CN102880802A - Fatal danger fountainhead analysis and evaluation method for safety production cloud service platform system facing industrial and mining enterprises - Google Patents

Abstract

Disclosed is a fatal danger fountainhead analysis and evaluation method for a safety production cloud service platform system. Evaluation of fatal danger fountainheads are performed from two aspects including inherent danger evaluation and real danger evaluation, offset factors for the external environment are added, an evaluation model is provided, relevant data of a database are searched according to codes of selected danger fountainheads and are analyzed so as to judge whether the varieties of substances contained in the data of the danger fountainheads are of danger or not, a circular area computed according to the formula is the range affected by the danger fountainheads, and the danger fountainheads are graded according the maximum death radius.

Description

A kind of for the assay method towards the major hazard source of industrial and mining enterprises safety in production cloud service platform system

Technical field

The invention belongs to the information acquisition of industrial and mineral system automation and control field, relate to electric system real-time data acquisition, processing and control system, especially based on industrial and mineral real time data integrated processing system and the method for designing of cloud computing.

Background technology

Used widely cloud in the internet, mainly comprised three kinds of different types, software is namely served SaaS, and platform is namely served PaaS, and architecture is namely served IaaS.Wherein PaaS provides the complete or application development partly that the user can access, and SaaS then provides the complete application program that can directly use, such as passing through Internet management enterprise resource.And the cloud among the IaaS is the most key with Mass Data Management technology, mass data distributed store technology, Intel Virtualization Technology, cloud computing platform administrative skill.Wherein Mass Data Management technology and distributed storage technology are the important component parts that data are processed, cloud computing need to be processed, analyze data that distribute, magnanimity, therefore, data management technique must be managed a large amount of data efficiently, needs in addition the mode of redundant storage to guarantee the reliability of the data.Widely used data-storage system is that the GFS of Google and the increasing income of GFS of Hadoop team develops are realized HDFS in the cloud computing system.

The safety in production of industrial and mining enterprises is in the production run of reality, must there be mass data (such as the data of various safety in production standards, monitoring, educational training knowledge etc.) to need to process, cloud service platform adopts the Distributed Calculation storage mode, with distribution of computation tasks parallel processing to many machines, improve arithmetic speed with this.To the keep the safety in production data message of enterprise practical productive accumulation of cloud platform carries out unified management and storage, for functions such as prediction that based on data is provided in the production run, abnormality detection, realizes the safety in production of enterprise.

The data acquisition function of tradition industrial and mining enterprises and concrete logic judging function are in conjunction with tight, usually finished by comparatively single hardware device, therefore, finishing different functions just needs configuration different equipment or system, and there are separately independently data acquisition unit in the equipment of these difference in functionalitys or system.Each data acquisition exists that overlapping collection, utilization factor are not high, data and the problems such as the information content is inconsistent, time disunity, formed " information island " that, horizontal system many with vertical level mostly is principal character, restricted further fusion and the application of information, the repeated acquisition of data and information and repetition transmission process certainly will cause the waste of various resources.Instrument in most of industrial and mining enterprises is divided into three classes: the protection class, and the observing and controlling class, the metering class, they have the data acquisition and processing (DAP) unit of oneself, can be connected with monitoring and display system, are used for underlying parameter and measure.But also there is following technical matters:

(1) precision of instrument is different, has caused data inconsistent, and is also uneconomical on the cost.

(2) the instrument to collect frequency is different, algorithm is different has caused same value also different in different instrument, data redundancy and confusion.

(3) system and equipment, communication difficulties between equipment and the equipment lacks unified standard.

(4) process data disappearance, in the process of sampling, produced a large amount of process datas, these data all are valuable for crash analysis or the professional expansion in later stage, but we from instrument, obtain all through N computing " secondary data ", " simplification value ".The problem that this separate configurations, independent calculating, the device of standalone feature bring is that information sharing is poor, the low and hardware and software wasting of resources problem of utilization factor.

For solving above-mentioned problem; especially data gather respectively, data are processed respectively and the key issues such as calculating, data are difficult to share; be necessary to set up the new framework of a cover and mechanism; the data that realize same type equipment once gather, and can utilize cloud computing technology to realize the multiple application functions such as observing and controlling, protection, phasor measurement, metering under the unified software application platform of the whole series.

Summary of the invention

The safety in production cloud service platform system that the purpose of this invention is to provide a kind of industrial and mining enterprises; poor with the information sharing that solves the equipment separate configurations, independently calculates, the device of standalone feature brings, utilization factor is low and the problem of the soft and hardware wasting of resources; realize the once collection of same type device data, reach the multiple application such as observing and controlling, protection, phasor measurement, metering.

The safety in production cloud service platform system of described industrial and mining enterprises comprises:

1) foundation is towards the safety in production cloud service platform of industrial and mining enterprises, this platform is integrated supports the functions such as mass data processing, production safety management business cooperation and system management of safety in production service, for the safety in production of industrial and mining enterprises and government regulation provide important technical support.

2) based on security incident event Analysis of Multidimensional Association Rule technology, can from the production data of magnanimity, have frequent factor and the potential rule that may cause accident and near miss incidents to occur by analysis mining, set up the dynamic monitoring early warning analysis method of safety in production standard effectiveness and shortage of standard situation, for government and supervision department formulate based on risk, reasonably operation or work standard provides the foundation of science.

3) mass data processing method in industrial and mining enterprises' safety in production has been proposed, specifically comprise safety in production abnormality detection technology based on support vector machine, based on the knowledge discovering technologies of generalization rule reasoning, can analyze and knowledge excavation the mass data of producing, thereby help enterprise to promote accident prevention early warning and emergency disposal ability, improve the enterprise safety operation level.

4) exploitation production safety management tool set makes things convenient for safety management that the industrial and mineral industrial security produces, supervision, configuration etc., realizes the quick investigation of enterprise, fast response.

The safety in production cloud service platform towards industrial and mining enterprises of exploitation wherein, the integrated functions such as mass data processing, production safety management business cooperation and system management of supporting safety in production service cloud realize Real-Time Monitoring and early warning to enterprise's production safety.

Particularly, a kind of for the assay method towards the major hazard source of industrial and mining enterprises safety in production cloud service platform system, it is characterized in that: described safety in production cloud service platform system towards industrial and mining enterprises, comprise secure cloud service platform portal sub-system, system management and related tool collection development subsystem, the application service layer subsystem, virtual resource straton system, platform service supporting layer subsystem, access and adaptation layer subsystem, security service resource subsystem, infrastructure services supporting layer subsystem; The major hazard source evaluation is estimated two aspects from natrual danger evaluation and real danger and is carried out in the described platform service supporting layer subsystem, increase external environment condition and offset the factor, following evaluation model is proposed: the numbering of selecting dangerous matter sources, then search the related data of database according to this numbering, obtaining data analyzes later on, judge the species number of the material that comprises in these dangerous matter sources data, whether toxic, according to

$A=\left\{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{\left(B_{111}\right)}_i{W}_{\mathrm{ij}}{\left(B_{112}\right)}_j\right\}{B}_{12}(1-B_{2k})$

Result of calculation determine dead radius, and then determine a border circular areas, (B111) i---the physical hazard evaluation of estimate of i kind material in the formula; The technique hazard assessment value of (B112) j---j kind material; The related coefficient of Wij---j item technique and i kind physical hazard; B12---the serious evaluation of estimate of accident; The factor is offset in B21---process equipment container building structure; B22---peopleware is offset the factor; The factor is offset in B23---safety management; N is natural number; Described border circular areas is exactly the scope of this dangerous matter sources impact, and comes to the dangerous matter sources classification by the dead radius of maximum.

Description of drawings

The safety in production cloud service platform system of Fig. 1 industrial and mining enterprises

Fig. 2 industrial and mining enterprises real time data integrated processing system

Fig. 3 data acquisition device clusters

Fig. 4 data acquisition equipment structure chart

Fig. 5 cloud storage system structural drawing

Fig. 6 cloud storage data center structural drawing

Fig. 7 device resource Real Time Monitoring model

Fig. 8 equipment performance Real Time Monitoring model

Fig. 9 SVM training and prediction process flow diagram

Figure 10 Immune Evolutionary Algorithm flow process

Figure 11 Association Rule Analysis flow process

Figure 12 Petri net mining model

Embodiment

Figure 1 shows that a kind of cloud service platform of the safety in production towards industrial and mining enterprises, the integrated functions such as real time data processing, production safety management business cooperation and system management of supporting safety in production service cloud realize Real-Time Monitoring and early warning to enterprise's production safety.

Particularly, described safety in production cloud service platform system towards industrial and mining enterprises, comprise secure cloud service platform portal sub-system, system management and related tool collection development subsystem, application service layer subsystem, virtual resource straton system, platform service supporting layer subsystem, access and adaptation layer subsystem, security service resource subsystem, infrastructure services supporting layer subsystem.

Described secure cloud service platform portal sub-system is responsible for information inquiry and management are realized in the outside based on the web2.0 technology, comprise government's supervision platform, enterprise's application platform, the trade supervision platform, described government supervision platform is responsible for device data and the property tax data that government supervises enterprise operation in real time, described enterprise application system is responsible for externally providing value added service and application query, and described trade supervision platform is used for monitoring and the feedback of product quality.

Described system management and related tool collection development subsystem comprise exploitation, deployment, monitoring, safety management, log management and the configuration of cloud service instrument.

Described application service layer, comprise mass data processing system and business cooperation operating system, wherein said mass data processing system is responsible for, to the data collection, Data Integration, data management and provide the standardized service interface to provide data stream for secure cloud service platform door, and described business cooperation operating system is responsible for collaborative task and cross-domain task are made up mathematical model, be specially business datum is gathered, the service content that the cloud platform is provided and the business datum of collection are integrated, the development of prediction follow-up business, described cross-domain task specifically comprises the management and supervision in different business field, and the work between the different departments of enterprise is realized task coordinate.

Described virtual resource layer, mainly be to provide digital resource after virtual for Cloud Server, specifically comprise the knowledge services resource pool, produce Service Source pond, data message resource pool, described virtual resource layer these original provided knowledge services has been provided, has produced the data message resource of service, pass through Intel Virtualization Technology, with this partial data content integration in the middle of application service layer, at any time can be by described mass data processing unit and business cooperation cell call.

Described access and adaptation layer subsystem, mainly from relevant criterion and verifying test system, obtain data resource, comprise in addition the security service resource, comprise that specifically production equipment data, standardization fundamental norms, laws and regulations, safety management system, accident source historical data, educational training knowledge, safety in production input, organizational structure access from third party's service adapter with responsible, hidden danger fox message.

Described platform service supporting layer is to above-mentioned application service layer, virtual resource layer, the service of access adaptation layer, the cloud service management is provided and supports engine, the collaborative logic engine of transaction, knowledge accumulating and classification engine, described cloud service management provides cloud service registration, issue, nullifies for described application service layer with the support engine, cloud service search, scheduling, combination, cloud service are carried out and monitoring; The collaborative logic engine of described transaction provides process management for business cooperation, cost accounting, credit evaluation, described knowledge accumulating and classification engine disperse knowledge acquisition for business cooperation provides the many resources of industry, the domain knowledge modeling, domain knowledge is assembled classification, and for other responsible operation management in the described platform service supporting layer, operation management, the terminal software exploitation, the module of platform development instrument is described virtual resource layer and access and adaptation layer service, many tenant's services, order management, delivery management are responsible in wherein said operation management, payment management, user management, integration managing, safety management is responsible in described operation management, performance management and optimization, system configuration, mass data is fault-tolerant manages with confidence level, described terminal software exploitation, comprise the fusion management of heat transfer agent, Service Source graphic interface and pervasive man-machine interaction instrument, and platform development instrument, described platform service supporting layer all by the infrastructure services supporting layer by cloud computing, the cloud network, the cloud storage is unified to be supported.

Industrial and mining enterprises' Real-Time Data Handling System (RTDHS) based on cloud computing, as shown in Figure 2, formed by data acquisition equipment, Data Integration equipment, data management apparatus, standardization service interface equipment and at a high speed reliable Networks of Fiber Communications five major parts, formation is based on industrial and mining enterprises' real time data integrated processing system of cloud computing

Data acquisition equipment: the network structure of this data integrated system shown in Figure 2, the wireless senser cluster is responsible for gathering the data at industrial and mineral scene, and transfer data to industrial and mineral Data Integration equipment, on the described Data Integration equipment, standard configuration HTTP Proxy module acting server and cloud management server, the two all uses ubuntu server blade plate, as the node of accepting management, node manager all has been installed on every node server, the major function of this manager is the management that realizes the KVM virtual machine, comprising: (1) receives the steering order of cloud manager and carries out the deployment of KVM virtual machine, the operations such as start and stop; (2) availability of the local KVM virtual machine of monitoring is made mistakes when unavailable at local KVM virtual machine, attempts starting the identical KVM virtual machine instance of another one; (3) behaviour in service of local each KVM resources of virtual machine of Real Time Monitoring, and the real-time resource operating position of KVM virtual machine sent to the cloud manager; (4) automatically perform the extended operation of KVM virtual machine according to the real-time resource operating position of virtual machine; (5) reception simultaneously, all also will be disposed a plurality of KVM virtual machines, and dispose node server in each virtual machine from the virtual machine (vm) migration instruction of cloud manager on every node server.All above-mentioned node servers are all shared same network data storer.And this locality of each node server storage only is used for each virtual machine self-operating data of buffer memory, and the image file of virtual machine and embodiment can be buffered in the shared network data storage, can more easily support like this migration operation of high availability and virtual machine.In addition, the data of industrial and mineral data acquisition, the control that also can be subjected to cloud Control Server and acting server by stochastic distribution in network memory, and by the application software finishing service on each node server collaborative and cross-domain task cooperation.Shown in Figure 3 is to be provided with wireless sensor data to gather cluster, described wireless sensor data harvester is according to application component function needs in the system, with by the industrial and mineral data equipment by data or analog interface, wired or wireless interface connects, and highest frequency and precision that frequency acquisition is set to function and needs are carried out data sampling, the data buffer storage that gathers is in the storer of self, and data compression becomes consolidation form to be transmitted into remote monitoring center by wireless transmit/receive units.Wherein also can carry out synchronous or asynchronous communication between each wireless sensor devices, composition data gathers cluster.The fundamental diagram of data acquisition equipment shown in Figure 4 comprises; Core processing unit, touch-screen, camera, microphone, clock, storer, power management, interface circuit, sensor circuit form; Described core processing unit comprises 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 circuitry; Described touch screen interface circuit, storer, audio interface circuit, multi-path serial interface, wireless communication module, debugging interface are connected with hi-speed USB interface with embedded controller is two-way and are connected; Described touch-screen is connected with the touch screen interface circuit is two-way; Described microphone and camera connect respectively the input end of audio interface circuit and video interface circuit; Described multi-path serial interface and hi-speed USB interface connect respectively the industrial and mineral data equipment; The respective input of the output termination embedded controller of described video interface circuit; Embedded controller is connected with the Internet; Power supply management device is used for providing electric energy, and power acquisition is with photovoltaic solar panel and the combined form of battery pack, and the equipment that can guarantee can effectively stable long-term work, also can avoid laying power lead; The charging-discharging controller that wherein uses the MC3063 chip to consist of, in order to stable voltage is provided, be provided with acceleration, pressure, temperature, optical sensor and load signal sensor in the sensor unit, realize calculating to displacement by acceleration signal and load signal, by the pressure and temperature sensor environment is monitored.Core processing unit is used for collection computing and the storage of sensor signal, and wireless communication module is realized reception and the transmission of wireless signal; Embedded controller can adopt the CC2530ZigBee chip, to realize easily the sending and receiving as the 2.4GHz ISM band signal on basis take ZigBee, in the time of for example on being used in trepan, trepan comprises motor, four-bar mechanism, walking beam, support, pumping rod; Motor drives the walking beam motion by four-bar mechanism, and walking beam drives with pumping rod and moves up and down, and acceleration signal sensor and load signal sensor on the data collector just can be arranged on the pumping rod, in order to gather the kinematic parameter of pumping rod.In addition, shown data acquisition unit further is arranged on the industrial and mineral electrical appliance equipment, be used for real-time video, audio frequency, digital monitoring running parameter, obtain duty, and the data communication device that gathers is crossed multi-protocol gateway upload onto the server, server is analyzed the working condition of relevant device according to the data that gather, in order to draw its running status.

Data storage layer: be provided with cloud data storage pool, for in the industrial and mining enterprises from the mass data in various different pieces of informations source, it is processed the subject data of the rear generation unified interface corresponding with different business, and these subject datas are stored in the distributed file system, when task requests is arranged, according to different task requests, the data of storing in the described distributed file system are carried out parallel computation and the analysis of multinode, multitask, analysis result is represented accordingly according to different application.As shown in Figure 5, the cloud storage pool that data provided by the invention are processed is drawn together at least one cloud storage administration node, at least one cloud storage space and at least one virtual unit, and cloud storage administration node, cloud storage space and virtual unit consist of privately owned cloud.Among the embodiment as shown in Figure 4, this cloud storage architecture comprises two cloud memory nodes, three cloud storage spaces and a plurality of virtual unit, and three cloud storage spaces and a plurality of virtual unit consist of a privately owned cloud.Cloud storage administration node a and cloud storage administration node b interconnect, mutually take between two cloud storage administration nodes, can equally loaded between two cloud storage administration nodes and therein a cloud storage administration node when breaking down, take over each other the unfailing performance when guaranteeing the system operation based on this cloud storage architecture.Wherein, two cloud storage administration nodes can manage three cloud storage spaces and a plurality of virtual unit respectively, its management comprises newly-built, deletion that three cloud storage spaces and a plurality of virtual unit are carried out and disposes, to carry out system backup, recovery and dilatation.In each cloud storage administration node data directory is arranged, data directory is used for the relevant information of record cloud storage space and virtual unit, and cloud storage administration node finds corresponding cloud storage space and virtual unit by the related data in its inner data directory.In addition, be provided with unified application access entrance on each cloud storage administration node, this application access entrance is application programming interfaces, and application program/service is by calling this application access entrance access cloud storage space.Each virtual unit among Fig. 4 all can be mapped as raw device, memory field, file system or the memory file system etc. in the operating system, and the disk array of virtual management physical memory, built-in disk and various interface, agreement.In addition, because a plurality of virtual unit c1, virtual unit c2 ..., virtual unit cn is the identical virtual unit of characteristic, so, in the present embodiment, virtual unit c1, virtual unit c2 ..., virtual unit cn can consist of a virtual unit group b, simplified management to a plurality of virtual units by this virtual unit group b.A plurality of virtual units can be distinguished a storage space in a corresponding physical storage device or the physical storage device.In the present embodiment, virtual unit a1, virtual unit a2 ..., virtual unit an respectively with physical storage device a1, physical storage device a2 ..., physical storage device an is connected, thereby will be from the data information memory of industrial and mineral collection in worksite above-mentioned physical storage device.Outside application program/service a, b is connected with cloud storage administration node a, b in the cloud storage architecture respectively.Owing to be provided with unified application access entrance at cloud storage administration node, therefore, thereby application program/service is accessed corresponding cloud storage space by the interface function of invokes application access entrance.Application program/service meeting indicates the corresponding information in access or any one cloud storage space of access by the application access entrance or cloud storage space and virtual unit is managed.Application program/service connects upper cloud storage administration node by the invokes application access entrance, then application program/service meeting indicates the cloud storage space, filename, side-play amount, accessing operation etc. of access by API, the information connecting inner data dictionary that cloud storage administration node imports into according to these API will finally operate to be assigned on one or more concrete physical storage devices finishes accessing operation, returns access results by cloud storage administration node at last.

Data computation layer: corresponding to cloud data computing platform, this cloud data computing platform be used for calling Real-time Collection that cloud data storage pool stores to data calculate its characteristic according to industrial and mineral system business public relation respectively, set up general data computational analysis model, such as numerical value such as computing equipment performance, load capacity, work efficiency, safe coefficient, environmental parameters, the data computation module of the virtual system on the Node Controller can realize adding flexibly and setting, upgrade the data analysis computing module, obtain corresponding calculated value.The structure of data center as shown in Figure 6.Comprise: the cloud control period of service is interconnected by network adapter and acting server and data network, internet, the monitoring and control a plurality of memory nodes and on virtual machine, the task control module of master controller is responsible for the downstream node controller is carried out the unified allocation of resources management, comprise the operations such as the phisical drive of any amount of readable data of interpolation, deletion and Migration Control System and storage medium, administrative model is responsible for load and the resource using information of collecting carried out analyzing and processing, then transfers to controller and controls.Computing node comprises the virtual machine of any amount, each intra-node comprises the resources of virtual machine control of a Node Controller responsible node inside, the coordinated management engine is responsible for the distribution management by synchronization of resource, the resource that runs application in the virtual machine, such as resource monitor, Network Performance Monitor, early warning monitoring device at interior a plurality of Network Performance Monitors.Data statistics flow process under take Network Performance Monitor as the example introduction, as shown in Figure 7, in real time load and monitoring resource model, according to the sampled data type, select default device type, and then call serviceability data for the sampling of the statistical model analyzing and processing of this equipment, generation equipment ideal Distribution data, carrying out model by controller in conjunction with the actual samples data again calculates, the data of output actual allocated, judgment device resource operating position and load level thereof, and then instruct the distribution of main controlled node control system equipment resource, and for the load monitoring module, as shown in Figure 8, in service in system, according to collected data message, the curve that Real Time Monitoring generates is observed the exceptional value appearance whether curve has catastrophe point or do not meet matched curve.If exist, illustrate that then Rush Hour appears in this application of current data center, the proof system resource requirement need to carry out larger change.Need this moment fast the system resource behaviour in service to be analyzed, when system resource reaches max cap., whether can satisfy resource requirement.Because be outlier, the current parameter that obtains can not represent relational model between whole load, performance and resource, if but untimely processing meeting the performance of system is had a great impact, so need in time carry out analyzing and processing to outlier.If the system resource Pooled resources satisfies current demand, then directly giving cloud controller processes, the current outlier of quick solution, if resource does not satisfy current demand, just start the alternate device resource, use during this time next 5 minutes operating load of simple linear regression model (LRM) prediction, simple linear regression model (LRM) can effectively catch the operating load change with time, even more complicated historical data also can be concluded its load of prediction easily.The operating load of prediction is assessed the performance that the required device resource requirements of existing workload and system can reach as the input of model.The factor of many complexity all can affect the performance of application program, such as change of environmental parameter, operating personnel's quantity etc., at this moment can adopt KCCA algorithm and remote related algorithm to realize the multivariate statistical analysis modeling, analyze simultaneously the impact that a plurality of influence factors are brought system performance, the adjustment model parameter generates desirable device resource allocation data in real time.The distribute data of calculating can be by standardization query interface or the communication interface on the main controlled node, is initiatively carried out query manipulation or passively is pushed to the user by the user.The data message that wherein needs real-time collecting for those, the data that need to upgrade in time, the service quality of guarantee data center.

Data access layer: the cloud service access modules is provided with the cloud service access interface, is used for the triggering according to application component, finds Cloud Server by fiber optic network, can obtain corresponding calculated value from cloud data computing platform in real time fast.The interface service that data access layer provides comprises the cloud service access interface that the Various types of data server provides.

More than four layers connect successively by reliable and light switched communication network at a high speed, this communication looped network is closely related with power circuit, the fibre circuit of high-speed communication is applied to all intelligent cells along power circuit, for enterprise provides powerful high speed information communication port.Compare the data acquisition in the traditional industrial and mineral system, data handling system of the present invention can be expanded data storage and the computing function in high in the clouds, and the logic judging function of concrete based on data model, finish the unified monitoring of different functions by disposing different hardware devices or system, there are separately independently data processing unit in the equipment of these difference in functionalitys or system.And traditional real-time data acquisition and processing unit are comprised of isolated a plurality of systems, and Data duplication collection and data are incomplete, and observing and controlling, metering, protection, automatic safety device are to realize by different hardware units is local.Its defective is that data acquisition repeats, and data are difficult to share, and the system applies function is difficult to effectively work in coordination with.The application function of the observing and controlling among the present invention, protection, metering, automatic safety device all is the clustering form realization by functional module; by the express network communication technology; real realization data are unified to be processed, calculate, shares, and is conducive to improve the reliability of system, installation cost and the maintenance cost of reduction system.

To the keep the safety in production data message of enterprise practical productive accumulation of above-mentioned magnanimity information processing cloud platform carries out unified management and storage, in addition, for functions such as prediction that based on data is provided in the production run, abnormality detection, realize the safety in production of enterprise, abnormal conditions during supvrs also wish in time to find to produce, find out reason, and in time propose counter-measure, keep normally carrying out of production.The judgement of the condition of production is divided into normal and undesired two kinds of situations, thus so long be classified as classification problem to it, adopt classifying quality preferably support vector machine carry out the condition of production and unusually judge.By the achievement data of keeping the safety in production in the analytical database, select product quality, composition, practical productivity to pass judgment on whether unusual supported data of the condition of production as index, to supporting raw data to carry out case chart analysis and correlation analysis, obtain influencing each other between the data.Select data sample, because the magnitude gap is larger between the variable, at first needs to carry out and need to carry out standardization to variable data,

$V_i^{\′}=\frac{V_i-\mathrm{\μ}}{\mathrm{\σ}},i=\mathrm{1,2,3},......,n$

Wherein, Vi is former variate-value, and μ is the mean value of former variate-value, and σ is former variable standard deviation.

Through after the standardization, can carry out training and the test of support vector machine decision model, as shown in Figure 9, use the training sample set data to carry out model training, its training algorithm adopts SMO algorithm (SequentialMinimal Optimization, the minimum optimization of sequence), draw the support vector of its disaggregated model, calculate the parameter of decision function f (x) according to support vector.The modified support vector machine that obtains after the process training is not optimum, and this is because some parameters in initial reference template and the algorithm arrange the result that can affect training.Select the model that more to be optimized by the design parameter in the selection layering nuclear with to software metrics.Tree form data structure corresponding to software module that input need to be predicted in the modified support vector machine that trains obtains the output between [1 ,+1], if export greater than 0, the product quality condition of production does not exist unusually; Otherwise output means that less than 0 the condition of production of product quality is unusual.

Also need to extract and produce relevant standard, knowledge, data etc. between virtual resource layer described in Fig. 1 and access and the adaptation layer and help to produce and decision-making, this just requires the cloud platform can search useful knowledge in database.For purpose and the requirement of the intension of knowledge and characteristic, Knowledge Conversion, we also are incorporated into Immunity Operator (Immune erator) in the standard Evolutionary Programming Algorithm.Knowledge, Knowledge Conversion and theory of immunity are combined, realize that based on the immune programming algorithm of Knowledge Conversion the reasoning of knowledge is found.In the operating process of reality, immune algorithm is a kind of global optimization approach that grows up on the genetic algorithm basis, the problem that most genetic algorithm can solve, immune algorithm can both effectively solve and efficient better than genetic algorithm, utilize the good optimizing ability of immune algorithm to access all nodes at virtual resource layer and wireless sensor network and finish Data Collection and the selecting paths scheme of total energy consumption minimum, finally realized reducing wireless sensor network energy consumption, reduce the network system time delay, improve the problem of virtual resource layer interactive efficiency.As described in Figure 10, at first, according to optimized target and condition, the problem of finding the solution is made a concrete analysis of and decomposed, extract the 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 the optimal constraint conditions; At last, this scheme is changed into Immunity Operator and is used for producing new individuality with suitable form.Based on the process of knowledge and Knowledge Conversion, on the basis of reasonable drawing immune vaccine, realize immunoevolution by vaccine inoculation and Immune Selection, effectively to treat the priori of asking problem, improve individual fitness.

Take immune collaborative fault diagnosis software workflow as example, in immunology research, mutually promoting and suppressing phenomenon between the various immunocytes can be understood as a kind of distinctive coevolution form-immune coevolution, can use for reference immune coevolution mechanism, find the solution characteristics and collaborative diagnosis pattern for the excitation system troubleshooting issue, a kind of many diagnostic models coevolution Diagnostic Strategy has been proposed, the main composition key element that the immunity collaborative diagnosis calculates is each immunodiagnosis cell colony, the diagnosis evolution algorithm of immunodiagnosis cell and cell population regulation mechanism etc., evolve and adopt basic immune algorithm, but immunity collaborative diagnosis strategy formalized description is as follows: ICED=(CPD, CPDN, CEDA, CPCM), wherein: CPD: immunocyte diagnosis population, CPDN: immunocyte diagnosis population number, the immune collaborative diagnosis algorithm of diagnosis population

$\mathrm{CEDA}=\left[\begin{array}{cccccc}{\mathrm{<figure-callout\; id="1"\; label="CED"\; filenames="HSA00000784655700081.png"\; state="\{\{state\}\}">\; <figure-callout\; id="1"\; label="CED"\; filenames="HSA00000784655700081.png"\; state="\{\{state\}\}">CED</figure-callout>\; </figure-callout>}}_{\mathrm{1,1}}& {\mathrm{<figure-callout\; id="1"\; label="CED"\; filenames="HSA00000784655700081.png"\; state="\{\{state\}\}">\; <figure-callout\; id="2"\; label="CED"\; filenames="HSA00000784655700041.png,HSA00000784655700081.png"\; state="\{\{state\}\}">CED</figure-callout>\; </figure-callout>}}_{\mathrm{1,2}}& ...& {\mathrm{<figure-callout\; id="1"\; label="CED"\; filenames="HSA00000784655700081.png"\; state="\{\{state\}\}">CED</figure-callout>}}_{1,\mathrm{CPN}}& {\mathrm{DA}}_1& {\mathrm{<figure-callout\; id="1"\; label="CN"\; filenames="HSA00000784655700081.png"\; state="\{\{state\}\}">CN</figure-callout>}}_1\\ {\mathrm{<figure-callout\; id="2"\; label="CED"\; filenames="HSA00000784655700041.png,HSA00000784655700081.png"\; state="\{\{state\}\}">\; <figure-callout\; id="1"\; label="CED"\; filenames="HSA00000784655700081.png"\; state="\{\{state\}\}">CED</figure-callout>\; </figure-callout>}}_{\mathrm{2,1}}& {\mathrm{<figure-callout\; id="2"\; label="CED"\; filenames="HSA00000784655700041.png,HSA00000784655700081.png"\; state="\{\{state\}\}">\; <figure-callout\; id="2"\; label="CED"\; filenames="HSA00000784655700041.png,HSA00000784655700081.png"\; state="\{\{state\}\}">CED</figure-callout>\; </figure-callout>}}_{\mathrm{2,2}}& ...& {\mathrm{CED}}_{2,\mathrm{CPN}}& {\mathrm{DA}}_2& {\mathrm{CN}}_2\\ .& .& .& .& .& .\\ .& .& .& .& .& .\\ .& .& .& .& .& .\\ {\mathrm{CED}}_{\mathrm{CPN},1}& {\mathrm{CED}}_{\mathrm{CPN},2}& ...& {\mathrm{CED}}_{\mathrm{CPN},\mathrm{CPN}}& {\mathrm{DA}}_{\mathrm{CPN}}& {\mathrm{CN}}_{\mathrm{CPN}}\end{array}\right]$

CED wherein _Ij={ PCM _Ij, CEDP _Ij, EDV _IjJj}, i, j=1,2,, CPDN: the Cooperative Mode of i same j the diagnosis cell population of diagnosis cell population, PCMij represent the synchronous task assessment, determine whether this task needs to finish with other intelligent body cooperations, and CEDPij represents synchronous evolutionary operator, and operator is estimated in the EDVij representative synchronously; DA _i={ s _i, g _i, p _i, f _i, d _i; i=1; 2 ..., CPN: the Immune Evolutionary Algorithm of i diagnosis cell population; si represents the selection strategy of i diagnosis cell population; gi represents the evolutional operation operator of i diagnosis cell population, comprises cell clone, clone inhibition etc., and pi represents the execution probability of the evolutional operation operator of i diagnosis cell population; fi represents the affinity degree function of i diagnosis cell population, and di represents the concentration function of i diagnosis cell population; CNi: the quantity of the contained individuality of i diagnosis cell population; CPCM={CPO, CPD, CPA, CPE}: immunocyte Advanced group species control model, CPO represents the population scale operator, and CPD represents the population concentration adjustment, and CPA represents the Evolution of Population objective function; CPE represents the population evaluation method.After adopting the foregoing description mode, if: CPD (k)={ CPD ₁ ^k, CPD ₂ ^k..., CPD _CPN ^kExpression k is for immune cell population, then k is expressed as for the immune cell population coevolution: CEDA{CPD (k) }=CEDA{{CPD ₁ ^k, CPD ₂ ^k..., CPD _CPN ^k, then evolving to k generation from k-1 generation in the immune collaborative diagnosis can be expressed as: CPCM (k): CEDA{CPD (k-1) } → CEDA{CPD (k) }, k=1,2 ...The abnormity diagnosis process roughly is divided into six stages: 1. genI=1, and the fault diagnosis territory is made as equipment failure set F={F1 ..., Fi ..., FM}, the diagnostic model set is FDM={FDM1 ..., FDMi, ..., FDMM}, M are the equipment failure pattern number, FDMi is i fault diagnosis model; 2. for certain diagnostic task FDM_TASK, the initialization fault diagnosis model FDM of colony distributes each diagnostic model weight w={w1 ..., wi ..., wM}, wi represent the diagnosis importance degree of diagnostic model FDMi in tracing trouble task FDM_TASK; 3. each fault model FDMi provides the sub-conclusion FDR={FDR1 of each self diagnosis to diagnostic task FDM_TASK ..., FDRi, ..., FDRM} calculates the individual FDMi affinity degree of each diagnostic model and concentration, estimate FDM_TASK fault diagnosis effect, if satisfy the fault diagnosis termination condition, then turn to 6.; 4. genI=genI+1 to the FDM of fault diagnosis model colony, chooses colony of new generation based on affinity degree and concentration value from previous generation colony; 5. Immunity Operator (clone, sudden change, inhibition) is applied in the individuality of colony, obtains the new FDM of fault diagnosis model colony, and distribute new diagnostic model weight w={w1 ..., wi ..., 3. wM} turns to; 6. fault diagnosis finishes.The failure diagnostic process in a certain moment, immune collaborative fault diagnosis software module 5 will be finished above-mentioned whole Diagnosis behavior, behind the diagnostic result of comparatively being satisfied with, just finish this diagnostic process.

In addition, very to the knowledge accumulating in the platform service supporting layer and classification engine, production and construction along with safety in production enterprise, the corresponding increase of the value volume and range of product of data information, relation day is also beneficial complicated between data, security incident is carried out traditional qualitative analysis to be not suitable with a large amount of and the needs of complex data, proposition utilizes the analysis tool of correlation rule to excavate the characteristics and rules of security incident data, find the Strong association rule between accident occurrence type and " people-machine-environment-management " each factor, to the early warning of carrying out of accident.Collect and excavate relevant historical disaster data and design prospecting data, obtain the data that construction accident is correlated with.The construction accident data are carried out data scrubbing, guarantee the accurate standard of data.Imperfect in the construction accident data recording, inconsistent, vicious information etc. also, therefore, for guarantee with data in the post analysis effectively, this one-phase need to be cleared up these type of data, mainly solves the personal error in the data file foundation, and some set up Multidimensional Data Model to the larger particular values of statistic analysis result impact in the data file.Figure 11 has provided a kind of data analysing method based on the Association Rule Analysis flow process, the reason that it occurs according to the characteristics accident of construction accident, data attribute is set, from database, extract casualty data and reconnoitre data formation accident sample data, set up its data cube from accident pattern, accident time of origin, three dimensions of equipment parameters, and utilize the Aggregation Query of sql like language and enable statement that this data cube is operated, and finish the search procedure of frequent predicate set and strong rule. produce frequent predicate set.Specifically need to satisfy minimum support and min confidence, it is 3 that the minimum support counting is set, thereby determine not have the predicate collection of Concept Hierarchies dimension, min confidence can collect the conditional probability that minimum support counting represents with item expresses, thereby utilizes improved apriori traditional to produce frequent predicate set.Produce Strong association rule.Further can satisfy by description the real implication of the correlation rule of minimum support threshold value and minimal confidence threshold, such as the time in month, equipment parameters carries out interaction prediction to contingent accident rank, to help the early warning of industrial and mineral accident, the auxiliary enterprises science decision, in concrete mining process, each threshold value can be set jointly by staff and domain expert, that is to say that whether rule is correct, applicable, depend on the setting of boundary condition.

The characteristics of, close in cooperation mutual frequent for industrial and mining enterprises' production safety management, the type of service of combing enterprise realizes that the safety management of supporting business cooperation is collaborative, also specifically comprises following technology:

According to external appreciative standard, combine again the different of the production technology level of China and each site surrounding simultaneously, can adopt the method for step analysis and fuzzy mathematics to carry out comprehensive evaluation, to judge that whether it is as major hazard source.The assay of major hazard source is comprised dangerous reason to various dangerous matter sources, accident occurrence probability, the coverage of consequence etc.The evaluation of major hazard source is one of key measure of the great industrial accident of control.The major hazard source evaluation should be estimated two aspects from natrual danger evaluation and real danger and carry out.The natrual danger evaluation has mainly reflected the material inherent characteristic, and the characteristics of dangerous substance production run and dangerous unit are inner, the external environment condition situation.The real danger evaluation is the counteracting factor of considering various danger on the former basis, has reflected the subjective initiative of people aspect the generation of control accident and the expansion of control damage sequence.Increase the external environment condition counteracting factor and proposed following evaluation model:

$A=\left\{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}{\left(B_{111}\right)}_i{W}_{\mathrm{ij}}{\left(B_{112}\right)}_j\right\}{B}_{12}(1-B_{2k})$

(B111) i---the physical hazard evaluation of estimate of i kind material in the formula; The technique hazard assessment value of (B112) j---j kind material; The related coefficient of Wij---j item technique and i kind physical hazard; B12---the serious evaluation of estimate of accident; The factor is offset in B21---process equipment container building structure; B22---peopleware is offset the factor; The factor is offset in B23---safety management.

Method when usually adopting the dangerous matter sources classification is determined a border circular areas by dead radius, and this border circular areas is exactly the scope of this dangerous matter sources impact.The scope of this border circular areas is suitable for for the impact of blast, but just inapplicable for some other situation, such as relating to Leakage Gas, just can not determine according to dead radius fully.Dangerous matter sources is judged accurately and classification then the result by its evaluation takes corresponding prevention, emergency treatment.The purpose of major hazard source Fast Evaluation classification is on the data message basis that the substantial risk source data is included, and major hazard source is carried out Fast Evaluation and classification, with convenient responsible departments of the government major hazard source is carried out macroscopical hierarchical monitoring and management.The dangerous matter sources fast appraisement method is mainly estimated the damage sequence that major hazard source may cause, and the dead radius that occurs take the prediction accident carries out the classification of major hazard source as main evaluation index with the size of dead radius.At first select the numbering of dangerous matter sources, then search the related data of database according to this numbering, obtain data and analyze later on, judge the species number of the material that comprises in these dangerous matter sources data, whether toxic, thus minute several approach are processed this dangerous matter sources.Wherein most important part is exactly to calculate noxious material, and is inflammable, the dead radius of the injury model of explosive material, and the dead radius by maximum comes to the dangerous matter sources classification at last.

In addition, enterprise can also strengthen by the cloud platform control of production scene safety management and production run.To the hidden danger that production run and material, installations and facilities, equipment, passage, operating environment etc. exist, should carry out analysis and control.The dangerous higher activity such as operation in fire operation, the restricted clearance, temporary electricity operation, high-rise working is implemented the work permit management, strictly fulfil examination and approval procedures.As shown in figure 12, take the process mining algorithm of Petri as the basis, the process model of non-Petri net modeling is converted to the Petri net, solves from daily record, to excavate and hide this open problem of task, comprise the not task node of tape label so that excavate in the Petri net that obtains.The safety in production process logs is analyzed, the data of production run are assessed, in time stop hidden danger.Can for the workflow engine of supporting a kind of composite service, comprise based on this Petr algorithm: interface layer, key-course, physical layer, accumulation layer and the database that is used for the flow instance of storage service; After described workflow engine was disposed, by the business information that described interface layer reception operation system or other interface systems send, described other interface systems comprised resource management system, service release management system, billing and accounting system; Described interface layer provides the interface of three kinds of modes, comprises api interface, Corba interface, WebService interface, is convenient to being connected of workflow engine and operation system; Described business information comprise professional flow instance, flow instance when the prosomite performance; After described key-course received the business information of described interface layer transmission, the flow process route control method according to supporting composite service came generation, scheduling, decomposition, merging, the end of the flow instance of service control; And determine that professional flow instance is that automatic flow arrives next link, still need the original place to wait for; The method that described key-course of while calls described physical layer to be provided records performance and the definite circulation result of described flow process route control method when prosomite of flow instance, and result of flow is returned by described interface layer; The method that described physical layer provides is: to the described management object in workflow engine inside newly-increased, revise, the operation of deletion and inquiry, described management object comprises: professional flow instance object, flow process routing object, flow instance when prosomite object, flow tasks object; Wherein, the specific tasks of the link of described each flow instance of flow tasks object factory execution; Described accumulation layer is by the flow instance information of the described business of the persistent preservation of described database.Can also merge a kind of verification method of Optimized Operation in the described flow operations process, concrete steps are:

(1) scheduling scheme generates: according to product marketing plan, purchasing of raw materials plan, equipment maintenance plan, product (intermediate, raw material) inventory information and equipment capacity, the production constraint informations such as the resources of production take, consumption, production cost, the data file that generates object module is set up the mathematical model of Optimized Operation, and the optimization aim of setting according to the user (maximum productivity, maximum profit or satisfy sales order), and the model solution device, calculate the Optimized Operation scheme.

(2) it is unusual to predict the production that may occur by reckoning: gather product marketing plan current in the enterprise resource planning, purchasing of raw materials plan (comprising the arrival situation), inventory information, Plant maintenance plan, energy supply plan etc., follow the tracks of reckoning by promiscuous device constraint (comprising batch production equipment and continuous producing apparatus), capacity-constrained, logistics Constraints of Equilibrium and energy constraint by the time, the production that precognition may occur is unusual.

(3) Optimized Operation scheme (data that step 1 is obtained) and the production that may occur unusual (data that step 2 is obtained) are combined, utilize the visual ESCPetri-Nets network technology emulation production run of figure.Take scheduling time as axis, dynamic graph display workshop equipment working condition, material balance (buying-stock-sale), simplation verification production scheduling situation.Centered by material, with the scheduling time variation, dynamically show the variation tendency equipment operation condition relevant with certain material production consumption with certain moment of certain material in the mode of figure.

Claims (1)

1. one kind is used for towards the assay method of the major hazard source of safety in production cloud service platform system of industrial and mining enterprises, it is characterized in that: described safety in production cloud service platform system towards industrial and mining enterprises, comprise secure cloud service platform portal sub-system, system management and related tool collection development subsystem, the application service layer subsystem, virtual resource straton system, platform service supporting layer subsystem, access and adaptation layer subsystem, security service resource subsystem, infrastructure services supporting layer subsystem; The major hazard source evaluation is estimated two aspects from natrual danger evaluation and real danger and is carried out in the described platform service supporting layer subsystem, increase external environment condition and offset the factor, following evaluation model is proposed: the numbering of selecting dangerous matter sources, then search the related data of database according to this numbering, obtaining data analyzes later on, judge the species number of the material that comprises in these dangerous matter sources data, whether toxic, according to

$A=\left\{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}{\left(B_{111}\right)}_i{W}_{\mathrm{ij}}{\left(B_{112}\right)}_j\right\}{B}_{12}(1-B_{2k})$

Result of calculation determine dead radius, and then determine a border circular areas, (B in the formula _{111) i}---the physical hazard evaluation of estimate of i kind material; (B ₁₁₂) _j---the technique hazard assessment value of j kind material; W _Ij---the related coefficient of j item technique and i kind physical hazard; B ₁₂---the serious evaluation of estimate of accident; B ₂₁---the factor is offset in the building structure of process equipment container; B ₂₂---peopleware is offset the factor; B ₂₃---the factor is offset in safety management; N is natural number; Described border circular areas is exactly the scope of this dangerous matter sources impact, and comes to the dangerous matter sources classification by the dead radius of maximum.

Images