbide wear-resistant blocks and a production parameter set corresponding to each production link;

[0048] a first generation module configured to generate a fluctuation analysis task based on the product quality set; wherein, the fluctuation analysis task is used to indicate a production link that needs to be analyzed at present;

[0049] an analysis module configured to analyze the fluctuation analysis task and the corresponding production parameter set to obtain a fluctuation parameter set; wherein, the fluctuation parameter set is used to reflect fluctuation features of production parameters that cause fluctuations in cemented carbide wear-resistant block production;

[0050] a second generation module configured to generate an analysis result based on the fluctuation parameter set; wherein, the analysis result is used to reflect a root cause of fluctuations in production parameters.

[0051] In a third aspect, an embodiment of the present application provides a cemented carbide wear-resistant block production fluctuation analysis device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method of any one of the first aspect when executing the computer program.

[0052] In a fourth aspect, an embodiment of the present application provides a computer readable storage medium, which stores a computer program, wherein the computer program is executable by a processor to implement the method of any one of the first aspect.

[0053] In a fifth aspect, an embodiment of the present application provides a computer program product, which, when executed on a cemented carbide wear-resistant block production fluctuation analysis device, causes the cemented carbide wear-resistant block production fluctuation analysis device to perform the cemented carbide wear-resistant block production fluctuation analysis method of any one of the first aspect.

[0054] It can be understood that the beneficial effects of the second aspect to the fifth aspect described above can be referred to the related description of the first aspect, which will not be repeated here. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0056] Figure 1 This is a schematic flowchart of the method for analyzing fluctuations in the production of cemented carbide wear-resistant blocks provided in the embodiments of this application;

[0057] Figure 2 This is a schematic diagram of the implementation process of the cemented carbide wear-resistant block production fluctuation analysis method provided in the embodiments of this application;

[0058] Figure 3 This is a schematic diagram of the structure of the cemented carbide wear-resistant block production fluctuation analysis system provided in the embodiments of this application;

[0059] Figure 4 This is a schematic diagram of the structure of the cemented carbide wear-resistant block production fluctuation analysis equipment provided in the embodiments of this application. Detailed Implementation

[0060] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0061] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0062] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0063] As used in the specification and appended claims herein, the term “if’ can be interpreted as meaning “when” or “upon” or “in response to a determination” or “in response to a detection” depending on the context. Similarly, the phrase “if it is determined” or “if a described condition or event is detected” can be interpreted as meaning “upon a determination” or “in response to a determination” or “upon a detection of the described condition or event” or “in response to a detection of the described condition or event” depending on the context.

[0064] In addition, in the description of the present application and the appended claims, the terms “first”, “second”, “third”, etc. are used only to distinguish descriptions, and cannot be understood as indicating or implying relative importance.

[0065] Reference in the specification to “one embodiment” or “some embodiments” means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the application. The appearances of the phrases “in one embodiment”, “in some embodiments”, “in other embodiments”, “in additional embodiments”, etc. in various places in the specification are not necessarily all referring to the same embodiment, although it can be so in some cases. The terms “including”, “containing”, “having” and variations thereof mean “including but not limited to”, unless expressly specified otherwise.

[0066] At present, the analysis method for the production fluctuation of hard alloy wear-resistant blocks has formed certain technical accumulation. The data-driven method widely used in the industry collects production parameters and product quality detection data at each link, and uses statistical analysis tools (such as correlation analysis, trend chart drawing, etc.) to find the correlation between parameter fluctuation and quality index change. For example, by comparing the sintering temperature data of different batches with the product hardness test results, the influence law of temperature fluctuation on hardness is identified; or based on historical data, a threshold range of parameter fluctuation is established, and when the real-time parameter exceeds the threshold, a warning is triggered. This method analyzes the fluctuation of each production link separately, not only the process is complex, the data to be processed is large, but also the root cause located may be biased, when multiple links parameters fluctuate at the same time, it is difficult to quickly determine the dominant influencing factor.

[0067] To solve the above problems, the embodiment of the present application provides a cemented carbide wear-resistant block production fluctuation analysis method. In the method, the cemented carbide wear-resistant block production information including a product quality set of the cemented carbide wear-resistant block and a production parameter set corresponding to each production link is obtained; then a fluctuation analysis task for indicating the current production link to be analyzed is generated based on the product quality set; then the fluctuation analysis task and the corresponding production parameter set are analyzed to obtain a fluctuation parameter set for reflecting the fluctuation characteristics of the production parameter causing the fluctuation of the cemented carbide wear-resistant block production; finally, the analysis result for reflecting the root cause of the fluctuation of the production parameter is generated based on the fluctuation parameter set. The method realizes the deep mining from "parameter value fluctuation" to "fluctuation rule and influence" by analyzing the fluctuation parameter set to obtain the fluctuation characteristics of the production parameter. Compared with the traditional method of only focusing on whether the parameter exceeds the standard, the method can capture the key characteristics such as frequency, duration and conduction path of the fluctuation, and provide more abundant basis for tracing the source; through the whole process correlation of "quality data-production link-parameter characteristics-fluctuation root", a closed-loop analysis link is formed, and the fluctuation tracing problem caused by multi-link parameter coupling is effectively solved. In the production of cemented carbide wear-resistant blocks, a certain quality index may be affected by multiple links such as mixing and sintering. The method locks the associated link through task generation, and then locates the core fluctuation parameter through parameter analysis, and finally traces to the specific root cause such as equipment, raw materials or operation. Compared with the one-sided judgment of simply equating "parameter fluctuation" to "fluctuation root", the method provides a clear improvement direction for production optimization, which is helpful for quickly developing targeted measures, thereby reducing quality fluctuation and improving product consistency, and is especially suitable for product production scenes such as cemented carbide wear-resistant blocks which have strict requirements on material performance.

[0068] The cemented carbide wear-resistant block production fluctuation analysis method provided by the embodiment of the present application can be applied to a cemented carbide wear-resistant block production fluctuation analysis device, and at this time, the cemented carbide wear-resistant block production fluctuation analysis device is the execution subject of the cemented carbide wear-resistant block production fluctuation analysis method provided by the embodiment of the present application, and the embodiment of the present application does not make any limitation on the specific type of the cemented carbide wear-resistant block production fluctuation analysis device.

[0069] For example, the hard alloy wear-resistant block production fluctuation analysis device can be a terminal device such as a mobile phone, a tablet computer, a wearable device, an augmented reality (AR) / virtual reality (VR) device, a notebook computer, an ultra-mobile personal computer (UMPC), a netbook, a personal digital assistant (PDA), a desktop computer, a smart big screen, a smart television, and the like, a handheld device, a computing device, or other processing device connected to a wireless modem having a wireless communication function, an Internet of Things terminal, a computer, a laptop computer, a handheld communication device, a handheld computing device, a satellite wireless device, a wireless modem card, customer premise equipment (CPE), and / or other devices for communicating over a wireless system, and a next-generation communication system, for example, a mobile terminal in a 5G network or a mobile terminal in a future evolved Public Land Mobile Network (PLMN), and the like.

[0070] In order to better understand the hard alloy wear-resistant block production fluctuation analysis method provided by the embodiments of the present application, the specific implementation process of the hard alloy wear-resistant block production fluctuation analysis method provided by the embodiments of the present application is exemplarily introduced as follows.

[0071] Figure 1 And Figure 2 A schematic flowchart of the hard alloy wear-resistant block production fluctuation analysis method provided by the embodiments of the present application is shown, and the hard alloy wear-resistant block production fluctuation analysis method comprises:

[0072] S100, obtaining hard alloy wear-resistant block production information; wherein the hard alloy wear-resistant block production information comprises a product quality set of the hard alloy wear-resistant block and a production parameter set corresponding to each production link.

[0073] It can be understood that the product quality set of the hard alloy wear-resistant block refers to a set of finished product quality data of the continuously produced hard alloy wear-resistant block, and includes multi-dimensional quality data such as hardness detection value, wear resistance data dynamic attribute data (such as weight loss amount, friction coefficient), dimensional accuracy, etc. The product quality set can be obtained by detecting the database, can be manually input, etc., but is not limited thereto. The detection database contains the detection data of each hard alloy wear-resistant block, and the corresponding detection quality data can be directly obtained by establishing a communication connection with the detection equipment.

[0074] The production parameter set refers to a set of process parameters corresponding to each production link (i.e., production process), for example, the mixing link includes powder ratio, stirring speed, and mixing time; the forming link includes pressing pressure and holding time; the sintering link includes heating rate, sintering temperature, holding time, and cooling rate, etc. The production parameter set can be obtained in real time by establishing a communication connection with the production equipment, or can be manually input by a person, etc., but is not limited thereto.

[0075] In S200, a fluctuation analysis task is generated based on the product quality set; wherein the fluctuation analysis task is used to indicate the production link that needs to be analyzed at present.

[0076] It can be understood that the quality data of different attributes may correspond to one or more production processes or production parameters, for example, if the wear resistance data dynamic attribute data fluctuates significantly, it may be associated with the holding time of the sintering link or the raw material particle size distribution; if the hardness and wear resistance fluctuate at the same time, the influence of the mixing uniformity or the forming pressure needs to be considered.

[0077] Exemplarily, the quality data that produces fluctuation or the quality data that produces the most serious fluctuation can be determined by analyzing the product quality set, and then a plurality of production links associated with the quality data are determined according to the fluctuating quality data, and the corresponding production link that mainly produces the fluctuation is determined according to the production link analysis, that is, the fluctuation analysis task; the product quality set can also be divided into data segments according to the time axis (such as every hour), the parameter change time stamp of each production link is recorded synchronously, and the time correlation rule (such as "sintering temperature fluctuation appears hardness fluctuation after 2 hours with support≥0.8") of "parameter fluctuation→quality fluctuation" is mined through the Apriori algorithm. The production link that meets the rule is listed as an analysis object to form a dynamic fluctuation analysis task, etc., but is not limited thereto.

[0078] In one possible implementation, in step S200, the fluctuation analysis task is generated based on the product quality set, including:

[0079] In S210, a key quality indicator is determined based on the product quality set; wherein the key quality indicator is used to indicate the main quality indicator of the quality data fluctuation exceeding the standard.

[0080] It can be understood that the key quality indicator is the analysis indicator for this fluctuation. Exemplarily, the product quality set can be divided into a plurality of hierarchical indicators according to the attribute, and then the fluctuation degree of each hierarchical indicator is analyzed, and finally the key quality indicator is determined according to the fluctuation degree of each hierarchical indicator; the "quality fluctuation entropy weight method" can also be used to quantify the importance of the indicator. That is, each indicator in the product quality set is regarded as a random variable, the objective weight is determined by calculating the information entropy, and then the indicators with high weight ranking are selected as the key quality indicators, etc., but are not limited thereto.

[0081] In a possible implementation, in step S210, the key quality indicators are determined based on the product quality set, including:

[0082] S211, the product quality set is divided into a basic indicator layer and a performance indicator layer; wherein the basic indicator layer is used to reflect the inherent attribute data of the hard alloy wear-resistant block, and the performance indicator layer is used to reflect the dynamic attribute data of the hard alloy wear-resistant block.

[0083] It can be understood that this hierarchical mode is a scientific classification logic based on the material characteristics and service requirements of the hard alloy wear-resistant block, and aims to accurately capture the source of quality fluctuations from different dimensions. The basic indicator layer focuses on the inherent attributes of the material itself, in addition to the hardness data, it can also extend to density, density, porosity and other parameters. These indicators are directly determined by the material composition and forming process, and are the "inherent basis" of product performance. For example, the hardness data is measured by a microhardness tester at 5 different points on the cross section of the product (edge, center and three equal points), and the average value is taken as the basic indicator value of the batch. The performance indicator layer focuses on the service performance of the product in the actual working condition. In addition to wear resistance, it also includes impact toughness, corrosion resistance and other dynamic performance indicators. For example, the wear resistance is tested by a pin-on-disk wear tester under a certain load (such as 50N) and rotation speed (such as 200r / min), and the wear amount (unit: mg) within 1 hour is recorded, and the wear resistance is replaced by the wear amount. After layering, each layer of indicators is standardized (such as converting different units of indicators to dimensionless values in the [0, 1] interval).

[0084] S212, the standard deviations of the basic indicator layer and the performance indicator layer are calculated respectively to obtain a first fluctuation coefficient and a second fluctuation coefficient; wherein the first fluctuation coefficient is used to reflect the data fluctuation degree corresponding to the basic indicator layer, and the second fluctuation coefficient is used to reflect the data fluctuation degree corresponding to the performance indicator layer.

[0085] It can be understood that the standard deviations of the basic indicator layer and the performance indicator layer are calculated to obtain the fluctuation coefficient, which is the core step of converting the fluctuation degree of quality data from qualitative description to quantitative analysis. The essence is to quantify the dispersion degree of data by statistical method. For example, the hardness data is obtained, and the standard deviation is compared with the process standard center value (such as 200HV) of the indicator, and the relative fluctuation proportion is obtained, such as σ 基础 =5HV, the fluctuation coefficient of hardness =5 / 200=2.5%.

[0086] S213, determining the key quality indicators based on the first fluctuation coefficient and the second fluctuation coefficient.

[0087] Exemplarily, the main fluctuation degree can be determined by the fluctuation degree of the first fluctuation coefficient and the second fluctuation coefficient, that is, compared with the preset threshold respectively, the key quality index is determined based on the main fluctuation degree; or the key quality index can be determined by constructing a "fluctuation contribution degree-process sensitivity" two-dimensional model. The model takes the first and second fluctuation coefficients as the vertical axis (representing the fluctuation contribution degree), and the sensitivity of the index to the process parameter as the horizontal axis (the higher the sensitivity, the larger the horizontal axis value). The index priority is sorted by quadrant division, that is, the first quadrant (high fluctuation contribution degree + high sensitivity): for example, the hardness fluctuation coefficient 5% (exceeding the threshold value 2%) in the basic index layer, and the sensitivity to sintering temperature reaches 0.4HV / ℃ (the temperature fluctuates by 1℃, and the hardness fluctuates by 0.4HV), which is listed as a key quality index; the second quadrant (high fluctuation contribution degree + low sensitivity): for example, the wear resistance fluctuation coefficient 6% (exceeding the threshold value 2%) in the performance index layer, but the sensitivity to fine grinding roughness is only 0.1mg / μm (the roughness changes by 1μm, and the wear amount fluctuates by 0.1mg), which is listed as a secondary key index; the third quadrant (low fluctuation contribution degree + low sensitivity), the index does not need to be listed as a key, only needs to be monitored regularly, and the fourth quadrant (low fluctuation contribution degree + high sensitivity) is included in the potential key index, and the like, but not limited thereto.

[0088] In this way, by dividing the product quality set into the basic index layer and the performance index layer, calculating the fluctuation coefficient and determining the key quality index, the hierarchical focusing of quality fluctuation tracing is realized: the basic index layer focuses on the influence of the front-end process on the inherent property, and the performance index layer locks the effect of the back-end process and the service scene on the dynamic property. The combination of the two not only accurately locates the fluctuation source of a single link, but also reveals the implicit correlation of "inherent property fluctuation transmission to dynamic property", greatly improves the targeting and efficiency of fluctuation analysis, provides a clear direction for differentiated process optimization, and reduces invalid analysis cost.

[0089] In a possible implementation, in step S213, the key quality index is determined based on the first fluctuation coefficient and the second fluctuation coefficient, including:

[0090] S2131, compare the first fluctuation coefficient with the first threshold value, and compare the second fluctuation coefficient with the second threshold value to obtain a comparison result; wherein the first threshold value is a critical value of the data fluctuation of the basic index layer, and the second threshold value is a critical value of the data fluctuation of the performance index layer.

[0091] It can be understood that the determination of the first threshold value and the second threshold value can be comprehensively set according to the application scene, process capacity and quality target of the product. Specifically, the first threshold value (the basic index layer) needs to refer to the process stability of the inherent attribute: by statistically analyzing the fluctuation data of the basic index of the qualified batches in the past three months, the 3σ value is taken as the initial threshold value (for example, the 3σ value of hardness is 6HV, and the first threshold value is 3%, that is, 6 / 200); if the product is used in a high-precision scene (such as the aerospace field), it can be tightened to the 2σ value (threshold value 2%). The second threshold value (the performance index layer) needs to be combined with the service requirement of the dynamic attribute: for example, the wear resistance of the wear-resistant block used in mines, submarine engineering construction and the like needs to be guaranteed to fluctuate by no more than 5% (otherwise, the service life is shortened by more than 20%), so the second threshold value is set to 5%; and the wear-resistant block used in ordinary machinery can be relaxed to 8%. In addition, the threshold value needs to be dynamically updated: when the process is improved (such as the introduction of an automatic mixing system), the fluctuation of the basic index is reduced, and the first threshold value and the second threshold value need to be recalculated and lowered.

[0092] S2132, determining the key quality index based on the comparison result.

[0093] Exemplarily, the corresponding quality index is determined as the housekeeper quality index by comparing the different situations, that is, when the first fluctuation coefficient > the first threshold value and the second fluctuation coefficient ≤ the second threshold value, it is indicated that the fluctuation of the inherent attribute is the main contradiction, at this time, the specific index with the most significant fluctuation (such as the hardness fluctuation coefficient 4.5% > 3%, while the density fluctuation coefficient 2.1% ≤ 3%, then “hardness” is locked) can be selected from the basic index layer; if the second fluctuation coefficient > the second threshold value and the first fluctuation coefficient ≤ the first threshold value, it is indicated that the fluctuation of the dynamic attribute is more critical, then the performance index (such as the wear resistance fluctuation coefficient 6% > 5%, then “wear resistance” is locked) can be selected from the performance index layer; if both exceed the threshold value, one of the two can be determined as the key quality index; or both can be taken as the key quality index, and the like.

[0094] In this way, by combining the threshold value comparison and the correlation analysis, the determination of the key quality index has both objectivity and logicality. The technical effects are as follows: first, the differentiated threshold values (the first and second threshold values) distinguish the control standards of the inherent attribute and the dynamic attribute, reducing the misjudgment caused by “one-size-fits-all” (such as adopting a more stringent threshold value for wear resistance, which directly affects the service life); second, through the scene-based decision (single-layer fluctuation / two-layer fluctuation), the key index can cover both the core problem of independent fluctuation and the hidden correlation of coordinated fluctuation; third, it provides a clear direction for subsequent analysis (such as the key index being “hardness”, focusing on forming / sintering, or being “wear resistance”, focusing on sintering / fining), reducing invalid analysis and improving the traceability efficiency. This method changes the quality control from “full-index monitoring” to “targeted management”.

[0095] In a possible implementation, in step S2132, the key quality indicator is determined based on the comparison result, including:

[0096] In S21321a, if the comparison result is that the first fluctuation coefficient is greater than the first threshold value and the second fluctuation coefficient is greater than the second threshold value, an associated quality indicator of the basic indicator layer and the performance indicator layer is generated, and the associated quality indicator is determined as the key quality indicator.

[0097] It can be understood that when the first fluctuation coefficient and the second fluctuation coefficient both exceed the corresponding threshold value, it indicates that there is a coordinated fluctuation between the basic indicator layer and the performance indicator layer, and the hidden association across the layers needs to be captured by generating the associated quality indicator. The generation of the associated quality indicator needs to be based on the process mechanism to excavate the internal relationship between the two: for example, when the hardness of the basic indicator layer and the wear resistance of the performance indicator layer both fluctuate, it is known through analysis that “insufficient mixing uniformity will lead to uneven hardness distribution, and then local wear resistance will decrease”, and therefore the associated quality indicator can be defined as “mixing uniformity-hardness-wear resistance coordinated fluctuation”. The specific generation process is as follows: the potential transmission link between the basic indicator layer and the performance indicator layer is combed through a process flow diagram, for example, from “raw material purity-mixing uniformity-hardness distribution-crystal boundary strength-wear resistance”, the causal relationship of each link is determined (for example, fluctuation of raw material purity will amplify uneven mixing), and in the confirmed transmission link, the core process parameters that have a significant impact on both layers are selected as the associated nodes. For example, in the “mixing uniformity-hardness-wear resistance” link, the mixing time (which affects uniformity) and the sintering temperature (which affects hardness and crystal boundary strength) are the key nodes, and the parameter fluctuation of the key nodes will simultaneously cause the abnormality of the two layers. Then, the core nodes and the associated two-layer indicators are combined to form the name of the associated quality indicator, such as “mixing time-sintering temperature-hardness-wear resistance coordinated fluctuation”.

[0098] In S21321b, if the comparison result is that only the first fluctuation coefficient is greater than the first threshold value, the quality indicator corresponding to the basic indicator layer is determined as the key quality indicator.

[0099] It can be understood that when only the first fluctuation coefficient exceeds the first threshold value, it indicates that the fluctuation is mainly caused by the inherent properties of the material, and at this time, the quality indicators corresponding to the basic indicator layer need to be listed as the key indicators as a whole and further refined to specific parameters. For example, when the basic indicator layer includes hardness, density, and density, the fluctuation contribution of each parameter needs to be calculated (for example, hardness fluctuation contribution is 60%, and density is 30%), and the parameter with the highest contribution (for example, hardness) is taken as the core key indicator.

[0100] In S21321c, if the comparison result is that only the second fluctuation coefficient is greater than the second threshold value, the quality indicator corresponding to the performance indicator layer is determined as the key quality indicator.

[0101] It can be understood that when only the second fluctuation coefficient exceeds the second threshold value, it indicates that the fluctuation is concentrated in the dynamic service performance of the product, and the quality index corresponding to the performance index layer needs to be determined as the key, and the rear-end process is associated with the use scene. For example, when the wear resistance of the performance index layer fluctuates, the wear resistance is determined as the key quality index.

[0102] In this way, by determining the key quality index according to the scene, when the fluctuations of the basic index layer and the performance index layer both exceed the standard, the associated quality index is generated, the implicit association of the cross-level coordinated fluctuation is captured, and when only a single index layer fluctuates, the corresponding level index is locked, which not only ensures accurate coverage of the common root of the coordinated fluctuation, but also realizes focused analysis of independent fluctuation, significantly improves the comprehensiveness and pertinence of key quality index identification, provides clear direction for subsequent production fluctuation tracing, and effectively reduces the problem of inefficient analysis caused by fragmented or missing indexes.

[0103] S220, establishing a quality-link mapping rule, and obtaining a candidate production link corresponding to the key quality index based on the quality-link mapping rule; wherein the candidate production link is used to reflect a set of multiple production links.

[0104] It can be understood that the quality-link mapping rule refers to a structured rule system for explicitly defining the correspondence between "quality index fluctuation" and "production link" by associating the process logic and historical data of the whole process of hard alloy wear-resistant block production through system analysis. The establishment process can be as follows: first, establish the control rule of the link and the parameter to clearly define the key parameters that can be adjusted in each link; then, through correlation analysis (such as Pearson correlation coefficient calculation), the correlation strength of parameter fluctuation and quality index exceeding the standard is quantified, and finally, the rationality of the correlation is verified combined with the process mechanism to form a structured quality-link mapping rule; or it can be established through a machine learning model, that is, the time sequence parameters (such as sintering temperature curve, pressing pressure change rate) and quality detection data of the whole production process are extracted to construct a data set containing 5000+ batches, the time domain / frequency domain features of the parameters are extracted, such as calculating the fluctuation entropy of the sintering temperature and the frequency spectrum features of the mixing power to enhance the model's perception ability of the fluctuation pattern, the production time sequence and static parameters are processed through feature engineering, a multi-modal fusion model is constructed to learn the influence of parameters on quality indexes, and the quality-link mapping rule is established, etc., but not limited to this. By inputting the key quality index into the quality mapping rule, the corresponding candidate production link can be obtained.

[0105] In one possible implementation, in step S220, the quality-link mapping rule is established, including:

[0106] S221, establishing a link-parameter control table; wherein the link-parameter control table is used to reflect the production process corresponding to each production process.

[0107] It can be understood that the establishment of the link-parameter correspondence table can adopt a hierarchical method of "process decomposition-parameter clustering", and the hard alloy wear-resistant block production chain is decomposed into four stages of raw material pretreatment (such as ball milling, drying), forming (such as die pressing, cold isostatic pressing), sintering (such as vacuum sintering, hot isostatic pressing), and post-processing (such as heat treatment, surface coating). Each stage is further divided into specific processes (such as the sintering stage containing heating, holding, and cooling three processes). For each process, key parameters are identified through process FMEA (failure mode and effects analysis) to form a link-parameter correspondence table.

[0108] S222, generating a correlation matrix based on historical production fluctuation data; wherein the correlation matrix is used to reflect the co-occurrence frequency of parameter fluctuation and quality index exceeding the standard.

[0109] It can be understood that the data is first standardized (such as converting the parameter fluctuation amplitude and quality index exceeding the standard value into dimensionless values in the [0, 1] interval), and then the co-occurrence frequency is calculated by the sliding window statistical method. Specifically, for example, the production data of the last 12 months (containing at least 500 batches of parameter records and corresponding quality detection results) can be extracted from the historical database, the parameter fluctuation events (such as "ball milling time deviates from the standard ± 10%") and the quality index exceeding the standard events (such as "wear resistance exceeds 5%") are aligned according to the production batch, and the number of times of their co-occurrence in the same batch is calculated. The proportion is used to form a "parameter-index" co-occurrence frequency matrix. For example, if 24 of the 30 events of ball milling time fluctuation are accompanied by wear resistance exceeding the standard, the co-occurrence frequency of the two is 80%.

[0110] S223, establishing a quality-link mapping rule based on the correlation matrix and the link-parameter correspondence table.

[0111] It can be understood that the high correlation parameters (such as sintering temperature, forming pressure) corresponding to a certain quality index (such as hardness) can be extracted from the correlation matrix; then, the production links to which these parameters belong (such as sintering temperature corresponding to sintering link, forming pressure corresponding to forming link) are queried in reverse through the link-parameter correspondence table; finally, the structured rules are integrated, for example, "when the hardness exceeds the standard, the associated parameters are sintering temperature (sintering link) and forming pressure (forming link), and the parameter stability of these two links should be checked first".

[0112] In this way, the corresponding relationship between the production process and the process parameters is determined through the link-parameter correspondence table, the correlation between the parameter fluctuation and the quality exceeding the standard is quantified by means of the correlation matrix, and finally the quality-link mapping rule realizes the traceability and verifiable correlation of the quality index to the production link, which not only reduces the subjectivity of relying on experience for judgment, but also solves the problem of lack of process interpretation in pure data correlation. When the quality index fluctuates, the associated production link and core parameter can be quickly located through the rule, greatly shortening the traceability time, and providing a clear direction for targeted process optimization (such as focusing on sintering temperature control to reduce hardness fluctuation), improving the accuracy and efficiency of production quality control.

[0113] In step S230, a fluctuation analysis task is obtained based on the candidate production link.

[0114] Exemplarily, a plurality of fluctuation transmission chains can be generated according to the process connection between the plurality of candidate production links, then the best one in terms of the influence degree on the current key quality index is determined according to the influence degree of each fluctuation transmission chain, and then the fluctuation analysis task is determined based on the fluctuation transmission chain; or the candidate production link and the key quality index can be input into a learning model, and the learning model outputs the corresponding fluctuation analysis task, etc., but not limited thereto. The learning model is trained by a plurality of sets of training data, and each set of training data in the plurality of sets of training data includes a candidate production link, a key quality index and a corresponding fluctuation analysis task.

[0115] In this way, the key quality index indicating the fluctuation exceeding the standard of the quality data is determined from the product quality set, then the quality-link mapping rule is established to obtain the candidate production link corresponding to the key quality index, and finally the fluctuation analysis task is generated based on the candidate production link, realizing the accurate traceability from the quality data to the production link: the determination of the key quality index focuses on the core fluctuation problem, the quality-link mapping rule establishes the correlation bridge between the quality and the production, and the candidate production link narrows the analysis range, which greatly improves the targeting and efficiency of the fluctuation analysis, provides a clear analysis direction and targeted solution for the quality fluctuation in the production of hard alloy wear-resistant blocks, reduces the invalid analysis cost, and helps the production process optimization and quality stability improvement.

[0116] In one possible implementation, in step S230, the fluctuation analysis task is obtained based on the candidate production link, including:

[0117] In step S231, a plurality of fluctuation transmission chains are generated based on the process connection relationship of the candidate production link; wherein the fluctuation transmission chain is used to reflect the conduction path of the parameter fluctuation of the previous link through the process coupling relationship to cause the parameter abnormality of the subsequent link.

[0118] It can be understood that the generation of the fluctuation transmission chain based on the process connection relationship of the candidate production link is to sort out the front-back sequence dependence and parameter coupling relationship between links to build a complete path of "front sequence fluctuation → middle sequence conduction → rear sequence anomaly". For example, the candidate production link is mixing, forming and sintering, and the process connection relationship is "mixing uniformity → forming green body density → sintering grain growth". The fluctuation transmission chain generated therefrom can be: mixing time fluctuation → uneven powder mixing → abnormal forming pressure distribution → green body density deviation → local grain overgrowth during sintering → hardness and wear resistance fluctuation, or mixing speed deficiency → uneven dispersion of WC and Co particles → local high cobalt content of the forming green body → low melting point phase (Co phase) aggregation during sintering → reduced grain boundary bonding strength → simultaneous fluctuation of impact toughness and wear resistance, and the like.

[0119] S232, calculating the fluctuation contribution degree of each fluctuation transmission chain; wherein the fluctuation contribution degree is used to reflect the cumulative influence degree of the co-occurrence probability of each production link in the fluctuation transmission chain.

[0120] It can be understood that the local contribution degree of each link in the chain is first calculated: the co-occurrence probability of parameter fluctuation of a certain link and parameter anomaly of the next link (such as the co-occurrence probability of 70% of mixing anomaly and forming density deviation), and then the cumulative contribution degree is calculated by chain multiplication (such as the contribution degree of mixing → forming → sintering = 70% × 65% × 80% = 36.4%). At the same time, a weight coefficient can be introduced to adjust the influence weight of each link (such as the weight of the sintering link to the final quality is 0.4, which is higher than the weight of the mixing link of 0.2), and the final contribution degree = cumulative co-occurrence probability × weight sum. For example, the cumulative co-occurrence probability of a certain transmission chain is 36.4%, and the weight sum is 1.2, then the fluctuation contribution degree = 36.4% × 1.2 = 43.7%. In this way, the fluctuation contribution degrees of multiple transmission chains can be compared horizontally (such as A chain 43.7%, B chain 28.5%).

[0121] S233, obtaining a fluctuation analysis task based on the fluctuation contribution degree.

[0122] Exemplarily, the main fluctuation transmission chain can be determined according to the size of the fluctuation contribution degree, and the fluctuation analysis task can be determined according to the main fluctuation transmission chain. The fluctuation transmission chain whose contribution degree exceeds the threshold value can be included in the analysis range according to the fluctuation contribution degree, and the analysis resources can be allocated according to the contribution degree proportion. For example, if the threshold value is set to 30%, and the contribution degrees of A chain, C chain and D chain are 43.7%, 39.2% and 32% respectively, all of which exceed the threshold value, then the three chains are listed as analysis objects.

[0123] In this way, the fluctuation conduction law of the production link is revealed through the fluctuation transmission chain, the influence weight of each path is quantified by means of the fluctuation contribution, and finally the generated fluctuation analysis task can focus on the core conduction path and key link, and reduce indiscriminate investigation on the whole process. The pertinence of the analysis task is improved, the root problem such as "mixing parameter abnormality → whole chain fluctuation" can be quickly located, and at the same time, through clear detection standards and priorities, the production personnel is guided to investigate in order, and the fluctuation analysis cycle is greatly shortened, which provides accurate task guidance for timely adjusting process parameters and stabilizing product quality.

[0124] In a possible implementation, in step S233, the fluctuation analysis task is obtained based on the fluctuation contribution, comprising:

[0125] S2331, sorting a plurality of fluctuation transmission chains based on the fluctuation contribution to obtain a transmission chain sequence.

[0126] It can be understood that when sorting, the transmission chain sequence can be generated in the manner of "numerical descending order + dynamic layering", specifically, all fluctuation transmission chains are first sorted in descending order of fluctuation contribution (such as A chain 43.7%, C chain 39.2%, D chain 32%, B chain 28.5%), and then the clustering algorithm (such as K-means) is used to divide the layers: the contribution degree ≥40% is the first layer (A chain), 30%-40% is the second layer (C chain, D chain), and <30% is the third layer (B chain), or the sorting can be directly performed according to the numerical descending order, etc., but not limited thereto.

[0127] S2332, determining a core fluctuation chain based on the transmission chain sequence.

[0128] It can be understood that all transmission chains in the first layer (such as A chain) are preferentially selected as the core chain; if the cumulative contribution degree of the transmission chains in the second layer exceeds 50% (such as C chain 39.2%+D chain 32%=71.2%), the second layer as a whole is included in the core chain; if a third layer transmission chain (such as B chain 28.5%) has low contribution degree, but is associated with a key performance index (such as impact toughness, which directly affects product safety), it is also upgraded to the core chain. At the same time, the "contribution degree difference test" is used to exclude the secondary chain: if the contribution degree difference of two adjacent chains is <5% (such as the difference between C chain 39.2% and D chain 32% is 7.2%, which is retained; the difference between D chain 32% and B chain 28.5% is 3.5%, only D chain is retained), so as to avoid that too many core chains cause dispersion of analysis resources. Finally, the core fluctuation chain needs to form a clear list, such as "core chain 1: A chain (43.7%, first layer); core chain 2: C chain (39.2%, second layer)", and mark the corresponding quality index influence range (such as A chain affects hardness and wear resistance, C chain affects density and compactness).

[0129] S2333, obtaining a fluctuation analysis task based on the core fluctuation chain.

[0130] Exemplarily, the candidate production link with the largest correlation degree can be determined as the fluctuation analysis task by calculating the correlation degree of each candidate production link in the core fluctuation chain and the key quality indicator; the core fluctuation chain can also be disassembled into three-level nodes of “link-parameter-quality indicator” according to the process sequence, and then the key quality indicator is matched to obtain the corresponding fluctuation analysis link, and the like, but not limited thereto.

[0131] In this way, by passing the chain to sort the priority level, focusing on the key fluctuation path by means of the core chain screening, and finally disassembling the fluctuation analysis task to realize the “from macro chain to micro parameter” accurate landing, the clear division of link and parameter level tasks can be performed step by step without redundant judgment, the analysis efficiency is improved, and the dynamic hierarchical and cross-validation mechanism ensures that the task can adapt to the changes of process fluctuation, thereby providing an iterative analysis framework for continuous and stable product quality.

[0132] In a possible implementation, in step S2333, the fluctuation analysis task is obtained based on the core fluctuation chain, including:

[0133] S23331, calculating the correlation degree of each candidate production link and the key quality indicator based on the order position of each candidate production link in the core fluctuation chain.

[0134] It can be understood that the order position of the core fluctuation chain is numbered according to the process flow direction (such as mixing is 1, molding is 2, and sintering is 3), and the smaller the order position, the closer to the front end, and the more basic the conduction influence on the subsequent link. The calculation formula is: correlation degree = (1 / order position) x link fluctuation transmission coefficient x key quality indicator sensitivity. For example, in the core chain “mixing (1) → molding (2) → sintering (3)”, the fluctuation transmission coefficient of the sintering link (order position 3) is 0.8 (i.e. the link fluctuation has a 80% probability of transmission to the next link), and the sensitivity of the key quality indicator hardness is 0.7 (70% of the hardness fluctuation can be explained by the link), then the correlation degree = (1 / 3) x 0.8 x 0.7 ≈ 0.187; the transmission coefficient of the molding link (order position 2) is 0.6, and the sensitivity is 0.5, the correlation degree = (1 / 2) x 0.6 x 0.5 = 0.15; the transmission coefficient of the mixing link (order position 1) is 0.5, and the sensitivity is 0.3, the correlation degree = 1 x 0.5 x 0.3 = 0.15.

[0135] S23332, determining the candidate production link with the largest correlation degree as the fluctuation analysis task.

[0136] In this way, by sequentially weighting the correlation degree, the conduction law of process fluctuation (front-end affecting back-end) is embodied, and the link that most directly affects the key quality indicators is highlighted, so that the targeting of the fluctuation analysis task is significantly improved. The correlation degree calculation can quantify the contribution of the link to the quality, reduce the misselection of the link caused by subjective judgment, focus on a single core link task setting, reduce the resource consumption of multi-link parallel analysis, improve the detection efficiency, and provide accurate task guidance for quickly positioning and solving production fluctuations.

[0137] S300, based on the fluctuation analysis task and the corresponding production parameter set, the fluctuation parameter set is obtained; wherein the fluctuation parameter set is used to reflect the fluctuation characteristics of the production parameters that cause the production of the hard alloy wear-resistant block to fluctuate.

[0138] Exemplarily, all production parameters in the fluctuation analysis task can be analyzed to obtain fluctuation data of each production parameter, and the fluctuation parameters and the corresponding fluctuation characteristics are filtered according to the fluctuation data; or

[0139] In one possible implementation, in step S300, based on the fluctuation analysis task and the corresponding production parameter set, the fluctuation parameter set is obtained, including:

[0140] S310, based on the fluctuation analysis task and the corresponding production parameter set, the parameter analysis target is obtained; wherein the parameter analysis target is used to indicate the production parameters in the production parameter set corresponding to the fluctuation analysis task.

[0141] It can be understood that the matching of the fluctuation analysis task and the corresponding production parameter set to obtain the parameter analysis target is to establish the correspondence between the task and the parameter, and accurately lock the production parameters that need to be analyzed. For example, if the fluctuation analysis task is "sintering link parameter fluctuation detection", the corresponding production parameter set includes sintering temperature, holding time, furnace atmosphere pressure and other parameters, and through matching, the parameter analysis target is determined as "sintering temperature, holding time, and furnace atmosphere pressure". The matching process needs to be based on the "link-parameter" association relationship (such as strong association between sintering link and temperature, time and other parameters) in the quality-link mapping rule.

[0142] S320, based on the parameter analysis target, the time sequence fluctuation data of the parameter analysis target in the preset time window is extracted; wherein the time sequence fluctuation data includes a parameter acquisition time point and a corresponding parameter value.

[0143] It can be understood that the setting of the preset time window needs to be combined with the cycle of the production process (for example, one production cycle of sintering is 4 hours, and the time window is set to 4 hours), and at the same time, the time window covers the quality fluctuation occurrence period corresponding to the fluctuation analysis task (for example, it is known that the hardness fluctuation of a batch of products occurs after 2 hours of sintering, and the time window includes the 2-hour period). The time granularity in the extraction of time sequence fluctuation data is consistent with the parameter acquisition frequency (for example, the temperature parameter is acquired every 5 minutes, and the data includes each 5-minute time point and the corresponding temperature value).

[0144] In S330, feature extraction is performed based on the time sequence fluctuation data to obtain a fluctuation parameter set; wherein the fluctuation parameter set includes time domain features and fluctuation pattern features.

[0145] It can be understood that the feature extraction based on the time sequence fluctuation data to obtain the fluctuation parameter set is to quantify the fluctuation law of the parameters and extract features that can reflect the essence of the fluctuation. The time domain features include the mean, standard deviation, range, fluctuation amplitude (such as the difference between the maximum and minimum values of temperature), fluctuation frequency (such as the number of times the fluctuation exceeds the threshold value per hour), etc., which are used to describe the overall fluctuation of the parameter in the numerical value; the fluctuation pattern features include the rising / falling rate of the fluctuation (such as the rate at which the temperature rises from 1400°C to 1420°C in 10 minutes), the duration of the fluctuation (such as the duration of the pressure being below the standard value), the periodicity of the fluctuation (such as the temperature fluctuating once every 20 minutes), etc., which are used to describe the dynamic change mode of the parameter fluctuation.

[0146] In this way, the analysis object is accurately locked by parameter matching, avoiding irrelevant parameter interference; the time sequence data extraction of the preset time window ensures the capture of the complete process of the fluctuation; the combination of time domain and pattern features comprehensively describes the essence of the parameter fluctuation, so that the fluctuation parameter set can not only quantify the degree of fluctuation, but also describe the mode of fluctuation. The accurate matching of the parameter analysis target reduces the invalid parameter analysis work; the targeted extraction of the time sequence data makes the capture of the fluctuation features more timely and accurate; the multi-dimensional feature extraction provides rich basis for subsequent fluctuation cause diagnosis, improves the accuracy of production fluctuation tracing, and provides specific feature reference for process parameter adjustment (such as determining the precision requirement of temperature control according to the fluctuation amplitude), which helps the stable control of production quality.

[0147] In S400, an analysis result is generated based on the fluctuation parameter set; wherein the analysis result is used to reflect the root cause of the fluctuation of the production parameter.

[0148] It can be understood that the generation of the analysis result based on the fluctuation parameter set is to analyze the characteristics of the fluctuation parameters in association with the production process mechanism, the equipment operation state, the environmental influence factors, etc., and trace back to the root cause of the production parameter fluctuation. Specifically, the fluctuation parameter set provides the time domain characteristics (such as fluctuation amplitude, frequency) and the morphological characteristics (such as rising rate, duration) of the parameters, which are the key clues for root cause analysis. For example, if the fluctuation parameter set shows that the sintering temperature "fluctuation amplitude ± 15℃, 3 times per hour over the threshold, rising rate 5℃ / min", it is necessary to make a comprehensive judgment in combination with the heating system principle of the sintering furnace (such as whether the heating wire power is stable), the equipment maintenance record (such as the time of the last furnace body calibration), the environmental factors (such as whether the workshop voltage fluctuates), etc. If it is found that the time sequence curve of voltage fluctuation and temperature fluctuation is highly consistent (correlation coefficient 0.9) at the same time, it can be determined that "unstable workshop voltage" is the root cause of temperature fluctuation; if the voltage is normal but the heating wire resistance detection value is out of standard, the root cause is "abnormal power output caused by heating wire aging". The analysis result needs to clearly indicate the type of the fluctuation root cause (such as equipment failure, parameter setting error, environmental interference), the specific location (such as the heating module of the sintering furnace, the raw material conveying pump) and the influence mechanism (such as the aging of the heating wire leading to the decrease of temperature control precision), which provides a direct basis for subsequent rectification measures.

[0149] In the analysis process, a matching rule library of "fluctuation characteristics-root cause type" can be established, for example: when the parameter fluctuation presents "sudden jump + no periodicity", it is preferred to associate with "sensor failure" or "instantaneous voltage impact"; when the fluctuation presents "gradual offset + continuous expansion", it tends to be "equipment wear" or "raw material batch difference".

[0150] It should be understood that the size of the serial number of each step in the above embodiments does not mean the order of execution, and the execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0151] The fluctuation analysis was performed on 30 batches of products of the hard alloy wear-resistant block production line with the model BAUCAST25x200x20, covering 8 core links such as mixing, molding and sintering. The data were obtained by automatic acquisition through the MES system combined with manual review, wherein the manual inspection adopted the "full parameter statistics + experience judgment" mode, and the specific data are shown in Table 1 below:

[0152] Table 1

[0153] Batch No. Key Quality Indicators Fluctuation Analysis Tasks Fluctuation Parameters and Characteristics Actual Fluctuation Sources Manual Inspection Time (H) Manual Analysis Sources Application Analysis Time (H) Application Analysis Sources 20250403 Hardness: fluctuation coefficient 4.2%; wear resistance: fluctuation coefficient 6.1% Mixing, molding, sintering Mixing: stirring time fluctuation ± 12 minutes, fluctuation frequency 2 times / hour; sintering: temperature fluctuation ± 8℃ (fluctuation form characteristics: rising rate 3℃ / min) Instability of mixing motor speed leads to abnormal stirring time, sintering furnace heating wire aging leads to temperature fluctuation 2 Mixing problems and abnormal molding pressure parameters 0.17 Precise positioning of mixing and sintering problems (accuracy 95%) 20250421 Density: fluctuation coefficient 3.8% Molding, raw material pretreatment Molding: press pressure fluctuation ± 18 MPa Mold wear leads to uneven pressure distribution 0.5 Positioning to the molding link, but analysis as unstable pressure of hydraulic system 0.08 Identify mold wear as the source (accuracy 99%) 20250511 Impact toughness: fluctuation coefficient 7.2% Sintering, post-processing Sintering: holding time fluctuation ± 20 minutes (fluctuation form characteristics: duration over 1 hour) Sintering furnace timer failure leads to inaccurate holding time 0.1 Positioning to the sintering link, but no root cause found 0.03 Accurate identification of timer failure (accuracy 95%) 20250515 Density: fluctuation coefficient 3.5%; friction coefficient: fluctuation coefficient 5.8% Raw material pretreatment, mixing, sintering Raw material pretreatment: powder particle size fluctuation ± 15μm; mixing: ball-to-material ratio fluctuation ± 0.3 (periodic fluctuation) Raw material supplier batch particle size is unstable, mixing machine ball-to-material ratio control system deviation 1.5 Only mixing problems are found, raw material problems are missed 0.2 Simultaneous positioning of raw material and mixing sources (accuracy 94%)

[0154] As shown in Table 1, compared with manual analysis, the fluctuation analysis method for the hard alloy wear-resistant block production of the application greatly shortens the time cost of fluctuation tracing in terms of analysis efficiency, especially in the multi-index coordinated fluctuation scene (such as the batch of 20250515); in terms of analysis accuracy, manual analysis is prone to misjudgment, partial positioning or omission of key factors, and the application greatly improves the accuracy through core fluctuation chain locking, parameter feature extraction and other logic, can accurately locate the fluctuation root of multi-link coupling (such as identifying the problems of mixing motor and sintering furnace at the same time), effectively reducing the subjectivity and one-sidedness of manual analysis. The closed-loop analysis link of "quality data-production link-parameter feature-fluctuation root" of the application: focus on the core problem through key quality indicators, narrow the analysis range through quality-link mapping rules, lock specific parameters through fluctuation transmission chain and parameter feature extraction, and finally realize accurate tracing from quality fluctuation to root, providing reliable support for rapid process optimization of hard alloy wear-resistant block production.

[0155] Corresponding to the hard alloy wear-resistant block production fluctuation analysis method described in the above embodiments, the embodiments of the application also provide a hard alloy wear-resistant block production fluctuation analysis system, and each module of the system can realize each step of the hard alloy wear-resistant block production fluctuation analysis method. Figure 3 The structure block diagram of the hard alloy wear-resistant block production fluctuation analysis system provided by the embodiments of the application is shown, and only the parts related to the embodiments of the application are shown for ease of description.

[0156] Referring to Figure 3 The hard alloy wear-resistant block production fluctuation analysis system comprises:

[0157] The acquisition module is configured to acquire hard alloy wear-resistant block production information; wherein the hard alloy wear-resistant block production information comprises a product quality set of the hard alloy wear-resistant block and a production parameter set corresponding to each production link.

[0158] The first generation module is configured to generate a fluctuation analysis task based on the product quality set; wherein the fluctuation analysis task is used to indicate the production link that needs to be analyzed at present.

[0159] The analysis module is configured to analyze based on the fluctuation analysis task and the corresponding production parameter set to obtain a fluctuation parameter set; wherein the fluctuation parameter set is used to reflect the fluctuation characteristics of the production parameter that causes the fluctuation of the hard alloy wear-resistant block production.

[0160] The second generation module is configured to generate an analysis result based on the fluctuation parameter set; wherein the analysis result is used to reflect the root of the fluctuation of the production parameter.

[0161] It should be noted that the information interaction and execution process between the above modules are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, which will not be repeated here.

[0162] Those skilled in the art will understand that, for the sake of convenience and brevity, the above-described module division is merely an example. In practical applications, the above functions can be assigned to different modules as needed, that is, the internal structure of the system can be divided into different modules to complete all or part of the functions described above. The modules in the embodiments can be integrated into one processing unit, or each module can exist physically separately, or two or more modules can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0163] This application also provides a device for analyzing fluctuations in the production of cemented carbide wear-resistant blocks, such as... Figure 4 As shown, the cemented carbide wear-resistant block production fluctuation analysis device 6 of this embodiment includes: at least one processor 60 ( Figure 4 Only one is shown in the image), at least one memory 61 ( Figure 4 (Only one is shown in the image) and a computer program 62 stored in the at least one memory 61 and executable on the at least one processor 60. When the processor 60 executes the computer program 62, it causes the cemented carbide wear-resistant block production fluctuation analysis device 6 to implement the steps in any of the above-described embodiments of the cemented carbide wear-resistant block production fluctuation analysis method, or causes the cemented carbide wear-resistant block production fluctuation analysis device 6 to implement the functions of each module in the above-described system embodiments.

[0164] Exemplarily, the computer program 62 may be divided into one or more modules / units, which are stored in the memory 61 and executed by the processor 60 to complete this application. The one or more modules / units may be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program 62 in the cemented carbide wear-resistant block production fluctuation analysis equipment 6.

[0165] The cemented carbide wear-resistant block production fluctuation analysis device 6 can be a desktop computer, laptop, handheld computer, or cloud server, etc. This cemented carbide wear-resistant block production fluctuation analysis device may include, but is not limited to, a processor 60 and a memory 61. Those skilled in the art will understand that...Figure 4 The example of the cemented carbide wear-resistant block production fluctuation analysis device 6 does not constitute a limitation on the cemented carbide wear-resistant block production fluctuation analysis device 6, and can include more or fewer components than those shown, or combine certain components, or different components, for example, can also include input / output devices, network access devices, buses, etc.

[0166] The processor 60 can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic components, discrete hardware components, etc. The general-purpose processor can be a microprocessor or can also be any conventional processor.

[0167] The memory 61 can be an internal storage unit of the cemented carbide wear-resistant block production fluctuation analysis device 6 in some embodiments, for example, a hard disk or a memory of the cemented carbide wear-resistant block production fluctuation analysis device 6. The memory 61 can also be an external storage device of the cemented carbide wear-resistant block production fluctuation analysis device 6 in other embodiments, for example, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. Further, the memory 61 can include both the internal storage unit and the external storage device of the cemented carbide wear-resistant block production fluctuation analysis device 6. The memory 61 is used to store an operating system, application programs, a boot loader, data, and other programs, for example, program codes of the computer program, etc. The memory 61 can also be used to temporarily store data that has been output or will be output.

[0168] The embodiments of the present application also provide a computer readable storage medium, which stores a computer program. The computer program is executed by a processor to implement the steps in any of the above method embodiments.

[0169] The embodiment of the present application provides a computer program product, when the computer program product runs on the hard alloy wear-resistant block production fluctuation analysis equipment, so that the hard alloy wear-resistant block production fluctuation analysis equipment realizes the steps in any respective method embodiment.

[0170] The integrated unit, if in the form of a software function unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on such understanding, the present application can realize all or part of the processes in the above-embodied methods by a computer program to instruct related hardware, and the computer program can be stored in a computer-readable storage medium. The computer program can realize the steps of each method embodiment when executed by a processor. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or some intermediate forms. The computer-readable medium at least includes any entity or device capable of carrying the computer program code to the hard alloy wear-resistant block production fluctuation analysis equipment, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunications signal, and a software distribution medium. For example, a U disk, a mobile hard disk, a magnetic disk, or an optical disk.

[0171] In the above embodiments, the description of each embodiment has its own focus, and the parts not described or recorded in detail in a certain embodiment can be referred to the relevant description of other embodiments.

[0172] Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0173] In the embodiments provided by the present application, it should be understood that the disclosed hard alloy wear-resistant block production fluctuation analysis device and system can be implemented in other manners. For example, the embodiments of the hard alloy wear-resistant block production fluctuation analysis system described above are merely illustrative. For example, the division of the modules is merely a logical function division. There can be another division manner in actual implementation. For example, a plurality of modules can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the modules shown or discussed can be indirect coupling or communication connection through some interfaces, devices or modules. It can be electrical, mechanical or other forms.

[0174] The modules described as separated components can or can not be physically separated, and the components shown as modules can or can not be physical modules, i.e. can be located in one place or distributed on a plurality of network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the embodiments.

[0175] The above-described embodiments are merely used to illustrate the technical solutions of the present application, but not limit it; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A method of analyzing production fluctuations of cemented carbide wear plates, characterized in that, The method comprises the following steps: obtaining cemented carbide wear-resistant block production information; wherein the cemented carbide wear-resistant block production information includes a product quality set of the cemented carbide wear-resistant block and a production parameter set corresponding to each production link; generating a fluctuation analysis task based on the product quality set; wherein the fluctuation analysis task is used to indicate the production link that needs to be analyzed at present; analyzing the fluctuation analysis task and the corresponding production parameter set to obtain a fluctuation parameter set; wherein the fluctuation parameter set is used to reflect the fluctuation characteristics of the production parameter that causes the fluctuation of the cemented carbide wear-resistant block production; generating an analysis result based on the fluctuation parameter set; wherein the analysis result is used to reflect the root cause of the fluctuation of the production parameter; the method of generating a fluctuation analysis task based on the product quality set comprises: determining a key quality indicator based on the product quality set; wherein the key quality indicator is used to indicate the main quality indicator of the quality data fluctuation exceeding the standard; establishing a quality-link mapping rule, and obtaining a candidate production link corresponding to the key quality indicator based on the quality-link mapping rule; wherein the candidate production link is used to reflect a set of multiple production links; obtaining a fluctuation analysis task based on the candidate production link; the method of obtaining a fluctuation analysis task based on the candidate production link comprises: generating a plurality of fluctuation transmission chains based on the process coupling relationship of the candidate production link; wherein the fluctuation transmission chain is used to reflect the conduction path that the parameter fluctuation of the previous link causes the parameter abnormality of the subsequent link through the process coupling relationship; calculating the fluctuation contribution degree of each fluctuation transmission chain; wherein the fluctuation contribution degree is used to reflect the cumulative influence degree of the co-occurrence probability of each production link in the fluctuation transmission chain; obtaining a fluctuation analysis task based on the fluctuation contribution degree; analyzing the fluctuation analysis task and the corresponding production parameter set to obtain a fluctuation parameter set, comprising: matching the fluctuation analysis task and the corresponding production parameter set to obtain a parameter analysis target; wherein the parameter analysis target is used to indicate the production parameter in the production parameter set corresponding to the fluctuation analysis task; extracting the time series fluctuation data of the parameter analysis target within a preset time window; wherein the time series fluctuation data includes a parameter collection time point and a corresponding parameter value; performing feature extraction based on the time series fluctuation data to obtain a fluctuation parameter set; wherein the fluctuation parameter set includes time domain features and fluctuation pattern features.

2. A cemented carbide wear block production fluctuation analysis method according to claim 1, c h a r a c t e r i s e d in that the method of determining a key quality indicator based on the product quality set comprises: dividing the product quality set into a basic indicator layer and a performance indicator layer; wherein the basic indicator layer is used to reflect the inherent attribute data of the cemented carbide wear-resistant block, and the performance indicator layer is used to reflect the dynamic attribute data of the cemented carbide wear-resistant block; calculating the standard deviation of the basic indicator layer and the performance indicator layer respectively to obtain a first fluctuation coefficient and a second fluctuation coefficient; wherein the first fluctuation coefficient is used to reflect the data fluctuation degree corresponding to the basic indicator layer, and the second fluctuation coefficient is used to reflect the data fluctuation degree corresponding to the performance indicator layer; Determine a key quality index based on the first fluctuation coefficient and the second fluctuation coefficient.

3. A cemented carbide wear block production fluctuation analysis method according to claim 2, c h a r a c t e r i s e d in that The determining of the key quality index based on the first fluctuation coefficient and the second fluctuation coefficient comprises: comparing the first fluctuation coefficient with a first threshold value and comparing the second fluctuation coefficient with a second threshold value to obtain a comparison result; wherein the first threshold value is a critical value of fluctuation of the basic index layer data, and the second threshold value is a critical value of fluctuation of the performance index layer data; determining the key quality index based on the comparison result.

4. The cemented carbide wear block production fluctuation analysis method according to claim 3, characterized in that, The determining of the key quality index based on the comparison result comprises: if the comparison result is that the first fluctuation coefficient is greater than the first threshold value and the second fluctuation coefficient is greater than the second threshold value, then generating an associated quality index of the basic index layer and the performance index layer, and determining the associated quality index as the key quality index; if the comparison result is that only the first fluctuation coefficient is greater than the first threshold value, then determining the quality index corresponding to the basic index layer as the key quality index; if the comparison result is that only the second fluctuation coefficient is greater than the second threshold value, then determining the quality index corresponding to the performance index layer as the key quality index.

5. The cemented carbide wear block production fluctuation analysis method according to claim 1, characterized in that, The establishing of the quality-link mapping rule comprises: establishing a link-parameter correspondence table; wherein the link-parameter correspondence table is used to reflect the production process corresponding to each production procedure; generating an association matrix based on historical production fluctuation data; wherein the association matrix is used to reflect the co-occurrence frequency of parameter fluctuation and quality index exceeding the standard; establishing a quality-link mapping rule based on the association matrix and the link-parameter correspondence table.

6. The cemented carbide wear block production fluctuation analysis method according to claim 1, characterized in that, The obtaining of the fluctuation analysis task based on the fluctuation contribution degree comprises: sorting a plurality of the fluctuation transmission chains based on the fluctuation contribution degree to obtain a transmission chain sequence; determining a core fluctuation chain based on the transmission chain sequence; obtaining a fluctuation analysis task based on the core fluctuation chain.

7. A cemented carbide wear block production fluctuation analysis method according to claim 6, c h a r a c t e r i z e d in that The obtaining of the fluctuation analysis task based on the core fluctuation chain comprises: calculating the association degree between each candidate production link and the key quality index based on the order position of each candidate production link in the core fluctuation chain; determining the candidate production link with the largest association degree as the fluctuation analysis task.

Images