CN115713234A - Equipment risk analysis early warning system and method based on industrial big data - Google Patents
Equipment risk analysis early warning system and method based on industrial big data Download PDFInfo
- Publication number
- CN115713234A CN115713234A CN202211435424.1A CN202211435424A CN115713234A CN 115713234 A CN115713234 A CN 115713234A CN 202211435424 A CN202211435424 A CN 202211435424A CN 115713234 A CN115713234 A CN 115713234A
- Authority
- CN
- China
- Prior art keywords
- industrial equipment
- data
- equipment data
- component
- industrial
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Images
Landscapes
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention discloses an equipment risk analysis early warning system and method based on industrial big data. According to the equipment risk analysis early warning system and method based on the industrial big data, provided by the invention, the industrial equipment data needing risk monitoring is automatically found through an algorithm, and the industrial equipment data risk monitoring rule is extracted through the machine characteristics, so that the rate of missing report and false report of industrial equipment data risk monitoring is reduced, the labor cost is reduced, and the risk monitoring efficiency is improved.
Description
Technical Field
The invention relates to the field of industrial equipment monitoring, in particular to an equipment risk analysis early warning system and method based on industrial big data.
Background
For industrial equipment, safe production is an important part, and it is important to predict in advance whether the equipment is operating normally, whether abnormalities and faults occur, or how high the probability of abnormalities and faults occurs, and to perform "predictive maintenance". Among them, the risk assessment is a heavy game called "predictive maintenance".
Generally, this judgment is "hard judgment", and all key components of the equipment have a set (one or more) of key indexes which normally operate, and once the fault is diagnosed to be contrary to the key indexes, the fault is judged to be a fault. However, these key indicators do not usually represent the operating state of the component itself, and the key indicators are not abnormal and cannot represent that the equipment does not have a fault or cannot indicate that the equipment does not have a fault risk.
In view of the above, it is necessary to provide a system and a method for analyzing and warning equipment risk based on industrial big data.
Disclosure of Invention
In order to overcome the defects and shortcomings in the prior art, the invention provides an equipment risk analysis early warning system and method based on industrial big data.
The technical scheme adopted by the invention is that the equipment risk analysis early warning system based on industrial big data comprises: the system comprises an industrial equipment operation data reading component, an industrial equipment data packet compression component, an abnormal operation discovery component, an industrial equipment risk monitoring standard setting component, an industrial equipment data real-time recording and updating component, an industrial equipment monitoring component, an industrial equipment transferring component, an industrial equipment data model algorithm component, a risk assessment result transmission component and a monitoring data early warning component, wherein,
the industrial equipment operation data reading component is used for reading and configuring data of industrial equipment to be monitored, and the industrial equipment operation data reading component is electrically connected with the industrial equipment data packet compression component;
the industrial equipment data packet compression component is used for reading industrial equipment data to be monitored from an industrial equipment database according to the industrial equipment data reading configuration of the industrial equipment operation data reading component, standardizing the industrial equipment data and compressing the standardized industrial equipment data into an agreed format, and is respectively and electrically connected with the industrial equipment operation data reading component, the abnormal operation finding component and the industrial equipment data model algorithm component;
the abnormal operation finding component is used for automatically finding which industrial equipment data in the standardized industrial equipment data table to be monitored in the industrial equipment data packet compression component need risk monitoring, and is respectively and electrically connected with the industrial equipment data packet compression component, the industrial equipment data risk monitoring rule generating and updating component and the industrial equipment monitoring component;
the industrial equipment risk monitoring standard setting component is used for setting standard parameters in a system, and the industrial equipment data rule configuration component is electrically connected with the industrial equipment data real-time recording and updating component;
the industrial equipment data real-time entry updating component is used for carrying out risk monitoring rule feature extraction on the industrial equipment data according to risk monitoring rule parameters defined or acquiescent by a monitoring rule configuration component user, historical industrial equipment data after the industrial equipment data packet compression component is standardized, and the abnormal operation discovery component automatically discovers (excavates) the need of carrying out risk monitoring on the industrial equipment data, and is respectively and electrically connected with the monitoring rule configuration component, the industrial equipment data packet compression component and the abnormal operation discovery component;
the industrial equipment monitoring component is used for supporting manual modification and adjustment of model parameters generated by the industrial equipment data real-time recording and updating component, perfecting monitoring rules, or manually defining new monitoring rules, and is respectively and electrically connected with the abnormal operation finding component and the industrial equipment data real-time recording and updating component;
the industrial equipment transferring component is used for completing the unified scheduling and execution of all functional components in the system, and is respectively and electrically connected with the monitoring rule configuration component and the risk assessment result transmission component;
the industrial equipment data model algorithm component is used for integrating the artificial self-defining rule of the industrial equipment monitoring component according to the industrial equipment data monitoring rule generated by the industrial equipment data real-time input updating component to calculate the newly added industrial equipment data to be monitored at risk and judging whether the newly added industrial equipment data triggers a risk abnormity alarm signal, and the industrial equipment data model algorithm component is respectively and electrically connected with the industrial equipment data real-time input updating component, the industrial equipment monitoring component and the risk assessment result transmission component;
the expression of the industrial equipment data model is as follows:
wherein A (x) represents an industrial equipment data early warning function, F s (x) Indicating industrial plant data set threshold function, R s Representing a data set of an industrial plant, eta representing a risk predictor, G s Representing a risk early warning level, and T representing a data acquisition period of industrial equipment;
the risk assessment result transmission component sends out related alarm signals to the industrial equipment data monitoring platform according to the calculation results of the industrial equipment data model algorithm component, and is respectively and electrically connected with the industrial equipment data model algorithm component and the monitoring data early warning component;
the monitoring data early warning assembly is used for receiving early warning of an industrial equipment data monitoring platform on alarm signal data, inputting the data early warning to the industrial equipment data real-time inputting and updating assembly, adjusting and optimizing a model according to manual early warning, and is respectively and electrically connected with the industrial equipment data real-time inputting and updating assembly and the risk assessment result transmission assembly.
Further, the industrial device data packet compression component comprises an industrial device data reading block subcomponent to be monitored and an industrial device data format standardization subcomponent, the industrial device data reading block subcomponent to be monitored is used for reading industrial device data to be monitored or historical normal industrial device data of the industrial device data to be monitored according to an industrial device data reading mode set by the industrial device operation data reading component, and the industrial device data format standardization subcomponent is used for standardizing the industrial device data read by the industrial device data reading block subcomponent to be monitored into a standard industrial device data format which can be recorded into the industrial device data real-time updating component and the industrial device data model algorithm component.
Furthermore, the abnormal operation discovery component comprises a numerical type industrial equipment data screening subassembly and a strongly related industrial equipment data pair screening subassembly, wherein the numerical type industrial equipment data screening subassembly is used for screening out numerical type data
And the strongly-relevant industrial equipment data pair screening subassembly is used for screening strongly-relevant industrial equipment data.
Further, the monitoring rule model of the monitoring rule configuration component is a monitoring model for updating the data of the industrial equipment in real time according to the parameters.
An equipment risk analysis early warning method based on industrial big data comprises the following steps:
step Q1, reading and configuring data of the industrial equipment to be monitored;
step Q2, reading the data of the industrial equipment to be monitored from the industrial equipment database according to the reading configuration, and standardizing the industrial equipment data into an agreed format;
step Q3, finding out which industrial equipment data in the standardized industrial equipment data table to be monitored need risk monitoring;
step Q4, setting standard parameters in the system;
step Q5, extracting the characteristics of the risk monitoring rule according to the industrial equipment data standardized in the step Q2, the industrial equipment data needing risk monitoring in the step Q3 and the hyper-parameters of the automatic monitoring rule model in the step Q4;
step Q6, modifying and adjusting the risk monitoring rule generated in step Q5 manually to perfect the monitoring rule;
and step Q7, carrying out risk calculation on the data of the industrial equipment to be monitored according to the risk monitoring rule generated by the step Q5 and the risk monitoring rule modified manually in the step Q6, and judging whether the newly added industrial equipment data trigger a risk abnormity alarm signal.
Step Q8: and sending out relevant alarm signals to the industrial equipment data monitoring platform according to the calculation result in the step Q7.
Further, the step Q8 is followed by a step Q9: and receiving the early warning of the industrial equipment data monitoring platform on the warning signal data, early warning the data in a step Q5, and adjusting and optimizing the risk monitoring feature extraction rule.
Further, the step Q2 specifically includes:
step Q21, reading the industrial equipment data to be monitored in the industrial equipment database or historical normal industrial equipment data of the industrial equipment data to be monitored according to the reading configuration;
and step Q22, standardizing the read industrial equipment data into a stipulated format.
Further, the step Q3 specifically includes:
q31, screening numerical industrial equipment data from the industrial equipment data table to be monitored;
step Q32, according to the screened numerical data historical industrial equipment data, calculating the covariance matrix of the relevant data industrial equipment data to obtain
Correlation metrics to data industrial equipment data;
and step Q33, screening out strongly relevant industrial equipment data according to the obtained correlation degrees of the different industrial equipment data.
Further, the step Q5 specifically includes:
step Q51, judging whether the data of the industrial equipment to be monitored is new data of the industrial equipment to be monitored;
step Q52, if the data is judged to be new industrial equipment data, risk monitoring rule model parameter learning is carried out;
and step Q53, if the data is not the new industrial equipment data, judging whether the current time point is the updating time point of the data of the industrial equipment to be monitored for risks. And if the updating time point is reached, updating and adjusting the model parameters of the monitoring rule to be subjected to risks.
Further, the step Q52 specifically includes:
step Q521, assuming industrial equipment data variables corresponding to the two data to be subjected to risk inspection;
step Q522, performing linear regression according to the historical industrial equipment data to obtain a linear model regression model;
step Q523, calculating the difference value between the linear regression model and the real industrial equipment data value;
and step Q524, calculating the mean value and the standard deviation of the historical difference values to obtain parameters of the whale algorithm.
Compared with the prior art, the equipment risk analysis early warning system and method based on the industrial big data provided by the invention have the advantages that the data risk of the industrial equipment is monitored through machine feature extraction, the rate of missing report and false report is reduced, meanwhile, the labor cost is greatly reduced, and the monitoring efficiency is improved.
Drawings
Fig. 1 is a schematic structural diagram of an equipment risk analysis early warning system based on industrial big data according to the present invention;
FIG. 2 is a flow chart of an equipment risk analysis early warning method based on industrial big data according to the present invention;
fig. 3 is a detailed flowchart of step Q200 of the method for analyzing and warning equipment risk based on industrial big data according to the present invention;
fig. 4 is a detailed flowchart of step Q300 of the method for analyzing and warning equipment risk based on industrial big data according to the present invention;
fig. 5 is a detailed flowchart of step Q500 of the method for analyzing and warning equipment risk based on industrial big data according to the present invention;
fig. 6 is a specific flowchart of step Q502 of the method for analyzing and warning equipment risk based on industrial big data according to the present invention.
Detailed Description
It should be noted that the embodiments and features of the embodiments can be combined with each other without conflict, and the present application will be further described in detail with reference to the drawings and specific embodiments.
As shown in fig. 1, a system and a method for analyzing and warning equipment risk based on industrial big data,
as shown in fig. 1, an embodiment of an equipment risk analysis early warning system based on industrial big data according to the present invention includes an industrial equipment operation data reading component 1, an industrial equipment data packet compression component 2, an abnormal operation discovery component 3, an industrial equipment risk monitoring standard setting component 4, an industrial equipment data real-time entry and update component 5, an industrial equipment monitoring component 6, an industrial equipment transferring component 7, an industrial equipment data model algorithm component 8, a risk assessment result transmission component 9, and a monitoring data early warning component 10, wherein,
the industrial equipment operation data reading assembly 1 is used for reading and configuring data of industrial equipment to be monitored, and the industrial equipment operation data reading assembly 1 is electrically connected with the industrial equipment data packet compression assembly 2. The industrial equipment operation data reading component 1 allows a user to configure an industrial equipment data source (including an industrial equipment database type, an IP address, a user, a password, an industrial equipment data table where the industrial equipment data to be monitored is located, a data name of the industrial equipment data to be monitored in the industrial equipment data table) of the industrial equipment data to be monitored or a front-end reading interface of the industrial equipment data to be monitored through a user interface (a user interface based on Web or mobile App); and simultaneously, the user can also be supported to input an industrial equipment data reading source code script (such as an industrial equipment data reading script written in python language) meeting the system design standard. After the user completes and confirms the setting on the user interface, the set content is written into an application industrial equipment database (which can adopt mysql, postgresql, monogo and other industrial equipment databases) through an application server (which can adopt a micro-service mode).
The industrial equipment data packet compression component 2 is used for reading industrial equipment data to be monitored from an industrial equipment database according to the industrial equipment data reading configuration of the industrial equipment operation data reading component 1, standardizing the industrial equipment data and compressing the standardized industrial equipment data into an agreed format, and the industrial equipment data packet compression component 2 is electrically connected with the industrial equipment operation data reading component 1, the abnormal operation discovery component 3 and the industrial equipment data model algorithm component 8 respectively. The industrial equipment packet compression component 2, generally comprises the following subcomponents: and (1) a data reading component 21 of the industrial equipment to be monitored. The component reads the data of the industrial equipment to be monitored or the historical normal industrial equipment data of the industrial equipment to be monitored according to the industrial equipment data reading mode set by the component 1. This component is typically accomplished through a compute services component. In the face of mass data of the industrial equipment to be monitored, the computing service component can be realized by distributed computing engines such as Apache Spark, apache Spark Streaming and the like, and can also finish real-time computing by adopting an elastically extensible micro-service architecture. And the industrial equipment data reading component calls the corresponding industrial equipment database reading substitution code base to read the industrial equipment data according to parameters such as the type of the industrial equipment database, the IP address, the user, the password, the industrial equipment data table where the industrial equipment data to be monitored is located, the data name of the industrial equipment data to be monitored in the table and the like input by the user. And if the user configures a front-end reading interface of the industrial equipment data to be monitored, the component acquires the industrial equipment data according to the configured industrial equipment data reading address and parameters. In order to support richer industrial equipment data reading modes, the component also supports the user to input an industrial equipment data reading code script which meets the specification for the computing service component to call to acquire industrial equipment data. And (2) an industrial equipment data format standardization component 22. The component normalizes the industrial device data read by the component 21 into a standard industrial device data format recognizable by the component 5, 8, such as (key, value), or n-dimensional industrial device data frame (DataFrame) format.
The industrial equipment data model has the expression:
wherein A (x) represents an industrial equipment data early warning function, F s (x) Indicating industrial plant data set threshold function, R s Representing a data set of an industrial plant, eta a risk predictor, G s And (4) representing a risk early warning level, and T representing a data acquisition period of the industrial equipment.
The abnormal operation discovery component 3 is configured to automatically discover which industrial device data in the standardized industrial device data table to be monitored in the industrial device data packet compression component 2 need to be risk monitored, and the abnormal operation discovery component 3 is electrically connected to the industrial device data packet compression component 2, the industrial device data risk monitoring rule generation and update component 5, and the industrial device monitoring component 6, respectively. The abnormal operation discovers the component 3, and the model can automatically discover which industrial equipment data in the standardized industrial equipment data table to be monitored in the component 2 need to be subjected to risk monitoring, so that the defects that the traditional manual searching and defining of the data needing risk monitoring are low in efficiency and easy to miss are overcome. The assembly mainly comprises the following two components: and (1) a numerical industrial equipment data screening subassembly. The invention mainly solves the problem of data risk monitoring of numerical industrial equipment, so the data type data of the industrial equipment needs to be screened out firstly, in the aspect of implementation, the data type of the industrial equipment is marked as the numerical data (2), the data of the industrial equipment is strongly related, the data screening subassembly 32 of the industrial equipment is mainly realized by the following steps:
step 1: calculating a covariance matrix of the relevant data industrial equipment data according to the numerical data historical industrial equipment data screened by the component 31 to obtain a correlation measurement between the data industrial equipment data;
step 2: and (4) screening out strongly-related industrial equipment data (the correlation degree is greater than a certain threshold) according to the correlation degrees of the different industrial equipment data obtained by calculation in the step (1).
The industrial equipment risk monitoring standard setting component 4 is used for setting standard parameters in a system, and the industrial equipment data rule configuration component 4 is electrically connected with the industrial equipment data real-time recording and updating component 5. The industrial equipment risk monitoring standard setting component 4 is mainly used for configuring monitoring parameters supported by default in the system or configuring customized industrial equipment data risk monitoring rules. When the two industrial equipment data are highly linearly related (coincident), one data is linearly regressed against the other, and then the difference between the regressed value and the value to be monitored is calculated. If the difference value is subject to normal distribution, a secondary alarm signal is sent out when the difference value between the value to be monitored and the regression value exceeds more than 2 times of the standard deviation of the historical difference value; and when the difference value is more than 4 times greater than the standard deviation of the historical difference value, a primary alarm signal is sent out.
The industrial equipment data real-time entry updating component 5 is used for performing risk monitoring rule feature extraction on industrial equipment data according to risk monitoring rule parameters customized or acquiescent by a user of the monitoring rule configuration component 4, historical industrial equipment data pair monitoring rules after standardization of the industrial equipment data packet compression component 2 and the requirement of automatic discovery (mining) of the abnormal operation discovery component 3, and the industrial equipment data real-time entry updating component 5 is respectively electrically connected with the monitoring rule configuration component 3, the industrial equipment data packet compression component 2 and the abnormal operation discovery component 4; the component performs feature extraction on the monitoring rule by utilizing standardized historical industrial equipment data to be monitored corresponding to the strong correlation data monitored by the component 3 according to a risk monitoring rule model defaulted by the system. Based on the characteristics of the industrial equipment data, we can set the following risk rule monitoring model.
Updating the monitoring model of the industrial equipment data in real time by using the parameters: two variables that are approximately linearly related often appear in numerical industrial equipment data of an industrial industry, such as a fund's rating score and a fund's rating level generally follow an approximately linear relationship. For two linear correlations (industrial plant data). Firstly, performing linear regression according to historical industrial equipment data to obtain a linear model; then, the difference between the predicted value of the linear model and the true value of the industrial equipment data is calculated, and a corresponding Gaussian model is established for the difference (namely, the mean value and the standard deviation of the difference are estimated).
The industrial equipment data risk monitoring rule self-production and updating component 5 can be designed to comprise the following steps:
step 1: judging whether the data of the industrial equipment to be monitored is new data of the industrial equipment to be monitored;
step 2: if the data in the step 1 is judged to be new industrial equipment data, risk monitoring rule model parameter learning 41 is carried out, and the calculation steps are as follows:
step 2_1: assuming that the data variables of the industrial equipment corresponding to the two data to be subjected to risk inspection are x and y;
step 2_2: and performing linear regression according to the historical industrial equipment data to obtain a linear model regression model. The parameter estimation method may be a least square method, a maximum likelihood method, or the like.
Step 2_3, calculating the difference value between the linear regression model and the real industrial equipment data value;
step 2_4: and calculating the mean value and the standard deviation of the historical difference values to obtain parameters of the whale algorithm.
And step 3: and if the current time point is not the new data of the industrial equipment to be monitored, judging whether the current time point is the updating time point of the data of the industrial equipment to be monitored. If the update time point is reached, the parameters of the monitoring rule model to be assessed for risk are updated and adjusted 42 (the calculation method can be updated according to the total historical industrial equipment data or the historical industrial equipment data in the latest period of time as described above). Otherwise, ending the current component.
The description is given here of the monitoring calculation method of the industrial equipment data risk monitoring rule after obtaining new industrial equipment data to be monitored from several common automatic monitoring rule models set forth in the production and update component 5:
when the data of the industrial equipment to be monitored is input, the linear model established in the front is used for calculating and calculating the predicted value, then the difference value between the predicted value and the data of the industrial equipment to be monitored is calculated, the difference value is input into a Gaussian model of the difference value between the predicted value and the true value obtained by calculation of the assembly 4, the probability of the difference value is obtained, and when the probability is lower than the set value (configured by the assembly 4), alarm signals of different levels are sent out.
If the component 4 establishes a gaussian mixture model in step 2 \4, the current difference value needs to be substituted into the trained gaussian mixture model, and when the probability of the input numerical value is lower than the set numerical value (configured by the component 4), alarm signals of different levels are sent out.
The above-described industrial device data monitoring calculation may be implemented by an Apache Spark calculation engine. The calculation with higher requirement on the real-time performance can be completed by Spark Streaming or Apache Spark flash.
The industrial equipment monitoring component 6 is used for supporting manual modification and adjustment of model parameters generated by the industrial equipment data real-time recording and updating component 5, perfecting monitoring rules, or manually defining new monitoring rules, and the industrial equipment monitoring component 6 is electrically connected with the abnormal operation discovery component 3 and the industrial equipment data real-time recording and updating component 5 respectively; the industrial equipment data risk monitoring rule generated by machine learning is adopted, under the condition that the historical normal industrial equipment data is less, the condition that the monitoring rule is not complete easily occurs, and the industrial equipment data real-time recording and updating component 5 supports manual modification and new rule addition on the automatically generated monitoring rule. The automatic generation rule is matched with the manual rule, so that the flexibility and the adaptability of the system are improved.
The industrial equipment transferring component 7 is used for completing unified scheduling and execution of all functional components in the system, and the industrial equipment transferring component 7 is electrically connected with the monitoring rule configuration component 4 and the risk assessment result transmission component 9 respectively;
the industrial equipment data model algorithm component 8 is used for calculating newly added industrial equipment data to be subjected to risk monitoring according to an industrial equipment data monitoring rule generated by the industrial equipment data real-time entry updating component 5 and a manual self-defining rule of the industrial equipment monitoring component 6, and judging whether the newly added industrial equipment data triggers a risk abnormity alarm signal, and the industrial equipment data model algorithm component 8 is respectively and electrically connected with the industrial equipment data real-time entry updating component 5, the industrial equipment monitoring component 6 and the risk assessment result transmission component 9;
the risk assessment result transmission component 9 sends out related alarm signals to the industrial equipment data monitoring platform according to the calculation results of the industrial equipment data model algorithm component 8, and the risk assessment result transmission component 9 is electrically connected with the industrial equipment data model algorithm component 8 and the monitoring data early warning component 10 respectively. The component is used for pushing the industrial equipment data quality monitoring alarm signal data output by the industrial equipment data model algorithm component 8, including the basic data of the industrial equipment data triggering the alarm signal, the grade of the alarm signal and the like, to a customer through channels such as WeChat, short message and application App by using a message pushing system. For example: app pushing can be realized through protocols such as MQTT and XMPP, and can also be realized by calling third-party platforms such as Huacheng pushing, aliyun mobile pushing and Tencent carrier pigeon pushing.
The monitoring data early warning component 10 is used for receiving the early warning of an industrial equipment data monitoring platform on alarm signal data, giving the data early warning to the industrial equipment data real-time recording and updating component 5, adjusting and optimizing a model according to manual early warning, and the monitoring data early warning component 10 is respectively and electrically connected with the industrial equipment data real-time recording and updating component 5 and the risk assessment result transmission component 9. The component mainly receives early warning data of an industrial equipment data monitoring alarm signal of an industrial equipment data monitoring platform, and gives early warning to an industrial equipment data monitoring rule updating component (component 4) to adjust and optimize the monitoring rule according to manual early warning. If the signal for sending out the alarm signal by the manual early warning is a false signal, the module 4 needs to be early warned to timely adjust the monitoring rule according to the current input value.
As shown in fig. 2, an embodiment of an apparatus risk analysis early warning method based on industrial big data according to the present invention includes:
step Q100, reading and configuring the data of the industrial equipment to be monitored;
step Q200, reading the data of the industrial equipment to be monitored from the industrial equipment database according to the reading configuration, and standardizing the industrial equipment data into an agreed format;
step Q300, which industrial equipment data in the standardized industrial equipment data table to be monitored need risk monitoring is found;
step Q400, setting standard parameters in the system;
step Q500, extracting the characteristics of the risk monitoring rule according to the industrial equipment data standardized in the step Q200, the industrial equipment data needing risk monitoring in the step Q300 and the hyper-parameters of the automatic monitoring rule model in the step Q400;
step Q600, modifying and adjusting the risk monitoring rule generated in step Q500 manually, and perfecting the monitoring rule;
step Q700, performing risk calculation on the data of the industrial equipment to be monitored according to the risk monitoring rule generated by the step Q500 and the risk monitoring rule modified manually in the step Q600, and judging whether the newly-added industrial equipment data triggers a risk abnormity alarm signal;
and step Q800, sending a relevant alarm signal to the industrial equipment data monitoring platform according to the calculation result in the step Q700.
And step Q900, receiving the early warning of the industrial equipment data monitoring platform on the warning signal data, early warning the data in the step Q500, and adjusting and optimizing the risk monitoring feature extraction rule.
As shown in fig. 3, the step Q200 specifically includes:
step Q201, reading the industrial equipment data to be monitored in the industrial equipment database or the historical normal industrial equipment data of the industrial equipment data to be monitored according to the reading configuration;
and step Q202, standardizing the read industrial equipment data into a stipulated format.
As shown in fig. 4, the step Q300 specifically includes:
q301, screening numerical industrial equipment data from the industrial equipment data table to be monitored;
step Q302, according to the screened numerical data historical industrial equipment data, calculating a covariance matrix of related data industrial equipment data to obtain correlation measurement among the data industrial equipment data;
and step Q303, screening out strongly-related industrial equipment data according to the obtained correlation degrees of the different industrial equipment data.
As shown in fig. 5, the step Q500 specifically includes:
step Q501, judging whether the data of the industrial equipment to be monitored is new data of the industrial equipment to be monitored;
step Q502, if the data is judged to be new industrial equipment data, risk monitoring rule model parameter learning is carried out;
and step Q503, if the current time point is not the new industrial equipment data, judging whether the current time point is the updating time point of the industrial equipment data to be monitored for risks. And if the updating time point is reached, updating and adjusting the model parameters of the monitoring rule to be subjected to risks.
As shown in fig. 6, the step Q502 specifically includes:
step Q5021, assuming the data variables of the industrial equipment corresponding to the two data to be subjected to risk inspection;
step Q5022, linear regression is carried out according to historical industrial equipment data, and a linear model regression model is obtained by 0;
step Q5023, calculating a difference value between the linear regression model and a real industrial equipment data value;
and step Q5024, calculating the mean value and the standard deviation of the historical difference value to obtain parameters of the whale algorithm.
The invention provides an equipment risk analysis early warning system and a monitoring method based on industrial big data, which have the following advantages:
1. the industrial equipment data needing risk monitoring can be automatically found, and the efficiency, accuracy and coverage rate of industrial equipment data risk monitoring are greatly improved.
2. After data needing risk monitoring is found, the industrial equipment data risk monitoring rule is automatically learned according to the historical value of the industrial equipment data, and therefore quality, efficiency and coverage of industrial equipment data risk monitoring are improved.
In the description of the present invention, it is to be noted that, unless otherwise explicitly specified or limited, the terms "disposed," "mounted," "connected," and "fixed" are to be construed broadly, e.g., as meaning either fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meaning of the above terms in the present invention can be understood by those of ordinary skill in the art through specific situations.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various equivalent changes, modifications, substitutions and alterations can be made herein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (9)
1. An equipment risk analysis early warning system based on industry big data, characterized in that, the system includes:
the industrial equipment operation data reading assembly is used for reading and configuring the data of the industrial equipment to be monitored;
the industrial equipment data packet compression component is used for reading the industrial equipment data reading configuration of the industrial equipment operation data reading component, reading the industrial equipment data to be monitored from an industrial equipment database, standardizing the industrial equipment data and compressing the industrial equipment data into an agreed format;
the abnormal operation discovering component is used for automatically discovering which industrial equipment data in the standardized industrial equipment data table to be monitored in the industrial equipment data packet compression component need risk monitoring;
the industrial equipment risk monitoring standard setting component is used for setting standard parameters in the system;
the industrial equipment data real-time entry updating component is used for configuring user-defined or default risk monitoring rule parameters of the component according to the monitoring rules, monitoring rules of historical industrial equipment data after the industrial equipment data packet compression component is standardized, and performing risk monitoring rule feature extraction on the industrial equipment data required to be subjected to risk monitoring when the abnormal operation discovery component automatically discovers;
the industrial equipment monitoring component is used for supporting manual modification and adjustment of model parameters generated by the industrial equipment data real-time recording and updating component, perfecting monitoring rules, or manually and autonomously defining new monitoring rules;
the industrial equipment transferring component is used for finishing the unified scheduling and execution of all functional components in the system;
the industrial equipment data model algorithm component is used for recording industrial equipment data monitoring rules generated by the updating component in real time according to the industrial equipment data, synthesizing manual self-defining rules of the industrial equipment monitoring component to calculate newly-added industrial equipment data to be monitored at risk, and judging whether the newly-added industrial equipment data triggers a risk abnormity alarm signal;
the industrial equipment data model has the expression:
wherein A (x) represents an industrial equipment data early warning function, F s (x) Indicating a threshold function, R, for industrial plant data s Representing a data set of an industrial plant, eta representing a risk predictor, G s Representing a risk early warning level, and T representing a data acquisition period of the industrial equipment;
the risk assessment result transmission component is used for sending a related alarm signal to the industrial equipment data monitoring platform according to the calculation result of the industrial equipment data model algorithm component;
and the monitoring data early warning component is used for receiving the early warning of the industrial equipment data monitoring platform on the alarm signal data, recording the data early warning to the industrial equipment data real-time recording and updating component, and adjusting and optimizing the model according to the manual early warning.
2. The equipment risk analysis and early warning system based on the industrial big data as claimed in claim 1, wherein the industrial equipment data packet compression component comprises an industrial equipment data reading block subcomponent to be monitored and an industrial equipment data format standardization subcomponent, the industrial equipment data reading block subcomponent to be monitored is used for reading industrial equipment data to be monitored or historical normal industrial equipment data of the industrial equipment data to be monitored according to an industrial equipment data reading mode set by the industrial equipment operation data reading component, and the industrial equipment data format standardization subcomponent is used for standardizing the industrial equipment data read by the industrial equipment data reading block subcomponent to be monitored into an industrial equipment data format recognizable by the industrial equipment data real-time recording updating component and the industrial equipment data model algorithm component.
3. The system of claim 1, wherein the abnormal operation discovery component comprises a numerical industrial equipment data screening subcomponent for screening out numerical data and a strongly correlated industrial equipment data pair screening subcomponent for screening out strongly correlated industrial equipment data.
4. The system of claim 1, wherein the monitoring rule model of the monitoring rule configuration component is a monitoring model for updating industrial equipment data in real time for parameters.
5. An equipment risk analysis early warning method based on industrial big data is characterized by comprising the following steps:
step Q1, reading and configuring data of the industrial equipment to be monitored;
step Q2, reading the data of the industrial equipment to be monitored from the industrial equipment database according to the reading configuration, and standardizing the industrial equipment data into an agreed format;
q3, searching and finding which industrial equipment data in the standardized industrial equipment data table to be monitored need risk monitoring;
step Q4, setting standard parameters in the system;
step Q5, extracting the characteristics of the risk monitoring rule according to the industrial equipment data standardized in the step Q2, the industrial equipment data needing risk monitoring in the step Q3 and the hyper-parameters of the automatic monitoring rule model in the step Q4;
step Q6, modifying and adjusting the risk monitoring rule generated in step Q5 manually to perfect the monitoring rule;
step Q7, carrying out risk calculation on the data of the industrial equipment to be monitored according to the risk monitoring rule generated by the step Q5 and the risk monitoring rule modified manually in the step Q6, and judging whether the newly-added industrial equipment data triggers a risk abnormity alarm signal or not;
step Q8: and sending out a relevant alarm signal to the industrial equipment data monitoring platform according to the calculation result in the step Q7.
6. The industrial big data-based equipment risk analysis and early warning method according to claim 5, wherein the step Q8 is followed by a step Q9: and receiving the early warning of the industrial equipment data monitoring platform on the warning signal data, early warning the data in a step Q5, and adjusting and optimizing the risk monitoring feature extraction rule.
7. The equipment risk analysis and early warning method based on industrial big data as claimed in claim 5, wherein said step Q2 specifically comprises:
step Q21, reading the industrial equipment data to be monitored in the industrial equipment database or historical normal industrial equipment data of the industrial equipment data to be monitored according to the reading configuration;
and step Q22, standardizing the read industrial equipment data into a convention format.
8. The equipment risk analysis and early warning method based on industrial big data as claimed in claim 5, wherein said step Q3 specifically comprises:
q31, screening numerical industrial equipment data from the industrial equipment data table to be monitored;
step Q32, calculating a covariance matrix of the related data industrial equipment data according to the screened numerical data historical industrial equipment data to obtain the correlation measurement among the data industrial equipment data;
and step Q33, screening out strongly related industrial equipment data according to the obtained correlation degrees of the different industrial equipment data.
9. The equipment risk analysis and early warning method based on industrial big data as claimed in claim 5, wherein said step Q5 specifically comprises:
step Q51, judging whether the data of the industrial equipment to be monitored is new data of the industrial equipment to be monitored;
step Q52, if the data is judged to be new industrial equipment data, learning risk monitoring rule model parameters;
and step Q53, if the data is not the new industrial equipment data, judging whether the current time point is the updating time point of the data of the industrial equipment to be monitored at the risk, and if the current time point is the updating time point, updating and adjusting the model parameters of the rule to be monitored at the risk.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211435424.1A CN115713234A (en) | 2022-11-16 | 2022-11-16 | Equipment risk analysis early warning system and method based on industrial big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211435424.1A CN115713234A (en) | 2022-11-16 | 2022-11-16 | Equipment risk analysis early warning system and method based on industrial big data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115713234A true CN115713234A (en) | 2023-02-24 |
Family
ID=85233417
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211435424.1A Withdrawn CN115713234A (en) | 2022-11-16 | 2022-11-16 | Equipment risk analysis early warning system and method based on industrial big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115713234A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116414097A (en) * | 2023-05-15 | 2023-07-11 | 广东思创智联科技股份有限公司 | Alarm management method and system based on industrial equipment data |
-
2022
- 2022-11-16 CN CN202211435424.1A patent/CN115713234A/en not_active Withdrawn
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116414097A (en) * | 2023-05-15 | 2023-07-11 | 广东思创智联科技股份有限公司 | Alarm management method and system based on industrial equipment data |
| CN116414097B (en) * | 2023-05-15 | 2023-09-29 | 广东思创智联科技股份有限公司 | Alarm management method and system based on industrial equipment data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10880181B2 (en) | Methods and apparatus for analysing performance of a telecommunications network | |
| CN109548057B (en) | Method and system for monitoring and maintaining base station | |
| CN115713234A (en) | Equipment risk analysis early warning system and method based on industrial big data | |
| US11586981B2 (en) | Failure analysis device, failure analysis method, and failure analysis program | |
| KR20130017352A (en) | Diagnostic system and method for home appliance | |
| CN110062393B (en) | Intelligent analysis method for network difference cell | |
| CN118054845B (en) | Distributed optical network terminal fault monitoring method and system | |
| Fortes et al. | Context-aware self-healing: User equipment as the main source of information for small-cell indoor networks | |
| CN109240882B (en) | Financial data consistency detection system and method | |
| JP4716657B2 (en) | Quality evaluation device | |
| CN117792864A (en) | Alarm processing method and device, storage medium and electronic device | |
| CN107105100B (en) | Method and system for monitoring mobile terminal game | |
| KR101249903B1 (en) | Diagnostic system and method for home appliance | |
| CN116991677A (en) | Timing anomaly detection method, apparatus, terminal device and storage medium | |
| US20230188440A1 (en) | Automatic classification of correlated anomalies from a network through interpretable clustering | |
| CN116963134A (en) | Fault diagnosis method and device, terminal equipment and network platform | |
| CN116453437A (en) | Display screen module testing method, device, equipment and storage medium | |
| FI128274B (en) | Diagnostic test prioritization based on accumulated diagnostic reports | |
| CN117591970B (en) | Fault early warning method and system for full-automatic assembly equipment | |
| CN115630949A (en) | Power equipment detection system and method based on artificial intelligence | |
| CN116701389A (en) | Efficient power distribution network data storage method and system based on artificial intelligence | |
| CN118035882B (en) | Electrical fault diagnosis method and system based on AI | |
| CN117113095A (en) | Diagnostic method, diagnostic device, electronic apparatus, and storage medium | |
| KR20210079655A (en) | Method, Device and Computer program for research the cause of wireless network | |
| CN111261271A (en) | Service availability diagnosis method and device for video monitoring environment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20230224 |