CN102903009B - Malfunction diagnosis method based on generalized rule reasoning and used for safety production cloud service platform facing industrial and mining enterprises - Google Patents

Malfunction diagnosis method based on generalized rule reasoning and used for safety production cloud service platform facing industrial and mining enterprises Download PDF

Info

Publication number
CN102903009B
CN102903009B CN201210370577.2A CN201210370577A CN102903009B CN 102903009 B CN102903009 B CN 102903009B CN 201210370577 A CN201210370577 A CN 201210370577A CN 102903009 B CN102903009 B CN 102903009B
Authority
CN
China
Prior art keywords
data
diagnosis
task
platform
service
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.)
Active
Application number
CN201210370577.2A
Other languages
Chinese (zh)
Other versions
CN102903009A (en
Inventor
陈明新
郑俊褒
方绪群
杨俊杰
张秀文
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
ZHEJIANG TOPINFO TECHNOLOGY Co Ltd
Original Assignee
ZHEJIANG TOPINFO TECHNOLOGY Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by ZHEJIANG TOPINFO TECHNOLOGY Co Ltd filed Critical ZHEJIANG TOPINFO TECHNOLOGY Co Ltd
Priority to CN201210370577.2A priority Critical patent/CN102903009B/en
Publication of CN102903009A publication Critical patent/CN102903009A/en
Application granted granted Critical
Publication of CN102903009B publication Critical patent/CN102903009B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Abstract

A malfunction diagnosis method based on generalized rule reasoning and used for a safety production cloud service platform facing industrial and mining enterprises includes extracting fundamental characteristic information or characteristic set, processing the characteristic information to converting the characteristic information into a scheme for resolving problems under local environment or optimal constraint condition, and converting the scheme into immune operator in proper mode for generating a new individual. Immune evolution is achieved through vaccination and immunoselection based on knowledge and knowledge transfer on the base that immune vaccine is reasonably extracted to effectively increase priori knowledge for malfunction diagnosis and improve individual fitness.

Description

A kind of safety production cloud service platform for towards industrial and mining establishment based on broad sense advise The then abnormality diagnostic method of reasoning
Technical field
The invention belongs to the information gathering of industrial and mineral system automation and control field, be related to power system real-time data acquisition, Process and control system, be based especially on industrial and mineral real time data integrated processing system and the method for designing of cloud computing.
Background technology
Cloud being widely used in the Internet, mainly having included three kinds of different types, software has serviced SaaS, puts down Platform services PaaS, and architecture services IaaS.Wherein PaaS provides the complete or partial application that user can access Program development, SaaS then provide the complete application program that can be used directly, and such as manage ERM by Internet. And the cloud in IaaS is with Mass Data Management technology, mass data distribution memory technology, Intel Virtualization Technology, cloud computing platform Management technique is the most key.Wherein Mass Data Management technology and distributed storage technology are the important composition portions of data processing Point, cloud computing needs distribution, magnanimity data are carried out processing, analyzed, therefore, data management technique is allowed for efficiently Substantial amounts of data are managed, the mode for also needing to redundant storage in addition ensures the reliability of data.Widely use in cloud computing system Data-storage system be Google GFS and Hadoop team exploitation increasing income for GFS realize HDFS.
The safety in production of industrial and mining establishment necessarily has mass data (such as various safety in production marks in actual production run The accurate, data of monitoring, educational training knowledge etc.) need to process, cloud service platform adopts Distributed Calculation storage mode, will calculate Task is assigned to parallel processing on multiple stage machine, improves arithmetic speed with this.Cloud platform will safety in production enterprise practical production product Tired data message carries out unified management and storage, be production during provide based on work(such as the prediction of data, abnormality detections Can, realize the safety in production of enterprise.
The data acquisition function of traditional industrial and mining establishment and specific logic judging function are tightly combined, generally by more single Hardware device complete, therefore, complete different functions and be accomplished by configuring different equipment or system, these difference in functionalitys set Standby or system has each independent data acquisition unit.Each data acquisition exist overlapping collection, utilization rate it is not high, Data and information content is inconsistent, time disunity the problems such as, define so that longitudinal level is more, horizontal system mostly is principal character " information island ", constrain information further fusion and application, data and information repeated acquisition and repeat transmission process The waste of various resources certainly will be caused.Instrument in most of industrial and mining establishment is divided into three classes:Protection class, observing and controlling class measure class, he Have the data acquisition and processing (DAP) unit of oneself, can be connected with monitoring and display system, for underlying parameter measure.But there is also Following technical problem:
(1) precision of instrument is different, and it is inconsistent to cause data, also uneconomical on cost.
(2) also different in different instrument, data that instrument to collect frequency is different, algorithm difference causes same value Redundancy and confusion.
(3) system and equipment, communication difficulties between equipment and equipment lack unified standard.
(4) process data disappearance, during sampling, generates substantial amounts of process data, and these data are for the later stage Accident analysiss or operation expanding be all valuable, but we are obtained from instrument is all through n times computing " second-hand number According to ", " simplifying value ".The problem that this separate configurations, independent calculating, the device of standalone feature bring is that information sharing is poor, is utilized Rate is low and hardware and software problem of resource waste.
For solving above-mentioned problem, especially data are gathered respectively, data are processed respectively and calculated, data are difficult to altogether The key issue such as enjoy, it is necessary to set up a set of new framework and mechanism, realize that the data of same type equipment are once gathered, can be Various application work(such as observing and controlling, protection, phasor measurement, metering are realized using cloud computing technology under a whole set of unified software application platform Energy.
The content of the invention
It is an object of the invention to provide a kind of safety production cloud service platform system of industrial and mining establishment, to solve equipment independence Configuration, it is independent calculate, information sharing that the device of standalone feature brings is poor, utilization rate is low and the problem of the soft and hardware wasting of resources, Realize the once collection of same type device data, and various applications such as observing and controlling, protection, phasor measurement, metering.
The safety production cloud service platform system of described industrial and mining establishment, including:
1) safety production cloud service platform towards industrial and mining establishment is set up, the platform intergration supports safety in production service The functions such as mass data processing, production safety management business cooperation and system administration, are the safety in production and political affairs of industrial and mining establishment Mansion supervision provides important technical support.
2) security incident event Analysis of Multidimensional Association Rule technology is based on, can be gone out from analysis mining in the creation data of magnanimity Frequent factor and the potential rule of accident and near miss incidents generation may be caused, safety in production standard effectiveness and shortage of standard is set up The dynamic monitoring early warning analysis method of situation, is that government and supervision department are formulated based on risk, rational operation or work standard The foundation of offer science.
3) mass data processing method in industrial and mining establishment's safety in production is proposed, is specifically included based on support vector machine Safety in production abnormality detection technology, the knowledge discovering technologies based on generalization rule reasoning, can be carried out point to the mass data for producing Analysis and knowledge excavation, so as to help enterprise to lift accident prevention early warning and emergency disposal ability, improve enterprise safety operation level.
4) production safety management tool set is developed, facilitates the safety management of industrial and mineral industrial security production, supervision, configuration etc., Realize quick investigation, the quick response of enterprise.
The safety production cloud service platform towards industrial and mining establishment wherein developed, is integrated with and supports safety in production service cloud The functions such as mass data processing, production safety management business cooperation and system administration, realize to the real-time of enterprise's production safety Monitoring and early warning.
Specifically, a kind of safety production cloud service platform for towards industrial and mining establishment based on generalization rule reasoning Abnormality diagnostic method, it is characterised in that:The safety production cloud service platform system towards industrial and mining establishment, takes including secure cloud Business platform portal subsystem, system administration develop subsystem, application service layer subsystem, virtual resource straton with related tool collection System, platform service support layer subsystem, access and are adapted to layer subsystem, security service resource subsystem, infrastructure services Support layer subsystem;Extracted by using Immunity Operator between described virtual resource layer and access and adaptation layer and produce relevant Standard, knowledge, data help produce and decision-making:Including extracting most basic characteristic information or feature set;Secondly, it is special to this Reference breath is processed, with a kind of scheme of Solve problems under being translated into local environment or optimal constraint conditionses;Finally, will This scheme changes into Immunity Operator the process for producing new individuality, knowledge based and Knowledge Conversion in a suitable form, On the basis of reasonable drawing immune vaccine, immunoevolution is realized by vaccination and immunoselection, with effectively to exception The priori of diagnosis, improves individual fitness.
It is described for being examined based on the exception of generalization rule reasoning towards the safety production cloud service platform of industrial and mining establishment Disconnected method, it is characterised in that:The main composition key element that the Immunity Operator is calculated by collaborative diagnosis is constituted, and the key element is that each is exempted from Epidemic disease diagnosis cell colony, the diagnosis evolution algorithm of immunologic diagnosises cell and cell population regulatory mechanism, evolve using immunity substantially Algorithm, abnormality diagnosis process are roughly divided into six stages:1. gen I=1, fault diagnosis domain is set to equipment fault set F= { F1 ..., Fi ..., FM }, diagnostic cast collection are combined into FDM={ FDM1 ..., FDMi ..., FDMM }, and M is equipment fault mould Type number, FDMi are i-th fault diagnosis model;2. certain diagnostic task FDM_TASK, initialization failure diagnostic cast group are directed to Body FDM, distributes each diagnostic cast weight w={ w1 ..., wi ..., wM }, and wi represents that diagnostic cast FDMi appoints in tracing trouble Diagnosis importance degree in business FDM_TASK;3. each fault model FDMi provides each self diagnosis knot to diagnostic task FDM_TASK By FDR={ FDR1 ..., FDRi ..., FDRM }, each diagnostic cast individuality FDMi affinities degree and concentration is calculated, evaluate FDM_ 6. TASK fault diagnosis effects, if meeting fault diagnosis termination condition, turn to;4. gen I=gen I+1, to fault diagnosis Model colony FDM, chooses colony of new generation from previous generation colonies based on affinity degree and concentration value;5. Immunity Operator is applied to In the individuality of colony, obtain new fault diagnosis model colony FDM, and distribute new diagnostic cast weight w=w1 ..., Wi ..., wM }, turn to 3.;6. fault diagnosis terminates.The failure diagnostic process at a certain moment, immune collaborative fault diagnosis software Module will complete above-mentioned full diagnostics behavior, after the diagnostic result being more satisfied with, just terminate this diagnostic process.
Description of the drawings
The safety production cloud service platform system of Fig. 1 industrial and mining establishment
Fig. 2 industrial and mining establishment real time data integrated processing system
Fig. 3 wireless data collection device clusters
Fig. 4 wireless data collection device structure charts
Fig. 5 cloud storage system structure charts
Fig. 6 cloud storage data center structure charts
Fig. 7 device resource monitor in real time models
Fig. 8 equipment performance monitor in real time models
Fig. 9 SVM are trained and prediction flow chart
Figure 10 Immune Evolutionary Algorithm flow processs
Figure 11 Association Rule Analysis flow processs
Figure 12 Petri network mining models
Specific embodiment
Fig. 1 show a kind of cloud service platform of the safety in production towards industrial and mining establishment, is integrated with support safety in production clothes The functions such as the real time data processing of business cloud, production safety management business cooperation and system administration, realize to enterprise's production safety Real-time monitoring and early warning.
Specifically, the described safety production cloud service platform system towards industrial and mining establishment, it is flat including safe cloud service Platform portal sub-system, system administration develop subsystem, application service layer subsystem, virtual resource straton system with related tool collection System, platform service support layer subsystem, access and are adapted to layer subsystem, security service resource subsystem, infrastructure services Support layer subsystem.
The safe cloud service platform portal sub-system is responsible for realizing outside information inquiry and pipe based on web2.0 technologies Reason, supervises platform including government, and enterprise's application platform, trade supervision platform, government's supervision platform are responsible for government and are supervised in real time Device data and the property tax data of enterprise operation are examined, the enterprise application system is responsible for externally providing value-added service and application is looked into Ask, the trade supervision platform is used for the monitoring of product quality and feedback.
The system administration and related tool collection develop subsystem includes the exploitation of cloud service instrument, deployment, monitoring, safely Management, log management and configuration.
The application service layer, including mass data processing system and business cooperation operating system, wherein described magnanimity number According to processing system be responsible for, to data acquisition, Data Integration, data management and standardized service interface is provided for secure cloud clothes Business platform portal provides data flow, and the business cooperation operating system is responsible for building mathematical modulo to collaborative task and cross-domain task Type, is specially acquired to business datum, and the service content provided by cloud platform is integrated with the business datum for gathering, after prediction The development of continuous business, the cross-domain task specifically include the management and monitoring in different business field, by the different departments of enterprise it Between work realize task coordinate.
The virtual resource layer, mainly provides the digital resource after virtualization for Cloud Server, specifically includes knowledge clothes Business resource pool, production Service Source pond, data message resource pool, the virtual resource layer include these original providing and know Know service, the data message resource of production service, by Intel Virtualization Technology, by the partial data content integration to application service layer It is central, at any time can be by the mass data processing unit and business cooperation cell call.
It is described to access and be adapted to layer subsystem, mainly data resource is obtained from relevant criterion and verifying test system, this It is outer also to include security service resource, specifically include production equipment data, standardization fundamental norms, laws and regulations, safety management system Degree, accident source historical data, educational training knowledge, safety in production input, organization and responsible, hidden danger check information from the 3rd Square service adapter is accessed.
The platform service supporting layer is to above-mentioned application service layer, virtual resource layer, access Adaptation Layer Service, there is provided cloud takes Business management and support engine, conclude the business collaborative logic$$$$ engine, knowledge accumulating and classification engine, the cloud service management and support engine Cloud service registration is provided, is issued, nullified for the application service layer, cloud service search, scheduling, combination, cloud service are performed and prison Control;The transaction collaborative logic$$$$ engine provides process management, cost accounting, credit evaluation, the knowledge accumulating for business cooperation The dispersion knowledge acquisition of industry multiple resource is provided with classification engine for business cooperation, domain knowledge modeling, domain knowledge aggregation are classified, And other responsible operation management, operation management, terminal software exploitation in the platform service supporting layer, platform development work The module of tool is the virtual resource layer and access and Adaptation Layer Service, and wherein described operation management is responsible for multi-tenant service, is ordered Menu manager, delivery management, payment management, user management, integration managing, the operation management are responsible for safety management, performance management With optimization, system configuration, mass data is fault-tolerant to be managed with credibility, and the terminal software exploitation, including the fusion of heat transfer agent Management, Service Source graphic interface and pervasive man-machine interaction instrument, and platform development instrument, the platform service supporting layer is all By infrastructure services supporting layer by cloud computing, cloud network, cloud storage unification are supported.
Based on industrial and mining establishment's real-time data processing system of cloud computing, as shown in Fig. 2 whole by data acquisition equipment, data The reliable five most of composition of Networks of Fiber Communications of conjunction equipment, data management apparatus, standardized service interface equipment and high speed, The industrial and mining establishment's real time data integrated processing system based on cloud computing is constituted,
Data acquisition equipment:The network structure of the data integrated system shown in Fig. 2, wireless senser cluster are responsible for collection The data at industrial and mineral scene, and industrial and mineral Data Integration equipment is transferred data to, on the Data Integration equipment, standard configuration HTTP Proxy Module proxy server and cloud management server, both of which use ubuntu server blade plates, used as the section for receiving management Point, is assembled with node manager on every node server, and the major function of this manager is to realize the pipe to KVM virtual machines Reason, including:(1) control instruction for receiving cloud manager carries out the operation such as the deployment of KVM virtual machines, start and stop;(2) monitor local The availability of KVM virtual machines, when the error of local KVM virtual machines is unavailable, attempts starting another identical KVM virtual machine Example;(3) behaviour in service of local each KVM resources of virtual machine of monitor in real time, and the real time resources of KVM virtual machines are used into feelings Condition is sent to cloud manager;(4) extended operation of KVM virtual machines is performed automatically according to the real-time resource service condition of virtual machine; (5) the virtual machine (vm) migration instruction from cloud manager is received, meanwhile, it will be empty all also multiple KVM will to be disposed on every node server Plan machine, and node server is disposed in each virtual machine.All above-mentioned node servers all share same network number According to memorizer.And locally stored of each node server is used for caching each virtual machine self-operating data, and virtual machine Image file and embodiment can all be buffered in shared network data storage, so can more easily support High Availabitity The migration operation of property and virtual machine.Additionally, the data of industrial and mineral data acquisition, also can be controlled server and proxy server by cloud Control is randomly distributed in network memory, and by the application software finishing service collaboration on each node server and cross-domain Business collaboration.Shown in Fig. 3 be provided with wireless sensor data collection cluster, the wireless sensor data harvester according to Need using assembly function in system, pass through data or analog interface with by industrial and mineral data equipment, wired or wireless interface connects, And frequency acquisition be set to into the highest frequency of function needs and precision carry out data sampling, the data buffer storage of collection itself In memorizer, data compression is transmitted into remote monitoring center into Jing wireless transmit/receive units after consolidation form.Wherein each wireless sensing Can also synchronize between device equipment or asynchronous communication, constitute data acquisition cluster.Wireless data collection device shown in Fig. 4 Fundamental diagram, including;Core processing unit, touch screen, photographic head, microphone, clock, memorizer, power management, interface electricity Road, sensor circuit composition;The core processing unit include embedded controller, audio interface circuit, video interface circuit, Touch screen interface circuit, multi-path serial interface, hi-speed USB interface, wireless communication module, debugging interface and sensor interface electricity Road;The touch screen interface circuit, memorizer, audio interface circuit, multi-path serial interface, wireless communication module, debugging interface It is bi-directionally connected with embedded controller with hi-speed USB interface respectively;The touch screen is bi-directionally connected with touch screen interface circuit;Institute State the input that microphone and photographic head connect audio interface circuit and video interface circuit respectively;The multi-path serial interface and high speed USB interface connects industrial and mineral data equipment respectively;The respective input of the output termination embedded controller of the video interface circuit; Embedded controller is connected with the Internet;Power supply management device is used for providing electric energy, and power supply adopts photovoltaic solar panel and storage The combined form of set of cells, both can guarantee that the long-term work that equipment can be effectively stable, can also avoid laying power line;Wherein make The charging-discharging controller constituted with MC3063 chips, to provide stable voltage, is provided with acceleration, pressure in sensor unit Power, temperature, optical sensor and load signal sensor, realize the calculating to displacement by acceleration signal and load signal, It is monitored by pressure and temperature sensors towards ambient.Core processing unit is used for the collection computing of sensor signal and deposits The reception and transmission of wireless signal is realized in storage, wireless communication module;Embedded controller can adopt CC2530ZigBee chips, with Sending and receiving for the 2.4GHz ISM band signals based on ZigBee is conveniently realized, such as when on trepan When, trepan includes motor, quadric chain, walking beam, support, pumping rod;Motor drives walking beam motion, trip by quadric chain Beam is driven and is moved up and down with pumping rod, and the acceleration signal sensor and load signal sensor on data acquisition unit just can be with It is arranged in pumping rod, to gather the kinematic parameter of pumping rod.Additionally, shown data acquisition unit can also be arranged on work On mine electrical appliance equipment, for real-time video, audio frequency, digital monitoring running parameter, working condition, and the number that will be gathered is obtained According to being uploaded onto the server by multi-protocol gateway, server is carried out to the working condition of relevant device point according to the data of collection Analysis, to draw its running status.
Data storage layer:Cloud data storage pond is provided with, for the magnanimity number in industrial and mining establishment from various different data sources According to, the subject data of generation unified interface corresponding with different business after processing to which, and these subject datas are stored In distributed file system, when there is task requests, according to different task requests, to depositing in the distributed file system The data of storage carry out multinode, the parallel computation of multitask and analysis, and analysis result is carried out according to different applications accordingly Represent.As shown in figure 5, the cloud storage pond of data processing that the present invention is provided, includes at least one cloud storage management node, at least one Individual cloud storage space and at least one virtual unit, cloud storage management node, cloud storage space and virtual unit constitute privately owned Cloud.In embodiment as shown in Figure 4, the cloud storage framework includes two cloud storage nodes, three cloud storage spaces and multiple void Propose standby, three cloud storage spaces and multiple virtual units constitute a private clound.Cloud storage management node a is managed with cloud storage Node b is connected with each other, and mutually takes between two cloud storage management nodes, can be with balanced negative between two cloud storage management nodes Carry and a cloud storage management node is taken over each other when breaking down wherein, to ensure based on the cloud storage framework to be Unfailing performance during system operation.Wherein, two cloud storage management nodes can respectively to three cloud storage spaces and multiple virtual Equipment is managed, its management include carrying out three cloud storage spaces and multiple virtual units it is newly-built, delete and configuration, with Carry out system backup, recovery and dilatation.There is data directory in each cloud storage management node, data directory is deposited for recording cloud Storage space and the relevant information of virtual unit, cloud storage management node are found by the related data in its internal data directory Corresponding cloud storage space and virtual unit.In addition, being both provided with unified application program in each cloud storage management node Access entrance, the application program access entrance are application programming interfaces, and application program/service is by calling application program access Entrance accesses cloud storage space.Each virtual unit in Fig. 4 can be mapped as raw device, an internal memory in operating system Area, file system or memory file system etc., and virtual management physical memory, built-in disk and various interfaces, the disk of agreement Array.Further, since multiple virtual unit c1, virtual unit c2 ..., virtual unit cn is characteristic identical and virtually sets It is standby, so, in the present embodiment, virtual unit c1, virtual unit c2 ..., virtual unit cn may be constructed one it is virtual Equipment group b, simplifies the management to multiple virtual units by virtual unit group b.Multiple virtual units can correspond to one respectively A memory space in individual physical storage device or a physical storage device.In the present embodiment, virtual unit a1, virtual Equipment a2 ..., virtual unit an respectively with physical storage device a1, physical storage device a2 ..., physical store sets Standby an is connected, so as to by the data information memory from industrial and mineral collection in worksite in above-mentioned physical storage device.Outside application Programs/services a, b are connected with cloud storage management node a, b in cloud storage framework respectively.Due in cloud storage management node On be provided with unified application program access entrance, therefore, application program/service is by calling connecing for application program access entrance Mouthful function is so as to accessing corresponding cloud storage space.Application program/service can by application program access entrance indicate access or Access the corresponding information in any one cloud storage space or cloud storage space and virtual unit are managed.Using Programs/services can be passed through by calling the upper cloud storage management node of application program access entrance connection, then application program/service API indicates the cloud storage space of access, filename, side-play amount, accessing operation etc., and cloud storage management node is according to these API institutes Incoming information connecting inner data dictionary is assigned to final operation on one or more specific physical storage devices and completes Accessing operation, returns access results finally by cloud storage management node.
Data computation layer:Corresponding to cloud data calculating platform, the cloud data calculating platform is used for calling cloud data storage pond The Real-time Collection of middle storage to data calculate its characteristic according to industrial and mineral system business public relation respectively, set up general data Calculate analysis model, the such as numerical value such as computing device performance, load capacity, work efficiency, safe coefficient, ambient parameter, node control The data computation module of the virtual system on device processed is capable of achieving flexible addition and arranges, and updates the data analytical calculation module, obtains To corresponding value of calculation.The structure of data center is as shown in Figure 6.Including:The cloud control period of service passes through network adapter and agency Server is interconnected with data network, internet, monitors and control multiple memory nodes and virtual machine thereon, master controller Task control module is responsible for carrying out unified allocation of resources management to downstream node controller, including addition, deletes and Migration Control System Any amount of readable data phisical drive and the operation such as storage medium, administrative model be responsible for the load to collecting and Resource using information is analyzed process, then transfers to controller to be controlled.Calculate node includes any number of virtual machine, Each intra-node includes the resources of virtual machine control inside a Node Controller responsible node, and coordinated management engine is responsible for Operation application resource, such as resource monitor, Network Performance Monitor, early warning monitoring in the distribution management by synchronization of resource, virtual machine Device is in interior multiple Network Performance Monitors.Lower data statisticss flow process is introduced by taking Network Performance Monitor as an example, as shown in fig. 7, real time load And monitoring resource model, according to sampled data type, default device type is selected, and then calls the statistical model for the equipment The service behaviour data of analyzing and processing sampling, generate equipment ideal Distribution data, then combine actual samples data by controller Model calculating is carried out, the data of the actual distribution of output judge device resource service condition and its load level, and then instruct master control The distribution of node control system equipment resource, and for load monitoring module, as shown in figure 8, in system operation, according to Whether collected data message, the curve that monitor in real time is generated, observation curve have catastrophe point or do not meet the different of matched curve Constant value occurs.If existing, illustrate that Rush Hour occurs in the application of current data center, it was demonstrated that system resource requirement needs to carry out Larger variation.Now need quickly to be analyzed system resource behaviour in service, when system resource reaches maximum capacity be It is no to meet resource requirement.Because being outlier, currently obtained parameter can not represent overall load, between performance and resource Relational model, but if can have a great impact to the performance of system if processing not in time, so needing in time to exception Value point is analyzed and processed.If system resource Pooled resources meet current demand, directly give cloud controller and processed, quickly Current outlier is solved, if resource is unsatisfactory for current demand, just starts alternate device resource, period is using simple linear The next 5 minutes live load of forecast of regression model, simple linear regression model (LRM) effectively can catch live load with Time Change, even increasingly complex historical data can also easily conclude prediction, which loads.The work of prediction The input as model is loaded assessing the performance that device resource requirements and system needed for existing workload can reach.Perhaps Many complicated factors can all affect the performance of application program, such as ambient parameter, the change of operator's quantity etc., at this moment can be with Realize that multi-variate statistical analyses are modeled using KCCA algorithms and remote related algorithm, while the multiple influence factors of analysis are to systematicness The impact that can be brought, real-time adjustment model parameter generate preferable device resource allocation data.The distribution data of calculating can be led to The standardized query interface crossed on main controlled node or communication interface, actively carry out inquiry operation by user or are passively pushed to use Family.Wherein for the data message that those need real-time collecting, the data that upgrade in time are needed, just can guarantee that the service of data center Quality.
Data access layer:Cloud service access modules are provided with cloud service access interface, for according to the triggering using component, leading to Cross fiber optic network and find Cloud Server, quickly corresponding value of calculation can be obtained from cloud data calculating platform in real time.Data access layer The interface service of offer includes the cloud service access interface that Various types of data server is provided.
Four layers of the above is by reliability and the light switched communication network of high speed is sequentially connected, and the communication looped network is tight with power circuit Close association, the fibre circuit of high-speed communication are applied to all intelligent cells along power circuit, provide powerful information for enterprise High-speed communication passage.The data acquisition in traditional industrial and mineral system is compared, the data handling system of the present invention can extend high in the clouds Data storage and processing function is calculated, and the specific logic judging function based on data model, it is different hard by configuring Part equipment or system complete the unified monitoring of different functions, and the equipment of these difference in functionalitys or system have each independent number According to processing unit.And traditional real-time data acquisition is made up of isolated multiple systems with processing unit, Data duplication collection And data are incomplete, observing and controlling, metering, protection, automatic safety device are realized by different hardware units is local.Its defect is Data acquisition repeats, and data are difficult to share, and system application function is difficult to effectively cooperate with.Observing and controlling, protection, meter in the present invention Amount, the application function of automatic safety device are realized by the clustered form of functional unit, by high- speed network communication skill Art, really realizes that data are uniformly processed, calculate, share, be conducive to raising system reliability, reduce system installation cost and Maintenance cost.
The data message of enterprise practical productive accumulation of keeping the safety in production is unified by above-mentioned magnanimity information processing cloud platform Management and storage, additionally, in order to provide in production process based on functions such as the prediction of data, abnormality detections, realize the peace of enterprise It is complete to produce, managers also want to find in time to produce in abnormal conditions, find out reason, and propose counter-measure in time, Keep being normally carried out for production.Judgement to the condition of production is divided into normal and abnormal two kinds of situations, so which is classified as point so long Class problem, is carried out the condition of production and is judged extremely using the preferable support vector machine of classifying quality.By pacifying in analytical database The achievement data of production, selects product quality, composition, practical productivity to prop up as whether the index judge condition of production is abnormal entirely Data are held, to supporting that initial data carries out box traction substation analysis and correlation analysiss, influencing each other between data is obtained.Select number According to sample, as between variable, magnitude gap is larger, it is necessary first to needed to be standardized variable data,
I=1,2,3 ..., n
Wherein, Vi is former variate-value, and μ is the meansigma methodss of former variate-value, and σ is former variable standard deviation.
After standardization, the training and test of vector machine decision model can be supported, as shown in figure 9, using Training sample set data carry out model training, and its training algorithm adopts SMO algorithms (Sequential Minimal Optimization, the minimum optimization of sequence), the supporting vector of its disaggregated model is drawn, judgement letter is calculated according to supporting vector The parameter of number f (x).The modified model support vector machine obtained after training are not optimum, and this is due to initial reference mould Some parameter settings in plate and algorithm can affect the result trained.By the design parameter in selection layering core and to software degree Amount is selected, the model that more can be optimized.In the modified model support vector machine for training, input needs to carry out pre- The corresponding tree form data structure of software module of survey, obtains the output between [- 1 ,+1], if output is more than 0, product quality production There is no exception in situation;Conversely, output means the condition of production exception of product quality less than 0.
Also need to extract and produce relevant standard between virtual resource layer and access and adaptation layer described in Fig. 1, know Knowledge, data etc. help produce and decision-making, and this requires that cloud platform can search useful knowledge in data base.For knowledge Intension and characteristic, the purpose of Knowledge Conversion and requirement, Immunity Operator (Immune erator) standard of being incorporated into also is entered by we Change in planning algorithm.Knowledge, Knowledge Conversion are combined with theory of immunity, the immune programming algorithm of knowledge based transfer is realized Knowledge-based reasoning finds.In actual operating process, immune algorithm is the one kind grown up on genetic algorithm basis Global optimization approach, the problem that most genetic algorithm can solve the problem that, immune algorithm can effectively solving and efficiency than heredity Algorithm is good, can access all nodes in virtual resource layer and wireless sensor network using the good optimizing ability of immune algorithm The selection route scheme of data collection and total energy consumption minimum is completed, the energy for finally realizing reduction wireless sensor network disappears Consumption, the problem for reducing network system time delay, improving virtual resource layer interactive efficiency.As described in Figure 10, first, according to optimized Target and condition, the problem to being solved are made a concrete analysis of and are decomposed, and extract most basic characteristic information or feature set;Its It is secondary, this characteristic information is processed, with one kind side of Solve problems under being translated into local environment or optimal constraint conditionses Case;Finally, this scheme is changed into into Immunity Operator in a suitable form and for producing new individuality.Knowledge based and knowledge turn The process of shifting, on the basis of reasonable drawing immune vaccine, realizes immunoevolution by vaccination and immunoselection, with effective Treat the priori for seeking problem, improve individual fitness.
By taking immune collaborative fault diagnosis software workflow as an example, in immunological investigation, between various immunocytes Mutually promote and a kind of one immune coevolution of distinctive coevolution form is can be understood as with suppression, can use for reference immune association Same evolutionary mechanism, solves feature and collaborative diagnosis pattern for excitation system troubleshooting issue, it is proposed that a kind of many diagnosis moulds Type coevolution diagnosis policy, the main composition key element that immune collaborative diagnosis are calculated is each immunologic diagnosises cell colony, immunity The diagnosis evolution algorithm of diagnosis cell and cell population regulatory mechanism etc., evolve using basic immune algorithm, immune collaborative diagnosis Strategy can formalized description it is as follows:ICED=(CPD, CPDN, CEDA, CPCM), wherein:CPD:Immunocyte diagnoses population, CPDN:Immunocyte diagnoses population number, diagnoses the immune collaborative diagnosis algorithm of population,
Wherein CEDij={ PCMij, CEDPij, EDVij, i, j=1,2 ..., CPDN:The same jth of i-th diagnosis cell population The Cooperative Mode of individual diagnosis cell population, PCMij represent synchronous task assessment, determine whether the task needs and other intelligent bodies Cooperation is completed, and CEDPij represents synchronous evolutionary operator, and EDVij represents synchronous evaluation operator;DAi={ si, gi, pi, fi, di, i= 1,2 ..., CPN:The Immune Evolutionary Algorithm of i-th diagnosis cell population, si represent the selection strategy of i-th diagnosis cell population, Gi represents the evolutional operation operator of i-th diagnosis cell population, and including cell clone, clone inhibition etc., pi represents i-th diagnosis The execution probability of the evolutional operation operator of cell population, fi represent the close fitness function of i-th diagnosis cell population, and di represents The concentration function of i diagnosis cell population;CNi:The contained individual quantity of i-th diagnosis cell population;CPCM=CPO, CPD, CPA, CPE }:Immunocyte Advanced group species control model, CPO represent population scale operator, and CPD represents that population concentration is adjusted Section, CPA represent Evolution of Population object function;CPE represents population evaluation method.After using foregoing description mode, if:CPD(k) ={ CPD1 k, CPD2 k..., CPDCPN kKth is represented for immune cell population, then kth is represented for immune cell population coevolution For:CEDA { CPD (k) }=CEDA { { CPD1 k, CPD2 k..., CPDCPN k, then evolve to from -1 generation of kth in immune collaborative diagnosis In kth generation, can be expressed as:CPCM(k):CEDA { CPD (k-1) } → CEDA { CPD (k) }, k=1,2 ....Abnormality diagnosis process is big Cause is divided into six stages:1. gen I=1, fault diagnosis domain is set to equipment fault set F={ F1 ..., Fi ..., FM }, examines Disconnected model set is FDM={ FDM1 ..., FDMi ..., FDMM }, and M is equipment fault pattern number, and FDMi examined for i-th failure Disconnected model;2. certain diagnostic task FDM_TASK, initialization failure diagnostic cast colony FDM is directed to, distributes each diagnostic cast power Weight w={ w1 ..., wi ..., wM }, wi represent that diagnosis of the diagnostic cast FDMi in tracing trouble task FDM_TASK is important Degree;3. each fault model FDMi is to diagnostic task FDM_TASK, be given each self diagnosis conclusion FDR=FDR1 ..., FDRi ..., FDRM }, each diagnostic cast individuality FDMi affinities degree and concentration is calculated, FDM_TASK fault diagnosis effects are evaluated, If meeting fault diagnosis termination condition, turn to 6.;4. gen I=gen I+1, to fault diagnosis model colony FDM, based on parent Right and concentration value chooses colony of new generation from previous generation colonies;5. Immunity Operator (clone, mutation, suppression) is applied to into group In the individuality of body, obtain new fault diagnosis model colony FDM, and distribute new diagnostic cast weight w=w1 ..., wi ..., WM }, turn to 3.;6. fault diagnosis terminates.The failure diagnostic process at a certain moment, immune collaborative fault diagnosis software module 5 will Above-mentioned full diagnostics behavior is being completed, after the diagnostic result being more satisfied with, is just terminating this diagnostic process.
Additionally, very to the knowledge accumulating and classification engine in platform service supporting layer, with the production of safety in production enterprise Build, the value volume and range of product of data information accordingly increases, between data also benefit is complicated relation day, security incident is carried out traditional to determine Property analysis be not suitable with a large amount of and needs of complex data, propose to excavate security incident data using the analytical tool of correlation rule The characteristics of and rule, find the Strong association rule between accident occurrence type and " people-machine-environment-management " each factor, to accident Carry out early warning.The historical disaster data relevant with excavation and design prospecting data is collected, construction accident correlation is obtained Data.Data scrubbing is carried out to construction accident data, it is ensured that the exact specification of data.It is imperfect in construction accident data record, It is inconsistent, also vicious information etc., therefore, be ensure with post analysis data it is effective, this stage is needed to such Data are cleared up, and mainly solve the personal error during data file is set up, and some are tied to statistical analysiss in data file Fruit affects larger particular values to set up Multidimensional Data Model.Figure 11 gives a kind of data based on Association Rule Analysis flow process Analysis method, which arranges data attribute the reason for accident occurs according to the characteristics of construction accident, extracts accident number from data base According to reconnoitre data and constitute accident sample data, set up from accident pattern, traffic injury time, three dimensions of equipment parameters Its data cube, and the Aggregation Query using sql like language and connection sentence operated to the data cube, is completed frequently The search procedure of predicate collection and strong rule. produce frequent predicate set.Specifically need to meet minimum support and min confidence, if Put minimum support and be counted as 3, so that it is determined that not having the predicate collection of Concept Hierarchies dimension, min confidence can be minimum with item collection The conditional probability that support counting is represented expressing, so as to produce frequent predicate set using improved apriori traditional.Produce Raw Strong association rule.The correlation rule of minimum support threshold value and minimal confidence threshold further can be met by description Real implication, the such as time in month, equipment parameters to may occur accident rank be associated prediction, it will help The early warning of industrial and mineral accident, auxiliary enterprises science decision, in specific mining process, each threshold value can be by staff and field Expert's common setups, that is to say, that regular whether correct, applicable, the setting of boundary condition to be depended on.
The characteristics of frequent, close in cooperation is interacted for industrial and mining establishment's production safety management, the type of service of combing enterprise are real The safety management collaboration of business collaboration is now supported, following technology is also specifically included:
According to external appreciative standard, while and combining the production technology level and each site surrounding of China not Together, step analysis and the method for fuzzy mathematics can be adopted to carry out overall merit, to judge which whether as major hazard source.Counterweight The assay of big dangerous matter sources includes the dangerous reason to various dangerous matter sources, accident occurrence probability, coverage of consequence etc..Weight The evaluation of big dangerous matter sources is to control one of key measure of substantial industrial accident.Major hazard source evaluation should be commented from natrual danger Carry out in terms of valency and real danger evaluation two.Natrual danger evaluation mainly reflects material inherent character, dangerous substance life Inside the characteristics of product process and dangerous unit, external ambient conditions.Real danger evaluation is to consider each on the basis of the former The dangerous counteracting factor is planted, people is reflected and the subjective initiative work in terms of damage sequence expansion is occurred and controls in control accident With.Increased external environment condition and offset the following evaluation model of factor proposition:
The physical hazard evaluation of estimate of i-th kind of material of (B111) i in formula;(B112) technique of j jth kind material Hazard assessment value;The correlation coefficient of Wij jth item techniques and i-th kind of physical hazard;B12 accidents are seriously evaluated Value;The factor is offset in the building structure of B21 process equipment containers;B22 peoplewares offset the factor;B23 bursting tubes Reason offsets the factor.
Method when being generally classified using dangerous matter sources, determines border circular areas, this circle by dead radius Domain is exactly the scope that the dangerous matter sources affect.The scope of this border circular areas for blast impact be it is applicable, but for which He does not just apply to certain situation, than such as relating to gas leakage, cannot determine fully according to dead radius.Dangerous matter sources are carried out Accurately judge and be classified, the result then evaluated by which is taking corresponding prevention, emergency treatment.Major hazard source is quick The purpose of Assessment for classification, is, on the basis of the data message included by substantial risk source data, major hazard source to be carried out quickly Evaluate and be classified, macroscopical hierarchical monitoring and management are carried out to major hazard source to facilitate responsible departments of the government.Dangerous matter sources are quickly commented Valency method mainly to major hazard source may caused by damage sequence evaluate, with predict accident occur dead radius based on Evaluation index is wanted, the classification of major hazard source is carried out with the size of dead radius.The numbering of dangerous matter sources is selected first, then basis The related data in this numbering searching data storehouse, is analyzed after obtaining data, judges the thing included in this dangerous source data The species number of matter, if toxic, processes the dangerous matter sources so as to point several approach.Most important of which part is exactly that calculating has Noxious material, inflammable, the dead radius of the injury model of explosive material, comes to dangerous matter sources point finally by maximum dead radius Level.
Additionally, enterprise can also strengthen production scene safety management and the control of production process by cloud platform.To production The hidden danger of the presence such as process and material, installations and facilities, equipment, passage, working environment, should be analyzed and control.To work of getting angry In industry, restricted clearance, the dangerous higher activity such as operation, temporary electricity operation, high-rise working implements work permit pipe Reason, strictly fulfils examination and approval procedures.As shown in figure 12, based on the process model mining algorithm of Petri, non-Petri network is modeled Process model is converted to Petri network, solves this open problem of the hiding task of excavation from daily record so that what excavation was obtained Task node comprising not tape label in Petri network.To keeping the safety in production, process logs are analyzed, and the data of production process are entered Row assessment, prevents hidden danger in time.Can be used for supporting a kind of workflow engine of composite service based on the Petr algorithms, Including:The data base of interface layer, key-course, physical layer, accumulation layer and the flow instance for storage service;The workflow After engine distribution, the business information that operation system or other interface systems send is received by the interface layer, described other connect Port system includes resource management system, service release management system, billing and accounting system;The interface layer provides three kinds of modes Interface, including api interface, Corba interfaces, web Service interface, be easy to being connected for workflow engine and operation system;Institute Stating business information includes the current link performance of the flow instance of business, flow instance;The key-course receives described After the business information of interface layer transmission, according to the flow process route control method for supporting composite service, carry out the flow process reality of control business The generation of example, scheduling, decomposition, merging, end;And determine that the flow instance of business is automatic flow to next link, or Original place is needed to wait;The key-course calls the method that the physical layer is provided to record the complete of the current link of flow instance simultaneously Into the circulation result that situation and the flow process route control method determine, result of flow is returned by the interface layer;It is described Physical layer provide method be:To the newly-increased of management object described by workflow engine inside, modification, the behaviour for deleting and inquiring about Make, the management object includes:The flow instance object of business, flow process routing object, the current link object of flow instance, stream Journey task object;Wherein, the specific tasks that the link of each flow instance of the flow tasks object factory is performed;The storage Layer is by the persistent flow instance information for preserving the business of the data base.Can also melt during the flow operations The verification method of unification kind Optimized Operation, comprises the concrete steps that:
(1) scheduling scheme is generated:It is (middle according to product marketing plan, purchasing of raw materials plan, equipment maintenance plan, product Body, raw material) inventory information and equipment capacity, the production constraint information such as resources of production occupancy, consumption, production cost generates The data file of object module sets up the mathematical model of Optimized Operation, and according to optimization aim (the largest production energy of user's setting Power, maximum profit meet sales order), and model solver, calculate Optimized Operation scheme.
(2) the production exception being likely to occur by reckoning precognition:Current product pin in collection enterprise resource planning The plan of selling, purchasing of raw materials plan (including arrival situation), inventory information, Plant maintenance plan, energy supply plan etc., by mixed Miscellaneous equipment is constrained (including Batch Process equipment and continuous producing apparatus), the constraint of capacity-constrained, logistical balancing and energy constraint are pressed Time is tracked reckoning, predicts the production exception being likely to occur.
(3) by Optimized Operation scheme (data that step 1 is obtained) and the abnormal (number that step 2 is obtained of production being likely to occur According to) be combined together, using the visualization ESCPetri-Nets network technology virtual production processes of figure.With scheduling time as axle Line, dynamic graph display workshop equipment working condition, material balance (buying-stock-sale), simplation verification production scheduling feelings Condition.Centered on material, change with scheduling time, graphically the variation tendency of certain material of Dynamic Announce and certain moment The equipment operation condition related to certain material production consumption.

Claims (1)

1. a kind of for being examined based on the exception of generalization rule reasoning towards the safety production cloud service platform system of industrial and mining establishment Disconnected method, it is characterised in that:The safety production cloud service platform system towards industrial and mining establishment, including safe cloud service platform Portal sub-system, system administration and related tool collection development subsystem, application service layer subsystem, virtual resource layer subsystem, Platform service supports layer subsystem, accesses and is adapted to layer subsystem, security service resource subsystem, infrastructure services supporting layer Subsystem;The safe cloud service platform portal sub-system is responsible for realizing outside information inquiry and pipe based on web2.0 technologies Reason, including government's supervision platform, enterprise's application platform, trade supervision platform, government's supervision platform is responsible for government and is supervised in real time Device data and the property tax data of enterprise operation are examined, enterprise's application platform is responsible for externally providing value-added service and application is looked into Ask, the trade supervision platform is used for the monitoring of product quality and feedback;The system administration develops subsystem with related tool collection System includes exploitation, deployment, monitoring, safety management, log management and the configuration of cloud service instrument;The application service layer subsystem Including mass data processing system and business cooperation operating system, the mass data processing system is responsible for data acquisition, number According to integrating, data management and standardized service interface is provided data is provided for the safe cloud service platform portal sub-system Stream, the business cooperation operating system are responsible for building mathematical model to collaborative task and cross-domain task;Described virtual resource layer Extract and produce relevant standard, knowledge, data to help produce and determine and access and adaptation layer between by using Immunity Operator Plan:Including extracting most basic characteristic information or feature set;Secondly, this characteristic information is processed, to be translated into A kind of scheme of Solve problems under local environment or optimal constraint conditionses;Finally, this scheme is changed in a suitable form and is exempted from Epidemic disease operator simultaneously is used for producing the process of new individuality, knowledge based and Knowledge Conversion, on the basis of reasonable drawing immune vaccine, Immunoevolution is realized by vaccination and immunoselection, effectively individual adaptation is improved to abnormality diagnostic priori Degree;
The main composition key element that the Immunity Operator is calculated by collaborative diagnosis is constituted, and the key element is each immunologic diagnosises cell mass Body, the diagnosis evolution algorithm of immunologic diagnosises cell and cell population regulatory mechanism, evolve using basic immune algorithm, abnormity diagnosis Process is roughly divided into six stages:1. gen I=1, fault diagnosis domain be set to equipment fault set F=F1 ..., Fi ..., FM }, diagnostic cast collection is combined into FDM={ FDM1 ..., FDMi ..., FDMM }, and M is equipment fault pattern number, and FDMi is i-th Fault diagnosis model;2. certain diagnostic task FDM_TASK, initialization failure diagnostic cast colony FDM is directed to, distributes each diagnosis Model Weight w={ w1 ..., wi ..., wM }, wi represent diagnostic cast FDMi examining in tracing trouble task FDM_TASK Disconnected importance degree;3. each fault model FDMi is to diagnostic task FDM_TASK, be given each self diagnosis conclusion FDR=FDR1 ..., FDRi ..., FDRM }, each diagnostic cast individuality FDMi affinities degree and concentration is calculated, FDM_TASK fault diagnosis effects are evaluated, If meeting fault diagnosis termination condition, turn to 6.;4. gen I=gen I+1, to fault diagnosis model colony FDM, based on parent Right and concentration value chooses colony of new generation from previous generation colonies;5. Immunity Operator is applied in the individuality of colony, is obtained New fault diagnosis model colony FDM, and distribute new diagnostic cast weight w={ w1 ..., wi ..., wM }, turn to 3.;6. event Barrier diagnosis terminates;Immune collaborative fault diagnosis software module will complete above-mentioned full diagnostics behavior, be more satisfied with After diagnostic result, just terminate this diagnostic process.
CN201210370577.2A 2012-09-25 2012-09-25 Malfunction diagnosis method based on generalized rule reasoning and used for safety production cloud service platform facing industrial and mining enterprises Active CN102903009B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201210370577.2A CN102903009B (en) 2012-09-25 2012-09-25 Malfunction diagnosis method based on generalized rule reasoning and used for safety production cloud service platform facing industrial and mining enterprises

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201210370577.2A CN102903009B (en) 2012-09-25 2012-09-25 Malfunction diagnosis method based on generalized rule reasoning and used for safety production cloud service platform facing industrial and mining enterprises

Publications (2)

Publication Number Publication Date
CN102903009A CN102903009A (en) 2013-01-30
CN102903009B true CN102903009B (en) 2017-03-22

Family

ID=47575229

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201210370577.2A Active CN102903009B (en) 2012-09-25 2012-09-25 Malfunction diagnosis method based on generalized rule reasoning and used for safety production cloud service platform facing industrial and mining enterprises

Country Status (1)

Country Link
CN (1) CN102903009B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103745005A (en) * 2014-01-24 2014-04-23 清华大学 Dynamic artificial immunity fault diagnosis method based on vaccine transplantation
US9870295B2 (en) 2014-07-18 2018-01-16 General Electric Company Automation of workflow creation and failure recovery
CN106143533A (en) * 2015-04-21 2016-11-23 深圳市神拓机电股份有限公司 A kind of mthods, systems and devices of industrial and mineral vehicle safety monitoring
CN111865899B (en) * 2020-06-02 2021-07-13 中国科学院信息工程研究所 Threat-driven cooperative acquisition method and device

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102508486A (en) * 2011-11-09 2012-06-20 李伟 Method and device for diagnosing excitation fault by using controller area network (CAN) communication data protocol
CN102592198B (en) * 2011-12-30 2015-09-30 福建富士通信息软件有限公司 A kind of workflow engine supporting composite service

Also Published As

Publication number Publication date
CN102903009A (en) 2013-01-30

Similar Documents

Publication Publication Date Title
CN102882969B (en) A kind of safety production cloud service platform of industrial and mining enterprises
CN102932419B (en) A kind of data-storage system for the safety production cloud service platform towards industrial and mining enterprises
CN102880802B (en) A kind of assay method for the major hazard source towards industrial and mining establishment's safety production cloud service platform system
CN102917032B (en) A kind of safety production cloud service platform of industrial and mining enterprises
CN102929827B (en) A kind of wireless sensor data for ore deposit enterprise safety in production cloud platform gathers cluster
CN104268695B (en) A kind of multicenter basin water environment distributed type assemblies management system and method
CN105005274B (en) Big data in management process control system
CN102930372A (en) Data analysis method for association rule of cloud service platform system orienting to safe production of industrial and mining enterprises
CN102903011A (en) Mass data processing system used for safety production cloud service platform facing industrial and mining enterprises
CN105407159B (en) Logistics transportation position service system
CN106709580A (en) Substation secondary system operation and maintenance cloud platform
CN103559562A (en) Power grid intelligent operation system and achieving method thereof
CN105893628A (en) Real-time data collection system and method
CN109976268A (en) Big data in Process Control System
CN105069025A (en) Intelligent aggregation visualization and management control system for big data
Wang et al. Conceptual design of remote monitoring and fault diagnosis systems
CN102903009B (en) Malfunction diagnosis method based on generalized rule reasoning and used for safety production cloud service platform facing industrial and mining enterprises
CN102147597A (en) Health monitoring system for structures of great building and bridge
CN102739420A (en) Distributed complex giant system based on digital drainage network operation management and maintenance
CN107065824B (en) A kind of Hydropower Unit remote fault diagnosis open platform
CN102917031A (en) Data computing system of safety production cloud service platform for industrial and mining enterprises
CN102903010A (en) Support vector machine-based abnormal judgment method for safety production cloud service platform orientating industrial and mining enterprises
CN101854652A (en) Telecommunications network service performance monitoring system
CN108985531A (en) A kind of multimode isomery electric power big data convergence analysis management system and method
CN105843182A (en) Power dispatching accident handling scheme preparing system and power dispatching accident handling scheme preparing method based on OMS

Legal Events

Date Code Title Description
PB01 Publication
C06 Publication
SE01 Entry into force of request for substantive examination
C10 Entry into substantive examination
GR01 Patent grant
C14 Grant of patent or utility model