TWI761975B - Device and method for monitoring abnormal machine process parameters, and storage medium - Google Patents

Abstract

A method for monitoring abnormal machine process parameters includes: selecting objective machine process parameters that affect measured values of each measurement point of a predetermined product from a pre-extracted of multiple product data sets; establishing a measured prediction model corresponding to each measurement point based on the objective machine process parameters and the measured values of each measurement point of multiple predetermined products, and calculating a goodness of fit of each measured prediction model; inputting the goodness of fit, a predicted value, the objective machine process parameters, and parameter coefficients of the measured prediction model corresponding to each measurement point into a problem index set; calculating an index value of influence degree of each machine process parameter based on elements of the problem index set; outputting warning information of machine process parameters that exceed a first predetermined index value of influence degree. A device for monitoring abnormal machine process parameters and a storage medium are also provided.

Description

機台製程參數的異常監測裝置、方法及可讀存儲介質 Device, method and readable storage medium for abnormality monitoring of machine process parameters

本發明涉及機台監控技術領域，尤其涉及一種機台製程參數的異常監測裝置、方法及電腦可讀存儲介質。 The invention relates to the technical field of machine monitoring, and in particular, to an abnormal monitoring device, method and computer-readable storage medium for machine process parameters.

在產品加工過程中，為了提升產品製程的良率與效率，降低產品製程設備的故障率具有重要意義。在現有產品生產過程中，雖然有些機台具備自動控制補償機制功能，可做部分製程參數的物理性自動修正，但往往不容易發現真實異常的製程參數，影響製程品質，甚至導致需進行停機維修。 In the process of product processing, in order to improve the yield and efficiency of the product process, it is of great significance to reduce the failure rate of the product process equipment. In the production process of existing products, although some machines have the function of automatic control compensation mechanism and can do physical automatic correction of some process parameters, it is often difficult to find the real abnormal process parameters, which affects the process quality and even leads to the need for downtime for maintenance. .

有鑑於此，有必要提供一種機台製程參數的異常監測裝置、方法及電腦可讀存儲介質，可實現早日發現異常機台製程參數，減少機台停機時間與尋找異常製程參數的時間，提升製程品質。 In view of this, it is necessary to provide an abnormality monitoring device, method and computer-readable storage medium for machine process parameters, which can realize early detection of abnormal machine process parameters, reduce machine downtime and time to find abnormal process parameters, and improve the process. quality.

本發明一實施方式提供一種機台製程參數的異常監測方法，多個所述機台製程參數用於加工多個預設產品，所述預設產品設置有第一量測點至第N量測點，N個所述量測點用於量測所述預設產品的同一產品參數，N為大於1的正整數，所述異常監測方法包括：基於預先提取的多筆產品資料集篩選得到影響所述預設產品在所述第一量測點的量測值的多個目的機台製程參數， 其中，多個所述目的機台製程參數為多個所述機台製程參數中的部分參數或全部參數，多筆所述產品資料集對應多個所述預設產品，每筆所述產品資料集包括多個所述機台製程參數及對應的預設產品在所述第一量測點的量測值；基於多個所述目的機台製程參數與多個所述預設產品在所述第一量測點的量測值構建與所述第一量測點對應的量測值預測模型，以根據所述量測值預測模型預測得到的所述第一量測點的預測值計算所述量測值預測模型的擬合度，其中，所述量測值預測模型根據每一所述目的機台製程參數及每一所述目的機台製程參數對應的參數係數計算得到所述第一量測點的預測值；將與所述第一量測點對應的量測值預測模型的擬合度、所述第一量測點的預測值，與所述第一量測點對應的目的機台製程參數及與所述第一量測點對應的目的機台製程參數的參數係數彙整至問題指標集；重複上述步驟直至將與所述第N量測點對應的量測值預測模型的擬合度、所述第N量測點的預測值，與所述第N量測點對應的目的機台製程參數及與所述第N量測點對應的目的機台製程參數的參數係數彙整至所述問題指標集；基於所述問題指標集中的元素計算得到每一所述機台製程參數的影響程度指標值；及輸出超過第一預設影響程度指標值的機台製程參數的警示資訊。 An embodiment of the present invention provides a method for monitoring abnormality of machine process parameters. A plurality of the machine process parameters are used to process a plurality of preset products, and the preset products are provided with a first measurement point to an Nth measurement point. The N measurement points are used to measure the same product parameter of the preset product, N is a positive integer greater than 1, and the abnormality monitoring method includes: filtering and obtaining influences based on multiple pre-extracted product data sets a plurality of target machine process parameters of the measurement value of the preset product at the first measurement point, Wherein, the plurality of target machine process parameters are some or all of the plurality of machine process parameters, and the plurality of product data sets correspond to a plurality of the preset products, and each product data set corresponds to a plurality of the preset products. The set includes a plurality of the machine process parameters and the measurement values of the corresponding preset products at the first measurement point; based on the plurality of the target machine process parameters and the plurality of the preset products in the The measurement value of the first measurement point constructs a measurement value prediction model corresponding to the first measurement point, so as to calculate the calculated value according to the prediction value of the first measurement point predicted by the measurement value prediction model. The fitting degree of the measurement value prediction model, wherein the measurement value prediction model calculates the first value according to each of the target machine process parameters and the parameter coefficients corresponding to each of the target machine process parameters. The predicted value of the measurement point; the purpose of predicting the fit of the model, the predicted value of the first measurement point, and the corresponding measurement value of the first measurement point to the first measurement point. The machine process parameters and the parameter coefficients of the target machine process parameters corresponding to the first measurement point are compiled into the problem index set; the above steps are repeated until the measurement value corresponding to the Nth measurement point is predicted by the model. The degree of fit, the predicted value of the Nth measurement point, the target machine process parameters corresponding to the Nth measurement point, and the parameter coefficient aggregation of the target machine process parameters corresponding to the Nth measurement point to the problem index set; calculating the influence degree index value of each of the machine process parameters based on the elements in the problem index set; and outputting the warning information of the machine process parameter exceeding the first preset influence degree index value .

可選地，所述方法還包括：收集加工所述預設產品的機台加工參數及所述預設產品的量測參數；從收集到的所述機台加工參數及所述量測參數提取指定資料存儲至分析資料庫，其中所述分析資料庫至少包括第一資料表、第二資料表及第三資料表，所述第一資料表用於存儲所述指定資料中的多個所述機台製程參數，所述第二資料表用於存儲所述指定資料中的N個所述量測點的量測值，所述第三資料表用於存儲多個所述機台製程參數與每個所述量測點的量測值的映射關係；從所述分析資料庫中提取多筆所述產品資料集。 Optionally, the method further includes: collecting machine processing parameters for processing the preset product and measurement parameters of the preset product; extracting from the collected machine processing parameters and the measurement parameters The specified data is stored in an analysis database, wherein the analysis database includes at least a first data table, a second data table and a third data table, and the first data table is used to store a plurality of the specified data. Machine process parameters, the second data table is used to store the measurement values of the N measurement points in the specified data, and the third data table is used to store a plurality of the machine process parameters and The mapping relationship of the measurement values of each of the measurement points; extracting a plurality of the product data sets from the analysis database.

可選地，所述基於收集到的所述機台加工參數及所述量測參數構 建分析資料庫的步驟，包括：利用預設擷取轉換與載入(Extract-Transform-Load，ETL)工具從收集到的所述機台加工參數及所述量測參數中提取多個所述機台製程參數及N個所述量測點的量測值，並載入至所述分析資料庫。 Optionally, the said machine tool processing parameter and the said measurement parameter structure based on the collection The step of building an analysis database includes: extracting a plurality of the said machine tool processing parameters and the measurement parameters collected by using a preset Extract-Transform-Load (ETL) tool The machine process parameters and the measurement values of the N measurement points are loaded into the analysis database.

可選地，所述基於多個所述目的機台製程參數與多個所述預設產品在所述第一量測點的量測值構建與所述第一量測點對應的量測值預測模型的步驟之前，包括：若所述第一量測點的量測值包括多個維度值，利用預設降維函數將所述第一量測點的量測值映射為一維的數值。 Optionally, the measurement value corresponding to the first measurement point is constructed based on a plurality of the target machine process parameters and a plurality of measurement values of the preset product at the first measurement point Before the step of predicting the model, it includes: if the measurement value of the first measurement point includes a plurality of dimension values, using a preset dimension reduction function to map the measurement value of the first measurement point to a one-dimensional value .

可選地，所述量測值預測模型為線性模型。 Optionally, the measurement value prediction model is a linear model.

可選地，所述線性模型包括多個線性係數，多個所述目的機台製程參數的參數係數與多個所述線性係數一一對應。 Optionally, the linear model includes a plurality of linear coefficients, and the parameter coefficients of the plurality of process parameters of the target machine are in one-to-one correspondence with the plurality of the linear coefficients.

可選地，所述基於所述問題指標集中的元素計算得到每一所述機台製程參數的影響程度指標值的步驟，包括：基於所述問題指標集中每一所述機台製程參數出現的次數及每一所述機台製程參數的參數係數換算得到每一所述機台製程參數的影響程度指標值。 Optionally, the step of obtaining the influence degree index value of each of the machine process parameters based on the elements in the problem index set includes: based on the problem index set, the occurrence of each of the machine process parameters. The number of times and the parameter coefficient of each of the machine process parameters are converted to obtain the influence degree index value of each of the machine process parameters.

可選地，所述基於所述問題指標集中每一所述機台製程參數出現的次數及每一所述機台製程參數的參數係數換算得到每一所述機台製程參數的影響程度指標值的步驟，包括：採用多種預設換算方式對所述問題指標集中每一所述機台製程參數出現的次數及每一所述機台製程參數的參數係數進行換算，得到每一所述機台製程參數的多個影響程度指標值；當每一所述機台製程參數存在多個超過所述第一預設影響程度指標值的影響程度指標值時，基於多個超過所述第一預設影響程度指標值的影響程度指標值得到一聚合機台異常指標，以描述每一所述機台製程參數的異常程度。 Optionally, the index value of the degree of influence of each of the machine process parameters is obtained by converting the number of occurrences of each of the machine process parameters and the parameter coefficient of each of the machine process parameters based on the problem index set. The step includes: using a variety of preset conversion methods to convert the number of occurrences of each of the machine process parameters in the problem index set and the parameter coefficient of each of the machine process parameters to obtain each of the machine tools. Multiple influence degree index values of process parameters; when there are multiple influence degree index values that exceed the first preset influence degree index value for each of the machine process parameters, based on the plurality of influence degree index values exceeding the first preset The influence degree index value of the influence degree index value obtains an aggregated machine abnormality index to describe the abnormality degree of each of the machine process parameters.

本發明一實施方式提供一種機台製程參數的異常監測裝置，所述裝置包括處理器及記憶體，所述記憶體上存儲有若干電腦程式，所述處理器用 於執行記憶體中存儲的電腦程式時實現上述機台製程參數的異常監測方法的步驟。 An embodiment of the present invention provides an abnormality monitoring device for machine process parameters, the device includes a processor and a memory, the memory stores a number of computer programs, the processor uses The steps of the abnormal monitoring method for the above-mentioned machine process parameters are realized when the computer program stored in the memory is executed.

本發明一實施方式還提供一種電腦可讀存儲介質，所述電腦可讀存儲介質存儲有多條指令，多條所述指令可被一個或者多個處理器執行，以實現上述的機台製程參數的異常監測方法的步驟。 An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a plurality of instructions, and the plurality of instructions can be executed by one or more processors, so as to realize the above-mentioned machine process parameters The steps of the anomaly monitoring method.

與現有技術相比，上述機台製程參數的異常監測裝置、方法及電腦可讀存儲介質，利用多點量測的量測資料與機台製程資料構建問題指標集並進行分析，以早日發現有異常的機台製程參數，減少機台停機時間，提升製程品質。 Compared with the prior art, the above-mentioned device, method and computer-readable storage medium for abnormality monitoring of machine process parameters use multi-point measurement data and machine process data to construct a problem index set and analyze it, so as to detect problems early. Abnormal machine process parameters reduce machine downtime and improve process quality.

10:記憶體 10: Memory

20:處理器 20: Processor

30:異常監測程式 30: Abnormal monitoring program

101:篩選模組 101: Screening Modules

102:構建模組 102: Building Mods

103:彙整模組 103: Assemble modules

104:計算模組 104: Computing Modules

105:輸出模組 105: Output module

100:異常監測裝置 100: Abnormal monitoring device

200:機台 200: machine

S300、S302、S304、S306、S308、S310:步驟 S300, S302, S304, S306, S308, S310: Steps

圖1是本發明一實施方式的機台製程參數的異常監測裝置的功能模組圖。 FIG. 1 is a functional block diagram of an abnormality monitoring device for machine process parameters according to an embodiment of the present invention.

圖2是本發明一實施方式的機台製程參數的異常監測程式的功能模組圖。 FIG. 2 is a functional block diagram of an abnormality monitoring program for machine process parameters according to an embodiment of the present invention.

圖3是本發明一實施方式的機台製程參數的異常監測方法的流程圖。 FIG. 3 is a flowchart of a method for abnormality monitoring of process parameters of a machine tool according to an embodiment of the present invention.

請參閱圖1，為本發明異常監測裝置較佳實施例的示意圖。 Please refer to FIG. 1 , which is a schematic diagram of a preferred embodiment of the abnormality monitoring device of the present invention.

異常監測裝置100可以實現對一個或多個機台200的機台製程參數進行監控，以實現提前發現異常的機台製程參數。機台200可以是指對零件、產品進行加工的機台，包括但不限於機床、數控設備、工業機器人等。機台製程參數可以是機台200的加工部件的相關參數，比如刀具的長度、轉速、切削 參數等。 The abnormality monitoring device 100 can monitor the machine process parameters of one or more machines 200, so as to detect abnormal machine process parameters in advance. The machine 200 may refer to a machine for processing parts and products, including but not limited to machine tools, numerical control equipment, industrial robots, and the like. The machine process parameters may be related parameters of the machined parts of the machine 200, such as the length, rotation speed, cutting parameters, etc.

異常監測裝置100可以包括記憶體10、處理器20以及存儲在記憶體10中並可在處理器20上運行的異常監測程式30。處理器20執行異常監測程式30時實現異常監測方法實施例中的步驟，例如圖3所示的步驟S300~S310。或者，所述處理器20執行異常監測程式30時實現圖2中各模組的功能，例如模組101~105。 The abnormality monitoring device 100 may include a memory 10 , a processor 20 , and an abnormality monitoring program 30 stored in the memory 10 and executable on the processor 20 . When the processor 20 executes the abnormality monitoring program 30 , the steps in the embodiment of the abnormality monitoring method are implemented, for example, steps S300 to S310 shown in FIG. 3 . Alternatively, when the processor 20 executes the abnormality monitoring program 30 , the functions of the modules shown in FIG. 2 , such as modules 101 to 105 , are implemented.

異常監測程式30可以被分割成一個或多個模組，所述一個或者多個模組被存儲在記憶體10中，並由處理器20執行，以完成本發明。所述一個或多個模組可以是能夠完成特定功能的一系列電腦程式指令段，所述指令段用於描述異常監測程式30在異常監測裝置100中的執行過程。例如，異常監測程式30可以被分割成圖2中的篩選模組101、構建模組102、彙整模組103、計算模組104及輸出模組105。各模組具體功能參見下圖2中各模組的功能。 The abnormality monitoring program 30 can be divided into one or more modules, and the one or more modules are stored in the memory 10 and executed by the processor 20 to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the abnormality monitoring program 30 in the abnormality monitoring device 100 . For example, the abnormality monitoring program 30 can be divided into a screening module 101 , a building module 102 , an aggregation module 103 , a calculation module 104 and an output module 105 in FIG. 2 . For the specific functions of each module, please refer to the function of each module in Figure 2 below.

本領域技術人員可以理解，所述示意圖僅是異常監測裝置100的示例，並不構成對異常監測裝置100的限定，可以包括比圖示更多或更少的部件，或者組合某些部件，或者不同的部件，例如異常監測裝置100還可以包括輸入顯示裝置、匯流排等。 Those skilled in the art can understand that the schematic diagram is only an example of the abnormality monitoring device 100, and does not constitute a limitation on the abnormality monitoring device 100. It may include more or less components than the one shown, or combine some components, or Various components such as the abnormality monitoring device 100 may also include input display devices, bus bars, and the like.

處理器20可以是中央處理單元(Central Processing Unit，CPU)，還可以是其他通用處理器、數位訊號處理器(Digital Signal Processor，DSP)、專用積體電路(Application Specific Integrated Circuit，ASIC)、現成可程式設計閘陣列(Field-Programmable Gate Array，FPGA)或者其他可程式設計邏輯器件、分立門或者電晶體邏輯器件、分立硬體元件等。通用處理器可以是微處理器或者處理器20也可以是任何常規的處理器等，處理器20可以利用各種介面和匯流排連接異常監測裝置100的各個部分。 The processor 20 may be a central processing unit (Central Processing Unit, CPU), or other general-purpose processors, digital signal processors (Digital Signal Processors, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), off-the-shelf processors Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor 20 can also be any conventional processor, etc. The processor 20 can use various interfaces and bus bars to connect various parts of the abnormality monitoring device 100 .

記憶體10可用於存儲異常監測程式30和/或模組，處理器20通過 運行或執行存儲在記憶體10內的電腦程式和/或模組，以及調用存儲在記憶體10內的資料，實現異常監測裝置100的各種功能。記憶體10可以包括高速隨機存取記憶體，還可以包括非易失性記憶體，例如硬碟、記憶體、插接式硬碟，智慧存儲卡(Smart Media Card,SMC)，安全數位(Secure Digital,SD)卡，快閃記憶體卡(Flash Card)、至少一個磁碟記憶體件、快閃記憶體器件、或其他易失性固態記憶體件。 The memory 10 can be used to store the abnormality monitoring program 30 and/or the module, and the processor 20 can use the Running or executing the computer programs and/or modules stored in the memory 10 , and calling the data stored in the memory 10 , realizes various functions of the abnormality monitoring device 100 . The memory 10 may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, Smart Media Card (SMC), Secure Digital (Secure) Digital, SD) card, flash memory card (Flash Card), at least one disk memory device, flash memory device, or other volatile solid state memory device.

圖2為本發明異常監測程式較佳實施例的功能模組圖。 FIG. 2 is a functional module diagram of a preferred embodiment of an abnormality monitoring program of the present invention.

參閱圖2所示，異常監測程式30可以包括篩選模組101、構建模組102、彙整模組103、計算模組104及輸出模組105。在一實施方式中，上述模組可以為存儲於記憶體10中且可被處理器20調用執行的可程式化軟體指令。可以理解的是，在其他實施方式中，上述模組也可為固化於處理器20中的程式指令或固件(firmware)。 Referring to FIG. 2 , the abnormality monitoring program 30 may include a screening module 101 , a construction module 102 , an aggregation module 103 , a calculation module 104 and an output module 105 . In one embodiment, the above-mentioned modules may be programmable software instructions stored in the memory 10 and invoked by the processor 20 for execution. It can be understood that, in other embodiments, the above-mentioned module can also be a program instruction or firmware solidified in the processor 20 .

篩選模組101用於基於預先提取的多筆產品資料集篩選得到影響預設產品在第一量測點的量測值的多個目的機台製程參數。 The screening module 101 is configured to screen out a plurality of target machine process parameters affecting the measurement value of the preset product at the first measurement point based on the pre-extracted plurality of product data sets.

在一實施方式中，機台200用於加工所述預設產品，假設機台200包括p個機台製程參數X₁~X_p，p為大於1的正整數。預設產品設置有N個量測點(定義為第一量測點Y₁至第N量測點Y_N)，N個量測點可選用於量測預設產品的同一產品參數，N為大於1的正整數。比如，預設產品為液晶面板，預設產品設置有N個量測點來量測面板厚度。多個目的機台製程參數為多個機台製程參數中的部分參數或全部參數，比如多個機台製程參數為Ω={X₁,X₂,X₃,...,X_p}，多個目的機台製程參數為{X'₁,X'₂,...,X'_k}，X'_k

Ω且k

p。 In one embodiment, the machine 200 is used to process the preset product, and it is assumed that the machine 200 includes p machine process parameters X ₁ to X _p , where p is a positive integer greater than 1. The preset product is set with N measurement points (defined as the first measurement point Y ₁ to the Nth measurement point Y _N ), and the N measurement points can be selected for measuring the same product parameters of the preset product, and N is A positive integer greater than 1. For example, the preset product is a liquid crystal panel, and the preset product is provided with N measuring points to measure the thickness of the panel. The process parameters of the multiple destination machines are some or all of the process parameters of the multiple machines. For example, the process parameters of the multiple machines are Ω={X ₁ , X ₂ , X ₃ ,...,X _p }, The process parameters of multiple destination machines are {X' ₁ ,X' ₂ ,...,X' _k }, X' _k

Ω and k

p.

多筆產品資料集對應多個預設產品，每筆產品資料集包括p個機台製程參數{X₁,X₂,X₃,...,X_p}及對應的預設產品在第一量測點Y₁的量測值。可以理解，對於多個預設產品而言，在第一量測點Y₁的量測值可能相同，也可能 不相同。 Multiple product data sets correspond to multiple preset products, and each product data set includes p machine process parameters {X ₁ , X ₂ , X ₃ ,..., X _p } and the corresponding preset products in the first Measurement value of measurement point Y ₁ . It can be understood that, for multiple preset products, the measurement values at the first measurement point Y ₁ may be the same or may be different.

在一實施方式中，可以定期或不定期收集機台200加工該預設產品所拋出的資料(如XML格式檔)及預設產品的量測資料，並手動或自動暫存到一指定存儲區。然後利用現有之ETL工具從該指定存儲區中提取指定資料轉存至分析資料庫，進而後續可以基於分析資料庫找出異常的機台製程參數。比如，該指定資料至少可以包括機台製程參數及量測點的量測值。所述分析資料庫可以至少包括第一資料表、第二資料表及第三資料表，所述第一資料表用於存儲多個所述機台製程參數，所述第二資料表用於存儲N個所述量測點的量測值，所述第三資料表用於存儲多個所述機台製程參數與每個所述量測點的量測值的映射關係。 In one embodiment, the data (such as XML format files) thrown out by the machine 200 for processing the preset product and the measurement data of the preset product can be collected periodically or irregularly, and temporarily stored in a designated storage manually or automatically. Area. Then, the existing ETL tool is used to extract the specified data from the specified storage area and transfer it to the analysis database, and then the abnormal machine process parameters can be found based on the analysis database later. For example, the specified data may at least include machine process parameters and measurement values of measurement points. The analysis database may at least include a first data table, a second data table and a third data table, the first data table is used to store a plurality of the machine process parameters, and the second data table is used to store N measurement values of the measurement points, and the third data table is used for storing a plurality of mapping relationships between the process parameters of the machine and the measurement values of each measurement point.

在一實施方式中，ETL工具同樣可以定期或不定期從該指定存儲區中提取指定資料轉存至分析資料庫。多筆產品資料集可以從該分析資料庫中提取得到。 In one embodiment, the ETL tool can also periodically or irregularly extract specified data from the specified storage area and transfer it to the analysis database. Multiple product datasets can be extracted from the analysis database.

在一實施方式中，篩選模組101可以基於預設篩預演演算法對預先提取的多筆產品資料集進行分析，以篩選得到影響預設產品在第一量測點的量測值的多個目的機台製程參數。該預設篩預演演算法通過確定多個機台製程參數的組合的差異程度指標來篩選目的機台製程參數，該差異程度指標可以包括均方誤差(mean-square error,MSE)，MSE可以反映估計量與被估計量之間差異程度。對於目的機台製程參數的篩選，差異程度指標越小越好，機台製程參數的個數越少越好。 In one embodiment, the screening module 101 can analyze the pre-extracted multiple product data sets based on a preset screening pre-algorithm, so as to obtain a plurality of items that affect the measurement value of the preset product at the first measurement point. Process parameters of the destination machine. The preset screening pre-calculation algorithm selects the process parameters of the target machine by determining the difference degree index of the combination of the process parameters of the multiple machines. The difference degree index may include the mean-square error (MSE), and the MSE may reflect The degree of difference between the estimator and the estimated quantity. For the screening of the process parameters of the target machine, the smaller the difference index, the better, and the fewer the number of machine process parameters, the better.

舉例而言，機台製程參數為10個，即機台製程參數{X₁,X₂,X₃,...,X₁₀}，對10個機台製程參數進行排列組合可以得到多個機台製程參數集，每個機台製程參數集可以包括至少一個機台製程參數，根據每個機台製程參數集預測的第一量測點的值與第一量測點的實際量測值的差值來確定每個機台製程參 數集對應的差異程度指標值。對於多個預設產品而言，如果在第一量測點的量測值存在一個或多個不相同，可以先計算第一量測點的量測值的平均值，再計算其與每個機台製程參數集預測的第一量測點的值的差值。假設，第一機台製程參數集{X₁,X₂}對應的差異程度指標值為0.3，第二機台製程參數集{X₁,X₂,X₃}對應的差異程度指標值為0.7，第三機台製程參數集{X₂,X₄,X₇}對應的差異程度指標值為0.1，第四機台製程參數集{X₂,X₄,X₇,X₉}對應的差異程度指標值為0.1。假設經過比對分析，多個機台製程參數集的最小差異程度指標值為0.1，且差異程度指標值為0.1的機台製程參數集為第三機台製程參數集與第四機台製程參數集，由於第三機台製程參數集中的元素個數少於第四機台製程參數集中的元素個數，進而第三機台製程參數集中包含的三個機台製程參數集{X₂,X₄,X₇}被定義為目的機台製程參數。 For example, if there are 10 machine process parameters, that is, machine process parameters {X ₁ , X ₂ , X ₃ ,..., X ₁₀ }, by arranging and combining 10 machine process parameters, multiple machines can be obtained. Machine process parameter set, each machine process parameter set may include at least one machine process parameter, according to the difference between the value of the first measurement point predicted by each machine process parameter set and the actual measurement value of the first measurement point The difference value is used to determine the index value of the degree of difference corresponding to the process parameter set of each machine. For multiple preset products, if there are one or more different measurement values at the first measurement point, the average value of the measurement values at the first measurement point can be calculated first, and then the average value of the measurement values at the first measurement point can be calculated. The difference between the values of the first measurement point predicted by the machine process parameter set. Suppose, the difference degree index value corresponding to the first machine process parameter set {X ₁ , X ₂ } is 0.3, and the difference degree index value corresponding to the second machine process parameter set {X ₁ , X ₂ , X ₃ } is 0.7 , the difference degree index value corresponding to the third machine process parameter set {X ₂ , X ₄ , X ₇ } is 0.1, and the fourth machine process parameter set {X ₂ , X ₄ , X ₇ , X ₉ } corresponds to the difference The degree index value is 0.1. Suppose that after comparative analysis, the minimum difference degree index value of multiple machine process parameter sets is 0.1, and the machine process parameter set with the difference degree index value of 0.1 is the third machine process parameter set and the fourth machine process parameter set Since the number of elements in the process parameter set of the third machine is less than the number of elements in the process parameter set of the fourth machine, the three machine process parameter sets included in the process parameter set of the third machine {X ₂ , X ₄ , X ₇ } is defined as the process parameter of the destination machine.

構建模組102用於基於多個所述目的機台製程參數與多個所述預設產品在所述第一量測點的量測值構建與所述第一量測點對應的量測值預測模型M1，以根據量測值預測模型M1預測得到的第一量測點的預測值計算量測值預測模型M1的擬合度。 The construction module 102 is configured to construct a measurement value corresponding to the first measurement point based on a plurality of the target machine process parameters and a plurality of measurement values of the predetermined product at the first measurement point The prediction model M1 calculates the fitting degree of the measurement value prediction model M1 based on the prediction value of the first measurement point predicted by the measurement value prediction model M1.

在一實施方式中，多個所述目的機台製程參數即是可以較準確地預測第一量測點的量測值的機台製程參數。量測值預測模型M1優選為線性模型，量測值預測模型M1可以根據每一目的機台製程參數及每一目的機台製程參數對應的參數係數計算得到第一量測點的預測值。 In one embodiment, the plurality of target machine process parameters are machine process parameters that can more accurately predict the measurement value of the first measurement point. The measurement value prediction model M1 is preferably a linear model, and the measurement value prediction model M1 can calculate the prediction value of the first measurement point according to the process parameters of each target machine and the parameter coefficients corresponding to the process parameters of each target machine.

在一實施方式中，構建模組102可以基於多個目的機台製程參數與多個所述預設產品在所述第一量測點的量測值構建得到與第一量測點對應的量測值預測模型M1。比如多個所述目的機台製程參數為機台製程參數{X₂,X₄,X₇}，第一量測點的量測值預測模型M1為Y₁=a* X₂+b*X₄+c*X₇，構建模組102可以基於目的機台製程參數{X₂,X₄,X₇}的參數值與多個預設產品在第一量測點 的量測值來進行回歸分析，以確定參數係數a、b、c的值。當確定參數係數a、b、c的值以後，即可以將機台製程參數{X₂,X₄,X₇}的一組參數值輸入至量測值預測模型M1，計算得到與該組參數值對應的第一量測點的預測值。 In one embodiment, the construction module 102 can construct a quantity corresponding to the first measurement point based on a plurality of target machine process parameters and a plurality of measurement values of the preset product at the first measurement point. Measured value prediction model M1. For example, a plurality of the target machine process parameters are machine process parameters {X ₂ , X ₄ , X ₇ }, and the measurement value prediction model M1 of the first measurement point is Y ₁ =a* X ₂ +b*X ₄ +c*X ₇ , the building module 102 can perform regression based on the parameter values of the target machine process parameters {X ₂ , X ₄ , X ₇ } and the measured values of multiple preset products at the first measuring point Analysis to determine the values of parameter coefficients a, b, c. After the values of the parameter coefficients a, b, and c are determined, a set of parameter values of the machine process parameters {X ₂ , X ₄ , X ₇ } can be input into the measurement value prediction model M1, and the calculation results are related to the set of parameters. The predicted value of the first measurement point corresponding to the value.

進一步地，當可以通過量測值預測模型M1預測第一量測點的值時，可根據量測值預測模型M1預測得到的第一量測點的預測值計算量測值預測模型M1的擬合度(模型的預測結果與實際結果的吻合程度)。具體地，可以對量測值預測模型M1預測得到的第一量測點的預測值與第一量測點的實際量測值進行分析，得到量測值預測模型M1的擬合度Rs1。 Further, when the value of the first measurement point can be predicted by the measurement value prediction model M1, the simulation of the measurement value prediction model M1 can be calculated according to the predicted value of the first measurement point predicted by the measurement value prediction model M1. Conformity (how well the predicted results of the model agree with the actual results). Specifically, the predicted value of the first measurement point predicted by the measurement value prediction model M1 and the actual measurement value of the first measurement point can be analyzed to obtain the fitting degree Rs1 of the measurement value prediction model M1.

在一實施方式中，可以預先構建多個線性模型，構建模組102可以根據每個線性模型的模型指標挑選合適的線性模型來構建得到與所述第一量測點對應的量測值預測模型M1。模型指標可以包括平均絕對誤差(Mean Absolute Error，MAE)、均方誤差(Mean Square Error，MSE)、均方根誤差(Root Mean Square Error，RMSE)。 In one embodiment, a plurality of linear models may be pre-built, and the building module 102 may select an appropriate linear model according to the model index of each linear model to construct a measurement value prediction model corresponding to the first measurement point. M1. Model indicators may include Mean Absolute Error (MAE), Mean Square Error (MSE), and Root Mean Square Error (RMSE).

在一實施方式中，由於線性模型的輸出一般為一維的數值。當N個量測點中的每個量測點都包含多個維度的量測值時，可以利用預設降維函數將每個量測點的量測值映射為一維的數值。比如，第一個量測點的量測值為三軸方向的偏差值Z₁、Z₂、Z₃，即包括三個維度值，預設降維函數可以根據實際 需求進行設定，如預設降維函數為g(x)，

。當將第一個量測點的量測值降維至一維後，構建模組102再基於多個目的機台製程參數與多個預設產品在第一量測點的量測值構建量測值預測模型M1。 In one embodiment, the output of the linear model is generally a one-dimensional numerical value. When each of the N measurement points includes measurement values of multiple dimensions, a preset dimension reduction function may be used to map the measurement values of each measurement point into a one-dimensional value. For example, the measurement value of the first measurement point is the deviation values Z ₁ , Z ₂ , and Z ₃ in the three-axis direction, which includes three dimension values. The preset dimension reduction function can be set according to actual needs, such as preset The dimensionality reduction function is g(x),

. After reducing the dimension of the measurement value of the first measurement point to one dimension, the construction module 102 constructs the quantity based on the plurality of target machine process parameters and the measurement value of the plurality of preset products at the first measurement point Measured value prediction model M1.

彙整模組103用於將與第一量測點對應的量測值預測模型M1的擬合度、第一量測點的預測值，與第一量測點對應的目的機台製程參數及與第一量測點對應的目的機台製程參數的參數係數彙整至問題指標集。 The integration module 103 is used for predicting the fit of the model M1 with the measurement value corresponding to the first measurement point, the prediction value of the first measurement point, the target machine process parameters corresponding to the first measurement point, and the The parameter coefficients of the target machine process parameters corresponding to the first measurement point are collected into the problem index set.

在一實施方式中，比如與第一量測點對應的量測值預測模型M1 為Y₁=0.2* X₂+0.3*X₄+0.1*X₇，則與第一量測點對應的量測值預測模型M1的擬合度為擬合度Rs1，第一量測點的預測值即為量測值預測模型M1預測得到的值，與第一量測點對應的目的機台製程參數為{X₂,X₄,X₇}，與第一量測點對應的目的機台製程參數的參數係數為{0.2,0.3,0.1}。 In one embodiment, for example, the measurement value prediction model M1 corresponding to the first measurement point is Y ₁ =0.2*X ₂ +0.3*X ₄ +0.1*X ₇ , then the quantity corresponding to the first measurement point is The degree of fit of the measured value prediction model M1 is the degree of fit Rs1, the predicted value of the first measurement point is the value predicted by the measured value prediction model M1, and the target machine process parameters corresponding to the first measurement point are {X ₂ , X ₄ , X ₇ }, the parameter coefficients of the process parameters of the target machine corresponding to the first measurement point are {0.2, 0.3, 0.1}.

可以理解，對於第二量測點，採用與上述第一量測點相同的處理方式，得到與第二量測點對應的量測值預測模型M2。彙整模組103可以將與第二量測點對應的量測值預測模型M2的擬合度、第二量測點的預測值，與第二量測點對應的目的機台製程參數及與第二量測點對應的目的機台製程參數的參數係數彙整至問題指標集。對於第N量測點，同樣可以採用與上述第一量測點相同的處理方式，得到與第N量測點對應的量測值預測模型MN。彙整模組103可以將與第N量測點對應的量測值預測模型MN的擬合度、第N量測點的預測值，與第N量測點對應的目的機台製程參數及與第N量測點對應的目的機台製程參數的參數係數彙整至問題指標集。 It can be understood that, for the second measurement point, the same processing method as the above-mentioned first measurement point is used to obtain the measurement value prediction model M2 corresponding to the second measurement point. The aggregation module 103 can predict the fitting degree of the model M2 corresponding to the measurement value corresponding to the second measurement point, the prediction value of the second measurement point, the process parameters of the target machine corresponding to the second measurement point, and the process parameters of the target machine corresponding to the second measurement point. The parameter coefficients of the target machine process parameters corresponding to the two measurement points are collected into the problem index set. For the Nth measurement point, the same processing method as the above-mentioned first measurement point can also be used to obtain the measurement value prediction model MN corresponding to the Nth measurement point. The aggregation module 103 can predict the fit of the model MN, the predicted value of the Nth measurement point, the target machine process parameters corresponding to the Nth measurement point, and the The parameter coefficients of the target machine process parameters corresponding to the N measurement points are aggregated into the problem index set.

計算模組104用於基於所述問題指標集中的元素計算得到每一機台製程參數的影響程度指標值。 The calculation module 104 is configured to calculate the influence degree index value of each machine process parameter based on the elements in the problem index set.

在一實施方式中，該影響程度指標值可反應出影響機台200加工預設產品的機台製程參數的重要程度，當計算出來的影響程度指標數值越大，表示該機台製程參數越重要，亦為有可能出現異常的機台製程參數。計算模組104可以選用預設換算函數及所述問題指標集中的元素計算得到每一機台製程參數的影響程度指標值。 In one embodiment, the influence degree index value can reflect the importance of the process parameters of the machine that affect the processing of the preset product by the machine 200. When the calculated influence degree index value is larger, it means that the process parameters of the machine are more important. , is also a machine process parameter that may appear abnormal. The calculation module 104 can select the preset conversion function and the elements in the problem index set to calculate the index value of the influence degree of the process parameter of each machine.

舉例而言，以10個機台製程參數{X₁,X₂,X₃,...,X₁₀}，5個量測點{Y₁,Y₂,Y₃,...,Y₅}為例進行說明，篩選模組101篩選得到影響預設產品在該5個量測點的量測值的目的機台製程參數如下表1所示。 For example, with 10 machine process parameters {X ₁ ,X ₂ ,X ₃ ,...,X ₁₀ }, 5 measurement points {Y ₁ ,Y ₂ ,Y ₃ ,...,Y ₅ } As an example to illustrate, the screening module 101 selects the target machine process parameters that affect the measurement values of the preset product at the five measurement points, as shown in Table 1 below.

表1

Table 1

如上表1所示，第一量測點Y₁對應的目的機台製程參數為機台製程參數X₂、X₆。第二量測點Y₂對應的目的機台製程參數為機台製程參數X₃、X₄、X₇。第五量測點Y₅對應的目的機台製程參數為機台製程參數X₂、X₆。當構建得到與每個量測點{Y₁,Y₂,Y₃,...,Y₅}的量測值預測模型M1~M5後，針對每個量測點或該每個量測點對應的目的機台製程參數一一建立線性模型，並進行存R-squared(可決係數)操作，可以得到如下表2所示的R-squared值。 As shown in Table 1 above, the target machine process parameters corresponding to the first measurement point Y ₁ are the machine process parameters X ₂ and X ₆ . The target machine process parameters corresponding to the second measurement point Y ₂ are the machine process parameters X ₃ , X ₄ , and X ₇ . The target machine process parameters corresponding to the fifth measurement point Y ₅ are the machine process parameters X ₂ and X ₆ . After constructing and obtaining the measurement value prediction models M1~M5 related to each measurement point {Y ₁ , Y ₂ , Y ₃ ,..., Y ₅ }, for each measurement point or each measurement point The corresponding target machine process parameters are established one by one to establish a linear model, and the operation of storing R-squared (coefficient of determination) is performed, and the R-squared value shown in Table 2 below can be obtained.

如上表2所示，對於上表X₂、Y₁而言，利用X₂和Y₁建立線性模型，並計算得到R-squared=0.43，後將0.43存入表2對應的欄位，對於上表X₆、Y₁而言，利用X₆和Y₁建立線性模型，並計算得到R-squared=0.1，後將0.1存入表2對應的欄位。對於不屬於與量測點對應的目的機台製程參數的其他機台製程參數，可以進行補0處理。 As shown in Table 2 above, for X ₂ and Y ₁ in the above table, use X ₂ and Y ₁ to establish a linear model, and calculate to obtain R-squared=0.43, and then store 0.43 into the column corresponding to Table 2. For the above For Tables X ₆ and Y ₁ , use X ₆ and Y ₁ to establish a linear model, and calculate to obtain R-squared=0.1, and then store 0.1 into the corresponding column of Table 2. For other machine process parameters that do not belong to the process parameters of the target machine corresponding to the measurement point, 0 can be supplemented.

可以理解，以上是以量測值預測模型為線性模型為例，因此對應的決策係數為R-squared，若量測值預測模型不是線性模型，則對應的決策係數不一定為R-squared，可以是其他的可決係數(coefficient of determination)。 It can be understood that the above is an example of the measurement value prediction model being a linear model, so the corresponding decision coefficient is R-squared. If the measurement value prediction model is not a linear model, the corresponding decision coefficient is not necessarily R-squared. are other coefficients of determination.

計算模組104可以基於每一機台製程參數在表2中出現的次數、R-squared值及每一機台製程參數的參數係數換算得到每一機台製程參數的影響程度指標值，如下表3所示。在實際情況中，對於每一機台製程參數而言，可能包含多個影響程度指標值。 The calculation module 104 can convert the index value of the influence degree of each machine process parameter based on the number of times each machine process parameter appears in Table 2, the R-squared value and the parameter coefficient of each machine process parameter, as shown in the following table 3 shown. In an actual situation, for each process parameter of a machine, there may be multiple influence degree index values.

如上表3所示，機台製程參數X₁僅出現一次，且機台製程參數X₁的R-squared值為0.08，經過預設換算函數可以換算得到機台製程參數X₁的影響程度指標值為0.08，機台製程參數X₂出現三次，且機台製程參數X₂對應的R-squared值包括0.43、0.28、0.22，經過預設換算函數可以換算得到機台製程參數X₂的影響程度指標值為1.597，機台製程參數X₅出現零次，經過預設換算函數可以換算得到機台製程參數X₅的影響程度指標值為0，機台製程參數X₇出現二次，且機台製程參數X₇對應的R-squared值包括0.1、0.11，經過預設換算函數可以換算得到機台製程參數X₇的影響程度指標值為0.398。該預設換算函數基於機台製程參數出現的次數及機台製程參數可以影響量測值的程度進行設定。 As shown in Table 3 above, the machine process parameter X ₁ appears only once, and the R-squared value of the machine process parameter X ₁ is 0.08. The influence degree index value of the machine process parameter X ₁ can be obtained by conversion through the preset conversion function. is 0.08, the machine process parameter X ₂ appears three times, and the R-squared value corresponding to the machine process parameter X ₂ includes 0.43, 0.28, and 0.22, and the influence degree index of the machine process parameter X ₂ can be converted through the preset conversion function. The value is 1.597, the machine process parameter X ₅ appears zero times, and the influence degree index value of the machine process parameter X ₅ can be converted through the preset conversion function. The value is 0, the machine process parameter X ₇ appears twice, and the machine process The R-squared value corresponding to the parameter X ₇ includes 0.1 and 0.11, and the influence degree index value of the machine process parameter X ₇ can be converted to 0.398 through a preset conversion function. The preset conversion function is set based on the number of occurrences of the machine process parameter and the degree to which the machine process parameter can affect the measurement value.

輸出模組105用於輸出超過第一預設影響程度指標值的機台製程參數的警示資訊。 The output module 105 is used for outputting the warning information of the process parameters of the machine tool exceeding the first preset influence degree index value.

在一實施方式中，第一預設影響程度指標值可以根據實際需求進行設定與調整。比如第一預設影響程度指標值設定為1.0，則在表3中，輸出模組105輸出機台製程參數X₂、X₆的警示資訊，以提醒機台用戶及時進行機台維 護，減少機台停機時間。比如，可以在異常監測裝置100的顯示裝置輸出機台製程參數X₂、X₆的警示資訊。 In one embodiment, the first preset influence degree index value can be set and adjusted according to actual needs. For example, the first preset influence degree index value is set to 1.0, then in Table 3, the output module 105 outputs the warning information of the machine process parameters X ₂ and X ₆ to remind the machine user to perform machine maintenance in time and reduce the number of machine tools. Desk downtime. For example, the warning information of the machine process parameters X ₂ and X ₆ may be output on the display device of the abnormality monitoring device 100 .

在一實施方式中，可以採用多種預設換算方式對所述問題指標集中每一所述機台製程參數出現的次數及每一所述機台製程參數的參數係數進行換算，得到每一所述機台製程參數的多個影響程度指標值。當每一所述機台製程參數存在多個影響程度指標值超過第一預設影響程度指標值時，可以根據該多個超過第一預設影響程度指標值的影響程度指標值輸出一聚合機台異常指標，該聚合機台異常指標可用於描述每一所述機台製程參數的異常程度，以便於機台用戶瞭解異常原因和程度。比如，聚合機台異常指標指示了機台200的一加工零件導致的異常。 In one embodiment, a variety of preset conversion methods can be used to convert the number of occurrences of each of the machine process parameters in the problem index set and the parameter coefficients of each of the machine process parameters to obtain each of the Multiple influence degree index values of machine process parameters. When each of the machine process parameters has a plurality of influence degree index values exceeding the first preset influence degree index value, an aggregator may be output according to the plurality of influence degree index values exceeding the first preset influence degree index value The abnormality index of the machine can be used to describe the abnormality degree of each of the process parameters of the machine, so that the machine user can understand the cause and degree of the abnormality. For example, the aggregation machine abnormality index indicates an abnormality caused by a machined part of the machine 200 .

圖3為本發明一實施方式中異常監測方法的流程圖。根據不同的需求，所述流程圖中步驟的順序可以改變，某些步驟可以省略。 FIG. 3 is a flowchart of an abnormality monitoring method in an embodiment of the present invention. According to different requirements, the order of the steps in the flowchart can be changed, and some steps can be omitted.

步驟S300，基於預先提取的多筆產品資料集篩選得到影響所述預設產品在所述第一量測點的量測值的多個目的機台製程參數。 In step S300, a plurality of target machine process parameters affecting the measurement value of the preset product at the first measurement point are obtained by screening based on the pre-extracted multiple product data sets.

在一實施方式中，機台200用於加工所述預設產品，假設機台200包括p個機台製程參數X₁~X_p，p為大於1的正整數。預設產品設置有N個量測點(定義為第一量測點Y₁至第N量測點Y_N)，N個量測點可選用於量測預設產品的同一產品參數，N為大於1的正整數。比如，預設產品為液晶面板，預設產品設置有N個量測點來量測面板厚度。多個目的機台製程參數為多個機台製程參數中的部分參數或全部參數，比如多個機台製程參數為Ω={X₁,X₂,X₃,...,X_p}，多個目的機台製程參數為{X'₁,X'₂,...,X'_k}，X'_k

Ω且k

p。 In one embodiment, the machine 200 is used to process the preset product, and it is assumed that the machine 200 includes p machine process parameters X ₁ to X _p , where p is a positive integer greater than 1. The preset product is set with N measurement points (defined as the first measurement point Y ₁ to the Nth measurement point Y _N ), and the N measurement points can be selected for measuring the same product parameters of the preset product, and N is A positive integer greater than 1. For example, the preset product is a liquid crystal panel, and the preset product is provided with N measuring points to measure the thickness of the panel. The process parameters of the multiple destination machines are some or all of the process parameters of the multiple machines. For example, the process parameters of the multiple machines are Ω={X ₁ , X ₂ , X ₃ ,...,X _p }, The process parameters of multiple destination machines are {X' ₁ ,X' ₂ ,...,X' _k }, X' _k

Ω and k

p.

多筆產品資料集對應多個預設產品，每筆產品資料集包括p個機台製程參數{X₁,X₂,X₃,...,X_p}及對應的預設產品在第一量測點Y₁的量測值。可以理解，對於多個預設產品而言，在第一量測點Y₁的量測值可能相同，也可能 不相同。 Multiple product data sets correspond to multiple preset products, and each product data set includes p machine process parameters {X ₁ , X ₂ , X ₃ ,..., X _p } and the corresponding preset products in the first Measurement value of measurement point Y ₁ . It can be understood that, for multiple preset products, the measurement values at the first measurement point Y ₁ may be the same or may be different.

在一實施方式中，可以定期或不定期收集機台200加工該預設產品所拋出的資料(如XML格式檔)及預設產品的量測資料，並手動或自動暫存到一指定存儲區。然後利用ETL工具從該指定存儲區中提取指定資料轉存至分析資料庫，進而後續可以基於分析資料庫找出異常的機台製程參數。比如，該指定資料至少可以包括機台製程參數及量測點的量測值。所述分析資料庫可以至少包括第一資料表、第二資料表及第三資料表，所述第一資料表用於存儲多個所述機台製程參數，所述第二資料表用於存儲N個所述量測點的量測值，所述第三資料表用於存儲多個所述機台製程參數與每個所述量測點的量測值的映射關係。 In one embodiment, the data (such as XML format files) thrown out by the machine 200 for processing the preset product and the measurement data of the preset product can be collected periodically or irregularly, and temporarily stored in a designated storage manually or automatically. Area. Then, the ETL tool is used to extract the specified data from the specified storage area and transfer it to the analysis database, and then the abnormal machine process parameters can be found based on the analysis database later. For example, the specified data may at least include machine process parameters and measurement values of measurement points. The analysis database may at least include a first data table, a second data table and a third data table, the first data table is used to store a plurality of the machine process parameters, and the second data table is used to store N measurement values of the measurement points, and the third data table is used for storing a plurality of mapping relationships between the process parameters of the machine and the measurement values of each measurement point.

在一實施方式中，ETL工具同樣可以定期或不定期從該指定存儲區中提取指定資料轉存至分析資料庫。多筆產品資料集可以從該分析資料庫中提取得到。 In one embodiment, the ETL tool can also periodically or irregularly extract specified data from the specified storage area and transfer it to the analysis database. Multiple product datasets can be extracted from the analysis database.

在一實施方式中，可以基於預設篩預演演算法對預先提取的多筆產品資料集進行分析，以篩選得到影響預設產品在第一量測點的量測值的多個目的機台製程參數。該預設篩預演演算法通過確定多個機台製程參數的組合的差異程度指標來篩選目的機台製程參數，該差異程度指標可以包括均方誤差(mean-square error,MSE)，MSE可以反映估計量與被估計量之間差異程度。對於目的機台製程參數的篩選，差異程度指標越小越好，機台製程參數的個數越少越好。 In one embodiment, the pre-extracted multiple product data sets can be analyzed based on a preset screening pre-calculation algorithm, so as to obtain multiple target machine manufacturing processes that affect the measurement value of the preset product at the first measurement point. parameter. The preset screening pre-calculation algorithm selects the process parameters of the target machine by determining the difference degree index of the combination of the process parameters of the multiple machines. The difference degree index may include the mean-square error (MSE), and the MSE may reflect The degree of difference between the estimator and the estimated quantity. For the screening of the process parameters of the target machine, the smaller the difference index, the better, and the fewer the number of machine process parameters, the better.

舉例而言，機台製程參數為10個，即機台製程參數{X₁,X₂,X₃,...,X₁₀}，對10個機台製程參數進行排列組合可以得到多個機台製程參數集，每個機台製程參數集可以包括至少一個機台製程參數，根據每個機台製程參數集預測的第一量測點的值與第一量測點的實際量測值的差值來確定每個機台製程參 數集對應的差異程度指標值。對於多個預設產品而言，如果在第一量測點的量測值存在一個或多個不相同，可以先計算第一量測點的量測值的平均值，再計算其與每個機台製程參數集預測的第一量測點的值的差值。假設，第一機台製程參數集{X₁,X₂}對應的差異程度指標值為0.3，第二機台製程參數集{X₁,X₂,X₃}對應的差異程度指標值為0.7，第三機台製程參數集{X₂,X₄,X₇}對應的差異程度指標值為0.1，第四機台製程參數集{X₂,X₄,X₇,X₉}對應的差異程度指標值為0.1。假設經過比對分析，多個機台製程參數集的最小差異程度指標值為0.1，且差異程度指標值為0.1的機台製程參數集為第三機台製程參數集與第四機台製程參數集，由於第三機台製程參數集中的元素個數少於第四機台製程參數集中的元素個數，進而第三機台製程參數集中包含的三個機台製程參數集{X₂,X₄,X₇}被定義為目的機台製程參數。 For example, if there are 10 machine process parameters, that is, machine process parameters {X ₁ , X ₂ , X ₃ ,..., X ₁₀ }, by arranging and combining 10 machine process parameters, multiple machines can be obtained. Machine process parameter set, each machine process parameter set may include at least one machine process parameter, according to the difference between the value of the first measurement point predicted by each machine process parameter set and the actual measurement value of the first measurement point The difference value is used to determine the index value of the degree of difference corresponding to the process parameter set of each machine. For multiple preset products, if there are one or more different measurement values at the first measurement point, the average value of the measurement values at the first measurement point can be calculated first, and then the average value of the measurement values at the first measurement point can be calculated. The difference between the values of the first measurement point predicted by the machine process parameter set. Suppose, the difference degree index value corresponding to the first machine process parameter set {X ₁ , X ₂ } is 0.3, and the difference degree index value corresponding to the second machine process parameter set {X ₁ , X ₂ , X ₃ } is 0.7 , the difference degree index value corresponding to the third machine process parameter set {X ₂ , X ₄ , X ₇ } is 0.1, and the fourth machine process parameter set {X ₂ , X ₄ , X ₇ , X ₉ } corresponds to the difference The degree index value is 0.1. Suppose that after comparative analysis, the minimum difference degree index value of multiple machine process parameter sets is 0.1, and the machine process parameter set with the difference degree index value of 0.1 is the third machine process parameter set and the fourth machine process parameter set Since the number of elements in the process parameter set of the third machine is less than the number of elements in the process parameter set of the fourth machine, the three machine process parameter sets included in the process parameter set of the third machine {X ₂ , X ₄ , X ₇ } is defined as the process parameter of the destination machine.

步驟S302，基於多個所述目的機台製程參數與多個所述預設產品在所述第一量測點的量測值構建與所述第一量測點對應的量測值預測模型，以根據所述量測值預測模型預測得到的所述第一量測點的預測值計算所述量測值預測模型的擬合度。 Step S302, building a measurement value prediction model corresponding to the first measurement point based on a plurality of the target machine process parameters and a plurality of measurement values of the preset product at the first measurement point, The degree of fit of the measurement value prediction model is calculated based on the predicted value of the first measurement point obtained according to the measurement value prediction model.

在一實施方式中，多個所述目的機台製程參數即是可以較準確地預測第一量測點的量測值的機台製程參數。量測值預測模型M1優選為線性模型，量測值預測模型M1可以根據每一目的機台製程參數及每一目的機台製程參數對應的參數係數計算得到第一量測點的預測值。 In one embodiment, the plurality of target machine process parameters are machine process parameters that can more accurately predict the measurement value of the first measurement point. The measurement value prediction model M1 is preferably a linear model, and the measurement value prediction model M1 can calculate the prediction value of the first measurement point according to the process parameters of each target machine and the parameter coefficients corresponding to the process parameters of each target machine.

在一實施方式中，可以基於多個目的機台製程參數與多個所述預設產品在所述第一量測點的量測值構建得到與第一量測點對應的量測值預測模型M1。比如多個所述目的機台製程參數為機台製程參數{X₂,X₄,X₇}，第一量測點的量測值預測模型M1為Y₁=a* X₂+b*X₄+c*X₇，可以基於目的機台製程參數{X₂,X₄,X₇}的參數值與多個預設產品在第一量測點的量測值來進行回歸分析， 以確定參數係數a、b、c的值。當確定參數係數a、b、c的值以後，即可以將機台製程參數{X₂,X₄,X₇}的一組參數值輸入至量測值預測模型M1，計算得到與該組參數值對應的第一量測點的預測值。 In one embodiment, a measurement value prediction model corresponding to the first measurement point can be constructed based on a plurality of target machine process parameters and a plurality of measurement values of the preset product at the first measurement point. M1. For example, a plurality of the target machine process parameters are machine process parameters {X ₂ , X ₄ , X ₇ }, and the measurement value prediction model M1 of the first measurement point is Y ₁ =a* X ₂ +b*X ₄ +c*X ₇ , regression analysis can be performed based on the parameter values of the target machine process parameters {X ₂ , X ₄ , X ₇ } and the measured values of multiple preset products at the first measurement point to determine Values of parameter coefficients a, b, c. After the values of the parameter coefficients a, b, and c are determined, a set of parameter values of the machine process parameters {X ₂ , X ₄ , X ₇ } can be input into the measurement value prediction model M1, and the calculation results are related to the set of parameters. The predicted value of the first measurement point corresponding to the value.

進一步地，當可以通過量測值預測模型M1預測第一量測點的值時，可根據量測值預測模型M1預測得到的第一量測點的預測值計算量測值預測模型M1的擬合度(模型的預測結果與實際結果的吻合程度)。具體地，可以對量測值預測模型M1預測得到的第一量測點的預測值與第一量測點的實際量測值進行分析，得到量測值預測模型M1的擬合度Rs1。 Further, when the value of the first measurement point can be predicted by the measurement value prediction model M1, the simulation of the measurement value prediction model M1 can be calculated according to the predicted value of the first measurement point predicted by the measurement value prediction model M1. Conformity (how well the predicted results of the model agree with the actual results). Specifically, the predicted value of the first measurement point predicted by the measurement value prediction model M1 and the actual measurement value of the first measurement point can be analyzed to obtain the fitting degree Rs1 of the measurement value prediction model M1.

在一實施方式中，可以預先構建多個線性模型，並根據每個線性模型的模型指標挑選合適的線性模型來構建得到與所述第一量測點對應的量測值預測模型M1。模型指標可以包括平均絕對誤差(Mean Absolute Error，MAE)、均方誤差(Mean Square Error，MSE)、均方根誤差(Root Mean Square Error，RMSE)。 In one embodiment, a plurality of linear models may be constructed in advance, and an appropriate linear model may be selected according to the model index of each linear model to construct a measurement value prediction model M1 corresponding to the first measurement point. Model indicators may include Mean Absolute Error (MAE), Mean Square Error (MSE), and Root Mean Square Error (RMSE).

在一實施方式中，由於線性模型的輸出一般為一維的數值。當N個量測點中的每個量測點都包含多個維度的量測值時，可以利用預設降維函數將每個量測點的量測值映射為一維的數值。比如，第一個量測點的量測值為三軸方向的偏差值Z₁、Z₂、Z₃，即包括三個維度值，預設降維函數可以根據實際 需求進行設定，如預設降維函數為g(x)，

。當將第一個量測點的量測值降維至一維後，再基於多個目的機台製程參數與多個預設產品在第一量測點的量測值構建量測值預測模型M1。 In one embodiment, the output of the linear model is generally a one-dimensional numerical value. When each of the N measurement points includes measurement values of multiple dimensions, a preset dimension reduction function may be used to map the measurement values of each measurement point into a one-dimensional value. For example, the measurement value of the first measurement point is the deviation values Z ₁ , Z ₂ , and Z ₃ in the three-axis direction, which includes three dimension values. The preset dimension reduction function can be set according to actual needs, such as preset The dimensionality reduction function is g(x),

. After the measurement value of the first measurement point is reduced to one dimension, a measurement value prediction model is constructed based on the process parameters of multiple target machines and the measurement values of multiple preset products at the first measurement point M1.

步驟S304，將與所述第一量測點對應的量測值預測模型的擬合度、所述第一量測點的預測值，與所述第一量測點對應的目的機台製程參數及與所述第一量測點對應的目的機台製程參數的參數係數彙整至問題指標集。 Step S304, predicting the fit of the model, the predicted value of the first measurement point, and the target machine process parameters corresponding to the first measurement point with respect to the measurement value corresponding to the first measurement point and the parameter coefficients of the process parameters of the target machine corresponding to the first measurement point are collected into a problem index set.

在一實施方式中，比如與第一量測點對應的量測值預測模型M1 為Y₁=0.2* X₂+0.3*X₄+0.1*X₇，則與第一量測點對應的量測值預測模型M1的擬合度為擬合度Rs1，第一量測點的預測值即為量測值預測模型M1預測得到的值，與第一量測點對應的目的機台製程參數為{X₂,X₄,X₇}，與第一量測點對應的目的機台製程參數的參數係數為{0.2,0.3,0.1}。 In one embodiment, for example, the measurement value prediction model M1 corresponding to the first measurement point is Y ₁ =0.2*X ₂ +0.3*X ₄ +0.1*X ₇ , then the quantity corresponding to the first measurement point is The degree of fit of the measured value prediction model M1 is the degree of fit Rs1, the predicted value of the first measurement point is the value predicted by the measured value prediction model M1, and the target machine process parameters corresponding to the first measurement point are {X ₂ , X ₄ , X ₇ }, the parameter coefficients of the process parameters of the target machine corresponding to the first measurement point are {0.2, 0.3, 0.1}.

可以理解，對於第二量測點，採用與上述第一量測點相同的處理方式，得到與第二量測點對應的量測值預測模型M2。可以將與第二量測點對應的量測值預測模型M2的擬合度、第二量測點的預測值，與第二量測點對應的目的機台製程參數及與第二量測點對應的目的機台製程參數的參數係數彙整至問題指標集。對於第N量測點，同樣可以採用與上述第一量測點相同的處理方式，得到與第N量測點對應的量測值預測模型MN。可以將與第N量測點對應的量測值預測模型MN的擬合度、第N量測點的預測值，與第N量測點對應的目的機台製程參數及與第N量測點對應的目的機台製程參數的參數係數彙整至問題指標集。 It can be understood that, for the second measurement point, the same processing method as the above-mentioned first measurement point is used to obtain the measurement value prediction model M2 corresponding to the second measurement point. The measurement value corresponding to the second measurement point can be used to predict the fit of the model M2, the predicted value of the second measurement point, the target machine process parameters corresponding to the second measurement point, and the second measurement point. The parameter coefficients of the corresponding destination machine process parameters are compiled into the problem index set. For the Nth measurement point, the same processing method as the above-mentioned first measurement point can also be used to obtain the measurement value prediction model MN corresponding to the Nth measurement point. The measurement value corresponding to the Nth measurement point can be used to predict the fit of the model MN, the predicted value of the Nth measurement point, the target machine process parameters corresponding to the Nth measurement point, and the Nth measurement point. The parameter coefficients of the corresponding destination machine process parameters are compiled into the problem index set.

步驟S306，重複上述步驟直至將與所述第N量測點對應的量測值預測模型的擬合度、所述第N量測點的預測值，與所述第N量測點對應的目的機台製程參數及與所述第N量測點對應的目的機台製程參數的參數係數彙整至所述問題指標集。 Step S306, repeating the above steps until the measurement value corresponding to the Nth measurement point is predicted to be the degree of fit of the model, the predicted value of the Nth measurement point, and the purpose corresponding to the Nth measurement point The machine process parameters and the parameter coefficients of the target machine process parameters corresponding to the Nth measurement point are collected into the problem index set.

步驟S308，基於所述問題指標集中的元素計算得到每一所述機台製程參數的影響程度指標值。 Step S308, calculating an index value of the degree of influence of each of the machine process parameters based on the elements in the problem index set.

在一實施方式中，該影響程度指標值可反應出影響機台200加工預設產品的機台製程參數的重要程度，當計算出來的影響程度指標數值越大，表示該機台製程參數越重要，亦為有可能出現異常的機台製程參數。可以選用預設換算函數及所述問題指標集中的元素計算得到每一機台製程參數的影響程度指標值。 In one embodiment, the influence degree index value can reflect the importance of the process parameters of the machine that affect the processing of the preset product by the machine 200. When the calculated influence degree index value is larger, it means that the process parameters of the machine are more important. , is also a machine process parameter that may appear abnormal. A preset conversion function and elements in the problem index set can be selected to calculate the influence degree index value of each machine process parameter.

步驟S310，輸出超過第一預設影響程度指標值的機台製程參數的警示資訊。 Step S310 , outputting warning information of the machine process parameter exceeding the first preset influence degree index value.

在一實施方式中，第一預設影響程度指標值可以根據實際需求進行設定與調整。比如第一預設影響程度指標值設定為1.0，則在表3中，可以輸出機台製程參數X₂、X₆的警示資訊，以提醒機台用戶及時進行機台維護，減少機台停機時間。比如，可以在異常監測裝置100的顯示裝置輸出機台製程參數X₂、X₆的警示資訊。 In one embodiment, the first preset influence degree index value can be set and adjusted according to actual needs. For example, if the first preset influence degree index value is set to 1.0, in Table 3, the warning information of the machine process parameters X ₂ and X ₆ can be output to remind the machine user to carry out the machine maintenance in time and reduce the machine downtime. . For example, the warning information of the machine process parameters X ₂ and X ₆ may be output on the display device of the abnormality monitoring device 100 .

在一實施方式中，可以採用多種預設換算方式對所述問題指標集中每一所述機台製程參數出現的次數及每一所述機台製程參數的參數係數進行換算，得到每一所述機台製程參數的多個影響程度指標值，當每一機台製程參數存在多個影響程度指標值超過第一預設影響程度指標值時，可以根據該多個超過第一預設影響程度指標值的影響程度指標值輸出一聚合機台異常指標，該聚合機台異常指標可用於描述每一所述機台製程參數的異常程度，以便於機台用戶瞭解異常原因和程度。比如，聚合機台異常指標指示了機台200的一加工零件導致的異常。比如，多個機台製程參數的聚合機台異常指標均與機台200的加工零件A相關時，該聚合機台異常指標可以是反映加工零件A可能出現異常的資訊。 In one embodiment, a variety of preset conversion methods can be used to convert the number of occurrences of each of the machine process parameters in the problem index set and the parameter coefficients of each of the machine process parameters to obtain each of the Multiple influence degree index values of the process parameters of the machine. When there are multiple influence level index values for each machine process parameter that exceed the first preset influence level index value, the multiple influence level index values that exceed the first preset influence level index can be used according to the The value of the influence degree index value outputs an aggregated machine abnormality index, which can be used to describe the abnormality degree of each of the machine process parameters, so that the machine user can understand the abnormality cause and degree. For example, the aggregation machine abnormality index indicates an abnormality caused by a machined part of the machine 200 . For example, when the aggregation machine abnormality index of multiple machine process parameters is related to the machined part A of the machine 200 , the aggregation machine abnormality index may be information reflecting that the machining part A may be abnormal.

上述機台製程參數的異常監測裝置、方法及電腦可讀存儲介質，利用多點量測的量測資料與機台製程資料構建問題指標集並進行分析，以早日發現有異常的機台製程參數，減少機台停機時間，提升製程品質。 The abnormal monitoring device, method and computer-readable storage medium for the above-mentioned machine process parameters use the measurement data of multi-point measurement and machine process data to construct a problem index set and analyze it, so as to find abnormal machine process parameters as soon as possible. , reduce machine downtime and improve process quality.

綜上所述，本發明符合發明專利要件，爰依法提出專利申請。惟，以上所述者僅為本發明之較佳實施方式，本發明之範圍並不以上述實施方式為限，舉凡熟悉本案技藝之人士爰依本發明之精神所作之等效修飾或變化，皆應涵蓋於以下申請專利範圍內。 To sum up, the present invention complies with the requirements of an invention patent, and a patent application can be filed in accordance with the law. However, the above descriptions are only the preferred embodiments of the present invention, and the scope of the present invention is not limited to the above-mentioned embodiments, and equivalent modifications or changes made by those who are familiar with the art of the present invention according to the spirit of the present invention are all applicable. Should be covered within the scope of the following patent applications.

S300、S302、S304、S306、S308、S310:步驟 S300, S302, S304, S306, S308, S310: Steps

Claims (10)

一種機台製程參數的異常監測方法，多個所述機台製程參數用於加工多個預設產品，所述預設產品設置有第一量測點至第N量測點，N個所述量測點用於量測所述預設產品的同一產品參數，N為大於1的正整數，所述異常監測方法包括：基於預先提取的多筆產品資料集篩選得到影響所述預設產品在所述第一量測點的量測值的多個目的機台製程參數，其中，多個所述目的機台製程參數為多個所述機台製程參數中的部分參數或全部參數，多筆所述產品資料集對應多個所述預設產品，每筆所述產品資料集包括多個所述機台製程參數及對應的預設產品在所述第一量測點的量測值；基於多個所述目的機台製程參數與多個所述預設產品在所述第一量測點的量測值構建與所述第一量測點對應的量測值預測模型，以根據所述量測值預測模型預測得到的所述第一量測點的預測值計算所述量測值預測模型的擬合度，其中，所述量測值預測模型根據每一所述目的機台製程參數及每一所述目的機台製程參數對應的參數係數計算得到所述第一量測點的預測值；將與所述第一量測點對應的量測值預測模型的擬合度、所述第一量測點的預測值，與所述第一量測點對應的目的機台製程參數及與所述第一量測點對應的目的機台製程參數的參數係數彙整至問題指標集；重複上述步驟直至將與所述第N量測點對應的量測值預測模型的擬合度、所述第N量測點的預測值，與所述第N量測點對應的目的機台製程參數及與所述第N量測點對應的目的機台製程參數的參數係數彙整至所述問題指標集；基於所述問題指標集中的元素計算得到每一所述機台製程參數的影響程度指標值；及輸出超過第一預設影響程度指標值的機台製程參數的警示資訊。 A method for monitoring abnormality of machine process parameters, wherein a plurality of the machine process parameters are used to process a plurality of preset products, and the preset products are provided with a first measurement point to an Nth measurement point, and N said The measurement point is used to measure the same product parameter of the preset product, N is a positive integer greater than 1, and the abnormality monitoring method includes: filtering based on a plurality of pre-extracted product data sets to obtain an impact on the preset product in the A plurality of target machine process parameters of the measurement value of the first measurement point, wherein the plurality of the target machine process parameters are some or all parameters of the plurality of the machine process parameters, and multiple The product data set corresponds to a plurality of the preset products, and each of the product data sets includes a plurality of the machine process parameters and the measurement values of the corresponding preset products at the first measurement point; based on A plurality of the target machine process parameters and a plurality of measurement values of the preset product at the first measurement point construct a measurement value prediction model corresponding to the first measurement point, so as to The prediction value of the first measurement point obtained by the measurement value prediction model is used to calculate the degree of fit of the measurement value prediction model, wherein the measurement value prediction model is based on the process parameters of each target machine. and the parameter coefficient corresponding to each of the target machine process parameters to obtain the predicted value of the first measurement point; The predicted value of the first measurement point, the process parameters of the target machine corresponding to the first measurement point and the parameter coefficients of the process parameters of the target machine corresponding to the first measurement point are assembled into the problem index set; repeat The above steps are until the measurement value corresponding to the Nth measurement point is predicted to fit the model, the predicted value of the Nth measurement point, and the target machine process parameters corresponding to the Nth measurement point. And the parameter coefficients of the target machine process parameters corresponding to the Nth measurement point are compiled into the problem index set; the influence degree index value of each of the machine process parameters is calculated based on the elements in the problem index set. ; and outputting warning information of the process parameters of the machine tool exceeding the first preset influence level index value. 如請求項1所述之機台製程參數的異常監測方法，還包括：收集加工所述預設產品的機台加工參數及所述預設產品的量測參數；從收集到的所述機台加工參數及所述量測參數提取指定資料存儲至分析資料庫，其中所述分析資料庫至少包括第一資料表、第二資料表及第三資料表，所述第一資料表用於存儲所述指定資料中的多個所述機台製程參數，所述第二資料表用於存儲所述指定資料中的N個所述量測點的量測值，所述第三資料表用於存儲多個所述機台製程參數與每個所述量測點的量測值的映射關係；從所述分析資料庫中提取多筆所述產品資料集。 The abnormal monitoring method for machine process parameters according to claim 1, further comprising: collecting machine processing parameters for processing the preset product and measurement parameters of the preset product; The specified data extracted from the processing parameters and the measurement parameters are stored in an analysis database, wherein the analysis database at least includes a first data table, a second data table and a third data table, and the first data table is used for storing all the data. a plurality of the machine process parameters in the specified data, the second data table is used to store the measurement values of the N measurement points in the specified data, and the third data table is used to store A mapping relationship between a plurality of the machine process parameters and the measurement value of each of the measurement points; and extracting a plurality of the product data sets from the analysis database. 如請求項2所述之機台製程參數的異常監測方法，其中所述基於收集到的所述機台加工參數及所述量測參數構建分析資料庫的步驟，包括：利用預設ETL工具從收集到的所述機台加工參數及所述量測參數中提取多個所述機台製程參數及N個所述量測點的量測值，並載入至所述分析資料庫。 The abnormality monitoring method for machine process parameters according to claim 2, wherein the step of constructing an analysis database based on the collected machine tool process parameters and the measurement parameters includes: using a preset ETL tool from A plurality of the machine process parameters and the measurement values of the N measurement points are extracted from the collected machine processing parameters and the measurement parameters, and loaded into the analysis database. 如請求項1所述之機台製程參數的異常監測方法，其中所述基於多個所述目的機台製程參數與多個所述預設產品在所述第一量測點的量測值構建與所述第一量測點對應的量測值預測模型的步驟之前，包括：若所述第一量測點的量測值包括多個維度值，利用預設降維函數將所述第一量測點的量測值映射為一維的數值。 The abnormality monitoring method for machine process parameters as claimed in claim 1, wherein the method is constructed based on a plurality of the target machine process parameters and a plurality of measurement values of the predetermined product at the first measurement point Before the step of predicting the measurement value corresponding to the first measurement point, the step includes: if the measurement value of the first measurement point includes a plurality of dimension values, using a preset dimension reduction function to The measurement value of the measurement point is mapped to a one-dimensional value. 如請求項1所述之機台製程參數的異常監測方法，其中所述量測值預測模型為線性模型。 The abnormality monitoring method for machine process parameters according to claim 1, wherein the measurement value prediction model is a linear model. 如請求項5所述之機台製程參數的異常監測方法，其中所述線性模型包括多個線性係數，多個所述目的機台製程參數的參數係數與多個所述線性係數一一對應。 The abnormality monitoring method for machine process parameters according to claim 5, wherein the linear model includes a plurality of linear coefficients, and the parameter coefficients of the plurality of the target machine process parameters are in one-to-one correspondence with the plurality of the linear coefficients . 如請求項1所述之機台製程參數的異常監測方法，其中所述基於所述問題指標集中的元素計算得到每一所述機台製程參數的影響程度指標值 的步驟，包括：基於所述問題指標集中每一所述機台製程參數出現的次數及每一所述機台製程參數的參數係數換算得到每一所述機台製程參數的影響程度指標值。 The method for monitoring abnormality of machine process parameters according to claim 1, wherein the index value of the influence degree of each of the machine process parameters is obtained by calculating based on the elements in the problem index set The step includes: converting the index value of the influence degree of each of the machine process parameters based on the number of occurrences of each of the machine process parameters and the parameter coefficient of each of the machine process parameters in the problem index set. 如請求項7所述之機台製程參數的異常監測方法，其中所述基於所述問題指標集中每一所述機台製程參數出現的次數及每一所述機台製程參數的參數係數換算得到每一所述機台製程參數的影響程度指標值的步驟，包括：採用多種預設換算方式對所述問題指標集中每一所述機台製程參數出現的次數及每一所述機台製程參數的參數係數進行換算，得到每一所述機台製程參數的多個影響程度指標值；當每一所述機台製程參數存在多個超過所述第一預設影響程度指標值的影響程度指標值時，基於多個超過所述第一預設影響程度指標值的影響程度指標值得到一聚合機台異常指標，以描述每一所述機台製程參數的異常程度。 The method for monitoring abnormality of machine process parameters according to claim 7, wherein the number of occurrences of each of the machine process parameters in the problem index set and the parameter coefficient of each of the machine process parameters are converted to obtain The step of the index value of the degree of influence of each of the machine process parameters includes: using a variety of preset conversion methods to set the number of occurrences of each of the machine process parameters and each of the machine process parameters in the problem index set. The parameter coefficients are converted to obtain a plurality of influence degree index values of each of the machine process parameters; when each of the machine process parameters has a plurality of influence degree indexes that exceed the first preset influence degree index value When the value is , an aggregated machine abnormality index is obtained based on a plurality of influence level index values exceeding the first preset influence level index value to describe the abnormality level of each of the machine process parameters. 一種機台製程參數的異常監測裝置，所述裝置包括處理器及記憶體，所述記憶體上存儲有若干電腦程式，所述處理器用於執行記憶體中存儲的電腦程式時實現如請求項1至8中任一項所述的機台製程參數的異常監測方法的步驟。 An abnormality monitoring device for machine process parameters, the device includes a processor and a memory, the memory stores a number of computer programs, and the processor is used to execute the computer program stored in the memory. Steps of the abnormal monitoring method for machine process parameters described in any one of to 8. 一種電腦可讀存儲介質，所述電腦可讀存儲介質存儲有多條指令，多條所述指令可被一個或者多個處理器執行，以實現如請求項1至8中任一項所述的機台製程參數的異常監測方法的步驟。 A computer-readable storage medium, the computer-readable storage medium stores a plurality of instructions, and a plurality of the instructions can be executed by one or more processors, so as to realize any one of the claim items 1 to 8. The steps of the abnormal monitoring method for the process parameters of the machine tool.

Images