CN111950577A

CN111950577A - Point inspection method and device

Info

Publication number: CN111950577A
Application number: CN201910413429.6A
Authority: CN
Inventors: 王昌明
Original assignee: Hitachi Ltd
Current assignee: Hitachi Ltd
Priority date: 2019-05-17
Filing date: 2019-05-17
Publication date: 2020-11-17

Abstract

The invention discloses a point inspection method and a point inspection device, relates to the technical field of data processing, and aims to solve the problem that the trigger precision of a point inspection task in the prior art is not high. The method comprises the following steps: acquiring historical spot inspection data; acquiring real-time point inspection data; analyzing the historical point inspection data and the real-time point inspection data to obtain an analysis result; and generating a point inspection task according to the analysis result and the real-time point inspection data. The embodiment of the invention can improve the triggering precision of the point inspection task.

Description

Point inspection method and device

Technical Field

The invention relates to the technical field of data processing, in particular to a point inspection method and a point inspection device.

Background

Point inspection is an important process in the manufacturing industry for determining whether a device or process flow is normal. The current point inspection method mainly depends on manual timing record. This approach relying on manual timing recording has the following problems: (1) the contradiction between the manpower expenditure and the timeliness of problem discovery brought by the fixed point inspection time interval; (2) the spot inspection report needs to be manually generated and distributed, and the delay time of abnormal feedback is long; (3) if the point inspection personnel do not operate according to the rules, the missed inspection or the error record can be caused; (4) it is difficult for the system to authenticate the operator.

CN108287529A describes an integrated inspection tour, remote precision diagnosis and maintenance system for industrial equipment. In the patent, a patrol task evaluation model is introduced to dynamically update a check period and push a check task.

However, in the above prior art, the cycle of the spot check task is changed only according to the spot check history and the equipment ledger, so that the precision of the trigger of the spot check task is not high.

Disclosure of Invention

The embodiment of the invention provides a point inspection method and a point inspection device, which aim to solve the problem that the triggering precision of a point inspection task in the prior art is not high.

In a first aspect, an embodiment of the present invention provides a spot inspection method, including:

acquiring historical spot inspection data;

acquiring real-time point inspection data;

analyzing the historical point inspection data and the real-time point inspection data to obtain an analysis result;

and generating a point inspection task according to the analysis result and the real-time point inspection data.

Wherein after the obtaining of the analysis results, the method further comprises:

acquiring point inspection record data; the spot inspection record comprises: manually recording information, the real-time point inspection data and the analysis result;

and generating a spot inspection report according to the analysis result and the spot inspection record data.

Wherein the real-time spot inspection data comprises: real-time operation parameters of production, real-time operation parameters of equipment, quality parameters and point inspection data input through a point inspection intelligent terminal; the quality parameters comprise at least one of online quality parameters and offline quality parameters;

the acquiring of the real-time spot inspection data comprises the following steps:

and respectively acquiring the production real-time operation parameters, the equipment real-time operation parameters, the quality parameters and the point inspection data input by the point inspection intelligent terminal according to a preset time interval.

acquiring the quality parameter;

acquiring quality detection data under the condition that the quality problem is determined to occur according to the quality parameters;

acquiring production parameters according to the quality detection data;

acquiring equipment parameters according to the production parameters;

and acquiring point inspection data input through the point inspection intelligent terminal.

Wherein, the acquiring real-time point inspection data further comprises:

and audio and video characteristic data of the equipment side sensor are obtained.

Analyzing the historical point inspection data and the real-time point inspection data to obtain an analysis result, wherein the analysis result comprises the following steps:

respectively inputting real-time operation parameters, audio and video characteristic values and quality parameters of equipment in the real-time point inspection data into a point inspection analysis model, wherein the point inspection analysis model is updated by using the historical point inspection data;

and analyzing the real-time operation parameters, the audio and video characteristic values and the quality parameters of the equipment by using the point inspection analysis model to obtain an analysis result.

Before analyzing the real-time operation parameters, the audio and video characteristic values and the quality parameters of the equipment by using the point inspection analysis model to obtain an analysis result, the method further comprises the following steps:

and inputting the production real-time operation parameters and the equipment real-time operation parameters in the real-time point inspection data into an equipment digital model to obtain an automatic analysis result.

The analyzing the real-time operation parameters, the audio and video characteristic values and the quality parameters by using the point inspection analysis model to obtain analysis results comprises the following steps:

and analyzing the real-time operation parameters, the audio and video characteristic values and the quality parameters of the equipment by using the point inspection analysis model to obtain a pre-analysis result, and combining the pre-analysis result with the automatic analysis result to obtain an analysis result.

The analyzing the real-time operation parameters, the audio and video characteristic values and the quality parameters by using the point inspection analysis model to obtain a pre-analysis result, and combining the pre-analysis result with the automatic analysis result to obtain an analysis result, comprising the following steps:

calling a fault and quality parameter correlation model to analyze the quality parameters;

under the condition that the fault reason is determined, determining the probability of the fault, the total loss estimated value and the emergency degree priority of the fault;

analyzing the real-time operation parameters of the equipment and the real-time operation parameters of the production under the condition that the fault reason is not determined; calling a correlation model of the fault and the equipment operation parameters to estimate the fault under the condition that the parameter set values of the equipment real-time operation parameters and the production real-time operation parameters are normal; and comparing the estimated faults with the automatic analysis result to determine the probability of the faults, the total loss prediction value and the emergency degree priority of the faults.

under the condition that the timing time is up, calling an audio and video characteristic value and a fault correlation model to analyze the audio and video characteristic value; determining the probability of the fault, the total loss predictive value and the emergency degree priority of the fault under the condition that the analysis result is abnormal;

and under the condition that the timing time is up, acquiring parameters of the equipment side sensor, and determining the probability of the fault, the total loss estimated value and the emergency degree priority of the fault when the parameters of the equipment side sensor exceed the preset parameter range.

Generating a point inspection task according to the analysis result and the real-time point inspection data, wherein the generating of the point inspection task comprises the following steps:

acquiring point inspection time and point inspection period required by each point inspection item according to the probability of the fault, the total loss pre-estimated value, the emergency degree priority of the fault and a point inspection item specification library;

and generating a point inspection task according to the point inspection time and the point inspection period required by each point inspection item.

The generating a point inspection task according to the analysis result and the real-time point inspection data further comprises:

reading a fault and abnormal knowledge base, and acquiring a related fault and a solution of the fault;

the generating of the point inspection task according to the point inspection time and the point inspection period required by each point inspection project comprises the following steps:

and generating a point inspection task according to the point inspection time and the point inspection period required by each point inspection project and the associated fault and solution.

Wherein, the generating of the spot inspection report according to the analysis result and the spot inspection record data comprises:

filling the point inspection record into a point inspection report template;

carrying out statistical analysis on the analysis result to obtain a report analysis result;

and generating a point inspection report according to the content of the point inspection report template and the report analysis result.

After the statistical analysis is performed on the spot inspection record and a report analysis result is obtained, the method further comprises the following steps:

and carrying out statistical analysis on the point inspection records to obtain an alarm report, wherein the alarm report comprises at least one of alarm level and alarm type.

In a second aspect, an embodiment of the present invention provides a spot inspection method, including:

acquiring an identifier of the point inspection equipment;

acquiring a point inspection task according to the identification of the point inspection equipment, wherein the point inspection task is generated according to historical point inspection data and real-time point inspection data;

and outputting the point inspection task.

Wherein the method further comprises at least one of:

authenticating the point inspection user to ensure that the point inspection user is positioned in a preset range of the point inspection equipment;

and receiving the point inspection data input by the point inspection user.

In a third aspect, an embodiment of the present invention provides a spot inspection apparatus, including:

the first acquisition module is used for acquiring historical point inspection data;

the second acquisition module is used for acquiring real-time point inspection data;

the analysis module is used for analyzing the historical point inspection data and the real-time point inspection data to obtain an analysis result;

and the task generating module is used for generating a point inspection task according to the analysis result and the real-time point inspection data.

Wherein the apparatus further comprises:

the third acquisition module is used for acquiring point inspection record data; the spot inspection record comprises: manually recording information, the real-time point inspection data and the analysis result;

and the report generation module is used for generating a point inspection report according to the analysis result and the point inspection record data.

In a fourth aspect, an embodiment of the present invention provides a spot inspection apparatus, including:

the first acquisition module is used for acquiring the identifier of the point inspection equipment;

the second acquisition module is used for acquiring a point inspection task according to the identification of the point inspection equipment, wherein the point inspection task is generated according to historical point inspection data and real-time point inspection data;

and the processing module is used for outputting the point inspection task.

Wherein the apparatus further comprises at least one of:

the authentication module is used for authenticating the point inspection user so as to ensure that the point inspection user is positioned in a preset range of the point inspection equipment;

and the receiving module is used for receiving the point inspection data input by the point inspection user.

In the embodiment of the invention, the historical point inspection data and the real-time point inspection data are combined to obtain an analysis result, and a point inspection task is generated according to the analysis result and the real-time point inspection data. Because the generated point inspection task simultaneously considers the historical point inspection data and the real-time point inspection data, the triggering precision of the point inspection task can be improved by utilizing the embodiment of the invention.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments of the present invention will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without inventive exercise.

FIG. 1 is a schematic diagram of a spot inspection system provided by an embodiment of the invention;

FIG. 2 is a flowchart of a spot inspection method according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of a data acquisition and storage module provided by an embodiment of the present invention;

fig. 4 is a schematic diagram of a data tracing process provided in the embodiment of the present invention;

FIG. 5 is a schematic diagram of a point inspection analysis engine provided by an embodiment of the invention;

FIG. 6 is a schematic diagram of a point inspection analysis model provided by an embodiment of the invention;

FIG. 7 is a schematic view of a spot check analysis process according to an embodiment of the present invention;

FIG. 8 is a second schematic view of a spot check analysis process according to an embodiment of the present invention;

FIG. 9 is a diagram of a task generation module for point inspection according to an embodiment of the present invention;

FIG. 10 is a flowchart illustrating the generation of a point inspection task according to an embodiment of the present invention;

FIG. 11 is a schematic diagram of a spot check report generation module provided by an embodiment of the invention;

FIG. 12 is a second flowchart of a point inspection method according to an embodiment of the present invention;

fig. 13 is a schematic diagram of a point inspection intelligent terminal provided in the embodiment of the present invention;

fig. 14 is a schematic workflow diagram of a point inspection intelligent terminal according to an embodiment of the present invention;

FIG. 15 is a schematic diagram of a data collection and storage process provided by an embodiment of the invention;

FIG. 16 is a schematic diagram of quality data out-of-tolerance trigger analysis provided by an embodiment of the present invention;

FIG. 17 is a schematic timing analysis flow chart provided by an embodiment of the present invention;

FIG. 18 is a schematic diagram of a task generation flow of a point inspection provided by an embodiment of the present invention;

FIG. 19 is a schematic view of a spot inspection process provided by an embodiment of the present invention;

FIG. 20 is a block diagram of a pointing device according to an embodiment of the present invention;

fig. 21 is a second structural diagram of a spot inspection apparatus according to an embodiment of the present invention;

FIG. 22 is a block diagram of a second acquisition module provided by embodiments of the present invention;

FIG. 23 is one of the block diagrams of an analysis module provided by the embodiments of the present invention;

FIG. 24 is a second block diagram of an analysis module according to an embodiment of the present invention;

FIG. 25 is one of the block diagrams of an analysis submodule provided in an embodiment of the present invention;

FIG. 26 is a second block diagram of an analysis submodule provided in an embodiment of the present invention;

FIG. 27 is one of the block diagrams of a task generation module provided by an embodiment of the present invention;

FIG. 28 is a second block diagram of a task generation module according to an embodiment of the present invention;

FIG. 29 is one of the block diagrams of a report generation module provided by an embodiment of the present invention;

FIG. 30 is a second block diagram of a report generation module according to an embodiment of the present invention;

FIG. 31 is a third block diagram of a pointing device according to an embodiment of the present invention;

FIG. 32 is a fourth block diagram of the spot inspection device according to the embodiment of the present invention;

fig. 33 is a structural diagram of a pointing device according to an embodiment of the present invention;

fig. 34 is a structural diagram of the intelligent point inspection terminal according to the embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Referring to fig. 1, a spot inspection system according to an embodiment of the present invention may include: the system comprises a point inspection management platform 11, a point inspection intelligent terminal 12, a production management system 13 and a field sensor 14.

The point inspection management platform is a core platform of the whole system. Its users are divided into the following categories: a checkman, an operator, a management layer, etc. Each type of user has a dedicated point inspection intelligent terminal (generally, a mobile phone APP).

The point inspection management platform 11 is respectively connected with the point inspection intelligent terminal 11, the production management system 13 and the field sensor 14 (such as noise, video and vibration sensors) and is used for collecting data related to production processes, equipment and quality. After being stored and analyzed, the data are used for automatic point inspection on one hand, automatic analysis pushing of manual point inspection tasks on the other hand and automatic point inspection report generation on the other hand.

The point inspection intelligent terminal 12 is mainly used for receiving a point inspection task, authenticating a point inspection person, automatically sending point inspection data to the point inspection person, receiving input of the point inspection person, and the like.

The production management system 13 is mainly used for inputting equipment, process, quality data and the like to the point inspection management platform.

The field sensors 14, which may include audio, video, shock sensors, etc., may have their data input to the checkstand management platform.

Wherein, the spot inspection management platform 11 may include: a data collection and storage module, a point inspection analysis engine, a point inspection task generation module, and a point inspection report generation module 114.

The data acquisition and storage module is mainly responsible for acquiring and storing relevant data such as equipment, production, quality, point inspection records and the like.

The point inspection analysis engine takes the real-time and historical data in the data acquisition and storage module as input and utilizes a built-in analysis model to output an analysis result. The analysis result is stored back to the data acquisition and storage module and is used by the point inspection task generation module and the point inspection report generation module. The point inspection analysis engine can also directly push alarm information to the operator client according to the analysis result.

The point inspection task generating module automatically generates a point inspection task according to the analysis of the point inspection analysis engine and the real-time data result, wherein the point inspection task comprises specific point inspection items, point inspection time and point inspection operation guidance, and then the information is pushed to a point inspection intelligent terminal of a point inspection worker.

And the point inspection report generating module automatically generates a point inspection report and statistical information according to the original input information of the data acquisition and storage module and the analysis result of the point inspection analysis engine, and pushes the point inspection report and the statistical information to the management layer client.

Referring to fig. 2, fig. 2 is a flowchart of a spot inspection method according to an embodiment of the present invention, and as shown in fig. 2, the method includes the following steps:

step 201, obtaining historical point inspection data.

In the embodiment of the present invention, the historical spot inspection data refers to data obtained in the historical spot inspection process, including but not limited to collected process (also called production), equipment, quality parameters, sensor data of the equipment, data collected by a spot inspection intelligent terminal, and the like.

Step 202, acquiring real-time spot inspection data.

In an embodiment of the present invention, the real-time spot inspection data includes: real-time operation parameters of production, real-time operation parameters of equipment, quality parameters and point inspection data input through a point inspection intelligent terminal; the quality parameters include at least one of online quality parameters and offline quality parameters.

In the embodiment of the invention, the real-time point inspection data can be acquired by the following two ways, so that the flexibility of acquiring the data is increased.

One mode is that the production real-time operation parameters, the equipment real-time operation parameters, the quality parameters and the point inspection data input by the point inspection intelligent terminal are respectively obtained according to preset time intervals. Wherein the predetermined time interval can be arbitrarily set.

Another way is to read the trace back data associated with quality issues. In this way, the quality parameter is acquired. And acquiring quality detection data under the condition that the quality problem is determined to occur according to the quality parameters. And then, acquiring production parameters according to the quality detection data, and acquiring equipment parameters according to the production parameters. Meanwhile, in this step, the point inspection data input through the point inspection intelligent terminal is also required to be acquired.

In addition, in order to improve the accuracy of the data, in the two modes, the audio and video characteristic data of the device side sensor can be acquired respectively.

With reference to fig. 1,

steps

201 and 202 are mainly performed by the data acquisition and storage module. As shown in fig. 3, the data collection and storage module may include:

equipment/process/quality parameter real-time database, historical database, point inspection database. The equipment/process/quality parameter real-time database is connected with the production management database, the equipment real-time database and the product quality database. Wherein the product quality database may in turn comprise one or both of an online product quality database and an offline product quality database. The equipment/process/quality parameter real-time database is timed or associated data is read from the database as needed. The historian periodically stores data from the equipment/process/quality database and the point check database for archiving and subsequent analysis. The check database stores check records of the equipment, including automatic records, manual records of check personnel, original images, videos, audios and the like.

The data acquisition and storage module is divided into two modes for acquiring data: and reading all data at fixed time and reading the tracing data related to the quality problem. The timing reading is to read all preset data according to a specified time interval. The flow of reading the trace back data is shown in fig. 4: first, the quality inspection analysis results of the work-in-process or manufactured product are read from the online and offline quality databases. If any result is abnormal, the detailed quality detection data in the online or offline quality database is read, and the corresponding retrospective batch number is read. And then reading the process parameters in the production management database according to the batch number, finding the machine number and the production time for producing the batch of products, reading the real-time equipment database according to the information, and acquiring the equipment parameters, the process parameters, the operation records and the like during production. If quality problems occur, these read data need to be labeled with the corresponding quality defect class. In addition, device side sensor data such as audio, video, vibration, etc. may also be read.

And 203, analyzing the historical point inspection data and the real-time point inspection data to obtain an analysis result.

In this step, the real-time operation parameters, the audio and video characteristic values and the quality parameters of the equipment in the real-time point inspection data are respectively input into a point inspection analysis model, wherein the point inspection analysis model is updated by using the historical point inspection data. And then, analyzing the real-time operation parameters, the audio and video characteristic values and the quality parameters of the equipment by using the point inspection analysis model to obtain an analysis result. In order to improve the accuracy of the analysis result, before the analysis result is obtained, the production real-time operation parameters and the equipment real-time operation parameters in the real-time point inspection data can be input into the equipment digital model to obtain an automatic analysis result. In this case, the point inspection analysis model is used to analyze the real-time operation parameters, the audio/video characteristic values and the quality parameters of the equipment to obtain a pre-analysis result, and the pre-analysis result and the automatic analysis result are combined to obtain an analysis result.

Specifically, in this step, the analysis result can be obtained in the following two ways:

in the first mode, a correlation model of the fault and the quality parameter is called to analyze the quality parameter. In the case where the cause of the fault is determined, the probability of the fault, the total loss estimate, and the urgency priority of the fault are determined. Analyzing the real-time operation parameters of the equipment and the real-time operation parameters of the production under the condition that the fault reason is not determined; calling a correlation model of the fault and the equipment operation parameters to estimate the fault under the condition that the parameter set values of the equipment real-time operation parameters and the production real-time operation parameters are normal; and comparing the estimated faults with the automatic analysis result to determine the probability of the faults, the total loss prediction value and the emergency degree priority of the faults.

In the second mode, under the condition that the timing time is up, calling an audio and video characteristic value and a fault correlation model to analyze the audio and video characteristic value; in the case of abnormal analysis results, the probability of failure, the total loss prediction value and the emergency degree priority of the failure are determined. Or, acquiring parameters of the equipment side sensor under the condition that the timing time is up, and determining the probability of the fault, the total loss estimated value and the emergency degree priority of the fault when the parameters of the equipment side sensor exceed the preset parameter range.

In conjunction with FIG. 1, step 203 is implemented using a point-check analysis engine. As shown in fig. 5, the spot inspection analysis engine includes: the system comprises an analysis model, an audio and video characteristic extraction module and a model updating engine. The analysis model is the core of the point inspection analysis engine, and reads effective parameters extracted before from the equipment/process/quality real-time database and imports the model for analysis. The audio and video feature extraction module reads original audio and video information collected by the point inspection intelligent terminal or the equipment side audio and video acquisition module from the point inspection database, and then extracts semantic information for the point inspection analysis model to use through audio analysis and video analysis. And finally, integrating the information by the point inspection analysis model to output an analysis result.

In order to ensure the accuracy of the analysis, the model updating engine is responsible for periodical updating of the point inspection analysis model. The point inspection analysis model continuously iterates itself according to historical data. The updating work of the analysis model can be triggered manually or at regular time, after the model updating operation is triggered, the model updating engine recalculates the analysis model through historical data by using a machine learning algorithm, verifies the analysis model, and updates the current analysis model after the verification is passed.

As shown in fig. 6, the spot inspection analysis model includes: the system comprises a fault and product quality parameter correlation model, a fault and equipment operation parameter correlation model and a fault and audio and video characteristic value correlation model. And the equipment operation parameters, the audio and video characteristic values and the quality parameters are respectively used as the input of the three models. The point inspection analysis model also comprises a comprehensive analysis module for carrying out comprehensive analysis according to the calculation result of the submodel, and the comprehensive analysis module predicts the possible loss according to the abnormal probability, the weight and the historical knowledge and automatically calculates the priority of the emergency degree. In addition, the point inspection analysis model may further include an equipment digital model, where the equipment digital model is a digital image of physical equipment, and the equipment digital model may output a simulation result, such as a threshold range of a normal output parameter, for providing a reference to the analysis model in response to a simulation request of the model according to input parameters (such as process parameters and equipment operation parameters).

The point inspection analysis process is divided into two types, one is quality out-of-tolerance triggered analysis, and the other is automatic analysis. The first analysis mode is triggered by the result that the quality detection data is unqualified, and the second analysis mode is completed according to the automatically collected data at regular time.

As shown in fig. 7, the first spot inspection analysis process includes:

and acquiring detailed quality data and calling a fault and quality correlation model. When the cause of the fault is determined, the process and equipment operating parameters are read. If the parameter set values of the process and equipment operation parameters are abnormal, the analysis result can be stored in the point inspection database and an alarm can be sent out. If the parameter setting values of the process and equipment operation parameters are normal, a correlation model of the fault and the equipment operation parameters is called. If the possible fault reason can not be determined, the reason is compared with the automatic analysis process result, and therefore the probability, the total loss value and the emergency degree priority of each fault are determined. And then, storing the analysis result into a point inspection database and giving an alarm. If possible failure reasons can be determined, the probability of each failure, the total loss value and the priority of the emergency degree are determined.

As shown in fig. 8, the second spot inspection analysis process includes:

and when the timing time is up, acquiring audio and video data and extracting audio and video characteristic values. And then calling an audio and video characteristic value and fault correlation model. And if the work is determined to be abnormal, determining the probability of each fault, the total loss value and the priority of the emergency degree. And then, storing the analysis result into a point inspection database and giving an alarm when the analysis result is abnormal. If the analysis result is normal, the analysis result is stored in the point inspection database and an alarm can be sent out. When the timing time is reached, the sensor parameters are read. It is determined whether the sensor parameter is out of range. If the range is exceeded, determining each fault probability, total loss value and emergency degree priority. Otherwise, the analysis result can be stored in the point inspection database and an alarm can be given out when the abnormality occurs.

And 204, generating a point inspection task according to the analysis result and the real-time point inspection data.

In the step, the point inspection time and the point inspection period required by each point inspection item are obtained according to the probability of the fault, the total loss estimated value, the emergency degree priority of the fault and the point inspection item specification library. And then, generating a point inspection task according to the point inspection time and the point inspection period required by each point inspection item. In order to make the spot inspection task more accurate, the fault and abnormal knowledge base can be read to obtain the associated fault and solution of the fault. Then, when the point inspection task is generated, the point inspection task is generated according to the point inspection time and the point inspection period required by each point inspection item, the associated fault and the solution.

In conjunction with fig. 1, step 204 is implemented by using the peer task generation module. As shown in fig. 9, the spot inspection task generating module includes: checking a project specification library; a knowledge base of faults and abnormal conditions; and (5) checking a task analysis model. The point inspection item specification library records point inspection items, standard point inspection periods, point inspection cautions and the like of specific equipment. The fault and exception knowledge base records the possible associated faults and solutions of the point inspection item exceptions. The point inspection task analysis model is combined with the abnormal probability and the emergency degree priority of each point inspection item finally output by the point inspection analysis engine module, loss estimation and information in the point inspection item specification library to recalculate the point inspection time and period required by each point inspection item.

Usually, the spot inspection task needs to be performed by people, such as checking the appearance of machine parts, lubrication, cleanliness, visual observation of key parts, noise judgment, and the like. The system is very important for saving labor cost and finding out the opportunity of problem spot inspection in time. Although the fault can be timely eliminated through the excessively frequent point inspection, the number of the devices is large, and a large amount of manpower is needed to complete the point inspection, which is generally unrealistic. Although this problem can be solved by reducing the frequency of spot inspection, a failure of the apparatus or a process variation cannot be found in time, resulting in a loss of production. In the system, automatic point inspection and manual point inspection are combined, and the manual point inspection is combined in a timing and on-demand mode to solve the problems. The automatic point inspection adopts video and audio acquisition equipment and other sensors installed beside the equipment to perform data acquisition and analysis at regular time. And if the automatic point inspection cannot completely determine the fault reason, the automatic point inspection is completed in cooperation with manual timing or point inspection according to requirements. The key of manual spot inspection is to determine when and what items to spot inspect, and meanwhile, the method can guide how to spot inspect each item, so that key factors are prevented from being omitted. The system well solves the problems through the point inspection task generating module.

With reference to fig. 10, the flow of generating the spot inspection task includes:

and reading the point inspection database, and judging whether a fault alarm occurs or whether preset time is reached. If yes, reading the automatically recorded original data, and automatically analyzing the result, the probability of each fault, the total loss value and the priority of the emergency degree. And according to the information, recalculating and updating the spot inspection task period. Meanwhile, the fault and abnormal knowledge base can be read. And finally, generating a point inspection scheme and pushing the point inspection task containing the automatic point inspection to the point inspection intelligent terminal.

On the basis of the above embodiment, a spot inspection report can also be generated. Specifically, in this step, spot inspection record data is acquired; the spot inspection record comprises: and manually recording information, the real-time point inspection data and the analysis result, and then generating a point inspection report according to the analysis result and the point inspection record data.

With reference to fig. 1, in the embodiment of the present invention, the generation of the spot check report is performed by the spot check report generation module.

As shown in fig. 11, the spot check report generation module includes: the system comprises an information recording model, a spot inspection report template, a statistical analysis engine and an alarm and report pushing module. The information recording model is a unified data management model and comprises manual recording information, automatic recording information and automatic analysis processing information. The manually recorded information includes written description of the spot inspector's record, the original picture taken, audio, video, etc. The automatic recording information includes automatically read device-related or process-related spot inspection parameters. The automatic analysis processing information comprises an audio and video analysis result, a fault probability, loss prediction, point inspection processing suggestions and the like. The spot inspection report template uses a unified template and automatically fills the content recorded by the information recording model. The template may take many forms, such as mail, office documents, etc. The statistical analysis engine can perform statistical analysis on the point inspection records to form a daily report, a weekly report, a monthly report and the like. In addition, the statistical analysis engine can also analyze the alarm level and the alarm type according to the point inspection report and output the alarm level and the alarm type together with the point inspection report. And the alarm and report pushing module determines whether to immediately distribute the spot inspection report and the target party of report distribution according to the alarm level and the type. Typically, the reports are distributed to managers in corresponding departments, such as equipment-related alarms distributed to equipment managers, production-related alarms distributed to production managers, process managers, and the like.

In combination with the above description, it can be seen that the triggering precision of the spot inspection task can be improved by using the embodiment of the invention. Meanwhile, specific point inspection items can be prompted in the point inspection task, so that the point inspection efficiency is improved. In addition, in the embodiment of the invention, the generated spot inspection report information is comprehensive, can be automatically pushed to spot inspection personnel, and gives an alarm when the spot inspection is abnormal. Through experiments, by pushing the spot inspection task and automatically performing spot inspection according to needs, 80% of manual spot inspection cost can be saved. Due to the fact that the spot inspection effectiveness is improved, the utilization rate of equipment can be improved by more than 30%, and the reject ratio of products can be reduced by more than 20%.

As shown in fig. 12, the spot inspection method according to the embodiment of the present invention may include:

step 1201, acquiring the identification of the point inspection equipment.

Step 1202, a point inspection task is obtained according to the identification of the point inspection equipment, and the point inspection task is generated according to historical point inspection data and real-time point inspection data.

And 1203, outputting the point inspection task.

In the embodiment of the present invention, the identifier of the pointing device is provided by the device side. Referring to fig. 1, the embodiment of the invention is applied to a point inspection intelligent terminal. After the identification of the point inspection equipment is obtained, the point inspection intelligent terminal can obtain the point inspection task corresponding to the point inspection equipment through the point inspection management platform according to the identification. The point inspection management platform generates a point inspection task according to the description of the foregoing embodiment. After the point inspection task is obtained, the point inspection intelligent terminal sends the point inspection task to a point inspection person.

As shown in fig. 13, the intelligent point inspection terminal includes: the system comprises a point inspection application program, a video acquisition module, an audio and video acquisition module, other sensor interfaces and a Bluetooth/WIFI (Wireless Fidelity)/4G interface. The video acquisition module utilizes the camera to photograph or record the equipment condition and the production condition. The audio acquisition module can shoot noise generated during the production of the equipment. The reserved other sensor interfaces can be connected with more sensors in a wireless mode, such as a temperature sensor and a vibration sensor. The Bluetooth/WIFI/4G interface is used for communicating with the point inspection management platform, the wireless sensor and the equipment side Bluetooth module.

A low-power Bluetooth beacon (beacon) module is installed on the device side, and the module can broadcast beacon packets containing unique IDs at regular time. The module can be used for identifying point inspection personnel in real time. Can measure the distance of point inspection intelligent terminal and equipment side bluetooth module through beacon package to guarantee that personnel are near this equipment all the time when the operation is examined to the point. In addition, the point inspection intelligent terminal can automatically log in after receiving the ID of the equipment, requests a point inspection task and data from the point inspection management platform, and fills the downloaded point inspection task with automatically acquired data in advance. The low-power-consumption Bluetooth beacon module can be powered by a battery, extra wiring is not needed, and the workload of installation and deployment is greatly reduced.

And after the point inspection personnel acquire the point inspection task, visually inspecting the point inspection part according to the task instruction, inputting a processing record in a text mode according to the prompting audio recording and video recording, and uploading a point inspection result.

With reference to fig. 14, the workflow of the point inspection intelligent terminal may include:

and the point inspection intelligent terminal detects the ID sent by the equipment side Bluetooth module and performs authentication. And if the authentication is passed, the point inspection intelligent terminal sends the ID to the point inspection management platform and receives the corresponding task sent by the point inspection management platform. Wherein in the task the system automatically fills in the acquired data. And then, prompting the spot inspection task to a spot inspection person. And the point inspection personnel acquire audio and video according to the point inspection task. The point inspection intelligent terminal sends the received audio and video, images, manually recorded data and the like to a point inspection management platform, the point inspection management platform analyzes the data, generates a point inspection record in a point inspection record library, and sends the point inspection record to a point inspection report generation module.

The following describes an implementation process of the embodiment of the present invention in detail with reference to specific embodiments.

The finish rolling is the last process of steel plate rolling and is also a key link for ensuring the quality of the steel plate. Here, a point inspection using a hydraulic system of a finishing mill AGC (Automatic Gain Control) will be described as an example. The hydraulic AGC system is an important component of a rolling mill and is also a key system for accurately controlling the thickness of a steel plate. The system collects feedback data of the sensor and then adjusts the roll gap value of the roller by controlling the up-and-down moving stroke of the two hydraulic cylinders through the electro-hydraulic servo valve, thereby controlling the thickness stability of the final rolled piece. The traditional point check items and check periods of the AGC system range from 24 hours to 1 week, and the point check method comprises visual inspection, audition, touch and instrument measurement. The embodiment of the invention improves the traditional point inspection mode, changes the periodic point inspection into the system pushing the point inspection task items and gives the suggestion of the point inspection. Table 1 is a conventional AGC hydraulic system point check table.

TABLE 1

In this example, the following process is mainly included:

1. data acquisition and storage process

And collecting related data by using a data collection module. Whether the AGC system normally operates or not directly affects the quality of a rolled piece, so that the quality detection parameters of the rolled piece are read to judge whether the AGC system normally operates, and meanwhile, the process parameters and the equipment operation parameters are read to analyze whether the AGC system normally operates and a spot inspection plan.

Referring to fig. 15, the quality test result of each batch of products is collected in real time, and the test result may be a real-time value of an on-line measuring device or a laboratory test result. If the detection result is abnormal, the system automatically triggers the reading process. The specific reading flow is shown in fig. 15.

The thickness out-of-tolerance is an important quality parameter of the steel plate, and means that the actual thickness of the steel plate on the longitudinal and transverse surfaces exceeds the allowable deviation value specified by the standard. If the quality inspection department concludes that the thickness of the steel plate is out of tolerance after inspecting the product, the system reads the thickness data and the batch number of each detection point of the detected steel plate, and then records the detection conclusion, such as whether the detection conclusion is caused by the incoming material characteristic error or the rolling process. If the rolling process is caused, the system reads production process parameters such as coil number, alloy mark, inlet thickness, outlet thickness, expected rolling force, uncoiling tension, coiling tension, incoming material width, coil rolling time, pass rolling time and the like in a production management system according to the batch number of the sample. And then reading the production running data stream of the equipment according to the production starting time and the production ending time, wherein the production running data stream comprises the given linear velocity of a host computer, the given rolling force, the given position of a transmission side, the given position of an operation side, the actual linear velocity of the host computer, the actual position of the transmission side, the actual position of the operation side, the actual tension of an inlet, the actual tension of an outlet, the measured value of an outlet thickness gauge, the measured value of temperature, the measured value of pressure of an oil cylinder. And finally storing the data in a real-time database.

In addition to reading parameters after a quality problem occurs, the system also periodically reads equipment parameters and equipment-side sensor parameters at regular time intervals and stores the equipment parameters and the equipment-side sensor parameters in a database so as to facilitate data analysis. The device side sensor in the example comprises a servo valve, an oil pump temperature sensor, an oil pump working video and noise acquisition device.

2. Point inspection task analysis workflow

And after the data acquisition process is finished, the point inspection analysis engine starts the analysis process. The analysis process is divided into timing analysis and quality data out-of-tolerance triggering analysis.

2.1. Quality data out-of-tolerance trigger analysis:

the main reasons for the thickness out-of-tolerance are: roll system factors: the eccentricity of the roller, the abrasion of the roller, the bending of the roller, the thermal expansion of the roller, the change of the thickness of an oil film and the overlarge ovality of a roller bearing; the material supply factors are as follows: the thickness, width and hardness of the incoming material are changed, and the friction coefficient of a rolling area is changed; the parameters of the rolling process are changed: rolling force, tension, rolling speed variation; detecting the error of the instrument: zero drift or too large measurement error of the thickness gauge and the thermometer; the automatic thickness control (AGC) system is maladjusted. The quality inspection department firstly eliminates the steel plate thickness out-of-tolerance caused by the variation of the incoming material characteristics through material analysis, and the remaining possible factors are four types of roll system factors, the variation of parameters in the rolling process, the error of a detection instrument and the action disorder of an automatic thickness control system. The thickness error caused by the roll system factor and the zero drift of the measuring instrument has obvious trend, the thickness change is obvious periodicity caused by the roll system factor error, the spacing distance is consistent with the perimeter of the roll, and the thickness error is obvious consistency, such as thin or thick, caused by the instrument zero drift error. So the analysis is first performed by a quality and fault correlation model.

The flow is shown in FIG. 16. And reading thickness data of each monitoring point, and calling a fault and quality correlation model. And if the fault reason is determined, reading the operation parameters of the process and the equipment. If the parameter set values of the process and equipment operation parameters are abnormal, the analysis result can be stored in the point inspection database and an alarm can be sent out. If the parameter setting values of the process and equipment operation parameters are normal, a correlation model of the fault and the equipment operation parameters is called. If the possible fault reason can not be determined, the reason is compared with the automatic analysis process result, and therefore the probability, the total loss value and the emergency degree priority of each fault are determined. And then, storing the analysis result into a point inspection database and giving an alarm. If possible failure reasons can be determined, the probability of each failure, the total loss value and the priority of the emergency degree are determined.

2.2 timing analysis:

for the items with the sensors, the system can perform regular automatic point inspection analysis according to a standard point inspection table, the period of the standard point inspection table is the conventional manual point inspection period, and automatic point inspection does not occupy manpower, so that the period can be set to be very short, for example, once per minute, and the probability that equipment faults cannot be found in time is greatly reduced. The sample system is provided with oil temperature sensors for the oil pump and the valve table in items 2-1 and 3-1 of the standard point inspection table, and the oil temperature is acquired and whether the temperature is normal or not is judged every time of automatic point inspection. And for the point inspection item 3-1, a high-definition camera and a professional microphone are installed near the oil pump, and the point inspection engine analyzes the running noise and the video of the oil pump every minute so as to find abnormal working conditions such as leakage, abnormal sound and cracks of the oil pump in time.

The flow is shown in FIG. 17. And when the timing time is up, reading the audio and video of the oil pump, extracting the audio and video characteristics, and calling an audio and video characteristic value and fault correlation model. And judging the leakage of the oil pump, and if the leakage occurs, determining the failure probability, total loss estimation, emergency degree priority and the like. And then storing the analysis result into a point inspection database. If abnormal, it can send out alarm. And if the leakage does not occur, storing the analysis result into the point inspection database. When the timing time is up, the temperature sensor parameter can be read. And if the oil temperature exceeds the range, determining the fault probability, total loss prediction, emergency degree priority and the like. And then storing the analysis result into a point inspection database. If abnormal, it can send out alarm. And if the range is not exceeded, storing the analysis result into the point inspection database.

Analyzing through a fault and quality correlation model:

(1) and an analysis model establishing stage, namely reading the thickness data of each detection point of the detected steel plate by a system, marking the data in a coordinate system of an X/Y axis, and then normalizing the data to obtain a complete three-dimensional model of the sample. The system uses a classification algorithm for data and adds labels to the data according to actual detection conclusions, such as thickness out-of-tolerance caused by roller abrasion, thickness out-of-tolerance caused by overlarge ovality of a roller bearing, out-of-tolerance caused by zero drift of a thickness meter, out-of-tolerance caused by other unknown reasons and the like. The well established model is then validated using more quality control data.

(2) Inputting the new quality inspection data into the model to predict the possible failure cause. The roll system factors and the zero drift error of the instrument can be accurately predicted, so that a preliminary conclusion can be obtained only by analyzing quality data. If the system gives the conclusion that the roller bearing ovality is too large in the analysis result. For the thickness out-of-tolerance distribution condition with unobvious regularity, the fault reasons which are difficult to accurately predict only by the quality detection data can be classified as other fault reasons.

3. Point inspection task analysis workflow

For other causes of failure, a combination of process and equipment data is needed for analysis. The system in the previous flow has stored the process parameters and the equipment parameters. The system firstly inputs the technological parameters into a rolling mill digital model to verify whether the dynamic set value of the equipment is in a threshold range or not, and thickness out-of-tolerance caused by parameter change in the rolling process is eliminated. If everything is in order, the remaining possibility is the thickness out-of-tolerance caused by the mis-action of the AGC automatic thickness control system. It is then necessary to perform an analysis using a model of the dependence of the fault on the operating parameters of the plant. The AGC system ensures that the thickness of an outlet steel plate is constant by controlling the rolling force to be constant, and the main influence factors of the rolling force are as follows: absolute reduction, chemical composition of a rolled piece, roller diameter, width of the rolled piece, thickness of the rolled piece, temperature of the rolled piece, friction coefficient and rolling speed. The AGC system controls the hydraulic oil cylinders on two sides of the roller to act through a built-in mathematical model, a set value of process parameters and a sensor acquisition value so as to control the absolute reduction to compensate the change of other parameters. The action of the oil cylinder is mainly controlled by the opening of the servo valve, and the AGC system achieves the purpose of controlling the displacement of the oil cylinder by giving instructions to the servo valve. The previous analysis has eliminated the process parameter setting problem, so that excessive thickness differences may be caused by malfunctions of the servo valves and hydraulic systems of the AGC system. Common faults of AGC servo and hydraulic systems are displacement sensor faults, leakage of an oil cylinder or a servo valve, zero offset of the servo valve and the like. Because the two hydraulic oil cylinders receive the same instruction, if the two systems work normally, the displacement of the two oil cylinders should be within a fixed error. If one of the servo valves or the cylinders has a potential failure, the synchronous performance of the actions of the two cylinders is deteriorated. When the overflow valve is not operated, the dynamic response characteristic of the oil cylinder is only related to the opening degree of the servo valve. The output current of the servo valve controls the opening degree of the servo valve. Therefore, normally, the synchronization error of the two cylinders and the servo valve current should have a fixed characteristic curve. Whether the AGC servo and the hydraulic system work normally or not is analyzed through a fault and equipment operation parameter correlation model.

Analyzing through a fault and equipment operation parameter correlation model:

(1) modeling using a historical data set: firstly, selecting a sampling time interval delta t, calculating the maximum displacement difference value of the two oil cylinders in the delta t time, and driving the average value of output currents of the servo valve in the delta t time and operating the servo valve in the delta t time. The data is marked as normal data, servo valve abrasion, oil cylinder leakage and the like. The model is then validated using the test data set.

(2) Inputting corresponding equipment data into a previous model, classifying by using a KNN (K-Nearest Neighbor) algorithm, and determining problem causes such as cylinder leakage probability p1, servo valve clamping stagnation probability p2 and oil pump leakage probability p 3.

Finally, a part of possible faults is eliminated by the results of the preceding automatic spot check. Such as video and noise analysis elimination through the operation of the oil pump, and the leakage fault probability of the oil pump is reduced or improved.

If the possibility of the fault still cannot be determined through the correlation model of the fault and the equipment operation parameters, the system enters other processing flows. Manual procedures are typically required for process analysis.

And finally, storing the analysis result and the original data into a point inspection database. If alarm occurs, according to the priority level, the alarm record is pushed to the relevant client and the point inspection task generation module. And a client user such as an operator judges whether to respond immediately according to the alarm condition. And the point inspection task generation module recalculates the point inspection period according to the alarm priority and the calculated loss value.

3. Point inspection task analysis engine workflow

Now, the probability of each fault possibly occurring is obtained, and then comprehensive analysis is carried out to estimate loss calculation and emergency degree evaluation.

The comprehensive analysis process comprises the following steps:

(1) the system calculates an estimated loss value which may be caused by each fault item through historical data. The loss prediction value comprises loss c1 caused by product defects, equipment loss c2 and personnel safety loss c3 … …, wherein each loss value is endowed with a weighted value a1 and a2 … … an, and finally multiplied by the probability p of fault occurrence. Thus, the total possible loss due to the first possible failure is calculated as: c_total1＝(a1*c1+a2*c2+…+an*cn)*p1。

(2) And adding the total loss values brought by all the possible fault terms to obtain a total loss estimated value: c_total＝C_total1+C_total2+…+C_totaln。

(3) And filling the occurrence probability into an emergency degree priority corresponding table, and then inquiring the corresponding emergency degree priority. If the urgency priority is high, an alert will be automatically sent to the operator client. The specific algorithm example refers to the urgency priority correspondence table of table 2, in which the urgency is calculated in advance based on historical experience.

TABLE 2

4. Point inspection task generation process

This flow is shown in fig. 18. If an alarm occurs in the analysis engine and the fault cannot be accurately located through automatic analysis, the method is carried outThe following process is used for updating the manual spot inspection task. And reading the automatically recorded original data, automatically analyzing the result, predicting the loss, and updating the point inspection period according to the grading result of the point inspection item. Specifically, the spot check period is dynamically adjusted through the calculated possible total loss value and the spot check item urgency priority: if C is present_total>C_limit(C_limitThe minimum total loss value representing the high-level definition of the emergency degree) directly sets the item emergency degree to be high level, and immediately generates a point inspection task; if C is present_total<C_limitAnd C_total>C_th1(C_th1A total loss value threshold value indicating the use of the cycle shortening factor f 1), and the cycle shortening factor f1 (0)<f1<1) The updated spot inspection period is as follows: scheduled spot check cycle f1, the updated next spot check time is updated as: last time point inspection time + updated point inspection period. And if the updated next point inspection time is earlier than the current time, the system immediately generates a point inspection task. The system administrator can set a plurality of threshold values and point inspection cycle shortening factors according to the situation.

After the system determines the push period and the point inspection items, historical records and processing methods are searched from the knowledge base and attached to the point inspection items. The intelligent terminal of the checking personnel receives the checking schedule and the detailed checking instruction book. If the spot inspection task item contains the automatically completed record and analysis, the spot inspection task item is also pushed to the spot inspection terminal along with the spot inspection task.

5. Point inspection process

This flow is shown in fig. 19. And the point inspection terminal automatically receives Beacon with the equipment ID sent by the Bluetooth module at the equipment side. After the ID of the equipment needing point inspection is compared by the point inspection terminal and is consistent with the ID sent by the Bluetooth module, the point inspection personnel are automatically authenticated, and a detailed point inspection interface appears. And performing video, audio acquisition, vibration measurement and the like on the position to be checked by the checking personnel according to the requirements of the terminal checking task specification. And the point inspection intelligent terminal sends the equipment ID to the point inspection management platform. And issuing the point inspection task book of the equipment by the point inspection management platform.

And the point inspection personnel push the received point inspection task. For the oil cylinder leakage, the clamping failure of a servo valve and the oil pump leakage which are possible to be failed, a spot inspection task specification requires to shoot videos (spot inspection standard table items 1-2) and valve working videos (spot inspection standard table items 2-1) of oil pipe joints of two oil cylinders, record valve working sounds (spot inspection standard table items 2-1) and read valve temperature values (spot inspection standard table items 2-1). The working temperature of the valve is measured by the equipment sensor in real time, the audio and video automatic analysis result of the oil pump is obtained, and the system automatically fills the read temperature value and the audio and video analysis result of the oil pump into a task to be checked. And the point inspection personnel holding point inspection terminal arrives at the point inspection equipment according to the preset time.

After the point inspection of the operator is completed, the system automatically analyzes the collected oil cylinder oil pipe joint video, valve working video and valve working sound by using an audio and video analysis engine, and provides an analysis conclusion whether the work is normal or not and whether the maintenance is needed or not.

The point inspection report analysis module integrates the point inspection original data record, the audio and video analysis results, the point inspection engine generates a final point inspection report for the analysis results of the fault probability, various potential loss values and the like.

Referring to fig. 20, fig. 20 is a structural diagram of a spot inspection apparatus according to an embodiment of the present invention, and as shown in fig. 20, a terminal device 2000 includes:

a first obtaining module 2001, configured to obtain historical spot inspection data; a second obtaining module 2002, configured to obtain real-time point inspection data; an analysis module 2003, configured to analyze the historical point inspection data and the real-time point inspection data to obtain an analysis result; and a task generating module 2004, configured to generate a point inspection task according to the analysis result and the real-time point inspection data.

Optionally, as shown in fig. 21, the apparatus further includes:

a third obtaining module 2005, configured to obtain spot inspection record data; the spot inspection record comprises: manually recording information, the real-time point inspection data and the analysis result; and a report generating module 2006, configured to generate a spot inspection report according to the analysis result and the spot inspection record data.

Optionally, the real-time spot inspection data includes: real-time operation parameters of production, real-time operation parameters of equipment, quality parameters and point inspection data input through a point inspection intelligent terminal; the quality parameters comprise at least one of online quality parameters and offline quality parameters; the second obtaining module is specifically configured to obtain the production real-time operation parameters, the equipment real-time operation parameters, the quality parameters, and the point inspection data input through the point inspection intelligent terminal, respectively, according to a predetermined time interval.

Optionally, the real-time spot inspection data includes: real-time operation parameters of production, real-time operation parameters of equipment, quality parameters and point inspection data input through a point inspection intelligent terminal; the quality parameters comprise at least one of online quality parameters and offline quality parameters; as shown in fig. 22, the second obtaining module 2002 includes:

a first obtaining submodule 2201, configured to obtain the quality parameter;

a second obtaining sub-module 2202, configured to obtain quality detection data when it is determined that a quality problem occurs according to the quality parameter;

a third obtaining submodule 2203, configured to obtain a production parameter according to the quality detection data;

a fourth obtaining submodule 2204, configured to obtain an apparatus parameter according to the production parameter;

the fifth obtaining submodule 2205 is configured to obtain the point inspection data input through the point inspection intelligent terminal.

Optionally, the second obtaining module may be further configured to obtain audio/video characteristic data of the device-side sensor.

Optionally, as shown in fig. 23, the analysis module 2003 includes:

the first processing submodule 2301 is configured to input a device real-time operation parameter, an audio/video characteristic value, and a quality parameter in the real-time point inspection data into a point inspection analysis model, where the point inspection analysis model is updated by using the historical point inspection data;

and the analysis submodule 2302 is used for analyzing the real-time operation parameters, the audio and video characteristic values and the quality parameters of the equipment by using the point inspection analysis model to obtain an analysis result.

Optionally, as shown in fig. 24, the analysis module 2003 further includes: the second processing sub-module 2303 is used for inputting the production real-time operation parameters and the equipment real-time operation parameters in the real-time point inspection data into an equipment digital model to obtain an automatic analysis result; the analysis submodule 2302 is specifically configured to analyze the real-time operation parameter, the audio/video characteristic value, and the quality parameter of the device by using the point inspection analysis model to obtain a pre-analysis result, and combine the pre-analysis result and the automatic analysis result to obtain an analysis result.

As shown in fig. 25, the analysis sub-module 2302 includes:

a first analysis unit 2501, configured to invoke a fault and quality parameter correlation model to analyze the quality parameter;

a first determination unit 2502, configured to determine, if the cause of the failure is determined, a probability of the failure, a total loss prediction value, and an urgency priority of the failure;

a second determining unit 2503, configured to analyze the real-time device operating parameter and the real-time production operating parameter when a cause of a fault is not determined; calling a correlation model of the fault and the equipment operation parameters to estimate the fault under the condition that the parameter set values of the equipment real-time operation parameters and the production real-time operation parameters are normal; and comparing the estimated faults with the automatic analysis result to determine the probability of the faults, the total loss prediction value and the emergency degree priority of the faults.

As shown in fig. 26, the analysis sub-module 2302 includes:

the third determining unit 2601 is configured to, when the timing time arrives, invoke an audio/video characteristic value and a fault correlation model to analyze the audio/video characteristic value; determining the probability of the fault, the total loss predictive value and the emergency degree priority of the fault under the condition that the analysis result is abnormal;

a fourth determining unit 2602, when the timing time arrives, obtains the parameter of the device-side sensor, and determines the probability of the failure, the predicted total loss value, and the priority of the urgency of the failure when the parameter of the device-side sensor exceeds the preset parameter range.

As shown in fig. 27, the task generation module 2004 may include:

the first obtaining sub-module 2701 is configured to obtain a point inspection time and a point inspection period required by each point inspection item according to the probability of the fault, the total loss prediction value, the emergency degree priority of the fault, and the point inspection item specification library;

the generating sub-module 2702 is configured to generate a spot check task according to the spot check time and the spot check period required by each spot check item.

As shown in fig. 28, the task generation module 2004 may further include:

a second obtaining sub-module 2703, configured to read a fault and exception knowledge base, and obtain a solution and a related fault of the fault;

the generating sub-module 2702 is specifically configured to generate a point inspection task according to the point inspection time and the point inspection period required by each point inspection item, and the associated fault and solution.

As shown in fig. 29, the report generation module 2006 includes:

a populating sub-module 2901 configured to populate the spot check record into a spot check report template;

a first analysis sub-module 2902, configured to perform statistical analysis on the analysis result to obtain a report analysis result;

a generating sub-module 2903, configured to generate a spot inspection report according to the content of the spot inspection report template and the report analysis result.

As shown in fig. 30, the report generating module 2006 further includes:

a second analysis sub-module 2904, configured to perform statistical analysis on the spot inspection record, and obtain an alarm report, where the alarm report includes at least one of an alarm level and an alarm type.

The working principle of the device of the embodiment of the invention can refer to the description of the embodiment of the method.

As shown in fig. 31, a spot inspection apparatus 3100 according to an embodiment of the present invention includes:

a first obtaining module 3101, configured to obtain an identifier of the spot inspection apparatus;

a second obtaining module 3102, configured to obtain a spot inspection task according to the identifier of the spot inspection device, where the spot inspection task is generated according to historical spot inspection data and real-time spot inspection data;

a processing module 3103, configured to output the spot inspection task.

As shown in fig. 32, the apparatus further includes at least one of the following modules:

an authentication module 3104, configured to authenticate the checkup user to ensure that the checkup user is located within a preset range of the checkup device;

a receiving module 3105, configured to receive the click data input by the click user.

As shown in fig. 33, the spot inspection apparatus according to the embodiment of the present invention includes: a processor 3301, a transceiver 3302, a memory 3303, and a bus interface, wherein:

in this embodiment of the present invention, the spot inspection apparatus 3300 further includes: a computer program stored on the memory 3303 and executable on the processor 3301, the computer program when executed by the processor 3301 implementing the steps of:

acquiring historical spot inspection data;

acquiring real-time point inspection data;

In FIG. 33, the bus architecture may include any number of interconnected buses and bridges, with various circuits being linked together, particularly one or more processors represented by processor 3301 and memory represented by memory 3303. The bus architecture may also link together various other circuits such as peripherals, voltage regulators, power management circuits, and the like, which are well known in the art, and therefore, will not be described any further herein. The bus interface provides an interface. The transceiver 3302 may be a number of elements including a transmitter and a receiver that provide a means for communicating with various other apparatus over a transmission medium.

The processor 3301 is responsible for managing the bus architecture and general processing, and the memory 3303 may store data used by the processor 3301 when performing operations.

Optionally, the computer program when executed by the processor 3303 may further implement the steps of:

The real-time point inspection data comprises: real-time operation parameters of production, real-time operation parameters of equipment, quality parameters and point inspection data input through a point inspection intelligent terminal; the quality parameters comprise at least one of online quality parameters and offline quality parameters;

The real-time point inspection data comprises: real-time operation parameters of production, real-time operation parameters of equipment, quality parameters and point inspection data input through a point inspection intelligent terminal; the quality parameters comprise at least one of online quality parameters and offline quality parameters; optionally, the computer program when executed by the processor 3303 may further implement the steps of:

acquiring the quality parameter;

acquiring production parameters according to the quality detection data;

acquiring equipment parameters according to the production parameters;

inputting the production real-time operation parameters and the equipment real-time operation parameters in the real-time point inspection data into an equipment digital model to obtain an automatic analysis result;

filling the point inspection record into a point inspection report template;

As shown in fig. 34, the intelligent point inspection terminal according to the embodiment of the present invention includes: the processor 3400 is used for reading the program in the memory 3420 and executing the following processes:

acquiring an identifier of the point inspection equipment;

and outputting the point inspection task.

A transceiver 3410 for receiving and transmitting data under the control of the processor 3400.

In fig. 34, among other things, the bus architecture may include any number of interconnected buses and bridges, with one or more processors represented by the processor 3400 and various circuits of memory represented by the memory 3420 being linked together. The bus architecture may also link together various other circuits such as peripherals, voltage regulators, power management circuits, and the like, which are well known in the art, and therefore, will not be described any further herein. The bus interface provides an interface. The transceiver 3410 may be a number of elements including a transmitter and a receiver that provide a means for communicating with various other apparatus over a transmission medium. The user interface 3430 may also be an interface capable of interfacing with a desired device for different user devices, including but not limited to a keypad, display, speaker, microphone, joystick, etc.

The processor 3400 is responsible for managing a bus architecture and general processing, and the memory 3420 may store data used by the processor 3400 in performing operations.

The processor 3400 is further configured to read the computer program and perform at least one of the following steps:

and receiving the point inspection data input by the point inspection user.

Furthermore, a computer-readable storage medium of an embodiment of the present invention stores a computer program, which is executable by a processor to implement the steps of any of the method embodiments described above.

In the several embodiments provided in the present application, it should be understood that the disclosed method and apparatus may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.

In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be physically included alone, or two or more units may be integrated into one unit. The integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.

The integrated unit implemented in the form of a software functional unit may be stored in a computer readable storage medium. The software functional unit is stored in a storage medium and includes several instructions to enable a computer device (which may be a personal computer, a server, or a network device) to execute some steps of the transceiving method according to various embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the appended claims.

Claims

1. A spot inspection method is characterized by comprising the following steps:

acquiring historical spot inspection data;

acquiring real-time point inspection data;

2. The method of claim 1, wherein after said obtaining an analysis result, the method further comprises:

3. The method of claim 1, wherein the real-time spot check data comprises: real-time operation parameters of production, real-time operation parameters of equipment, quality parameters and point inspection data input through a point inspection intelligent terminal; the quality parameters comprise at least one of online quality parameters and offline quality parameters;

4. The method of claim 1, wherein the real-time spot check data comprises: real-time operation parameters of production, real-time operation parameters of equipment, quality parameters and point inspection data input through a point inspection intelligent terminal; the quality parameters comprise at least one of online quality parameters and offline quality parameters;

acquiring the quality parameter;

acquiring production parameters according to the quality detection data;

acquiring equipment parameters according to the production parameters;

5. The method of claim 3 or 4, wherein the obtaining real-time spot check data further comprises:

6. The method of claim 5, wherein analyzing the historical point inspection data and the real-time point inspection data to obtain an analysis result comprises:

7. The method of claim 6, wherein before analyzing the real-time operation parameters, the audio/video characteristic values and the quality parameters of the equipment by using the point inspection analysis model to obtain an analysis result, the method further comprises:

8. The method according to claim 7, wherein the analyzing the real-time operation parameters, the audio/video characteristic values and the quality parameters of the equipment by using the point inspection analysis model to obtain an analysis result comprises:

9. The method according to claim 8, wherein the analyzing the real-time operation parameters, the audio/video characteristic values and the quality parameters of the equipment by using the point inspection analysis model to obtain a pre-analysis result, and combining the pre-analysis result and the automatic analysis result to obtain an analysis result comprises:

10. The method according to claim 8, wherein the analyzing the real-time operation parameters, the audio/video characteristic values and the quality parameters of the equipment by using the point inspection analysis model to obtain a pre-analysis result, and combining the pre-analysis result and the automatic analysis result to obtain an analysis result comprises:

11. The method according to claim 9 or 10, wherein generating a spot check task according to the analysis result and the real-time spot check data comprises:

12. The method of claim 11, wherein generating a point inspection task based on the analysis results and the real-time point inspection data further comprises:

13. The method of claim 2, wherein generating a spot check report based on the analysis results and the spot check record data comprises:

filling the point inspection record into a point inspection report template;

14. The method of claim 13, wherein after said performing a statistical analysis on said spot check record and obtaining a report analysis result, further comprising:

15. A spot inspection method is characterized by comprising the following steps:

acquiring an identifier of the point inspection equipment;

and outputting the point inspection task.

16. The method of claim 15, further comprising at least one of:

and receiving the point inspection data input by the point inspection user.

17. A spot inspection device, comprising:

18. The apparatus of claim 17, further comprising:

19. A spot inspection device, comprising:

and the processing module is used for outputting the point inspection task.

20. The apparatus of claim 19, further comprising at least one of: