CN104854524A - Excavator abnormality determination method, management device, and excavator - Google Patents

Abstract

In a period in which an excavator is being operated, and predetermined movements are being performed, a plurality of reference waveforms, which represent changes over time in detection values of target physical quantities obtained from the excavator, are prepared. The presence or absence of abnormalities in an excavator being evaluated, which is of the same type as the aforementioned excavator, is determined on the basis of the reference waveforms, by executing steps (a) and (b) below. (a) In a period in which an excavator being evaluated is being operated, and similar movements to the predetermined movements are being performed, target physical quantities obtained from the excavator being evaluated are detected, and evaluation waveforms, which represent changes over time in the detected values, are acquired. (b) The presence or absence of abnormalities in the excavator being evaluated is determined on the basis of a plurality of reference waveforms and the evaluation waveforms. Abnormalities in the excavator can thus be determined by detecting abnormal changes over a short period.

Description

The abnormality determination method of navvy, management devices and navvy

Technical field

The present invention relates to based on certain physical quantity obtained from navvy detected value judge navvy exception with presence or absence of method and judge navvy exception with presence or absence of management devices and navvy.

Background technology

The Work machines such as navvy are used in various construction site, civil engineering work scene etc., when there occurs fault, require fault repair rapidly.Develop the evaluation system (patent documentation 1,2) of the various parameter detecting exceptions changed based on the state according to Work machine.Such as, abnormal based on multiple parameter detecting such as engine speed, operating oil pressure.As an example, utilize the time integral value etc. of the various parameters of collecting from Work machine.By carrying out time integral, the impact of noise can be got rid of.

In patent documentation 3, disclosing a kind of administrative authority of management and service to carrying out navvy, supplying the fuel supplying part door of fuel and working oil, lease the taxi dealer of hydraulic actuated excavator, check civil engineering work amount and manage the device information transmitting system of the transmissions such as the working-yard superintendent office of the development situation of constructing about the information of navvy.In the device information transmitting system disclosed in patent documentation 3, by the information classification about navvy be the information of the management of Related Work time, the information of the management in Related Work place, relevant periodic maintenance service information, about antitheft information, information etc. about consumables Exchange Service.

Display is had about the monitor scope of the information of navvy according to Department formation.The monitor scope of all departments shows the information useful to this department.

Conventional art document

Patent documentation

Patent documentation 1: Japanese Unexamined Patent Publication 2006-53818 publication

Patent documentation 2: Japanese Unexamined Patent Publication 2007-257366 publication

Patent documentation 3: Japanese Unexamined Patent Publication 2002-203066 publication

Summary of the invention

Invent problem to be solved

If carry out the time integral of the various parameters detected, even if then there is abnormal variation in the short time, this variation can not be detected.The object of this invention is to provide a kind of abnormality determination method that can detect the variation of the exception of short time, the exception of judgement navvy.

According to a technical scheme of the present invention, a kind of abnormality determination method of navvy is provided, prepare to represent and navvy operate and multiple reference waveforms of the time variations of the detected value of the concern physical quantity obtained from above-mentioned navvy during carrying out certain set action, the presence or absence with the exception of the evaluation object navvy of above-mentioned navvy same model is judged with reference to waveform based on above-mentioned, have: above-mentioned evaluation object navvy is operating and during carrying out the action similar with above-mentioned set action by (a), detect the above-mentioned concern physical quantity obtained from above-mentioned evaluation object navvy, obtain the operation of the evaluation waveform of the time variations as detected value, b () judges operation with presence or absence of the exception of above-mentioned evaluation object navvy based on multiple above-mentioned reference waveforms and above-mentioned evaluation waveform.

According to another technical scheme of the present invention, a kind of management devices of navvy is provided, have: memory storage, preserve and to represent at running navvy and the detected value of the concern physical quantity obtained from above-mentioned navvy during carrying out certain set action is time dependent multiple with reference to waveform; Communicator, communicates with evaluation object navvy; And treating apparatus, above-mentioned treating apparatus is during above-mentioned evaluation object navvy carries out the action similar with above-mentioned set action, obtain the time dependent evaluation waveform of detected value as the above-mentioned concern physical quantity obtained from above-mentioned evaluation object navvy, based on multiple above-mentioned presence or absence judging the exception of above-mentioned evaluation object navvy with reference to waveform and above-mentioned evaluation waveform.

According to another technical scheme of the present invention, a kind of navvy is provided, have to preserve and represent the time dependent multiple memory storage with reference to waveform of the detected value of the concern physical quantity obtained during carrying out certain set action and treating apparatus, above-mentioned treating apparatus obtains the time dependent evaluation waveform of detected value of the above-mentioned concern physical quantity obtained during carrying out the action similar with above-mentioned set action, based on multiple above-mentioned presence or absence judging exception with reference to waveform and above-mentioned evaluation waveform.

The effect of invention

The variation of the exception of the short time paying close attention to physical quantity can be detected, judge the exception of navvy.

Accompanying drawing explanation

Fig. 1 is the management devices used in the abnormality determination method of the navvy of embodiment, the block figure judging the navvy of object and of navvy.

Fig. 2 is the process flow diagram of the preparatory stage of the abnormality determination method of embodiment.

The curve map that Fig. 3 is the first pilot of swing arm increase instruction of the set action represented for navvy is described, swing arm reduces the first pilot of instruction and the time dependent example of engine speed.

Fig. 4 A is the curve map of the example represented with reference to waveform, and Fig. 4 B and Fig. 4 C is the curve map of the example representing the evaluation waveform obtained from evaluation object navvy.

Fig. 5 is the curve map of the example represented with reference to waveform.

Fig. 6 A is the chart represented with reference to the characteristic quantity of waveform and an example of typical value, and Fig. 6 B is the chart of the example representing the characteristic quantity evaluating waveform.

Fig. 7 is the process flow diagram of the abnormality determination method of embodiment.

Fig. 8 is the detailed process flow diagram of the step SB2 shown in Fig. 7.

Fig. 9 is the figure of the definition representing the mahalanobis distance MD evaluating waveform.

Figure 10 is the process flow diagram of the step SB2 of the abnormality determination method of another embodiment.

Figure 11 is the curve map of the example representing standardized multiple reference vectors and standardized pricing vector.

Figure 12 be represent standardized abnormal time vector and the curve map of an example of standardized pricing vector.

Figure 13 be represent standardized abnormal time vector and standardized pricing vector, the example different from Figure 12 curve map.

Figure 14 is the chart of the example representing abnormality juding object information.

Figure 15 is the process flow diagram of the process that the treating apparatus of of navvy performs.

Figure 16 is the figure of the example representing the image shown in the display device of of navvy.

To be the figure of another example representing the image shown in the display device of of navvy, Figure 17 B be Figure 17 A represents that icon at navvy is by the figure of an example of the image of display after touching.

The figure of the image that Figure 18 shows when being the scale compression representing map compared with the state of Figure 17 A on the display apparatus.

Figure 19 is the figure of the example representing display image on the display apparatus.

Figure 20 is the figure of the example representing display image on the display apparatus.

Figure 21 is the process flow diagram of the process making cause-effect relationship information.

Figure 22 is the chart representing the measured value of operating variable and another example of abnormal class obtained in step SD1 (Figure 21).

Figure 23 is the histogram of the A duration of runs.

Figure 24 is the chart of the guide look of operating variable after representing sliding-model control and abnormal class.

Figure 25 is the chart representing the prior probability of abnormal presumption model and an example of SNNP probability.

Figure 26 is the process flow diagram using cause-effect relationship information to carry out the method for abnormality juding.

Figure 27 is the chart of the example representing the posterior probability calculated.

Figure 28 is the navvy of another embodiment and the block figure of navvy management devices.

Embodiment

Represent in FIG use in the abnormality determination method of the navvy of embodiment management devices 45, as the block figure judging the navvy 30 of object and 50 of navvy.

In navvy 30, possess vehicle control device 31, communicator 32, GPS (full earth location system) receiver 33, display device 34 and sensor 35.The various operating variables of navvy measured by sensor 35.The measured value of sensor 35 is inputted to vehicle control device 31.In operating variable, such as, comprise the duration of runs, prexxure of the hydraulic pump, cooling water temperature, hydraulic pressure load, working time etc.The measured value of the identification information of navvy, various operating variable and the current location information that calculated by gps receiver 33 send to management devices 45 from communicator 32 via communication line 40 by vehicle control device 31.And then, vehicle control device 31 by the various information displaying about navvy in display device 34.

Management devices 45 comprises communicator 46, treating apparatus 47, memory storage 48 and display device 49.The various information sent via communication line 40 from navvy 30 are inputted to treating apparatus 47 via communicator 46.In memory storage 48, store program, various management information that treating apparatus 47 performs.Treating apparatus 47, based on the measured value of the identification information received from navvy 30, various operating variable, current location information and the management information that is stored in memory storage 48, carries out the abnormality juding of navvy 30.In abnormality juding process, utilize the reference waveform etc. be stored in memory storage 48.Abnormality juding result is exported to display device 49.And then identification information, current location information and abnormality juding object information send from communicator 46 via 50 of communication line 40 to navvy by treating apparatus 47.

50 of navvy comprises transmission circuit 51, treating apparatus 52, memory storage 53, display device 54 and input media 55.In 50 of navvy, use the panel computer terminal of such as touch panel formula.In the case, display device 54 also plays a role as input media 55.

Represent the process flow diagram of the preparatory stage of the abnormality determination method of embodiment in fig. 2.In the preparatory stage, carry out the collection with reference to waveform used in abnormality determination method, and calculate various numerical value subsidiary in reference waveform.

In step SA1, in the set action of the navvy 30 (Fig. 1) of normal state, obtain the concern physical quantity measured by navvy 30.Specifically, the detected value of the concern physical quantity detected by navvy 30 (Fig. 1) is sent to management devices 45 via communication line 40.Set action refers to an action from the various Action Selection the running of navvy.

With reference to Fig. 3, set action is described.Fig. 3 represents the first pilot of swing arm increase instruction, swing arm reduces the first pilot of instruction and the time dependent example of engine speed.At moment t1, if the key that will operate is opened, then engine starts to rotate.Engine speed is now such as about 1000rpm.At moment t2, if engine speed is set as 1200rpm by operator, then engine speed rises to about 1200rpm.

At moment t3, if operator carries out the operation of swing arm rising, then produce the first pilot of swing arm increase instruction.If in moment t4 shut-down operation, then swing arm increase instruction guide pushes back initial value.Now, engine speed is such as maintained 1200rpm.At moment t5, if operator carries out the operation of swing arm reduction, then produce swing arm and reduce the first pilot of instruction.At moment t6, if operation is stopped, then swing arm reduction instruction guide pushes back initial value.Between moment t4 and t5, engine speed rises to about 1800rpm.Automatically engine speed is adjusted according to the operational situation of navvy.

Select the idle running action from moment t1 to t2, the swing arm from moment t3 to t4 raises action and reduce 1 action action as set action from the swing arm of moment t5 to t6.In addition, in addition, hydraulic pressure release movement, revolution action, forward motion, backward movement etc. also can be selected as set action.

As concern physical quantity, such as, adopt engine speed.In addition, also the action according to navvy can be conceived to and other physical quantitys changed.Such as, as concern physical quantity, also can adopt prexxure of the hydraulic pump, be used for the advance, retrogressing, revolution etc. that control navvy action pressure, be used for the action pressure of the hydraulic cylinder controlling swing arm etc.

In step SA2 (Fig. 2), obtain the reference waveform as the time variations paying close attention to physical quantity.When adopt engine speed as concern physical quantity, have selected idle running action as set action, the time variations of the engine speed during obtaining in idle running action in Ta (Fig. 3) is as with reference to waveform.When have selected swing arm as set action and raising action or swing arm reduction action, obtain the time variations of the engine speed during Tb (Fig. 3) or swing arm reduce in action during swing arm raises in action in Tc (Fig. 3) respectively as reference waveform.Obtaining with reference to the length during waveform is such as about 10 seconds.Represent the example with reference to waveform in Figure 4 A.

In step SA3 (Fig. 2), calculate multiple characteristic quantity for 1 with reference to waveform.So-called " characteristic quantity " is the various statistics of the shape imparting feature showing waveform.In the above-described embodiments, as characteristic quantity, the maximal value (hereinafter referred to as characteristic quantity E) of the quantity (hereinafter referred to as characteristic quantity D) of calculating mean value (hereinafter referred to as characteristic quantity A), standard deviation (hereinafter referred to as characteristic quantity B), maximum crest value (hereinafter referred to as characteristic quantity C), spike, signal not life period.

With reference to Fig. 5, the maximal value (characteristic quantity E) of the number (characteristic quantity D) of spike and signal not life period is described.Represent the example with reference to waveform in Figure 5." quantity of spike " is defined as the quantity at the position of waveform crosscut threshold value Pth0.During shown in Fig. 5, at crossover sites H1 ~ H4, waveform crosses threshold value Pth0.Therefore, the quantity of spike is calculated as 4.

Be that signal does not exist interval by section definition lower than threshold value Pth1 for waveform.In the example as shown in fig. 5, occur that signal does not exist interval T1 ~ T4.The maximal value of life period " signal not " refers to that multiple signal does not exist the maximum time-amplitude in interval time-amplitude.In the example as shown in fig. 5, the time-amplitude adopting signal to there is not interval T3 is as the maximal value of signal not life period.Generally speaking, if the fluctuating having the cycle longer in waveform, then the maximal value of signal not life period becomes large.

In Fig. 4 B and Fig. 4 C, represent an example of the waveform of the concern physical quantity (engine speed) when there occurs exception.The standard deviation of the waveform shown in Fig. 4 B is larger than the standard deviation with reference to waveform shown in Fig. 4 A.The maximal value of the signal of the waveform shown in Fig. 4 C not life period is larger than the maximal value of the not life period of the signal with reference to waveform shown in Fig. 4 A.

Repeat above-mentioned steps SA1 to step SA3 (Fig. 2), until obtain the reference waveform of sufficient amount.

If obtain the reference waveform of sufficient amount, then, in step SA4 (Fig. 2), about to the multiple characteristic quantities calculated respectively with reference to waveform, calculate their typical value and standard deviation.Such as mean value, median etc. is adopted as " typical value ".In step SA5, preserve to memory storage 48 (Fig. 1) with reference to waveform, characteristic quantity, typical value and standard deviation.

Represent an example of multiple respective characteristic quantity A ~ characteristic quantity E of reference waveform WF (i) and the typical value of each characteristic quantity and standard deviation in fig. 6.Here, parameter i is natural number.Characteristic quantity A ~ characteristic quantity E with reference to waveform WF (i) represents with a (i) ~ e (i) respectively.The typical value (such as mean value) of characteristic quantity A ~ characteristic quantity E is represented with Xa ~ Xe respectively.The standard deviation of characteristic quantity A ~ characteristic quantity E is represented with σ a ~ σ e respectively.

Represent the process flow diagram of the abnormality determination method of embodiment in the figure 7.In step SB1, during carrying out set action by evaluation object navvy, obtain the time variations of the detected value paying close attention to physical quantity from evaluation object navvy.The time variations of the detected value of the concern physical quantity obtained from evaluation object navvy is called evaluation waveform.Here, set action and set action when paying close attention to physical quantity and obtain reference waveform and to pay close attention to physical quantity identical.In addition, obtain set action when evaluating waveform and obtain and need not be identical action with reference to set action during waveform.Such as, when set action is swing arm rising action, even if differences such as the move angles of swing arm ramp-up rate and swing arm, alternatively these two actions are identical set actions.Under various parameters in action do not need identical meaning, obtaining set action when evaluating waveform and obtaining with reference to set action during waveform is alternatively mutually similar action.In addition, evaluation object navvy is the navvy with the navvy same model as the object obtained with reference to waveform.

In step SB2, based on the reference waveform be stored in memory storage 48 (Fig. 1) and the evaluation waveform obtained in step SB1, judge the presence or absence of the exception of evaluation object navvy.About decision method with presence or absence of exception, be described with reference to Fig. 8 below.In step SB3, result of determination is exported to display device 49 (Fig. 1).

Represent the process flow diagram of the step SB2 shown in Fig. 7 in fig. 8.In step SB21, multiple characteristic quantities of Calculation Estimation waveform.As shown in Figure 6B, the value calculated of the characteristic quantity A ~ characteristic quantity E evaluating waveform is expressed as ao ~ eo.

In step SB22, will there is the reference waveform of multiple characteristic quantity A ~ characteristic quantity E as variable as unit space, the mahalanobis distance of Calculation Estimation waveform.

Represent the definition of the mahalanobis distance MD evaluating waveform in fig .9.In this definition, ao ~ eo (Fig. 6 B) is the value of the characteristic quantity A ~ characteristic quantity E evaluating waveform respectively, Xa ~ Xe is the typical value (such as mean value) (Fig. 6 A) of multiple characteristic quantity A ~ characteristic quantity E with reference to waveform respectively, and σ a ~ σ e is the standard deviation (Fig. 6 A) of multiple characteristic quantity A ~ characteristic quantity E with reference to waveform respectively.Comprise the correlation matrix that r (A, A) ~ r (E, E) is the characteristic quantity A ~ characteristic quantity E with reference to waveform as the matrix of key element.

In step SB23 (Fig. 8), the mahalanobis distance evaluating waveform is compared with decision threshold.Decision threshold is pre-stored in memory storage 48 (Fig. 1).In step SB24, based on the comparative result of mahalanobis distance and decision threshold, judge the presence or absence of the exception of evaluation object navvy.Such as, if mahalanobis distance MD is more than decision threshold, then judges that evaluation object navvy is abnormal, be judged to be normal in other cases.If it is abnormal that evaluation object navvy is determined, then, determine the candidate of abnormal class in step SB25 after, perform step SB3 (Fig. 7).If it is normal that evaluation object navvy is determined, does not then perform step SB25 and perform step SB3 (Fig. 7).

Below, an example of the defining method of the candidate of the abnormal class in step SB25 is described.First, the waveform of concern physical quantity under various abnormal state and the characteristic quantity of this waveform are associated with abnormal class Parallel database by navvy.These characteristic quantities are carried out principal component analysis (PCA) as the factor.The waveform of same abnormality can be seen in major component coordinate system to the tendency that specific region (hereinafter referred to as known exception concentrated area) is concentrated.

The position of the characteristic quantity of the evaluation waveform in major component coordinate system is included in certain known exception concentrated area, can infers and there occurs this known exception in evaluation object navvy.

Such as, its standard deviation of the waveform shown in Fig. 4 B becomes more greatly a key factor, and mahalanobis distance becomes large.In addition, the maximal value of its signal of the waveform shown in Fig. 4 C not life period becomes more greatly a key factor, and mahalanobis distance becomes large.Therefore, it is abnormal for judging into the evaluation object navvy achieving the waveform shown in Fig. 4 B, Fig. 4 C.

In contrast, the integrated value of the integrated value of the waveform shown in Fig. 4 B, Fig. 4 C or the waveform shown in mean value and Fig. 4 A or mean value roughly equal.Thus, when carry out based on the integrated value of waveform or mean value abnormal with presence or absence of judge, have achieving Fig. 4 B, the evaluation object navvy of waveform of Fig. 4 C be judged to be normal situation.In the abnormality determination method of embodiment, the evaluation object navvy achieving the waveform shown in Fig. 4 B, Fig. 4 C can be judged to be exception.

Then, with reference to Figure 10 ~ Figure 12, the abnormality determination method of another embodiment is described.Below, the difference with the embodiment shown in Fig. 1 ~ Fig. 9 is described, about identical incomplete structure explanation.The step SB2 of the abnormality determination method of the embodiment shown in Fig. 7 is changed to the process flow diagram shown in Figure 10 by the abnormality determination method of the embodiment shown in Figure 10 ~ Figure 12.

Represent the process flow diagram of the step SB2 of the abnormality determination method of the present embodiment in Fig. 10.In step SB21, the characteristic quantity of Calculation Estimation waveform.The step SB21 of the embodiment shown in this operation with Fig. 8 is identical.In step SB221, respectively will carry out standardization with reference to vector about characteristic quantity with what be key element with reference to respective multiple characteristic quantities of waveform, be 0 to make mean value, standard deviation is for 1.Multiple mean vector with reference to vector (standardization is with reference to vector) after standardization are zero vector.In the example shown in Fig. 6 A, the standardized characteristic quantity A with reference to waveform WF (i) represents with (a (i)-Xa)/σ a.

Represent the example of multiple standardization with reference to vector in fig. 11.In fig. 11, by standardization with reference to vector representation be 2 n dimensional vector ns with characteristic quantity A and these two key elements of characteristic quantity B.The front end of standardization with reference to vector is represented with the circle mark of hollow.Due to standardization with reference to the mean vector of vector be zero vector, the standard deviation of each characteristic quantity is 1, so standardization is distributed in the region 70 of the circle of the vicinity of initial point with reference to vector.Region 70 is called " reference area ".

In step SB222 (Figure 10), use the mean value of the characteristic quantity of reference waveform WF (i) and standard deviation to carry out standardization the pricing vector being key element with the characteristic quantity evaluating waveform, generate standardization pricing vector.The mean vector (i.e. zero vector) of this standardization pricing vector and standardization reference vector is contrasted.Represent an example of standardization pricing vector 71 in fig. 11.In the example shown in Figure 11, standardization pricing vector 71 departs from significantly from reference area 70.

In step SB223, based on standardization with reference to the mean vector (zero vector) of vector and the comparing result of standardization pricing vector 71 (Figure 11), judge the presence or absence of the exception of evaluation object navvy.As an example, when standardization pricing vector 71 (Figure 11) is positioned at the inner side of reference area 70 (Figure 11), being judged to be that evaluation object navvy is normal, when being positioned at outside, being judged to be that evaluation object navvy is abnormal.Also the presence or absence of the exception of evaluation object navvy can be judged based on the similar degree (Euclidean distance, manhatton distance etc.) of mean vector and standardization pricing vector 71.

When being judged to be abnormal in step SB223, in step SB224, determine the candidate of abnormal class.Then, in step SB3 (Fig. 7), result of determination is exported.Be judged to be normal situation in step SB223 under, the candidate of uncertain abnormal class, and result of determination is exported in step SB3 (Fig. 7).

With reference to Figure 12, to determining that in step SB224 (Figure 10) example of the method for the candidate of abnormal class is described.

Characteristic quantity is calculated over time in advance, vector when obtaining exception based on the concern physical quantity obtained from the navvy distinguishing abnormal class.During abnormal class identical multiple abnormal, vector has intensive to the tendency in specific region.

An example of vector and standardization pricing vector when representing that standardization is abnormal in fig. 12.Intensive specific region (X abnormal area) 80 from vector during the standardization exception that the navvy of the exception that there occurs certain abnormal class X obtains, intensive another specific region (Y abnormal area) 82 from vector during the standardization exception that the navvy of the exception that there occurs another abnormal class Y obtains.Obtain the vector average value of vector during intensive standardization exception in X abnormal area 80, mean vector 81 when determining that X is abnormal.Equally, obtain the vector average value of vector during intensive standardization exception in Y abnormal area 82, and mean vector 83 when determining that Y is abnormal.

Obtain in advance X abnormal time mean vector 81 and Y abnormal time mean vector 83, be stored in memory storage 48 (Fig. 1).Also obtain the mean vector during exception corresponded to from which all different abnormal class of abnormal class X and abnormal class Y in advance, and be stored in memory storage 48 (Fig. 1).

By standardization pricing vector 84,85 with various abnormal time mean vector compare.At standardization pricing vector with the difference of mean vector is less time abnormal, be speculated as and there occurs the exception corresponding with mean vector during this exception.In the example shown in Figure 12, when standardization pricing vector 85 is abnormal with X, the difference of mean vector 81 is less.Therefore, be speculated as there occurs the exception that abnormal class is X in the evaluation object navvy achieving standardization pricing vector 85.

When standardization pricing vector 84 is abnormal from which, mean vector is all left.Be speculated as and there occurs unknown exception in the evaluation object navvy achieving standardization pricing vector 84.In addition, based on the length of standardization pricing vector 84,85, the importance degree of abnormality juding result is judged.Standardization pricing vector 84,85 longer, the importance degree of corresponding exception is higher.

In the embodiment shown in Figure 10 ~ Figure 12, also same with the embodiment shown in Fig. 1 ~ Fig. 9, the exception of the time variations of the short time paying close attention to physical quantity can be detected.

Then, with reference to Figure 13, the abnormality determination method of another embodiment is described.Below, the point different from the embodiment shown in Figure 10 ~ Figure 12 is described, about identical incomplete structure explanation.In the embodiment shown in the embodiment shown in Figure 13 and Figure 10 ~ Figure 12, the process of the candidate of the determination abnormal class in step SB224 (Figure 10) is different, and other process are identical.

In the embodiment shown in fig. 12, the profile of X abnormal area 80 and Y abnormal area 82 is circular.In fact, as shown in figure 13, X abnormal area 80, Z abnormal area 90 etc. is had to have the situation of the longer profile along the straight line through initial point.The example of candidate of the abnormal class determining two standardization pricing vectors 86,87 is under these circumstances described.

In the embodiment shown in fig. 13, preserve in memory storage 48 (Fig. 1) mean vector 81 and Z when making X abnormal abnormal time mean vector 88 length be 1 X abnormal time unit vector 81u and Z abnormal time unit vector 88u.Equally, about other abnormal class, the unit vector when length of also preserving mean vector when making exception in memory storage 48 is the exception of 1.Compared by unit vector and standardization pricing vector 86,87 when these are abnormal, determine the candidate of abnormal class.

Specifically, obtain X abnormal time unit vector 81u and standardization pricing vector 86,87 angulation.When this angle is less than decision threshold, the candidate as the exception occurred in the navvy of evaluation object can enumerate abnormal class X.In the example shown in Figure 13, when standardization pricing vector 86 is abnormal with X, unit vector 81u angulation is less than decision threshold.Therefore, be speculated as there occurs the exception that abnormal class is X in the navvy achieving the evaluation waveform corresponding with standardization pricing vector 86.In contrast, unit vector 81u angulation is larger than decision threshold when the standardization pricing vector 87 of the opposing party is abnormal with X.Therefore, the exception that to there occurs in the navvy achieving the evaluation waveform corresponding with standardization pricing vector 87 beyond abnormal class X is speculated as.

Then, to adopting the effect of the embodiment shown in Figure 13 to be described.As an example, in X abnormal area 80 and Z abnormal area 90, be distributed with the standardization exception vector that standardization exception vector that abnormal class is X and abnormal class are Z respectively.When X is abnormal, the length D1 of mean vector 81 and the differential vector of standardization pricing vector 86 and X is abnormal, mean vector 81 is roughly equal with the length D2 of the differential vector of the standardization pricing vector 87 of the opposing party.When standardization pricing vector 87 is abnormal with Z, the length D3 of the differential vector of mean vector 88 is longer than length D1.When standardization pricing vector 87 is abnormal with Z, mean vector 88 angulation is less than decision threshold.

Only based on as evaluation object standardization pricing vector and various abnormal time mean vector the length of differential vector determine in the method for the candidate of abnormal class, as the abnormal class of standardization pricing vector 87 candidate and extract abnormal class X.Resolve known by various abnormal data, standardization pricing vector 87 is roughly irrelevant with abnormal class X, represents that the situation of the omen that abnormal class Z occurs is more.

In the embodiment shown in fig. 13, based on standardization pricing vector 87 with various abnormal time unit vector angulation, extract the candidate of abnormal class.In the example shown in Figure 13, when when standardization pricing vector 87 is abnormal with Z, unit vector 88u angulation is more abnormal with X than standardization pricing vector 87, unit vector 81u angulation is little.Therefore, as standardization pricing vector 87 abnormal class and extract abnormal class Z.Like this, in the embodiment shown in fig. 13, the extraction precision of the candidate of abnormal class can be improved.

After the candidate extracting abnormal class, can abnormal relative to Z based on the length of standardization pricing vector 87 time mean vector 88 the ratio of length infer abnormal degree.When both smaller, can infer that abnormal degree is lower, when both larger, can infer that abnormal degree is higher.And then, when inferring abnormal degree, unit vector 81u, 88u when also can use exception.

Then, with reference to Figure 14 ~ Figure 20, the embodiment of the process of 50 (Fig. 1) about navvy is described.In this embodiment, by the information displaying relevant with the degree (importance degree) of the exception calculated in the embodiment illustrated with reference to Fig. 2 ~ Figure 13 on 50 of navvy.

Represent an example of abnormality juding object information in fig. 14.Abnormality juding object information comprises the importance degree of abnormal class, part name, abnormal position, abnormal part, countermeasure and exception.Abnormal class is the identification code determining to be speculated as the exception occurred in object navvy 30.Abnormal importance degree such as represents with " severe ", " moderate ", " slightly " and " normally " this 4 stage.As an example, will the anomaly classification of engine stop be brought for " severe ", will the anomaly classification of the significant hydraulic performance decline of engine be brought for " moderate ", be " slightly " by the anomaly classification that can be worked on by emergency function.To not there is abnormal state classification for " normally ".Such as engine controller exception etc. is comprised in the exception of " severe ".Fuel leakage, fuel blocking, engine distribution broken string etc. are comprised in the exception of " moderate ".Abnormality of temperature sensors, boosting sensor abnormality etc. is comprised in the exception of " slightly ".

Represent the process flow diagram of the process that the treating apparatus 52 of 50 (figure 01) of navvy performs in fig .15.If start the status display routine of navvy, then in step SC1, treating apparatus 52 receives the abnormality juding object information (Figure 14) of the respective identification information of the multiple navvys 30 as management object, the respective current location information of navvy 30 and the respective of navvy 30 via transmission circuit 51 from management devices 45 (figure 01).

In step SC2, treating apparatus 52 (figure 01) based on the current location information of the multiple navvys 30 received from management devices 45 (figure 01), the scope of the map that decision will show to display device 54 (figure 01).Such as, the engineer's scale of map is determined, with the current location making the map of display comprise whole navvy 30 of management object.In addition, also can determine the scope of the map that will show, to comprise the current location of at least 1 navvy 30 of management object.

In step SC3, in the display device 54 (figure 01) of 50 of navvy, be presented at the map of the scope determined in step SC2.And then, the position corresponding with the current location of the navvy 30 of management object on the map of display, the icon of display navvy.By the icon of navvy so that the form display of the importance degree of the abnormality juding result based on abnormality juding object information can be identified.

Represent an example of the image be presented in the display device 54 of 50 (figure 01) of navvy in figure 16.In display frame, ensure that map display area 60, icon declare area 61 and navvy information display area 62.In map display area 60, show map, show the icon 63 of navvy at the position corresponding with the current location of navvy.The icon 63 of navvy has the flat shape corresponding with the profile of navvy, and is the importance degree of exception occurred in navvy and distinguish color display by inference.Such as, be " severe ", " moderate " by importance degree, the icon 63 of the navvy of " slightly " and " normally " distinguish respectively color be red, pink, yellow and blue.In icon declare area 61, the color of icon of display navvy and the corresponding relation of importance degree.

In order to identify the importance degree of abnormality juding result, the mode beyond also the icon of navvy can being distinguished with color shows.Such as, the thickness of pie graph target line also can be made different, also can make varying in size of icon.Or, also can make abnormality juding result be the navvy of " severe " icon flicker.

In navvy information display area 62, the information of navvy is shown with sheet form.Such as, the information of navvy comprises the model of navvy, body number, location, the value of timer and the importance degree of exception.And then, according to the body number of navvy, show the button being used for being linked to details.If select this button by touch etc., then the details of the navvy of the body number that display is corresponding with selected button.In details, comprise the information relevant with the abnormal class shown in Figure 14, part name, abnormal position, abnormal part and countermeasure.

Maintenance management personnel, according to the information be presented on 50 (figure 01) of navvy, easily can identify the distribution of the navvy of management object and be presumed to the current location that there occurs abnormal navvy.

Another example of the image be presented in map display area 60 is represented in Figure 17 A.In the example shown in Figure 17 A, the icon 63A for 1 navvy imparts and draws note portion 64.In the number of units of the navvy that the numeric representation of drawing display in note portion 64 exists at the position of the icon 63A showing 1 navvy.Figure 17 A means that the place on the map of the icon 63A showing navvy exists 3 navvys.

When there is multiple navvy in certain the narrow and small partition on map, if shown by the icon of whole navvys, then icon is overlapping and be difficult to the number of units and the importance degree that identify navvy.In the example shown in Figure 17 A, display is present in the icon of the highest navvy of importance degree abnormal in the multiple navvys in narrow and small partition, the display of the icon of other navvys is omitted.Even without the icon of other navvys of display, owing to showing the icon of the highest navvy of abnormal importance degree, so also the attention of maintenance management personnel can be caused.In addition, according in the numeral of drawing display in note portion 64, the number of units of navvy can easily be grasped.

If touched by the icon 63A with the navvy drawing note portion 64, then centered by the place of icon 63A showing navvy, map is amplified display.

In Figure 17 B, represent the image be presented at after being touched by the icon 63A (Figure 17 A) of navvy in map display area 60.Under the state of Figure 17 A, icon 63B, 63C of two navvys of not display is shown.Like this, the icon display not having the navvy shown under the state of Figure 17 A can easily be made.Thus, maintenance management personnel can identify the current location of whole navvys and the importance degree of exception.

In Figure 17 A, by the display of the icon beyond the navvy that the importance degree of omission abnormality juding result is the highest, can more easily identify relative to other navvys the navvy that the importance degree of abnormality juding result is the highest.Also navvy the highest for the importance degree of abnormality juding result can be carried out display icon in other modes that can easily identify relative to other navvys.Such as, also the icon of navvy lower for importance degree be configured in relative lower floor, the icon of navvy higher for importance degree be configured in the mode on relative upper strata, multiple icon overlap can be shown.

In figure 18, the image be presented at when making the engineer's scale of map less than the state of Figure 17 A in map display area 60 is represented.If the engineer's scale of map diminishes, then the number of units being present in the navvy in identical scope in display frame increases.The icon 63A of the navvy shown in Figure 17 A is present in icon will be gathered in the identical partition of expression with the icon apart from its nearest position in the map of small scale.In the case, in the example shown in Figure 18, on the icon 63A of navvy with the numerical value drawn in note portion 64 increase to " 4 " from " 3 ".Equally, in other places in map, also have and multiple icons of the navvy of display independent in Figure 17 A are represented, by the display abridged situation of other icons with 1 icon in figure 18.In the case, what in the icon of representative, show the number of units of expression navvy draws note portion.Like this, the number of the icon of the navvy in the partition of the reference area be presented on map is adjusted according to the engineer's scale of the map of display.

As shown in figure 19, also the current location bearing the service car of the maintenance of navvy can be shown in map display area 60.Management devices 45 (figure 01) receives current location information from service car.50 (figure 01) of this current location information to navvy is sent.If of navvy 50 receives the current location information of service car, be then presented on the map in map display area 60, corresponding with the current location of service car position, the icon 65 of display service car.Also the icon 63A ~ 63C of navvy is shown in identical map.Thus, the maintenance management personnel taking service car easily can grasp the position relationship between the position of the current location of oneself and the navvy of management object.Like this, the current location of the multiple navvys as management object be dispersed on a large scale and the state of navvy can easily be grasped.

As shown in figure 20, the path 66 from the current location of service car to the current location of specific navvy can also be shown.The icon 63A of the navvy as object touches by maintenance management personnel.If treating apparatus 52 detects that the icon 63A of navvy is touched, then obtain the path 66 from service car to the current location by the navvy represented by the icon 63A touched, and be shown on map.Thus, maintenance management personnel easily can move to the navvy as object.

In the above-described embodiments, management devices 45 (Fig. 1) has the function of the abnormality juding carrying out navvy, and 50 of navvy has the function of the importance degree display of the exception occurred in navvy.As another example, 50 of navvy also can be made to have the function of the abnormality juding carrying out navvy.In other words, management devices 45 also can be made to have the function of 50 of navvy, by the realizations such as panel computer terminal of this management devices 45.

In the case, do not need management devices 45, directly communicate with between navvy 30 at 50 of navvy.In the memory storage 53 of 50 of navvy, store program, various management information that treating apparatus 52 performs.Treating apparatus 52, based on the measured value of the identification information received from navvy 30, various operating variable, current location information and the management information that is stored in memory storage 53, carries out the abnormality juding of navvy 30.

With reference to Figure 21 ~ Figure 27, the abnormality determination method of the navvy of another embodiment is described.The process that the abnormality juding of navvy carries out abnormality juding by the process and use cause-effect relationship information that are constructed for the cause-effect relationship information of carrying out abnormality juding is formed.

Represent the process flow diagram being constructed for the process of the cause-effect relationship information of carrying out abnormality juding in figure 21.In step SD1, management devices 45 (figure 01) obtains the measured value of operating variable and the abnormal class of generation during collecting this measured value from multiple navvys 30 (figure 01) of management object.

Represent the measured value of the operating variable obtained in step SD1 and an example of abnormal class in fig. 22.The measured value of operating variable and obtaining the body number (identification information) according to navvy and carrying out according to during certain collection of abnormal class.To such as be set as during collection 1 day (24 hours).1 evaluation object is formed from the ensemble collected in 1 body is during 1 is collected.

In fig. 22, as an example, the information of evaluation object No.1 is the information obtained from the navvy of body number a on July 1st, 2011, and the duration of runs, A was 24, and pump pressure B is 19, and cooling water temperature C is 15, and hydraulic pressure load D is 11, and working time E is 14." duration of runs " refers to from the starting switch of navvy and is pressed into time till shutdown switch is pressed, i.e. navvy time of having started." working time " refers to that operator operates the time of navvy.In addition, the abnormal class X of evaluation object No.1 is X1.This means, on July 1st, 2011, in the navvy of body number a, there occurs the exception of abnormal class X1.Abnormal class X0 shown in Figure 22 means and exception does not occur.

Then, in step SD2 (Figure 21), carry out the sliding-model control of operating variable, each operating variable is replaced into finite discrete type item.

With reference to Figure 23, the method A duration of runs being replaced into finite discrete type item is described.In addition, finite discrete type item can be replaced into too about other operating variables.

Figure 23 represents an example of the histogram of the A duration of runs.The transverse axis of Figure 23 represents the A duration of runs, and the longitudinal axis represents the quantity (frequency) of evaluation object.If the duration of runs, the mean value of A was μ, standard deviation is σ.Scope till μ-3 σ to μ+3 σ is carried out 3 deciles.That is, transverse axis is divided into μ-3 σ ~ μ-σ, μ-σ ~ μ+σ, these 3 regions of μ+σ ~ μ+3 σ.The partition being below μ-σ by the A duration of runs is set to A1, the partition of μ-σ ~ μ+σ is set to A2, the partition of more than μ+σ is set to A3.

About the A duration of runs, there occurs measured value and get the item of the value in partition A1, the item of getting the value in partition A2 and certain item of getting in the item of the value in partition A3.Represent the guide look of the operating variable after sliding-model control and abnormal class in fig. 24.A duration of runs partition A1, A2, A3 belonging to its measured value is represented.Equally, other operation informations are also replaced into finite discrete type item.

Then, in step SD3 (Figure 21), make cause-effect relationship information, and be kept in memory storage 48 (figure 01).

Shown in Figure 24 by the operating variable A of finite discrete type item, B, C ... establish with abnormal class X the complete list associated to can be described as and be reason item with abnormal class X, take operating variable as the cause-effect relationship information of result item.

Represent the prior probability of abnormal presumption model and an example of conditional probability in fig. 25.With abnormal class X for reason item, with each operating variable for imagining the result item occurred by reason, the cause-effect relationship information according to Figure 24 prior probability P (X) can be calculated.And then, about each operating variable A, B, C ..., the conditional probability P (A|X) of condition premised on the item that causes by abnormal class X respectively can be calculated, P (B|X) ...Represent an example of prior probability P (X) and conditional probability P (A|X), the P (B|X) calculated in fig. 25.

Represent the process flow diagram using cause-effect relationship information to carry out the method for abnormality juding in fig. 26.In step SE1, management devices 45 (figure 01) obtains the measured value of operating variable from the navvy 30 of management object.In step SE2, carry out the sliding-model control of acquired operating variable.This sliding-model control carries out based on the benchmark identical with the sliding-model control carried out in the step SD2 of Figure 21.Represent an example of the operating variable after sliding-model control in figure 27.Such as, the duration of runs, the discretized values of A was A2, and the discretized values of pump pressure B is B3, and the discretized values of cooling water temperature C is C1, and the discretized values of hydraulic pressure load D is D2, and the discretized values of working time E is E2.

In step SE3 (Figure 26), prior probability P (X), the conditional probability P (A|X) etc. that use the cause-effect relationship information according to Figure 22 to obtain, obtain the posterior probability (carrying out Bayesian inference) of each abnormal class.

As an example, there occurs under the duration of runs, A was the condition of the item of A2, the posterior probability P (X=X1|A=A2) (following, to be expressed as P (X1|A2)) that the exception of abnormal class X1 occurs can calculate by following formula.

[numerical expression 1]

$P\left(X1\right|A2)=\frac{P\left(A2\right|X1)P\left(X1\right)}{\underset{X}{\mathrm{\Σ}}P\left(X\right)P\left(A2\right|X)}$

Equally, the posterior probability P (X2|A2) of the exception that there occurs abnormal class X2, X3 etc. can be calculated, P (X3|A2) ...

And then, by the posterior probability P (X1|A2) calculated, P (X2|A2), P (X3|A2) ... again dispose as prior probability, under the condition of the item of B3 in the discretized values that there occurs pump pressure B, the posterior probability P (X1|A2, B3) that the exception of abnormal class X1 occurs can calculate by following formula.In addition, assuming that the duration of runs A and pump pressure B be independently.

[numerical expression 2]

$P\left(X1\right|A2,B3)=\frac{P\left(B3\right|X1,A2)P\left(X1\right|A2)}{\underset{X}{\mathrm{\Σ}}P\left(X\right|A2)P\left(B3\right|X,A2)}$

The P (B3|X1, A2) on the right can the cause-effect relationship information according to Figure 22 obtain.Equally, can obtain and there occurs the abnormal posterior probability P (X2|A2, B3) such as abnormal class X2, X3, P (X3|A2, B3) ...

And then, other operating variables such as cooling water temperature C, hydraulic pressure load D, working time E being added as new result, by calculating posterior probability, the objectivity of the posterior probability calculated can be improved further.

An example of the posterior probability calculated is represented in Figure 27.In this embodiment, infer and as in the navvy of evaluation object, it is 50% that abnormal probability does not occur, and the probability that the exception of abnormal class X1 occurs is 5%, and the probability that the exception of abnormal class X2 occurs is 20%.

In addition, in above-described embodiment 1, add item as a result successively, recalculate posterior probability by stages, but might not need to calculate posterior probability by stages.Also can use the conditional probability P (A|X) of the prior probability P shown in Figure 25 (X) and each operating variable, P (B|X) etc., consider whole operating variables item as a result, calculate the posterior probability of abnormal class.

As described above, by with the discretized values of the measured value of the operating variable shown in Figure 27 item as a result, the cause-effect relationship information shown in Figure 22 of use carries out Bayesian inference, can calculate the posterior probability of the abnormal class as reason item.

Then, in step SE4 (Figure 26), by inferring that the abnormal class that and posterior probability thereof are associated with body number, store to memory storage 48 (figure 01).

In the method shown in Figure 26, there is the situation being derived multiple abnormal class by abnormality juding.In the example shown in Figure 27, infer that the possibility of the exception generation abnormal class X1 is 5%, the possibility of the exception of generation abnormal class X2 is 20%.Like this, when being derived multiple abnormal possibility, as long as adopt the importance degree of the highest exception of posterior probability as inferring that the importance degree of the exception occurred in this navvy is just passable.Or, posterior probability also can be adopted to be the highest abnormality degree in the importance degree of the exception of certain reference value such as more than 20%, as the importance degree inferring the exception occurred in this navvy.

Also the method for carrying out abnormality juding shown in Figure 26 be can replace and method, the method for the embodiment shown in Figure 10 ~ Figure 12, the method etc. of the embodiment shown in Figure 13 of the embodiment shown in Fig. 1 ~ Fig. 9 used.

The navvy of another embodiment and the block figure of navvy management devices is represented in Figure 28.In the embodiment shown in the embodiment shown in Fig. 1 ~ Fig. 9 and Figure 10 ~ Figure 12, in step SA1 (Fig. 2), the detected value paying close attention to physical quantity is sent to management devices 45 from navvy 30 via communication line 40.In the embodiment shown in Figure 28, management devices 45 is mounted on navvy 30.Be mounted in management devices 45 on navvy 30 by the method identical with the abnormality determination method of the embodiment shown in the embodiment shown in Fig. 1 ~ Fig. 9 or Figure 10 ~ Figure 12, judge the presence or absence of the exception of navvy 30.

Result of determination is sent to navvy management devices 25 from navvy 30 via communication line 40.Navvy management devices 25 by the result of determination received from navvy 30 can identify that the mode of the individuality of navvy 30 exports to output unit 26.

In the embodiment shown in Figure 28, via the information just result of determination with presence or absence of exception that communication line 40 is received and dispatched.Therefore, compared with the embodiment shown in Fig. 1 ~ Fig. 9 paying close attention to the detected value of physical quantity with transmitting-receiving and the embodiment shown in Fig. 1 ~ Figure 12, the data volume via communication line 40 transmitting-receiving can be cut down.

In addition, in the embodiment shown in Figure 28,50 (Fig. 1) from the management devices 45 be mounted in navvy 30 to navvy sends various data.In addition, the management devices 45 be mounted on navvy 30 also can be made to have the function of 50 of navvy.In the case, the respective current location of the respective identification information of the multiple navvys as evaluation object and multiple navvys of evaluation object is received with the management devices 45 of 1 navvy 30.Be mounted in the map denotation of the current location of at least 1 of multiple navvys that management devices 45 on navvy 30 will comprise as evaluation object.And then, the position corresponding with the current location of the navvy of evaluation object on the map of display, so that the mode of the importance degree of the exception based on result of determination with presence or absence of exception can be identified, the icon of display navvy.

Describe the present invention according to embodiment above, but the present invention does not limit by them.Such as, for a person skilled in the art, obviously various change, improvement, combination etc. can be carried out.

Label declaration

25 navvy management devices

26 output units

30 navvys

31 vehicle control devices

32 communicators

33 gps receivers

34 display device

35 sensors

40 communication lines

45 management devices

46 communicators

47 treating apparatus

48 memory storages

49 display device

Of 50 navvys

51 transmission circuits

52 treating apparatus

53 memory storages

54 display device

55 input medias

60 map display area

61 icon declare areas

62 navvy information display area

63, the icon of 63A ~ 63C navvy

64 draw note portion

The icon of 65 service cars

66 paths

70 reference areas

80 X abnormal areas

Mean vector when 81 X are abnormal

Unit vector when 81u X is abnormal

82 Y abnormal areas

Mean vector when 83 Y are abnormal

86,87 standardization pricing vectors

Mean vector when 88 Z are abnormal

Unit vector when 88u Z is abnormal

90 Z abnormal areas

Claims (15)

1. the abnormality determination method of a navvy, prepare to represent at running navvy and the detected value of the concern physical quantity obtained from above-mentioned navvy during carrying out certain set action is time dependent multiple with reference to waveform, the presence or absence with the exception of the evaluation object navvy of above-mentioned navvy same model is judged with reference to waveform based on above-mentioned, it is characterized in that there is following operation:

A () at running above-mentioned evaluation object navvy during carrying out the action similar with above-mentioned set action, detect the above-mentioned concern physical quantity obtained from above-mentioned evaluation object navvy, and obtains as the time dependent evaluation waveform of detected value;

B () is based on multiple above-mentioned presence or absence judging the exception of above-mentioned evaluation object navvy with reference to waveform and above-mentioned evaluation waveform.

2. the abnormality determination method of navvy as claimed in claim 1, is characterized in that,

Multiple above-mentioned reference waveform obtains when above-mentioned navvy is normal state.

3. the abnormality determination method of navvy as claimed in claim 1 or 2, is characterized in that,

Obtain multiple characteristic quantity according to multiple above-mentioned reference waveform, obtain typical value according to multiple above-mentioned characteristic quantity;

Operation with presence or absence of the operation obtaining the above-mentioned characteristic quantity of above-mentioned evaluation waveform above-mentioned operation (b) comprising and the exception judging above-mentioned evaluation object navvy based on the above-mentioned characteristic quantity of above-mentioned typical value and above-mentioned evaluation waveform.

4. the abnormality determination method of navvy as claimed in claim 1 or 2, is characterized in that,

Multiple characteristic quantity is obtained according to multiple above-mentioned reference waveform;

Above-mentioned operation (b) comprises operation with presence or absence of the operation obtaining the mahalanobis distance of above-mentioned evaluation waveform in space in units of multiple above-mentioned above-mentioned characteristic quantity with reference to waveform and the exception judging above-mentioned evaluation object navvy based on above-mentioned mahalanobis distance.

5. the abnormality determination method of navvy as claimed in claim 1 or 2, is characterized in that,

Multiple characteristic quantity is obtained according to multiple above-mentioned reference waveform;

In above-mentioned operation (b), obtain the above-mentioned characteristic quantity of above-mentioned evaluation waveform, based on the comparing result of the pricing vector that the multiple reference mean vector of vector being key element with the respective multiple above-mentioned characteristic quantity of multiple above-mentioned reference waveform and the above-mentioned characteristic quantity of above commentary valency waveform are key element, judge the presence or absence of the exception of above-mentioned evaluation object navvy.

6. the abnormality determination method of navvy as claimed in claim 1 or 2, is characterized in that,

Also there is following operation:

Of navvy receives the abnormality juding object information as respective identification information, the respective current location of above-mentioned navvy and the result of determination as above-mentioned operation (b) of multiple navvys of evaluation object;

Of above-mentioned navvy will comprise the map denotation of the current location of at least 1 of multiple navvys of above-mentioned evaluation object in display device, the position corresponding with the current location of the navvy of management object on a displayed map, shows the icon of navvy in the mode of the importance degree that can identify the abnormality juding result based on above-mentioned abnormality juding object information.

7. the abnormality determination method of navvy as claimed in claim 6, is characterized in that,

When there is multiple navvy in certain partition being presented at the map in above-mentioned display device, the icon of navvy the highest for the importance degree of abnormality juding result is shown in the mode identified easier than the icon of other navvys.

8. the abnormality determination method of navvy as claimed in claim 6, is characterized in that,

According to the engineer's scale of the map be presented in above-mentioned display device, adjustment is shown to the quantity of the icon of the above-mentioned navvy in above-mentioned display device.

9. a management devices for navvy, is characterized in that,

Have:

Memory storage, preserves and to represent at running navvy and the detected value of the concern physical quantity obtained from above-mentioned navvy during carrying out certain set action is time dependent multiple with reference to waveform;

Communicator, communicates with evaluation object navvy; With

Treating apparatus;

Above-mentioned treating apparatus is during above-mentioned evaluation object navvy carries out the action similar with above-mentioned set action, obtain the time dependent evaluation waveform of detected value as the above-mentioned concern physical quantity obtained from above-mentioned evaluation object navvy, based on multiple above-mentioned presence or absence judging the exception of above-mentioned evaluation object navvy with reference to waveform and above-mentioned evaluation waveform.

10. the management devices of navvy as claimed in claim 9, is characterized in that,

Above-mentioned reference waveform obtains when above-mentioned navvy is normal state.

The management devices of 11. navvys as described in claim 9 or 10, is characterized in that,

The multiple characteristic quantity obtained according to multiple above-mentioned reference waveform and the typical value obtained according to multiple above-mentioned characteristic quantity is preserved in above-mentioned memory storage;

The above-mentioned characteristic quantity of above-mentioned evaluation waveform obtained by above-mentioned treating apparatus, and the above-mentioned characteristic quantity based on above-mentioned typical value and above-mentioned evaluation waveform judges the presence or absence of the exception of above-mentioned evaluation object navvy.

The management devices of 12. navvys as described in claim 9 or 10, is characterized in that,

Also connect on the display apparatus;

Above-mentioned display device receives as the respective identification information of multiple navvys of evaluation object and the respective current location of above-mentioned navvy, display comprises the map of the current location of at least 1 of multiple navvys of above-mentioned evaluation object, the position corresponding with the current location of the navvy of above-mentioned evaluation object on a displayed map, so that the mode of the importance degree of the exception based on result of determination with presence or absence of exception can be identified, the icon of display navvy.

13. 1 kinds of navvys, have the memory storage and treating apparatus preserving and represent the time dependent multiple reference waveform of the detected value of the concern physical quantity obtained during carrying out certain set action, it is characterized in that,

Above-mentioned treating apparatus obtains the time dependent evaluation waveform of detected value as the above-mentioned concern physical quantity obtained during carrying out the action similar with above-mentioned set action, based on multiple above-mentioned presence or absence judging exception with reference to waveform and above-mentioned evaluation waveform.

14. navvys as claimed in claim 13, is characterized in that,

Above-mentioned reference waveform obtains when above-mentioned navvy is normal state.

15. navvys as described in claim 13 or 14, is characterized in that,

Be connected with display device;

Above-mentioned display device receives as the respective identification information of multiple navvys of evaluation object and the respective current location of above-mentioned navvy, display comprises the map of the current location of at least 1 of multiple navvys of above-mentioned evaluation object, the position corresponding with the current location of the navvy of above-mentioned evaluation object on a displayed map, so that the mode of the importance degree of the exception based on result of determination with presence or absence of exception can be identified, the icon of display navvy.