CN117172425A

CN117172425A - Intelligent monitoring-based ceramic riving knife production system operation analysis method and system

Info

Publication number: CN117172425A
Application number: CN202311430731.5A
Authority: CN
Inventors: 张俊堂; 庞吉宏
Original assignee: Suzhou Xinhe Semiconductor Materials Co ltd
Current assignee: Suzhou Xinhe Semiconductor Materials Co ltd
Priority date: 2023-10-31
Filing date: 2023-10-31
Publication date: 2023-12-05

Abstract

The invention provides an operation analysis method and an operation analysis system for a ceramic riving knife production system based on intelligent monitoring, and relates to the technical field of artificial intelligence. In the invention, the target upstream data corresponding to the target ceramic riving knife production system is extracted; loading the target upstream data into an equipment operation analysis network obtained by at least secondary network optimization, and analyzing downstream response parameters corresponding to each preset first downstream data in a plurality of preset first downstream data; marking preset first downstream data corresponding to the downstream response parameter with the maximum value, so that the preset first downstream data is marked as target downstream data corresponding to target upstream data; and determining the operation state of the target ceramic riving knife production system based on the predicted operation process reflected by the target downstream data and the actual operation process corresponding to the operation process. Based on the above, the efficiency of the operation state analysis can be improved to some extent.

Description

Intelligent monitoring-based ceramic riving knife production system operation analysis method and system

Technical Field

The invention relates to the technical field of artificial intelligence, in particular to an operation analysis method and an operation analysis system of a ceramic riving knife production system based on intelligent monitoring.

Background

The running state of the ceramic riving knife production system directly influences the excellent performance degree of the produced ceramic riving knife. Therefore, there is a need to monitor the operating conditions of a ceramic riving knife production system. However, in the prior art, when the operation state analysis is performed based on the operation data obtained by the monitoring, there is a problem in that the efficiency of the operation state analysis is relatively low.

Disclosure of Invention

Therefore, the invention aims to provide an operation analysis method and an operation analysis system for a ceramic riving knife production system based on intelligent monitoring so as to improve the efficiency of operation state analysis to a certain extent.

In order to achieve the above purpose, the embodiment of the present invention adopts the following technical scheme:

an operation analysis method of a ceramic riving knife production system based on intelligent monitoring comprises the following steps:

extracting target upstream data corresponding to a target ceramic riving knife production system, wherein the target upstream data belongs to an equipment operation log of upstream production equipment in the target ceramic riving knife production system;

loading the target upstream data into an equipment operation analysis network obtained by at least twice network optimization, and analyzing downstream response parameters corresponding to each preset first downstream data in a plurality of preset first downstream data, wherein the downstream response parameters corresponding to the preset first downstream data are used for reflecting the possibility that the operation process corresponding to the preset first downstream data occurs after the operation process corresponding to the target upstream data occurs;

Marking preset first downstream data corresponding to the downstream response parameter with the maximum value, so that the preset first downstream data is marked as target downstream data corresponding to the target upstream data;

and determining the operation state of the target ceramic riving knife production system based on the predicted operation process reflected by the target downstream data and the actual operation process corresponding to the operation process, wherein the degree of the excellent operation state is negatively related to the distinguishing information between the predicted operation process reflected by the target downstream data and the actual operation process corresponding to the predicted operation process.

In some preferred embodiments, in the above method for analyzing operation of a ceramic riving knife production system based on intelligent monitoring, the step of loading the target upstream data into an equipment operation analysis network obtained by at least two times of network optimization to analyze a downstream response parameter corresponding to each preset first downstream data in the plurality of preset first downstream data includes:

loading the target upstream data to a device operation analysis network obtained by at least secondary network optimization;

performing feature space mapping operation on the target upstream data to form upstream data feature distribution corresponding to the target upstream data, wherein the upstream data feature distribution comprises local upstream data feature distribution corresponding to each data segment in the target upstream data, and the local upstream data feature distribution is used for reflecting segment semantic information and segment distribution information of the corresponding data segment;

For each local upstream data characteristic distribution in the upstream data characteristic distribution, taking the local upstream data characteristic distribution as a target upstream data characteristic distribution, and determining relevant upstream data characteristic distribution corresponding to the target upstream data characteristic distribution in other local upstream data characteristic distribution, wherein the number of the relevant upstream data characteristic distribution is a target number;

sorting a target number of related upstream data feature distributions based on a degree of correlation with the target upstream data feature distribution to form a corresponding feature distribution sequence in which the degree of correlation corresponding to each of the related upstream data feature distributions has a distribution relationship from small to large;

extracting a first relevant upstream data characteristic distribution from the characteristic distribution sequence by using a first characteristic processing unit in a target number of characteristic processing units, carrying out association analysis operation on the target upstream data characteristic distribution based on the first relevant upstream data characteristic distribution, and then carrying out aggregation operation on the result of the association analysis operation and the first relevant upstream data characteristic distribution to form an aggregation characteristic distribution corresponding to the characteristic processing unit, wherein the target number of characteristic processing units are connected successively;

Extracting relevant upstream data characteristic distribution of a corresponding distribution position from the characteristic distribution sequence by utilizing the characteristic processing units for each characteristic processing unit except the first characteristic processing unit in the target number of characteristic processing units, carrying out association analysis operation on the aggregation characteristic distribution corresponding to the previous characteristic processing unit based on the relevant upstream data characteristic distribution, and carrying out aggregation operation on the result of the association analysis operation and the relevant upstream data characteristic distribution to form the aggregation characteristic distribution corresponding to the characteristic processing unit;

and performing aggregation operation on the aggregate characteristic distribution corresponding to the last characteristic processing unit and the target upstream data characteristic distribution, outputting aggregate upstream data characteristic distribution corresponding to the target upstream data characteristic distribution, merging aggregate upstream data characteristic distribution corresponding to each local upstream data characteristic distribution to form optimized upstream data characteristic distribution corresponding to the target upstream data, and analyzing downstream response parameters corresponding to each preset first downstream data in the plurality of preset first downstream data according to the optimized upstream data characteristic distribution.

In some preferred embodiments, in the above method for analyzing operation of a ceramic riving knife production system based on intelligent monitoring, the method for analyzing operation of a ceramic riving knife production system further comprises:

extracting a plurality of first example production operation information, wherein the first example production operation information is formed based on operation monitoring operation on an example ceramic riving knife production system, the first example production operation information comprises first example upstream data and first example downstream data, the first example upstream data is used for reflecting an operation process of upstream production equipment of the example ceramic riving knife production system, the first example downstream data is used for reflecting an operation process of downstream production equipment of the example ceramic riving knife production system, and the upstream production equipment and the downstream production equipment refer to two equipment with relevance and precedence in production;

determining a candidate equipment operation analysis network, wherein the equipment operation analysis network is used for analyzing downstream response parameters corresponding to a plurality of preset first downstream data based on any one first upstream data;

network optimizing the plant operation analysis network based on first example upstream data and first example downstream data included in the plurality of first example production operation information;

Extracting a plurality of second example production operation information, wherein the second example production operation information is formed by performing operation monitoring operation on the example ceramic riving knife production system, and comprises second example upstream data and second example downstream data;

and performing secondary network optimization on the equipment operation analysis network obtained by performing network optimization based on the plurality of second example production operation information and target response parameters corresponding to the plurality of second example production operation information, wherein the target response parameters corresponding to the second example production operation information are determined by performing analysis operation on second example upstream data and second example downstream data in the second example production operation information based on a determination rule using target parameters, and the target response parameters are used for reflecting the possibility that the operation process corresponding to the second example downstream data reappears after the operation process corresponding to the second example upstream data analyzed by using the target parameter determination rule appears.

In some preferred embodiments, in the above method for analyzing operation of a ceramic riving knife production system based on intelligent monitoring, the step of optimizing the device operation analysis network based on the first example upstream data and the first example downstream data included in the plurality of first example production operation information includes:

For each first example production operation information, loading first example upstream data in the first example production operation information to enable the first example upstream data to be loaded into the equipment operation analysis network, and analyzing downstream response parameters corresponding to the plurality of preset first downstream data;

marking preset first downstream data with a maximum value, so that the preset first downstream data is marked as matching first downstream data corresponding to the first example upstream data;

and performing network optimization operation on the equipment operation analysis network based on the distinguishing information between the matched first downstream data and the first example downstream data in the first example production operation information, so that the degree of distinction between new matched first downstream data analyzed by the equipment operation analysis network based on the first example upstream data and the first example downstream data is reduced after the network optimization operation.

In some preferred embodiments, in the above method for analyzing operation of a ceramic riving knife production system based on intelligent monitoring, the step of performing a network optimization operation on the device operation analysis network based on the difference information between the matched first downstream data and the first example downstream data in the first example production operation information, so that the degree of distinction between the new matched first downstream data and the first example downstream data analyzed by the device operation analysis network based on the first example upstream data after the network optimization operation is reduced includes:

Sequentially performing the cyclic network optimization operation on the equipment operation analysis network based on the distinguishing information between the matched first downstream data corresponding to each of the plurality of first example production operation information and the first example downstream data in the first example production operation information, and stopping continuing the cyclic network optimization operation when the number of the performed network optimization operations is greater than or equal to the preset first optimization number; or alternatively

And sequentially performing the cyclic network optimization operation on the equipment operation analysis network based on the distinguishing information between the matched first downstream data corresponding to each of the plurality of first example production operation information and the first example downstream data in the first example production operation information, and stopping continuing the cyclic network optimization operation when the preset first downstream data with the maximum response parameter analyzed by the equipment operation analysis network based on any one of the first example upstream data and the first example downstream data corresponding to any one of the first example upstream data have the distinguishing degree smaller than or equal to the first distinguishing degree determined in advance.

In some preferred embodiments, in the above method for analyzing operation of a ceramic riving knife production system based on intelligent monitoring, the step of performing secondary network optimization on the device operation analysis network obtained by performing network optimization based on the plurality of second example production operation information and the target response parameters corresponding to the plurality of second example production operation information includes:

Loading second example upstream data in the second example production operation information aiming at each second example production operation information, so that the second example upstream data in the second example production operation information is loaded into the equipment operation analysis network obtained by network optimization based on the plurality of first example production operation information, and the downstream response parameters corresponding to the preset first downstream data in the plurality of preset first downstream data are analyzed;

extracting a plurality of preferred first downstream data from the plurality of preset first downstream data based on the downstream response parameters corresponding to the plurality of preset first downstream data, wherein the downstream response parameters corresponding to each of the plurality of preferred first downstream data exceeds the downstream response parameters corresponding to each of the other preset first downstream data in the plurality of preset first downstream data;

based on the target parameter determining rule, respectively analyzing the second example upstream data and each preferred first downstream data, outputting a rule analysis output parameter corresponding to each preferred first downstream data, and marking a rule analysis output parameter with the maximum value in the rule analysis output parameters corresponding to the preferred first downstream data so as to be marked as a corresponding target rule analysis output parameter;

Based on the target parameter determining rule, analyzing the second example upstream data and the second example downstream data in the second example production operation information, and outputting a target response parameter corresponding to the second example production operation information;

and carrying out secondary network optimization operation on the equipment operation analysis network based on the distinguishing information between the target rule analysis output parameters and the target response parameters, so that the degree of distinction between the target rule analysis output parameters and the target response parameters, which are analyzed based on the second example upstream data, of the equipment operation analysis network after network optimization is reduced.

In some preferred embodiments, in the above method for analyzing the operation of a ceramic riving knife production system based on intelligent monitoring, the step of performing a secondary network optimization operation on the device operation analysis network based on the distinguishing information between the target rule analysis output parameter and the target response parameter, so that the device operation analysis network after network optimization has a reduced degree of distinction between the target rule analysis output parameter and the target response parameter analyzed based on the upstream data of the second example, includes:

Analyzing distinguishing information between output parameters and target response parameters based on the target rules corresponding to the second example production operation information respectively, sequentially performing cyclic network optimization operations on the equipment operation analysis network, and stopping continuing the cyclic network optimization operations when the number of the performed network optimization operations is greater than or equal to a second preset optimization number; or,

and sequentially performing the cyclic network optimization operation on the equipment operation analysis network based on the distinguishing information between the target rule analysis output parameters and the target response parameters corresponding to the second example production operation information, and stopping continuing the cyclic network optimization operation when the distinguishing degree between the target rule analysis output parameters analyzed by the equipment operation analysis network based on any one second example production operation information and the target response parameters corresponding to the any one second example production operation information is smaller than or equal to a second predetermined distinguishing degree.

In some preferred embodiments, in the above method for analyzing operation of a ceramic riving knife production system based on intelligent monitoring, the step of extracting a plurality of first example production operation information includes:

Performing operation monitoring operation on the ceramic riving knife production system to form a plurality of first example upstream data and first example downstream data of each first example upstream data;

determining an example upstream data amount corresponding to each of the plurality of first example downstream data;

expanding first example upstream data corresponding to any one first example downstream data such that the example upstream data corresponding to any one first example downstream data is greater than a standard data amount when the example upstream data amount corresponding to the any one first example downstream data is less than or equal to the standard data amount configured in advance; or alternatively

Calculating a multiplication value between a preconfigured corresponding adjustment parameter and a target upstream data amount to form a corresponding preset data amount, and expanding first example upstream data corresponding to any one first example downstream data under the condition that the example upstream data amount corresponding to the any one first example downstream data is smaller than or equal to the preset data amount, so that the example upstream data amount corresponding to the any one first example downstream data is larger than the preset data amount, wherein the target upstream data amount is used for reflecting the total data amount of the plurality of first example upstream data.

The embodiment of the invention also provides an operation analysis method of the ceramic riving knife production system based on intelligent monitoring, which comprises the following steps:

extracting target upstream data corresponding to a target ceramic riving knife production system;

loading the target upstream data into an equipment operation analysis network obtained by secondary network optimization, and analyzing downstream response parameters corresponding to each preset first downstream data in a plurality of preset first downstream data;

determining a plurality of preferred first downstream data from the plurality of preset first downstream data based on a downstream response parameter corresponding to each of the plurality of preset first downstream data, wherein the downstream response parameter corresponding to each of the plurality of preferred first downstream data exceeds the downstream response parameter corresponding to each of the other preferred first downstream data in the plurality of preset first downstream data;

based on a target parameter determining rule, analyzing the target upstream data and each of the preferred first downstream data, outputting a rule analysis output parameter corresponding to each of the preferred first downstream data, and marking the preferred first downstream data corresponding to the rule analysis output parameter with the maximum value among the analyzed rule analysis output parameters so as to be marked as target downstream data corresponding to the target upstream data;

The embodiment of the invention also provides an operation analysis system of the ceramic riving knife production system based on intelligent monitoring, which comprises a processor and a memory, wherein the memory is used for storing a computer program, and the processor is used for executing the computer program so as to realize the operation analysis method of the ceramic riving knife production system based on intelligent monitoring.

The method and the system for analyzing the operation of the ceramic riving knife production system based on intelligent monitoring can firstly extract the target upstream data corresponding to the target ceramic riving knife production system; loading the target upstream data into an equipment operation analysis network obtained by at least secondary network optimization, and analyzing downstream response parameters corresponding to each preset first downstream data in a plurality of preset first downstream data; marking preset first downstream data corresponding to the downstream response parameter with the maximum value, so that the preset first downstream data is marked as target downstream data corresponding to target upstream data; and determining the operation state of the target ceramic riving knife production system based on the predicted operation process reflected by the target downstream data and the actual operation process corresponding to the operation process. Based on the foregoing, since the target downstream data can be predicted first, the operation state of the target ceramic riving knife production system can be determined based on the predicted operation process reflected by the target downstream data and the actual operation process corresponding to the predicted operation process, so that, compared with the conventional technical scheme of directly performing state analysis on the operation data of upstream and downstream production equipment, the data processing capacity of the neural network can be reduced to a certain extent (i.e., specific mining analysis on the operation data of downstream production equipment is not required), the efficiency of operation state analysis can be improved to a certain extent, and in addition, the reliability is higher due to the full utilization of the production relevance and the sequencing of the equipment, the reliability of operation state analysis can be improved to a certain extent, thereby improving the problem of low reliability of state analysis.

In order to make the above objects, features and advantages of the present invention more comprehensible, preferred embodiments accompanied with figures are described in detail below.

Drawings

FIG. 1 is a block diagram of an operation analysis system of a ceramic riving knife production system based on intelligent monitoring according to an embodiment of the invention.

Fig. 2 is a schematic flow chart of steps included in an operation analysis method of a ceramic riving knife production system based on intelligent monitoring according to an embodiment of the present invention.

Fig. 3 is a schematic diagram of each module included in an operation analysis device of a ceramic riving knife production system based on intelligent monitoring according to an embodiment of the present invention.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are only some embodiments of the present invention, but not all embodiments of the present invention. The components of the embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.

Thus, the following detailed description of the embodiments of the invention, as presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

As shown in fig. 1, the embodiment of the invention provides an operation analysis system of a ceramic riving knife production system based on intelligent monitoring. The ceramic riving knife production system operation analysis system based on intelligent monitoring can comprise a memory and a processor.

In detail, the memory and the processor are electrically connected directly or indirectly to realize transmission or interaction of data. For example, electrical connection may be made to each other via one or more communication buses or signal lines. The memory may store at least one software functional module (computer program) that may exist in the form of software or firmware. The processor can be used for executing the executable computer program stored in the memory, so that the ceramic riving knife production system operation analysis method based on intelligent monitoring provided by the embodiment of the invention is realized.

It should be appreciated that in some possible embodiments, the memory may be, but is not limited to, random access memory (RandomAccessMemory, RAM), read-only memory (ReadOnlyMemory, ROM), programmable read-only memory (Programmable read-OnlyMemory, PROM), erasable read-only memory (ErasableProgrammable read-OnlyMemory, EPROM), electrically erasable read-only memory (ElectroErasableProgrammable read-OnlyMemory, EEPROM), and the like.

It should be appreciated that in some possible embodiments, the processor may be a general purpose processor, including a central processor (CentralProcessingUnit, CPU), a network processor (NetworkProcessor, NP), a system on chip (SystemonChip, soC), and the like; but also Digital Signal Processors (DSPs), application Specific Integrated Circuits (ASICs), field Programmable Gate Arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

It should be appreciated that in some possible embodiments, the intelligent monitoring-based ceramic riving knife production system operation analysis system may be a server with data processing capabilities.

With reference to fig. 2, the embodiment of the invention also provides an operation analysis method of the ceramic riving knife production system based on intelligent monitoring, which can be applied to the operation analysis system of the ceramic riving knife production system based on intelligent monitoring. The method comprises the steps defined by the flow related to the intelligent monitoring-based ceramic riving knife production system operation analysis method, wherein the method steps can be realized by the intelligent monitoring-based ceramic riving knife production system operation analysis system. The specific flow shown in fig. 2 will be described in detail.

And step S100, extracting target upstream data corresponding to a target ceramic riving knife production system.

In the embodiment of the invention, the ceramic riving knife production system based on intelligent monitoring can extract the target upstream data corresponding to the target ceramic riving knife production system, and the target upstream data belongs to the equipment operation log of the upstream production equipment in the target ceramic riving knife production system. It is assumed that the target ceramic riving knife production system includes two production devices: such as a material cutter (apparatus a) as an upstream production apparatus and a grinding apparatus (apparatus B) as a downstream production apparatus. For device a, the following information may be collected: cutting speed (measured in length of material cut per unit time), cutting angle, cutting pressure, cutting time. For device B, the following information may be collected: polishing rate (measured as surface area polished per unit time), polishing pressure, polishing time. For example, operational parameter data for device a and device B may be recorded over a period of time, such as, for example, operational log data for device a: timestamp 1, cutting speed=10m/s, cutting angle=45°, cutting pressure=200n, cutting time=5 seconds; timestamp 2 cutting speed=12 m/s, cutting angle=50°, cutting pressure=180n, cutting time=4 seconds; timestamp 3 cutting speed=11 m/s, cutting angle=48°, cutting pressure=190N, cutting time=5 seconds. Log data of device B: timestamp 1, grinding speed=8 mm/s, grinding pressure=150n, grinding time=10 seconds; timestamp 2, grinding speed=7mm%/s, grinding pressure=160n, grinding time=9sec; timestamp 3 grinding speed=9 mm/s, grinding pressure=155N, grinding time=11 seconds.

Step S200, loading the target upstream data into an equipment operation analysis network obtained by at least secondary network optimization, and analyzing a downstream response parameter corresponding to each preset first downstream data in a plurality of preset first downstream data.

In the embodiment of the invention, the ceramic riving knife production system based on intelligent monitoring can load the target upstream data to the equipment operation analysis network obtained by at least secondary network optimization, and analyze the downstream response parameters corresponding to each preset first downstream data in a plurality of preset first downstream data, wherein the downstream response parameters corresponding to the preset first downstream data are used for reflecting the possibility that the operation process corresponding to the preset first downstream data occurs after the operation process corresponding to the target upstream data occurs, such as 0.1, 0.5 or 0.8.

And step S300, marking the preset first downstream data corresponding to the downstream response parameter with the maximum value so as to be marked as target downstream data corresponding to the target upstream data.

In the embodiment of the present invention, the ceramic riving knife production system based on intelligent monitoring may mark preset first downstream data corresponding to a downstream response parameter having a maximum value, so that the preset first downstream data is marked as target downstream data corresponding to the target upstream data, that is, the most matched downstream data, where the target downstream data may be: timestamp 1, grinding speed=8 mm/s, grinding pressure=150n, grinding time=10 seconds; timestamp 2 grinding speed=7mm%/s, grinding pressure=165N, grinding time=8 seconds; timestamp 3 grinding speed=9 mm/s, grinding pressure=155N, grinding time=11 seconds.

Step S400, determining the operation state of the target ceramic riving knife production system based on the predicted operation process reflected by the target downstream data and the actual operation process corresponding to the operation process.

In the embodiment of the invention, the ceramic riving knife production system based on intelligent monitoring can determine the operation state of the target ceramic riving knife production system based on the predicted operation process reflected by the target downstream data and the actual operation process corresponding to the operation process, and the degree of the excellent operation state is negatively related to the distinguishing information between the predicted operation process reflected by the target downstream data and the actual operation process corresponding to the predicted operation process. In addition, the actual running process corresponding to the running process may refer to actual downstream data corresponding to the target downstream data. That is, since the cutting is generally performed based on the material cutter (apparatus a) and then the grinding is performed based on the grinding apparatus (apparatus B), wherein the cutting result of the material cutter (apparatus a) affects the performance of the grinding apparatus (apparatus B), the grinding process of the grinding apparatus (apparatus B) can be predicted based on the cutting process (e.g., the target upstream data) of the material cutter (apparatus a), and then the predicted grinding process and the actual grinding process of the grinding apparatus (apparatus B) can be compared to determine the operation state, that is, the greater the difference in the characterization of the comparison result, the worse the operation state, that is, the surface may deviate from the normal operation, and the smaller the difference in the characterization of the comparison result, the better the operation state. That is, it is possible to "timestamp 1: grinding speed=8 mm/s, grinding pressure=150n, grinding time=10 seconds; timestamp 2, grinding speed=7mm%/s, grinding pressure=160n, grinding time=9sec; timestamp 3: grinding speed=9 mm/s, grinding pressure=155N, grinding time=11 seconds "and timestamp 1: grinding speed=8 mm/s, grinding pressure=150N, grinding time=10 seconds; timestamp 2 grinding speed=7mm%/s, grinding pressure=165N, grinding time=8 seconds; timestamp 3. Grinding speed=9mm/s, grinding pressure=155N, grinding time=11sec "difference calculation is performed, for example, differences of timestamp 1, timestamp 2 and timestamp 3 in two groups of data are calculated respectively, and then average calculation is performed on each difference to obtain a final difference.

In addition, based on the determined operating conditions, corresponding measures may be taken to optimize and improve the ceramic riving knife production system, for example, to repair equipment failures or to update production flows, parameters, and the like.

Based on the foregoing, since the target downstream data can be predicted first, the operation state of the target ceramic riving knife production system can be determined based on the predicted operation process reflected by the target downstream data and the actual operation process corresponding to the predicted operation process, so that, compared with the scheme of performing state analysis on the operation data of upstream and downstream production equipment, the data processing capacity of the neural network can be reduced to a certain extent (i.e., specific mining analysis on the operation data of the downstream production equipment is not required), the efficiency of operation state analysis can be improved to a certain extent, and in addition, the reliability is higher due to the full utilization of the production relevance and the sequencing of the equipment, the reliability of operation state analysis can be improved to a certain extent, thereby improving the problem of low reliability of state analysis.

It should be appreciated that, in some possible embodiments, the step of loading the target upstream data into the device operation analysis network obtained by using at least two times of network optimization to analyze the downstream response parameter corresponding to each of the plurality of preset first downstream data may further include the following specific implementation procedure described below:

Loading the target upstream data into a device operation analysis network obtained by at least secondary network optimization (for subsequent processing);

performing feature space mapping operation on the target upstream data to form an upstream data feature distribution corresponding to the target upstream data, wherein the upstream data feature distribution comprises a local upstream data feature distribution corresponding to each data segment in the target upstream data, and the local upstream data feature distribution is used for reflecting segment semantic information and segment distribution information of the corresponding data segment, that is, the target upstream data can be mapped into a feature space (such as ebedding) to represent discrete target upstream data by using continuous feature vectors, so as to obtain an upstream data feature distribution, thus being beneficial to finding potential relations and similarities among data, supporting subsequent data analysis and processing, and in addition, text granularity of the data segments is not limited, such as words, sentences and the like, the segment distribution information can refer to distribution coordinates of the data segments in the target upstream data, such as belonging to a first data segment, a second data segment or a third data segment and the like;

extracting a first relevant upstream data characteristic distribution (namely, the characteristic distribution with the minimum degree of correlation) from the characteristic distribution sequence by using the characteristic processing unit, carrying out association analysis operation on the target upstream data characteristic distribution based on the first relevant upstream data characteristic distribution, carrying out aggregation operation on the result of the association analysis operation and the first relevant upstream data characteristic distribution, for example, superposing the result of the association analysis operation and the first relevant upstream data characteristic distribution, and carrying out normalization processing to realize the aggregation operation of the characteristic distribution so as to form an aggregation characteristic distribution corresponding to the characteristic processing unit, wherein the target quantity of characteristic processing units are connected successively;

Extracting relevant upstream data characteristic distribution corresponding to a distribution position from the characteristic distribution sequence by utilizing the characteristic processing units for each characteristic processing unit except the first characteristic processing unit in the target number of characteristic processing units, carrying out association analysis operation on the aggregation characteristic distribution corresponding to the previous characteristic processing unit based on the relevant upstream data characteristic distribution, carrying out aggregation operation on the result of the association analysis operation and the relevant upstream data characteristic distribution, such as superposition of the result of the association analysis operation and the relevant upstream data characteristic distribution, and carrying out normalization processing to form the aggregation characteristic distribution corresponding to the characteristic processing unit;

performing aggregation operation, such as splicing, on the aggregate feature distribution corresponding to the last feature processing unit and the target upstream data feature distribution, so as to output aggregate upstream data feature distribution corresponding to the target upstream data feature distribution, and merging aggregate upstream data feature distribution corresponding to each local upstream data feature distribution, for example, merging corresponding aggregate upstream data feature distribution according to a distribution relation between each local upstream data feature distribution, so as to form optimized upstream data feature distribution corresponding to the target upstream data, and analyzing a downstream response parameter corresponding to each preset first downstream data in the plurality of preset first downstream data according to the optimized upstream data feature distribution, for example, performing feature reduction operation on the optimized upstream data feature distribution to form corresponding reduced data, and then calculating text similarity between the reduced data and the preset first downstream data, so as to obtain a downstream response parameter; alternatively, in other embodiments, the feature space mapping operation may be performed on the preset first downstream data, and then similarity calculation, such as cosine similarity, is performed on each preset feature distribution of the preset feature distributions corresponding to the preset first downstream data and the optimized upstream data feature distribution, so as to obtain corresponding downstream response parameters; that is, or, the analysis results of the respective feature processing units may be comprehensively considered, thereby obtaining a more comprehensive and accurate data feature representation and analysis result.

In general, the above steps combine feature space mapping, correlation analysis, aggregation operations, and data analysis techniques for processing the target upstream data to obtain a more comprehensive and accurate feature representation and analysis result (i.e., downstream response parameters corresponding to each preset first downstream data).

Wherein, it should be understood that, in some possible embodiments, the step of regarding, for each local upstream data feature distribution in the upstream data feature distribution, the local upstream data feature distribution as a target upstream data feature distribution, and determining, among other local upstream data feature distributions, a relevant upstream data feature distribution corresponding to the target upstream data feature distribution may further include the following specific implementation procedure described below:

for each of the upstream data feature distributions, taking the local upstream data feature distribution as a target upstream data feature distribution, and marking each other of the upstream data feature distributions as a candidate local upstream data feature distribution;

for each candidate local upstream data feature distribution, analyzing the distribution position distance of the data segment corresponding to the candidate local upstream data feature distribution and the data segment corresponding to the target upstream data feature distribution to determine an importance characterization parameter corresponding to the candidate local upstream data feature distribution, for example, the importance characterization parameter is inversely related to the distribution position distance;

Clustering (any existing clustering method can be adopted) is carried out on each local upstream data characteristic distribution in the upstream data characteristic distribution so as to form a plurality of clustering centers; for example, based on the K-means algorithm, clustering is performed according to the following steps: initializing parameters of a K-means algorithm, including the number (K value) of clustering centers, inputting each local upstream data characteristic distribution in the upstream data characteristic distribution into the K-means algorithm for clustering, dividing the K-means algorithm into K different clustering groups according to the similarity of data, and optimizing a clustering effect by iteratively adjusting the clustering centers, wherein after each iteration, the clustering centers are updated to better reflect the characteristic distribution of the data, and when the algorithm reaches a convergence state (the clustering centers are not changed any more), the clustering process is ended, so that a plurality of clustering centers can be formed;

determining a cluster center of each candidate local upstream data feature distribution as a corresponding target cluster center, calculating a center distance between the target cluster center and the cluster center of the target upstream data feature distribution, determining an initial correlation parameter of negative correlation based on the center distance, and adjusting (such as multiplication operation) the initial correlation parameter based on the corresponding importance characterization parameter to output a corresponding target correlation parameter;

And marking the local upstream data characteristic distribution with the largest target number of the corresponding target related parameters as the related upstream data characteristic distribution corresponding to the target upstream data characteristic distribution.

It should be understood, that in some possible embodiments, the steps of extracting, for a first feature processing unit of the target number of feature processing units, a first relevant upstream data feature distribution from the feature distribution sequence by using the feature processing unit, and performing, based on the first relevant upstream data feature distribution, a correlation analysis operation on the target upstream data feature distribution may further include the following specific implementation procedure described below:

extracting a first related upstream data feature distribution from the feature distribution sequence by using a first feature processing unit in the target number of feature processing units;

performing a first mapping operation on the target upstream data feature distribution, for example, multiplying a first mapping parameter distribution and the target upstream data feature distribution to form a corresponding first mapping feature distribution, where the first mapping parameter distribution is used as a network parameter to be adjusted in an optimization process;

Performing a second mapping operation and a third mapping operation on the first related upstream data feature distribution respectively to form a corresponding second mapping feature distribution and a corresponding third mapping feature distribution, for example, a second mapping parameter distribution and the first related upstream data feature distribution may be multiplied to form a corresponding second mapping feature distribution, the second mapping parameter distribution is used as a network parameter to be adjusted in the optimization process, and for example, a third mapping parameter distribution and the first related upstream data feature distribution may be multiplied to form a corresponding third mapping feature distribution, and the third mapping parameter distribution is used as a network parameter to be adjusted in the optimization process;

multiplying the transposed results of the first and second mapping feature distributions to output corresponding correlation parameters, and multiplying the correlation parameters and the third mapping feature distribution to output the results of an initial correlation analysis operation corresponding to the target upstream data feature distribution;

replacing the first relevant upstream data feature distribution based on the result of the initial association analysis operation corresponding to the target upstream data feature distribution, so as to perform the second mapping operation and the third mapping operation on the first relevant upstream data feature distribution in a revolving way to form a corresponding second mapping feature distribution and third mapping feature distribution, wherein the revolving execution times of the step can be configured according to actual demands, such as values of 3, 5 and the like;

And multiplying the transposed results of the first and second mapping feature distributions by the last execution to output a corresponding correlation parameter, and multiplying the correlation parameter by the third mapping feature distribution to output a result of an initial correlation analysis operation corresponding to the target upstream data feature distribution, wherein the output result of the initial correlation analysis operation is used as the result of the correlation analysis operation corresponding to the target upstream data feature distribution.

On the basis of the above, the intelligent monitoring-based ceramic riving knife production system operation analysis method can further comprise the following steps of optimizing the equipment operation analysis network.

Step S110, a plurality of first example production run information is extracted.

In the embodiment of the invention, the ceramic riving knife production system operation analysis system based on intelligent monitoring can extract a plurality of first example production operation information. The first example production operation information is formed based on operation monitoring operation of the example ceramic riving knife production system, the first example production operation information includes first example upstream data and first example downstream data, the first example upstream data is used for reflecting operation process of upstream production equipment of the example ceramic riving knife production system, the first example downstream data is used for reflecting operation process of downstream production equipment of the example ceramic riving knife production system, the upstream production equipment and the downstream production equipment refer to two equipment with relativity and precedence in production, that is, the first equipment performs processing (such as cutting of equipment A) after processing materials of the ceramic riving knife and then performs processing (such as grinding of equipment B) through the second equipment, and processing of the second equipment has relativity with processing of the first equipment, for example, the first equipment is used for performing rough cutting or grinding of materials of the ceramic riving knife, the second equipment user performs fine cutting or grinding of materials of the ceramic riving knife, and the second equipment has influence on cutting or grinding result of the rough cutting or grinding has relativity.

Step S120, determining a candidate device operation analysis network.

In the embodiment of the invention, the intelligent monitoring-based ceramic riving knife production system operation analysis system can determine candidate equipment operation analysis networks. The device operation analysis network is used for analyzing a plurality of downstream response parameters corresponding to preset first downstream data based on any one first upstream data, and the downstream response parameters corresponding to the preset first downstream data are used for reflecting the possibility that the operation process corresponding to the preset first downstream data appears after the operation process corresponding to the first upstream data appears, which is analyzed based on the device operation analysis network. The plurality of preset first downstream data may be configured by a user according to a running process in a normal state. Additionally, the candidate device operation analysis network may be a convolutional neural network.

Step S130, performing network optimization on the device operation analysis network based on the first example upstream data and the first example downstream data included in the plurality of first example production operation information.

In the embodiment of the invention, the ceramic riving knife production system operation analysis system based on intelligent monitoring can perform network optimization on the equipment operation analysis network based on first example upstream data and first example downstream data included in the plurality of first example production operation information.

Step S140, extracting a plurality of second example production run information.

In the embodiment of the invention, the ceramic riving knife production system operation analysis system based on intelligent monitoring can extract a plurality of second example production operation information. The second example production run information is formed based on performing a run monitoring operation on the example ceramic riving knife production system, the second example production run information including second example upstream data and second example downstream data. The plurality of second example production operation information may be obtained by screening the plurality of first example production operation information, or the plurality of first example production operation information may be formed by performing operation monitoring operation on a system having a higher degree of excellence in the operation state. The first and second have no further meaning and are only used to distinguish between example information used in two phases.

Step S150, performing secondary network optimization on the device operation analysis network obtained by performing network optimization based on the plurality of second example production operation information and the target response parameters corresponding to the plurality of second example production operation information.

In the embodiment of the invention, the operation analysis system of the ceramic riving knife production system based on intelligent monitoring can perform secondary network optimization (namely, perform secondary network optimization) on the equipment operation analysis network obtained by performing network optimization based on the plurality of first example production operation information based on the plurality of second example production operation information and target response parameters corresponding to the plurality of second example production operation information. And analyzing the second example upstream data and the second example downstream data in the second example production operation information based on a target parameter determination rule to determine the target response parameter corresponding to the second example production operation information, wherein the target response parameter is used for reflecting the possibility that the operation process corresponding to the second example downstream data reappears after the operation process corresponding to the second example upstream data analyzed by the target parameter determination rule appears, namely, the parameter representing the possibility height.

It should be appreciated that, in some possible embodiments, step S110 in the foregoing description may further include the following specific implementation procedure described below:

Performing operation monitoring operation on an example ceramic riving knife production system to form a plurality of first example upstream data and first example downstream data of each first example upstream data, for example, the example ceramic riving knife production system may be subjected to operation log extraction;

in the case that the example upstream data amount corresponding to any one of the first example downstream data is smaller than or equal to the preconfigured standard data amount, expanding the first example upstream data corresponding to the any one of the first example downstream data, for example, the duration of the running process corresponding to the first example upstream data corresponding to the any one of the first example downstream data is from time a to time B, and can be expanded from time C to time B, wherein time C is earlier than time a, so that the duration of the running process is expanded, so that the example upstream data amount corresponding to the any one of the first example downstream data is larger than the standard data amount, and the specific value of the standard data amount is not limited and can be configured according to requirements; or alternatively

Calculating a multiplication value between a preconfigured corresponding adjustment parameter and a target upstream data volume to form a corresponding preset data volume, and expanding first example upstream data corresponding to any one first example downstream data under the condition that the example upstream data volume corresponding to any one first example downstream data is smaller than or equal to the preset data volume, wherein the expansion mode is as described before, so that the example upstream data volume corresponding to any one first example downstream data is larger than the preset data volume, the target upstream data volume is used for reflecting the total data volume of the plurality of first example upstream data, and the corresponding adjustment parameter can be configured according to actual demands, such as numerical values of 0.1, 0.15, 0.2, 0.25, 0.3 and the like.

It should be appreciated that, in some possible embodiments, step S130 in the above description may further include the following specific implementation procedure:

for each first example production operation information, loading first example upstream data in the first example production operation information to enable the first example upstream data to be loaded into the equipment operation analysis network, and analyzing downstream response parameters corresponding to the plurality of preset first downstream data, as described in the related description below;

based on the distinguishing information between the matched first downstream data and the first example downstream data in the first example production operation information, performing network optimization operation on the equipment operation analysis network, so that the degree of distinction between the new matched first downstream data analyzed by the network operation analysis network based on the first example upstream data and the first example downstream data is reduced, specifically, a corresponding network optimization index can be determined based on the distinguishing information, the network optimization index can be positively correlated with the distinguishing information, and then, in the direction of reducing the network optimization index, network parameters of the equipment operation analysis network can be optimized and adjusted, such as the number of included convolution kernels, window size, step size and the like.

It should be appreciated that, in some possible embodiments, the step of performing the network optimization operation on the device operation analysis network based on the distinguishing information between the matching first downstream data and the first example downstream data in the first example production operation information, so that the device operation analysis network after the network optimization operation analyzes the new matching first downstream data and the first example downstream data based on the first example upstream data to reduce the degree of distinction between the new matching first downstream data and the first example downstream data may further include the following specific implementation procedure described in the following:

Sequentially performing a cyclic network optimization operation on the equipment operation analysis network based on distinguishing information between the matched first downstream data corresponding to each of the plurality of first example production operation information and the first example downstream data in the first example production operation information, and stopping continuing the cyclic network optimization operation when the number of the performed network optimization operations is greater than or equal to a preset first optimization number, wherein the specific numerical value of the first optimization number is not limited, such as 2, 4, 8 and the like; or alternatively

And sequentially performing the cyclic network optimization operation on the equipment operation analysis network based on the distinguishing information between the matched first downstream data corresponding to each of the plurality of first example production operation information and the first example downstream data in the first example production operation information, and stopping continuing the cyclic network optimization operation when the equipment operation analysis network has the preset first downstream data with the maximum value based on the response parameter analyzed by any one of the first example upstream data, and the distinguishing degree between the first example downstream data corresponding to any one of the first example upstream data is smaller than or equal to the first distinguishing degree determined in advance, wherein the specific value of the first distinguishing degree is not limited.

It should be appreciated that, in some possible embodiments, step S150 in the above description may further include the following specific implementation procedure described below:

extracting a plurality of preferred first downstream data from the plurality of preset first downstream data based on the downstream response parameters corresponding to the plurality of preset first downstream data, wherein the downstream response parameters corresponding to each of the plurality of preferred first downstream data exceed the downstream response parameters corresponding to each other preset first downstream data in the plurality of preset first downstream data, for example, a specified number of preset first downstream data with the largest corresponding downstream response parameters may be marked as preferred first downstream data, or a configured screening parameter may be extracted, and then each of the preset first downstream data with the corresponding downstream response parameters larger than the screening parameter may be marked as preferred first downstream data to obtain a plurality of preferred first downstream data;

Based on the target parameter determining rule, respectively performing analysis operation on the second example upstream data and each preferred first downstream data, outputting a rule analysis output parameter corresponding to each preferred first downstream data, marking a rule analysis output parameter with the maximum value in the rule analysis output parameters corresponding to the preferred first downstream data so as to be marked as a corresponding target rule analysis output parameter, wherein the target parameter determining rule can be configured according to actual requirements, for example, text similarity calculation can be performed on the second example upstream data and the preferred first downstream data, during calculation, importance degree, such as positive correlation between occurrence frequency and importance degree, of each keyword can be determined according to a corresponding expected database, and then semantic similarity fusion can be performed by combining importance degrees, so that corresponding text similarity is obtained;

based on the target parameter determining rule, analyzing the second example upstream data and the second example downstream data in the second example production operation information, and outputting a target response parameter corresponding to the second example production operation information, as described above;

It should be appreciated that, in some possible embodiments, the step of performing the second network optimization operation on the device operation analysis network based on the distinguishing information between the target rule analysis output parameter and the target response parameter, so that the device operation analysis network after network optimization, based on the target rule analysis output parameter and the target response parameter analyzed by the second example upstream data, has a reduced degree of distinction, may further include the following specific implementation procedure described below:

analyzing the distinguishing information between the output parameters and the target response parameters based on the target rules corresponding to the second example production operation information, sequentially performing the cyclic network optimization operation on the equipment operation analysis network, and stopping continuing the cyclic network optimization operation when the number of the performed network optimization operations is greater than or equal to a second preset optimization number, wherein the specific numerical value of the second optimization number is not limited, such as 10, 50 and the like; or,

And sequentially performing a cyclic network optimization operation on the equipment operation analysis network based on the distinguishing information between the target rule analysis output parameters and the target response parameters corresponding to the second example production operation information, and stopping continuing the cyclic network optimization operation when the distinguishing degree between the target rule analysis output parameters analyzed by the equipment operation analysis network based on any one second example production operation information and the target response parameters corresponding to the any one second example production operation information is smaller than or equal to a second predetermined distinguishing degree, wherein the specific numerical value of the second distinguishing degree is not limited, such as 0.7, 0.5, 0.2, 0.1 and the like.

The embodiment of the invention also provides another intelligent monitoring-based ceramic riving knife production system operation analysis method which can be applied to the intelligent monitoring-based ceramic riving knife production system operation analysis system. The method steps defined by the flow related to the intelligent monitoring-based ceramic riving knife production system operation analysis method can be realized by the intelligent monitoring-based ceramic riving knife production system operation analysis system, and the intelligent monitoring-based ceramic riving knife production system operation analysis method comprises the following steps:

Extracting target upstream data corresponding to a target ceramic riving knife production system, wherein the target upstream data can be current operation log data of upstream production equipment to be analyzed;

loading the target upstream data into an equipment operation analysis network obtained by utilizing secondary network optimization, and analyzing downstream response parameters corresponding to each preset first downstream data in the plurality of preset first downstream data, wherein the downstream response parameters are described in the related description;

determining a plurality of preferred first downstream data from the plurality of preset first downstream data based on the downstream response parameter corresponding to each of the plurality of preset first downstream data, wherein the downstream response parameter corresponding to each of the plurality of preferred first downstream data exceeds the downstream response parameter corresponding to each of the other preferred first downstream data in the plurality of preset first downstream data, as described in the related description;

performing analysis operation on the target upstream data and each of the preferred first downstream data based on the target parameter determining rule, outputting a rule analysis output parameter corresponding to each of the preferred first downstream data, and marking the preferred first downstream data corresponding to the rule analysis output parameter having the maximum value among the analyzed rule analysis output parameters so as to be marked as target downstream data corresponding to the target upstream data, as described in the previous related description;

With reference to fig. 3, the embodiment of the invention also provides an operation analysis device of the ceramic riving knife production system based on intelligent monitoring, which can be applied to the operation analysis system of the ceramic riving knife production system based on intelligent monitoring. The ceramic riving knife production system operation analysis device based on intelligent monitoring can include:

the upstream data extraction module is used for extracting target upstream data corresponding to a target ceramic riving knife production system, and the target upstream data belongs to equipment operation logs of upstream production equipment in the target ceramic riving knife production system;

the device operation analysis module is used for loading the target upstream data to an device operation analysis network obtained by at least secondary network optimization, and analyzing downstream response parameters corresponding to each preset first downstream data in a plurality of preset first downstream data, wherein the downstream response parameters corresponding to the preset first downstream data are used for reflecting the possibility that the operation process corresponding to the preset first downstream data occurs after the operation process corresponding to the target upstream data occurs;

The downstream data determining module is used for marking preset first downstream data corresponding to the downstream response parameter with the maximum value so as to be marked as target downstream data corresponding to the target upstream data;

the running state determining module is used for determining the running state of the target ceramic riving knife production system based on the predicted running process reflected by the target downstream data and the actual running process corresponding to the running process, and the degree of the excellent running state is negatively related to the distinguishing information between the predicted running process reflected by the target downstream data and the actual running process corresponding to the predicted running process.

In summary, the method and the system for analyzing the operation of the ceramic riving knife production system based on intelligent monitoring provided by the invention can firstly extract the target upstream data corresponding to the target ceramic riving knife production system; loading the target upstream data into an equipment operation analysis network obtained by at least secondary network optimization, and analyzing downstream response parameters corresponding to each preset first downstream data in a plurality of preset first downstream data; marking preset first downstream data corresponding to the downstream response parameter with the maximum value, so that the preset first downstream data is marked as target downstream data corresponding to target upstream data; and determining the operation state of the target ceramic riving knife production system based on the predicted operation process reflected by the target downstream data and the actual operation process corresponding to the operation process. Based on the foregoing, since the target downstream data can be predicted first, the operation state of the target ceramic riving knife production system can be determined based on the predicted operation process reflected by the target downstream data and the actual operation process corresponding to the predicted operation process, so that, compared with the conventional technical scheme of directly performing state analysis on the operation data of upstream and downstream production equipment, the data processing capacity of the neural network can be reduced to a certain extent (i.e., specific mining analysis on the operation data of downstream production equipment is not required), the efficiency of operation state analysis can be improved to a certain extent, and in addition, the reliability is higher due to the full utilization of the production relevance and the sequencing of the equipment, the reliability of operation state analysis can be improved to a certain extent, thereby improving the problem of low reliability of state analysis.

The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. The operation analysis method of the ceramic riving knife production system based on intelligent monitoring is characterized by comprising the following steps of:

2. The method for analyzing the operation of the ceramic riving knife production system based on intelligent monitoring according to claim 1, wherein the step of loading the target upstream data into the equipment operation analysis network obtained by at least two times of network optimization to analyze the downstream response parameters corresponding to each preset first downstream data in the plurality of preset first downstream data comprises the steps of:

3. The method for analyzing the operation of the ceramic riving knife production system based on intelligent monitoring according to claim 1 or 2, wherein the method for analyzing the operation of the ceramic riving knife production system further comprises the steps of:

4. The method for analyzing the operation of the ceramic riving knife production system based on intelligent monitoring according to claim 3, wherein the step of optimizing the equipment operation analysis network based on the first example upstream data and the first example downstream data included in the plurality of first example production operation information comprises the steps of:

5. The method for analyzing the operation of the ceramic riving knife production system based on intelligent monitoring according to claim 4, wherein the step of performing the network optimization operation on the equipment operation analysis network based on the difference information between the matched first downstream data and the first example downstream data in the first example production operation information so that the degree of distinction between the new matched first downstream data analyzed by the equipment operation analysis network based on the first example upstream data and the first example downstream data after the network optimization operation is reduced comprises the steps of:

6. The method for analyzing the operation of a ceramic riving knife production system based on intelligent monitoring according to claim 5, wherein the step of performing secondary network optimization on the equipment operation analysis network obtained by performing network optimization based on the plurality of second example production operation information and the target response parameters corresponding to the plurality of second example production operation information comprises the steps of:

7. The method for analyzing the operation of the ceramic riving knife production system based on intelligent monitoring according to claim 6, wherein the step of performing the network optimization operation on the equipment operation analysis network for the second time based on the distinguishing information between the target rule analysis output parameter and the target response parameter so that the degree of distinction between the target rule analysis output parameter and the target response parameter analyzed based on the second example upstream data is reduced in the equipment operation analysis network after the network optimization comprises the steps of:

8. The method for analyzing the operation of a ceramic riving knife production system based on intelligent monitoring according to claim 3, wherein the step of extracting a plurality of first example production operation information comprises the steps of:

9. The operation analysis method of the ceramic riving knife production system based on intelligent monitoring is characterized by comprising the following steps of:

10. An intelligent monitoring-based ceramic riving knife production system operation analysis system, characterized by comprising a processor and a memory, wherein the memory is used for storing a computer program, and the processor is used for executing the computer program so as to realize the method of any one of claims 1-9.