AMC threshold dynamic adjustment method and system

Abstract

The present invention is applicable to the field of threshold adjustment technology, and in particular, relates to a method and system for dynamically adjusting an AMC threshold. The method comprises: S100: collecting historical measurement values ​​of AMC components, traversing influencing factors, wherein the influencing factors at least include: AMC residue, production task and temperature, using a preset detection device, obtaining historical readings of the influencing factors, and determining the deviation weight of each influencing factor, and deviating from the historical measurement value to calculate a predicted value; S200: constructing a number block corresponding to the influencing factor one by one, migrating the deviation weight to the corresponding number block, and recording the real-time measurement value of the AMC component. The present invention can preliminarily determine the AMC component threshold by using the historical threshold, and can dynamically adjust the threshold, optimize resource utilization, and verify the change number block by using a secondary verification mechanism to avoid false alarms, further improve the accuracy of the dynamic threshold, and also greatly improve the user experience.

Description

Technical Field

[0001] The present invention relates to the technical field of pollutant detection, and in particular to an AMC threshold dynamic adjustment method and system. Background Art

[0002] AMC refers to the contaminated gas mixture that affects wafer processing during the semiconductor production process. The sources of AMC in the semiconductor production workshop are divided into external sources and internal sources. The external sources mainly come from the air purification and filtration system supplied to the clean room, while the internal sources may come from multiple sources, including process residues, use of cleaning solvents, escape of materials in the clean room, and equipment maintenance.

[0003] AMC will alarm only when it reaches a certain fixed concentration (threshold). However, in actual applications, there may be influences from adjacent detection behaviors. For example, if there is residual AMC in the equipment from the previous time, the threshold needs to be adjusted higher to prevent false alarms. Therefore, "how to use historical data to determine the dynamic threshold and reduce the probability of false alarms" is the technical problem that the present invention needs to solve. Summary of the invention

[0004] The object of the present invention is to provide a method and system for dynamically adjusting an AMC threshold, so as to solve the problem of "how to determine a dynamic threshold and reduce the probability of false alarms" raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A method for dynamically adjusting an AMC threshold, the method comprising:

[0007] S100: Collect historical measurement values ​​of AMC components, traverse influencing factors, and the influencing factors at least include: AMC component residue, production task and temperature, use preset detection equipment to obtain historical readings of the influencing factors, and determine the deviation weight of each influencing factor, and deviate from the historical measurement value to calculate the predicted value;

[0008] S200: constructing blocks corresponding to the influencing factors one by one, migrating the deviation weights to the corresponding blocks, recording the real-time measurement values ​​of the AMC components, comparing the difference between the real-time measurement values ​​and the predicted values, and using the difference to index the hidden items and superimpose them into the influencing factors;

[0009] S300: Integrate all digital blocks to generate a digital block chain, insert a head node and a tail node into both ends of the digital block chain respectively, synchronize the historical measurement value to the head node, synchronize the real-time measurement value to the tail node, and embed an adjustment mechanism into the digital block chain;

[0010] S400: trace back the historical threshold corresponding to the historical measurement value, transfer the historical threshold to the head node, determine the real-time threshold through the block chain and the deviation weight, and use the real-time reading of the detection device to correct the real-time threshold, compare the historical threshold and the real-time threshold, traverse the variable block, and use the preset secondary verification mechanism to verify the variable block. If the verification passes, define the real-time threshold as the final threshold. If the verification fails, activate the preset communication link, transfer the backup value in the detection device to the variable block, and recalculate the predicted value and the real-time threshold.

[0011] Furthermore, the S100 includes:

[0012] Divide the AMC component into a number of individual items, create a number of blockchains equal to the number of the individual items, and configure a mapping relationship;

[0013] Traversing the influencing factors of each of the single items, generating labels, and inserting the labels into the corresponding digital block chain;

[0014] A timestamp is added to the historical measurement value, and the reading of the detection device at the same time as the timestamp is defined as the historical reading.

[0015] Furthermore, the S100 further includes:

[0016] Establish a weight comparison table, wherein the weight comparison table includes: real-time reading items and weight items, and each of the number blocks corresponds to a weight comparison table;

[0017] The real-time reading is used to query the weight comparison table, traverse the weight items of each of the blocks, and offset the historical measurement value to obtain the predicted value.

[0018] Further, the S200 includes:

[0019] Using the preset processing nodes, a distributed processing architecture is established to process all the items in parallel and determine the predicted value of each item;

[0020] With time as the horizontal axis and the predicted value and real-time measurement value as the vertical axis, a control curve and a change curve are drawn, and the critical line is marked.

[0021] Further, the S300 includes:

[0022] Inserting a pointer into the digital block chain, configuring the processing flow of the historical measurement value, and defining the head node as the input interface and the tail node as the output interface;

[0023] The adjustment mechanism is triggered, wherein the adjustment mechanism is: crossing the processing flow, traversing all the blocks in sequence, fine-tuning the deviation weights in the blocks, and converging the predicted values ​​to the real-time measured values.

[0024] Furthermore, the S400 includes:

[0025] Input the historical threshold into the head node, and output the real-time threshold through the blockchain;

[0026] The historical readings and the real-time readings are compared to determine the fluctuation value of each block, and the blocks whose fluctuation values ​​exceed a preset threshold are defined as changing blocks.

[0027] Furthermore, the S400 includes:

[0028] If the change block fails verification, a communication link between the change block and the detection device is established, the communication link is activated, and a backup value is determined using the detection device;

[0029] The reserve value is used to cover the changed block, and the predicted value and the real-time threshold are recalculated.

[0030] Furthermore, the method further comprises:

[0031] Clustering all of the blocks into several categories, and after integration, constructing a scene library, and dynamically updating the scene library using real-time measurement values;

[0032] The AMC components under different categories are modeled using existing simulation programs, scenario models are constructed, and the deviation weights are corrected.

[0033] Furthermore, the system comprises:

[0034] A calculation module is used to collect historical measurement values ​​of AMC components, traverse influencing factors, and the influencing factors include at least: AMC component residue, production task and temperature, use preset detection equipment to obtain historical readings of the influencing factors, and determine the deviation weight of each influencing factor, and deviate from the historical measurement value to calculate the predicted value;

[0035] A superposition module is used to construct a number block corresponding to the influencing factors one by one, transfer the deviation weight to the corresponding number block, record the real-time measurement value of the AMC component, compare the difference between the real-time measurement value and the predicted value, and use the difference to index the hidden item and superimpose it to the influencing factor;

[0036] An embedding module, used to integrate all digital blocks, generate a digital block chain, insert a head node and a tail node into both ends of the digital block chain, synchronize the historical measurement value into the head node, synchronize the real-time measurement value into the tail node, and embed an adjustment mechanism into the digital block chain;

[0037] The transfer module is used to trace back the historical threshold corresponding to the historical measurement value, transfer the historical threshold to the head node, determine the real-time threshold via the block chain and the deviation weight, and use the real-time reading of the detection device to correct the real-time threshold, compare the historical threshold and the real-time threshold, traverse the change block, and use the preset secondary verification mechanism to verify the change block. If the verification passes, the real-time threshold is defined as the final threshold. If the verification fails, the preset communication link is activated, the backup value in the detection device is transferred to the change block, and the predicted value and the real-time threshold are recalculated.

[0038] Furthermore, the calculation module includes:

[0039] A configuration unit, used to divide the AMC component into a number of single items, create a number of block chains equal to the number of the single items, and configure a mapping relationship;

[0040] An insertion unit, used to traverse the influencing factors of each of the single items, generate a label, and insert the label into the corresponding digital block chain;

[0041] A definition unit, configured to add a timestamp to the historical measurement value, and define a reading of the detection device at the same time as the timestamp as a historical reading;

[0042] An establishing unit, used for establishing a weight comparison table, wherein the weight comparison table includes: real-time reading items and weight items, and each of the number blocks corresponds to a weight comparison table;

[0043] The obtaining unit is used to use the real-time reading to query the weight comparison table, traverse the weight items of each of the number blocks, and offset the historical measurement value to obtain the predicted value.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] By collecting historical measurement values ​​of AMC components, the changing trends of AMC components in different time periods can be determined, which is helpful to predict the future concentrations of AMC components and take effective preventive measures. By comparing historical measurement values ​​and predicted values, the deviation weights can be adjusted, which greatly improves the accuracy of the deviation weights. By constructing blocks, the influencing factors can be accurately quantified and better monitored. By using historical thresholds, the AMC component thresholds can be preliminarily determined, and the thresholds can be dynamically adjusted to optimize resource utilization. By using the secondary verification mechanism to verify the changing blocks, the deviation weights can be calibrated to further improve the accuracy of the dynamic thresholds, while also greatly improving the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention.

[0047] Figure 1 A flowchart of a method for dynamically adjusting an AMC threshold provided by an embodiment of the present invention;

[0048] Figure 2 A first sub-flow chart of the AMC threshold dynamic adjustment method provided by an embodiment of the present invention;

[0049] Figure 3 A second sub-flow chart of the AMC threshold dynamic adjustment method provided by an embodiment of the present invention;

[0050] Figure 4 A third sub-flow chart of the AMC threshold dynamic adjustment method provided in an embodiment of the present invention;

[0051] Figure 5 A fourth sub-flow chart of the AMC threshold dynamic adjustment method provided by an embodiment of the present invention;

[0052] Figure 6 A block diagram of the composition of the AMC threshold dynamic adjustment system provided by an embodiment of the present invention;

[0053] Figure 7 A block diagram of the composition of a calculation module in the AMC threshold dynamic adjustment system provided by an embodiment of the present invention;

[0054] Figure 8 A block diagram of the composition of the superposition module in the AMC threshold dynamic adjustment system provided by an embodiment of the present invention;

[0055] Fig. 9 A block diagram of the composition of the embedded module in the AMC threshold dynamic adjustment system provided by an embodiment of the present invention;

[0056] Fig.10 A block diagram of the composition of the transfer module in the AMC threshold dynamic adjustment system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0058] In Example 1, Figure 1 The implementation process of the AMC threshold dynamic adjustment method provided by the embodiment of the present invention is shown and described in detail below:

[0059] S100: Collect historical measurement values ​​of AMC components, traverse influencing factors, and the influencing factors include at least: AMC component residue, production task and temperature. Use preset detection equipment to obtain historical readings of the influencing factors, and determine the deviation weight of each influencing factor, and deviate from the historical measurement value to calculate the predicted value.

[0060] The historical measurement values ​​of AMC components are collected by using the AMC online monitoring system, wherein the AMC online monitoring system is composed of acid analysis equipment, alkali analysis equipment, sulfide analysis equipment, organic matter analysis equipment, multi-point samplers and online monitoring software, etc. The historical measurement value is the concentration data of each component measured by the AMC online monitoring system at a certain moment in the past, that is, the number of historical measurement values ​​is multiple, and the influencing factors of each historical measurement value are found out. The influencing factors are the factors that may cause the historical measurement values ​​to fluctuate, wherein the influencing factors include at least: AMC component residues, production tasks and temperature, etc.; for example, when the production task is increased, while maintaining the original power of the processing equipment, some AMC components in the production workshop may exceed the standard, and the production task is the influencing factor.

[0061] Use the preset detection equipment to read the value of the influencing factor, and define this value as the historical reading, where the preset detection equipment is the temperature sensor, humidity sensor, etc. installed in the production workshop. Determine the deviation weight corresponding to each historical reading, and deviate the historical measurement value to get the predicted value.

[0062] S200: constructing blocks corresponding to the influencing factors one by one, transferring the deviation weights to the corresponding blocks, recording the real-time measurement values ​​of the AMC components, comparing the differences between the real-time measurement values ​​and the predicted values, and using the differences to index hidden items and superimpose them into the influencing factors.

[0063] Each influencing factor corresponds to a number block. The deviation weight is transferred to the corresponding number block, the real-time measurement value of the AMC component is recorded, and the difference between the real-time measurement value and the predicted value is found. Using the difference, continue to traverse the number blocks to find the hidden items, where the hidden items are the factors that affect the AMC components except the number blocks, and the hidden items are superimposed on the influencing factors. The advantage of doing this is that the factors that affect the threshold can be fully located.

[0064] S300: Integrate all digital blocks to generate a digital block chain, insert a head node and a tail node into both ends of the digital block chain respectively, synchronize the historical measurement value to the head node, synchronize the real-time measurement value to the tail node, and embed an adjustment mechanism into the digital block chain.

[0065] Integrate all digital blocks, generate a digital block chain, insert the head node and the tail node into the digital block chain, transfer the historical measurement values ​​to the head node, transfer the real-time measurement values ​​to the tail node, establish the connection between the historical measurement values ​​and the real-time measurement values, and embed the adjustment mechanism into the digital block chain; the adjustment mechanism is the specific method of adjusting the deviation weight in the digital block.

[0066] S400: trace back the historical threshold corresponding to the historical measurement value, transfer the historical threshold to the head node, determine the real-time threshold through the block chain and the deviation weight, and use the real-time reading of the detection device to correct the real-time threshold, compare the historical threshold and the real-time threshold, traverse the variable block, and use the preset secondary verification mechanism to verify the variable block. If the verification passes, define the real-time threshold as the final threshold. If the verification fails, activate the preset communication link, transfer the backup value in the detection device to the variable block, and recalculate the predicted value and the real-time threshold.

[0067] After determining the deviation weight of each block, the historical threshold is transferred to the head node, and all blocks are traversed in turn. The real-time threshold is calculated using the deviation weight in the block. Then, according to the changes in the influencing factors in the production workshop, the influencing factors with large fluctuations are found, and the corresponding blocks are determined as variable blocks. The variable blocks are verified using a secondary verification mechanism. The secondary verification mechanism is to measure the variable blocks again using backup detection equipment. For example, when it is determined that the block corresponding to the AMC component residue is a variable block, the AMC component residue is measured again using a backup detection equipment. If the measurement result is significantly different from the previous real-time reading, the verification fails, the preset communication link is activated, the backup value of the detection equipment is transferred to the variable block, the deviation weight of the block is readjusted, and the predicted value and the real-time threshold are recalculated. The backup value of the detection equipment is the reading of the backup detection equipment.

[0068] If the AMC component residue measured by the backup detection equipment is the same as the previous real-time reading, the real-time threshold is directly determined as the final threshold, and the deviation weight is dynamically adjusted according to the changes in influencing factors.

[0069] In Example 2, Figure 2 The implementation process of the AMC threshold dynamic adjustment method provided by the embodiment of the present invention is shown, and S100 is described in detail as follows:

[0070] S101: Divide the AMC component into a number of single items, create a number of block chains equal to the number of the single items, and configure a mapping relationship.

[0071] The AMC components are divided into several single items. In actual production, AMC is generally composed of NH3+NO+NO2, HCL+HF and VOC components. The NH3+NO+NO2 single item, HCL+HF single item and VOC single item are obtained by dividing, and a block chain is created for each single item. The block chain is a collection of influencing factors that affect the concentration of the single item.

[0072] S102: Traverse out the influencing factors of each of the single items, generate a label, and insert the label into the corresponding block chain.

[0073] Traverse all the factors that affect the concentration of each single item, generate labels, and insert the labels into the digital block chain corresponding to each single item.

[0074] S103: Add a timestamp to the historical measurement value, and define the reading of the detection device at the same time as the timestamp as the historical reading.

[0075] The historical measurement value of each single item is recorded, and a timestamp is added thereto, and the reading of the detection device of the single item influencing factor within the same period of time is determined as the historical reading.

[0076] It should be noted that historical measurement values, real-time measurement values ​​and predicted values ​​are all relative to a single item, that is, these measurement values ​​show the concentration of the corresponding gas in the production workshop, and historical readings and real-time readings are the measurement results of the factors affecting the concentration of these gases; for example, the temperature in the production workshop will have a greater impact on NH3+NO+NO2. At this time, the measurement result of NH3+NO+NO2 is also the historical measurement value or the real-time measurement value, and the measurement result of temperature is also the historical reading or the real-time reading.

[0077] In Example 3, Figure 2 The implementation process of the AMC threshold dynamic adjustment method provided by the embodiment of the present invention is shown, and S100 is further described in detail below, as follows:

[0078] S104: Establish a weight comparison table, wherein the weight comparison table includes: real-time reading items and weight items, and each of the number blocks corresponds to a weight comparison table.

[0079] Although each influencing factor will affect the measurement results of a single item, the degree of their influence on the single item is not the same, and the size of the historical readings of the influencing factor will also affect the measurement results; therefore, it is necessary to determine the weight of each influencing factor and integrate them to construct a weight comparison table, in which each number block should have a weight comparison table.

[0080] For example, temperature, humidity, etc. will affect the concentration of NH3+NO+NO2 in the production workshop. The impact of the two on the concentration of NH3+NO+NO2 is also different, and different temperatures have different effects on the concentration of NH3+NO+NO2; therefore, it is necessary to construct a weight comparison table for temperature and humidity respectively.

[0081] S105: Utilizing the real-time reading, querying the weight comparison table, traversing the weight items of each of the number blocks, and offsetting the historical measurement value to obtain a predicted value.

[0082] After determining the real-time reading, the weight comparison table can be queried to determine the weight item of each block. Using the historical measurement values ​​and weight items, the concentration of each AMC component in the production workshop at the preset time can be calculated.

[0083] In Example 4, Figure 3 The implementation process of the AMC threshold dynamic adjustment method provided by the embodiment of the present invention is shown, and S200 is described in detail below, as follows:

[0084] S201: using preset processing nodes to form a distributed processing architecture, processing all the items in parallel, and determining a predicted value for each item;

[0085] A processing node is a terminal that can process data. By using processing nodes to build a distributed processing structure and calculating the predicted value of each single item, the predicted value of the AMC component can be determined.

[0086] S202: With time as the horizontal axis and the predicted value and the real-time measured value as the vertical axis, a control curve and a change curve are drawn, and a critical line is marked.

[0087] A control curve is drawn with time as the horizontal axis and the predicted value as the vertical axis, and a change curve is drawn with the real-time measurement value as the vertical axis, and a critical line is marked in it. The advantage of this is that the concentration changes of each component in AMC can be intuitively displayed through the change curve, and the control curve and critical line can be used to provide decision support for users, such as increasing the power of processing equipment, reducing production intensity, etc.

[0088] In Example 5, Figure 4 The implementation process of the AMC threshold dynamic adjustment method provided by the embodiment of the present invention is shown, and S300 is described in detail below, as follows:

[0089] S301: inserting a pointer into the digital block chain, configuring a processing flow of the historical measurement value, and defining the head node as an input interface and the tail node as an output interface.

[0090] Insert a direction into the block. By setting the direction, the influencing factors can be integrated to avoid data disorder. Determine the processing flow of the historical measurement values. For example, the processing flow can be, according to the direction, using the deviation weights in the block to adjust the historical measurement values ​​in turn; determine the head node and the tail node in the block chain as the input interface and the output interface respectively.

[0091] S302: triggering the adjustment mechanism, wherein the adjustment mechanism is: crossing the processing flow, traversing all the number blocks in sequence, fine-tuning the deviation weights in the number blocks, and converging the predicted value to the real-time measurement value.

[0092] After adjusting the historical measurement values ​​using all the blocks, the predicted values ​​are obtained, and the predicted values ​​are compared with the real-time measurement values. Then, based on the comparison results, the deviation weights in the blocks are fine-tuned to make the predicted values ​​the same as the real-time measurement values.

[0093] In Example 6, Figure 5 The implementation process of the AMC threshold dynamic adjustment method provided by the embodiment of the present invention is shown, and S400 is described in detail below, as follows:

[0094] S401: Input the historical threshold into the head node, and output the real-time threshold through the blockchain.

[0095] The historical threshold is input into the head node, the real-time threshold is calculated using the deviation weight in the block, and the real-time threshold is output through the tail node.

[0096] S402: Compare the historical readings with the real-time readings to determine the fluctuation value of each data block, and define the data blocks whose fluctuation values ​​exceed a preset threshold as variable data blocks.

[0097] The fluctuation value of each influencing factor is determined. If the fluctuation value exceeds a preset threshold, the corresponding number block is defined as a variable number block.

[0098] In Example 7, Figure 5 The implementation process of the AMC threshold dynamic adjustment method provided by the embodiment of the present invention is shown, and S400 is further described in detail below, as follows:

[0099] S403: If the variable number block fails verification, a communication link between the variable number block and the detection device is established, the communication link is activated, and the backup value is determined using the detection device. If the verification passes, the real-time threshold is determined as the final threshold.

[0100] Perform secondary verification on the changing number block. If the verification passes, the real-time threshold is determined as the final threshold. The final threshold is dynamically adjusted using the real-time reading of the detection equipment. If the verification fails, the real-time reading of the detection equipment is determined as the backup value.

[0101] S404: Using the backup value to cover the changed block, recalculate the predicted value and the real-time threshold.

[0102] The backup value is used to recalculate the deviation weight of the corresponding block, and the predicted value and the real-time threshold are recalculated using the deviation weight.

[0103] In Example 8, different from Example 1, in this embodiment of the present invention, the method further includes:

[0104] Clustering all of the blocks into several categories, and after integration, constructing a scene library, and dynamically updating the scene library using real-time measurement values;

[0105] The AMC components under different categories are modeled using existing simulation programs, scenario models are constructed, and the deviation weights are corrected.

[0106] Divide the blocks into several categories and build a scenario library, which is a collection of all factors affecting the AMC components. According to the changes in real-time measurement values, a more detailed traversal is performed in the production workshop to determine whether there are other influencing factors, and the scenario library is dynamically updated based on the judgment results. Use the simulation program to model the AMC components and build a scenario model; in other words, input the historical measurement values ​​of the AMC and the real-time readings of the detection equipment into the scenario model, and the predicted value of the AMC can be output; the advantage of this is that it can greatly shorten the calculation process of the predicted value and ensure the timeliness of the predicted value.

[0107] In actual production, AMC components are composed of a variety of polluting gases, and the measurement results of each polluting gas will be affected by multiple influencing factors. That is to say, the concentration of each gas (historical measurement value) and the changes in influencing factors (historical readings) on the previous working day can be used to judge the gas concentration (real-time measurement value) on the current working day. On the other hand, the historical measurement values ​​and real-time measurement values ​​can also be used to determine the degree of influence of each influencing factor (deviation weight), and then the threshold of the previous working day (historical threshold) is used to integrate the deviation weight to determine the threshold of the current working day (real-time threshold); at the same time, the real-time threshold is dynamically adjusted according to the specific changes in the influencing factors (historical readings and real-time readings).

[0108] Figure 6 The structure block diagram of the AMC threshold dynamic adjustment system provided by an embodiment of the present invention is shown. The AMC threshold dynamic adjustment system 1 includes:

[0109] The calculation module 11 is used to collect historical measurement values ​​of AMC components, traverse the influencing factors, and the influencing factors at least include: AMC residue, production task and temperature, use the preset detection equipment to obtain the historical readings of the influencing factors, and determine the deviation weight of each influencing factor, and deviate from the historical measurement value to calculate the predicted value;

[0110] The superposition module 12 is used to construct a number block corresponding to the influencing factors one by one, transfer the deviation weight to the corresponding number block, record the real-time measurement value of the AMC component, compare the difference between the real-time measurement value and the predicted value, and use the difference to index the hidden item and superimpose it to the influencing factor;

[0111] The embedding module 13 is used to integrate all the digital blocks, generate a digital block chain, insert a head node and a tail node into both ends of the digital block chain, synchronize the historical measurement value into the head node, synchronize the real-time measurement value into the tail node, and embed an adjustment mechanism into the digital block chain;

[0112] The transfer module 14 is used to trace back the historical threshold corresponding to the historical measurement value, transfer the historical threshold to the head node, determine the real-time threshold through the block chain and the deviation weight, and use the real-time reading of the detection device to correct the real-time threshold, compare the historical threshold and the real-time threshold, traverse the change block, and use the preset secondary verification mechanism to verify the change block. If the verification passes, the real-time threshold is defined as the final threshold. If the verification fails, the preset communication link is activated, the backup value in the detection device is transferred to the change block, and the predicted value and the real-time threshold are recalculated.

[0113] Figure 7The structure block diagram of the AMC threshold dynamic adjustment system provided by the embodiment of the present invention is shown, and the calculation module 11 includes:

[0114] A configuration unit 111 is used to divide the AMC component into a number of individual items, create a number of block chains equal to the number of the individual items, and configure a mapping relationship;

[0115] An inserting unit 112, used to traverse the influencing factors of each of the single items, generate a label, and insert the label into the corresponding block chain;

[0116] A definition unit 113, configured to add a timestamp to the historical measurement value, and define the reading of the detection device at the same time as the timestamp as the historical reading;

[0117] An establishing unit 114 is used to establish a weight comparison table, wherein the weight comparison table includes: real-time reading items and weight items, and each of the number blocks corresponds to a weight comparison table;

[0118] The obtaining unit 115 is used to use the real-time reading to query the weight comparison table, traverse the weight items of each of the blocks, and offset the historical measurement value to obtain the predicted value.

[0119] Figure 8 The structure block diagram of the AMC threshold dynamic adjustment system provided by the embodiment of the present invention is shown, and the superposition module 12 includes:

[0120] The parallel processing unit 121 is used to use the preset processing nodes to form a distributed processing architecture, process all the single items in parallel, and determine the predicted value of each single item;

[0121] The marking unit 122 is used to draw a control curve and a change curve with time as the horizontal axis and the predicted value and the real-time measured value as the vertical axis, and mark the critical line.

[0122] Fig. 9 The structure block diagram of the AMC threshold dynamic adjustment system provided by the embodiment of the present invention is shown, and the embedded module 13 includes:

[0123] The linking unit 131 is used to insert a pointer into the digital block chain, configure the processing flow of the historical measurement value, and define the head node as an input interface and the tail node as an output interface;

[0124] The fine-tuning unit 132 is used to trigger the adjustment mechanism, wherein the adjustment mechanism is: crossing the processing flow, traversing all the blocks in sequence, fine-tuning the deviation weights in the blocks, and converging the predicted values ​​to the real-time measured values.

[0125] Fig.10The structure block diagram of the AMC threshold dynamic adjustment system provided by the embodiment of the present invention is shown, and the transfer module 14 includes:

[0126] The output unit 141 is used to input the historical threshold into the head node, and output the real-time threshold through the block chain;

[0127] A comparison unit 142, used for comparing the historical readings with the real-time readings, determining the fluctuation value of each data block, and defining the data block whose fluctuation value exceeds a preset threshold as a changing data block;

[0128] A building unit 143 is used to build a communication link between the changing number block and the detection device when the changing number block fails to be verified, activate the communication link, and use the detection device to determine a backup value;

[0129] The covering unit 144 is used to cover the changed block with the backup value and recalculate the predicted value and the real-time threshold.

[0130] The calculation module 11 is mainly used to complete step S100, the superposition module 12 is mainly used to complete step S200, the embedding module 13 is mainly used to complete step S300, and the transfer module 14 is mainly used to complete step S400;

[0131] The configuration unit 111 is mainly used to complete step S101, the insertion unit 112 is mainly used to complete step S102, the definition unit 113 is mainly used to complete step S103, the establishment unit 114 is mainly used to complete step S104, and the acquisition unit 115 is mainly used to complete step S105;

[0132] The parallel processing unit 121 is mainly used to complete step S201, and the marking unit 122 is mainly used to complete step S202;

[0133] The linking unit 131 is mainly used to complete step S301, and the fine-tuning unit 132 is mainly used to complete step S302;

[0134] The output unit 141 is mainly used to complete step S401, the comparison unit 142 is mainly used to complete step S402, the construction unit 143 is mainly used to complete step S403, and the covering unit 144 is mainly used to complete step S404.

[0135] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for dynamically adjusting an AMC threshold, characterized in that: The method comprises: S100: Collect historical measurement values ​​of AMC components, traverse influencing factors, and the influencing factors at least include: AMC component residue, production task and temperature, use preset detection equipment to obtain historical readings of the influencing factors, and determine the deviation weight of each influencing factor, and deviate from the historical measurement value to calculate the predicted value; S200: constructing blocks corresponding to the influencing factors one by one, migrating the deviation weights to the corresponding blocks, recording the real-time measurement values ​​of the AMC components, comparing the difference between the real-time measurement values ​​and the predicted values, and using the difference to index the hidden items and superimpose them into the influencing factors; S300: Integrate all digital blocks to generate a digital block chain, insert a head node and a tail node into both ends of the digital block chain respectively, synchronize the historical measurement value to the head node, synchronize the real-time measurement value to the tail node, and embed an adjustment mechanism into the digital block chain; The S300 includes: Inserting a pointer into the digital block chain, configuring the processing flow of the historical measurement value, and defining the head node as the input interface and the tail node as the output interface; Triggering the adjustment mechanism, wherein the adjustment mechanism comprises: crossing the processing flow, sequentially traversing all the number blocks, fine-tuning the deviation weights in the number blocks, and converging the predicted value to the real-time measurement value; S400: trace back the historical threshold corresponding to the historical measurement value, transfer the historical threshold to the head node, determine the real-time threshold through the block chain and the deviation weight, and use the real-time reading of the detection device to correct the real-time threshold, compare the historical threshold and the real-time threshold, traverse the change block, and use the preset secondary verification mechanism to verify the change block. If the verification passes, define the real-time threshold as the final threshold. If the verification fails, activate the preset communication link, transfer the spare value in the detection device to the change block, and recalculate the predicted value and the real-time threshold; The S100 further includes: Establish a weight comparison table, wherein the weight comparison table includes: real-time reading items and weight items, and each of the number blocks corresponds to a weight comparison table; The real-time reading is used to query the weight comparison table, traverse the weight items of each of the blocks, and offset the historical measurement value to obtain the predicted value.

Images