Summary of the invention
The present invention, in order to solve problems of the prior art, provides a kind of industry control signal detection system.
In order to realize above-mentioned object, technical scheme of the present invention is: a kind of industry control signal detection system, comprising:
Data acquisition module, the real-time status data of collecting device electrical specification;
And the historical process data storehouse being used for the real-time status data that storage data acquisition module collects is connected with data acquisition module;
Also comprise be connected with historical process data storehouse and set up the signal process forecast model of signal input and output corresponding relation in electrical specification by the data in historical process data storehouse; Described signal process forecast model, according to the real-time input signal collected in data acquisition module, obtains prediction output signal, and this prediction output signal is sent to
Traffic filter, described traffic filter receive from the prediction output signal of signal process forecast model and data collecting module collected to real time output and compare, if corresponding real time output outputs signal feature conform to prediction, then regarding as effectively can executive signal; Otherwise then regard as invalid refusal executive signal.
Preferably, described data acquisition module comprises:
Digital data acquisition submodule, gathers the digital information of industry control signal;
Analog acquisition submodule, gathers the analog quantity information of industry control signal;
Logical signal gathers submodule, gathers the data message in industry control signal logic bus.
Preferably, the fault analysis judgment models be connected with traffic filter is also comprised.
Preferably, described signal process forecast model is according to neural network, support vector machine or PCA method establishment.
Present invention also offers a kind of detection method of industry control signal, comprise the following steps:
Step S1: the real-time status data of data collecting module collected equipment electrical specification; And stored in historical process data storehouse;
Step S2: with the data in historical process data storehouse for foundation, sets up signal process forecast model;
Step S3: input signal to be checked is in real time sent to signal process forecast model, and prediction of output output signal;
Step S4: prediction output signal feature and real time output feature are carried out consistance contrast by traffic filter; If corresponding real time output outputs signal feature conform to prediction, then regarding as effectively can executive signal; If it is inconsistent that corresponding real time output and prediction output signal feature, then regard as invalid refusal executive signal.
Preferably, also comprise after step s4:
Step S5: incorrect signal is uploaded to fault analysis judge module, fault analysis judges that diagnostic module failure judgement occurs and carries out fault analysis.
Preferably, in described step S1, the data message on the digital information in data collecting module collected equipment electrical specification, analog quantity information, logic bus.
Detection system of the present invention and detection method, the final output function detection of subsystems from each industrial control system, pass through off-line training, with prediction output signal, believable consistance contrast is carried out to the real time output of detected system, the system invasion that Timeliness coverage is caused by information security and fault.Fundamentally prevent information security to cause industrial control system safety, change the method for putting prevention first that industrial control information safety is traditional, with the method for the reverse backstepping of the result of function accuracy, fundamentally solve the information security issue of industrial control system.
Embodiment
The technical matters solved to make the present invention, the technical scheme of employing, the technique effect easy to understand obtained, below in conjunction with concrete accompanying drawing, be described further the specific embodiment of the present invention.
With reference to figure 1, the invention provides a kind of industry control signal detection system, comprising: data acquisition module 1, be used for the real-time status data of collecting device electrical specification; The real-time status of the electrical specification of hardware device in this data acquisition module acquisition system, mainly comprise the voltage of equipment, electric current, frequency etc., these values can be gathered by PLC or DCS system, only need train control system to provide acquisition interface just passable.Preferably, described data acquisition module 1 comprises digital data acquisition submodule 10, analog acquisition submodule 11 and logical signal and gathers submodule 12 in the present invention.Wherein, digital data acquisition submodule 10 gathers digital information, as switch closed condition etc.; Analog acquisition submodule 11 gathers analog quantity information, as voltage, electric current etc.; Logical signal gathers the data message etc. on submodule 12 data acquisition bus.
Industry control signal detection system of the present invention also comprises the historical process data storehouse 2 being connected the real-time status data arrived for storage data acquisition module collection 1 with data acquisition module 1; Foundation for following model provides data basis.Also comprise be connected with historical process data storehouse 2 and set up the signal process forecast model 3 of signal input and output corresponding relation in electrical specification by the data in historical process data storehouse 2.
These data collected and output have corresponding relation, in this case, as long as provide a large amount of historical datas, signal process prediction module carries out off-line training modeling to it, the relation wherein between input and output can be found out, represent with y=f (x), wherein y represents output quantity; X represents input variable.Y may have multiple output quantity y1, y2...ym; X also may have multiple input quantity x1, x2...xn.
By different types of historical data source, can rapid build model, analyze continuous, discrete, batch production process.Data encasement, visual check data can be carried out, can also rule model be set up.Draw the interact relation between parameters, predict the value of inconvenient measurement parameter, carry out off-line or online optimal control, the knowledge excavated from these models, help to process improve income assess, whole process also very simple and convenient, workload is little.Use this model system, the reason of production run fluctuation can be found out, and adjust, reach stabilized product quality, improve the target of output.
The signal process forecast model 3 of the corresponding relation of input and output can be built by existing technology.Such as according to neural network, support vector machine or PCA method.
Artificial neural network (Artificial Neural Networks, be abbreviated as ANNs) also referred to as neural network (NNs) or be called link model (Connection Model), it is a kind of imitation animal nerve network behavior feature, carries out the algorithm mathematics model of distributed parallel information processing.This network relies on the complexity of system, by adjusting interconnective relation between inner great deal of nodes, thus reaches the object of process information.
The algorithm that neural network is used is exactly vector multiplication, and extensive symbolization function and variously to approach.Parallel, fault-tolerant, can hardware implementing and self-teaching characteristic, being several principal advantages of neural network, is also the difference place of neural computing method and classic method.
Support vector machine method is that the VC being based upon Statistical Learning Theory ties up on theoretical and Structural risk minization basis, between the complicacy (namely to the study precision of specific training sample) and learning ability (namely identifying the ability of arbitrary sample error-free) of model, optimal compromise is sought, in the hope of obtaining best Generalization Ability according to limited sample information.
DUAL PROBLEMS OF VECTOR MAPPING in the space of a more higher-dimension, is set up and is had a largest interval lineoid by support vector machine in this space.Two lineoid parallel to each other are had on the both sides of the separately lineoid of data.Setting up the suitable separating hyperplane in direction makes the distance between two parallel with it lineoid maximize.It is assumed to, the distance between parallel lineoid or gap larger, the total error of sorter is less.
Principal component analysis (PCA) (Principal Component Analysis, PCA) or pivot analysis.Be a kind of statistical analysis technique grasping things principal contradiction, it can parse major influence factors from polynary things, discloses the essence of things, simplifies complicated problem.The object calculating major component is that high dimensional data is projected to comparatively lower dimensional space.M observed value of a given n variable, form the data matrix of a n ' m, n is usually larger.If the main aspect of things is just embodied on several primary variables, we only need by these separating variables out, to carry out labor.But, in the ordinary course of things, can not directly find out such key variables.At this moment we can represent the main aspect of things with the linear combination of original variable, and PCA is exactly so a kind of analytical approach.
Industry control signal detection system of the present invention, described signal process forecast model 3 is according to the real-time input signal collected in data acquisition module 1, obtain prediction output signal, and this prediction output signal is sent to traffic filter 4, described traffic filter 4 receives real time output that prediction output signal and data acquisition module 1 from signal process forecast model 3 collect and compares, if corresponding real time output outputs signal feature conform to prediction, then regarding as effectively can executive signal; Otherwise then regard as invalid refusal executive signal.
The present invention preferably, also comprise the fault analysis judgment models 5 be connected with traffic filter 4, when traffic filter 4 judges real time output and predict that outputing signal feature does not conform to, by this signal transmission in fault analysis judgment models 5, fault analysis judgment models 5 failure judgement occurs and carries out fault analysis.Fault analysis judgment models 5 pairs of data analysis, utilize neuron algorithm to set up operational model, find out the various factors relevant to fault.Fault can be reported to the police after occurring, and user defines the rank of warning according to the threshold value of data.Alert levels is divided into critical, senior, intermediate, rudimentary, normal condition and unknown accident.
Present invention also offers a kind of detection method of industry control signal, with reference to figure 2, comprise the following steps:
Step S1: the real-time status data of data collecting module collected equipment electrical specification; And stored in historical process data storehouse;
Step S2: with the data in historical process data storehouse for foundation, sets up signal process forecast model;
Step S3: input signal to be checked is in real time sent to signal process forecast model, and prediction of output output signal;
Step S4: prediction output signal feature and real time output feature are carried out consistance contrast by traffic filter; If corresponding real time output outputs signal feature conform to prediction, then regarding as effectively can executive signal; If it is inconsistent that corresponding real time output and prediction output signal feature, then regard as invalid refusal executive signal.
Preferably, also comprise after step s4:
Step S5: incorrect signal is uploaded to fault analysis judge module, fault analysis judges that diagnostic module failure judgement occurs and carries out fault analysis.
Preferably, in described step S1, the data message on the digital information in data collecting module collected equipment electrical specification, analog quantity information, logic bus.
Detection system of the present invention and detection method, the final output function detection of subsystems from each industrial control system, pass through off-line training, with prediction output signal, believable consistance contrast is carried out to the real time output of detected system, the system invasion that Timeliness coverage is caused by information security and fault.Fundamentally prevent information security to cause industrial control system safety, change the method for putting prevention first that industrial control information safety is traditional, with the method for the reverse backstepping of the result of function accuracy, fundamentally solve the information security issue of industrial control system.
The present invention is by preferred embodiment having carried out detailed explanation.But, by studying carefully above, concerning the change of each embodiment with to increase be apparent for one of ordinary skill in the art.Being intended that these changes all and increasing of applicant has all dropped in scope that the claims in the present invention protect.