CN101941649A - Automatic detection and fault diagnosis method of lifting states of travelling vehicles - Google Patents

Abstract

The invention relates to an automatic detection and fault diagnosis method of the lifting states of travelling vehicles. The method has the following beneficial effects: a weighing sensor is adopted to detect the weight of the materials; a current sensor is adopted to detect the current of main hook motors of the travelling vehicles; after acquiring multiple groups of samples obtained by the two sensors, travelling vehicle terminal software carries out fitting by a least square method to obtain a weight-current prediction model; in the process of travelling vehicle lifting, after median filtering on the detected weight of the materials and the corresponding current of the main hooks, intelligent analysis is carried out according to the weight-current prediction model to obtain the accurate weight information and the lifting states of the materials lifted by the travelling vehicles; and timely alarm is carried out in case the weighing sensor or the current sensor has faults, thus improving the accuracy and the real-time property of lifting state monitoring of the travelling vehicles. The automatic identification technology related in the invention can make up for the randomness and the orderless property of the personnel operation warehouse areas and create conditions for full-automatic operation of the travelling vehicles.

Description

Driving handling state detects and method for diagnosing faults automatically

Technical field

The present invention relates to a kind of driving handling state detects and method for diagnosing faults automatically, the variation of this method energy accurate recording weight of material, driving handling state, symmetry retransmits sensor arrangement can carry out the actv. fault detection, for the full-automatic operation of driving a vehicle creates conditions.

Background technology

Driving is that the important tool of production that production material transported and went out to put in storage task is carried out in the material reservoir area.Existing driving degree of automation is lower, can not obtain the weight information and the handling state of handling material accurately and reliably, has become a big technical bottleneck of the full-automatic operation of driving.

The mono-LOAD CELLS is fragile inefficacy under the industrial environment of complexity, poor anti jamming capability, and work life is short.In case damage weight information is made mistakes, thereby bring potential danger to moving operation.Utilize the mono-current sensor to come sense weight, because precision can not provide delicate information inadequately.Simultaneously travelling crane hook often has examination to hang or adjust the action of suspension hook position when the handling material, the shake that all can produce sensor signal this moment, and the mono-sensor is difficult to judge for such instantaneous signal and slings or put down action.Therefore the weighting apparatus of single-sensor mode, because its poor reliability, fault is not easy to identify etc., and reason can not improve the degree of automation of reservoir region management.

Visibly different interval is being sling, had when putting down material to driving main hook current of electric, can auxiliary judgment sling, put down operation according to this characteristic.Can also increase redundancy with being used of LOAD CELLS, fault is effectively reported to the police.

Summary of the invention

Technical matters: in order to improve the accuracy of handling status information identification, the reliability of the information of gaining in weight and handling state recognition and the maintainability of weighting apparatus.The invention provides a kind of driving handling state detects and method for diagnosing faults automatically, under the situation of the normal operation of weighting apparatus, do not need manual intervention just can judge weight of material information and driving handling state automatically, and weighing or current sensor is in time reported to the police when fault is arranged.

Technical scheme: driving handling state of the present invention detects automatically and method for diagnosing faults comprises that specifically step is as follows:

Step 1: under driving handling state automatic detection system off-line state, sling, put down operation,, obtain a plurality of weight of material W by LOAD CELLS and current sensor are gathered with various criterion weight material _kWith the cooresponding electric current I of slinging _Uk, put down electric current I _DkPerhaps steady current I _Sk, utilize method of least square to carry out weight-electric current prediction model that match obtains electric current under weight of material and the different handling states;

Step 2: in the online actual measurement continuous several times electric current and weight information are carried out filtering with median average filter method;

Step 3: by in the actual on-line operation, the intelligent comparative analysis in different handling state current interval under this weight under continual and steady electric current and weight-electric current prediction model during a certain material of handling, judge that accurately driving is sling, down state, and weighing or current sensor in time reports an error when being out of order;

Step 4: show weight information, handling state, or fault alarm.

Described median average filter method is removed a maxim and a minimum value for N data of continuous sampling, calculates the center line average values of N-2 data then.

Described weight of material W _kWith the electric current I of slinging _UkObtain I ' by the least square fitting formula fitting _Uk=a ₀W _k+ a ₁, a wherein ₀And a ₁Be fitting coefficient, the matched curve and the electric current I of slinging _UkMaximum error of measuring be I _{λ umax}, can estimate out at weight of material W _kThe time, the interval I ' of being of the current model of slinging _Uk± I _{λ umax}

Described weight of material W _kWith put down electric current I _DkObtain I ' by the least square fitting formula fitting _Dk=a ₂W _k+ a ₃, a wherein ₂And a ₃Be fitting coefficient, matched curve with put down electric current I _DkMaximum error of measuring be I _{λ dmax}, can estimate out at weight of material W _kThe time, put down the interval I ' of being of current model _Dk± I _{λ dmax}

During a certain material of handling, when when vertical direction is static, electric current does not become I substantially _Sk, obtaining average current after taking multiple measurements is I ' _Sk, average current and vertical direction quiescent current I _SkMaximum error of measuring be I _{λ smax}, can estimate out vertical direction when static, between its Current Zone I ' _Sk± I _{λ smax}

In the described online actual measurement process, obtain weight of material information W ₀With cooresponding current information I ₀, in weight-electric current prediction model that match obtains with I ₀With W in the model ₀Compare between cooresponding Current Zone,

If I ₀∈ I ' _Sk± I _{λ smax}, then do not deal with, wait for observed reading next time,

If I ₀∈ I ' _Dk± I _{λ dmax}, then show down state,

If I ₀Drop on the I ' between Current Zone that slings _Uk± I _{λ umax}, then show the state of slinging,

If I ₀, work as I not in these two intervals ₀More close I ' _Dk, and | I ₀-I ' _Dk|/I ' _Dk≤ α, wherein α determines according to engineering is actual, shows down state, writes down such situation number of times simultaneously, display alarm information when reaching certain number of times; Otherwise work as | I ₀-I ' _Dk|/I ' _Dk＞α, demonstration is weighed or current sensor faults need be overhauled,

If I ₀More close I ' _Uk, and | I ₀-I ' _Uk|/I ' _Uk≤ α shows down state, writes down such situation number of times simultaneously, display alarm information when reaching certain number of times; Otherwise work as | I ₀-I ' _Uk|/I ' _Uk＞α, demonstration is weighed or current sensor faults need be overhauled,

Thereby automatically judge it is the operation of slinging/put down, and weighing or current sensor in time reports an error when being out of order.

Beneficial effect: this method adopts LOAD CELLS to detect weight of material information, current sensor senses driving main hook current of electric information.After the driving terminal software is gathered above-mentioned two kinds of many groups that sensor obtains samples, adopt method of least square to carry out match and obtain weight-electric current prediction model.In driving handling process, after detected weight of material, corresponding main hook electric current carried out the median average filter, carry out intellectual analysis according to weight-electric current prediction model, driven a vehicle accurately handling weight of material information, handling state, and weighing or in time reporting to the police during current sensor faults, improve the accuracy and the real-time of driving handling condition monitoring, avoided the suspension hook examination to hang the error that causes.The automatic identification technology that relates in the invention can remedy the randomness of personnel operation reservoir area, messy property, provides technical foundation for realizing the operation of driving full automaticity.

Description of drawings

The present invention is further described below in conjunction with drawings and Examples.

Fig. 1 is a software flow pattern of the present invention.

Fig. 2 is a driving schematic top plan view involved in the present invention.

Fig. 3 is weight involved in the present invention-current model fitting algorithm diagram of circuit.

Fig. 4 is an intellectual analysis subroutine flow chart involved in the present invention.

Have among the figure: driving cart 1, driving dolly 2, driving worker operator's compartment 3, travelling crane hook 4, LOAD CELLS 5, current sensor 6.

The specific embodiment

The present invention is further detailed explanation below in conjunction with accompanying drawing.

Figure 1 shows that DFD of the present invention, this method adopts LOAD CELLS to detect weight of material information, current sensor senses driving main hook current of electric information.After the driving terminal software is gathered above-mentioned two kinds of many groups that sensor obtains samples, adopt method of least square to carry out match and obtain weight-electric current prediction model.In driving handling process, after detected weight of material, corresponding main hook electric current carried out the median average filter, carry out intellectual analysis according to weight-electric current prediction model, driven a vehicle accurately handling weight of material information, handling state, and weighing or in time reporting to the police during current sensor faults.

Bridge crane structure involved in the present invention as shown in Figure 2.The driving of material reservoir area mainly partly is made up of driving cart 1, driving dolly 2, driving worker operator's compartment 3, travelling crane hook 4 etc.LOAD CELLS 5 is installed in the suspension hook place, and current sensor 6 is installed in driving main hook motor place, and weight-electric current prediction model that two sensing datas and match obtain compares to obtain driving a vehicle slings/put down action.When driving handling material, weight information, the driving states information of determining according to location information and the present invention of driving cart dolly just can be determined the operation of driving a vehicle automatically, thereby provides technical foundation for the full automaticity operation of driving a vehicle.

Figure 3 shows that weight of the present invention-electric current match diagram of circuit, obtain weight of material and corresponding current by repeatedly sampling.Utilize fitting formula (1) and (2) to obtain weight-electric current prediction model I ' _Xk=a ₀W ' _k+ a ₁For different k, calculate maximum error of measuring Δ I _Xk=I _Xk-I ' _Xk, relatively draw | Δ I _Xk| middle maxim is designated as I _{X λ max}, the electric current maximum error of measuring of representing to sling when wherein x is for u is when x represents to put down the electric current maximum error of measuring during for d.Therefore for the different material weight W _k, its electric current can be estimated and be I ' _Xk± I _{X λ max}Amperemeter/ammtr is shown I ' specific to slinging _Uk± I _{λ umax}, putting down electric current is I ' _Dk± I _{λ dmax}, the vertical direction quiescent current is I ' _Sk± I _{λ smax}

$a_0=\frac{\left(\underset{k=1}{\overset{N}{\mathrm{\Σ}}}{W}_k^2\right)\left(\underset{k=1}{\overset{N}{\mathrm{\Σ}}}{I}_{\mathrm{xk}}\right)-\left(\underset{k=1}{\overset{N}{\mathrm{\Σ}}}{W}_k\right)\left(\underset{k=1}{\overset{N}{\mathrm{\Σ}}}{W}_k{I}_{\mathrm{xk}}\right)}{N\underset{k=1}{\overset{N}{\mathrm{\Σ}}}{W}_k^2-{\left(\underset{k=1}{\overset{N}{\mathrm{\Σ}}}{W}_k\right)}^2}---\left(1\right)$

$a_1=\frac{N\underset{k=1}{\overset{N}{\mathrm{\Σ}}}{W}_k{I}_{\mathrm{xk}}-\left(\underset{k=1}{\overset{N}{\mathrm{\Σ}}}{W}_k\right)\left(\underset{k=1}{\overset{N}{\mathrm{\Σ}}}{I}_{\mathrm{xk}}\right)}{N\underset{k=1}{\overset{N}{\mathrm{\Σ}}}{W}_k^2-{\left(\underset{k=1}{\overset{N}{\mathrm{\Σ}}}{W}_k\right)}^2}---\left(2\right)$

Wherein N represents sampling number in formula (1), (2), and k represents which time sampling, I _XkX represent the electric current of slinging during for u, x represents to put down electric current during for d, x represents the vertical direction quiescent current during for s.

Figure 4 shows that the present invention's intelligence comparative analysis figure.Obtain weight of material W by time sampling to sensing data ₀With cooresponding state current I ₀, counting machine 1 adds the total sampling number of a record after once sampling.

If I ₀∈ I ' _Sk± I _{λ smax}, then do not deal with, wait for observed reading next time.

If I ₀∈ I ' _Dk± I _{λ dmax}, then show down state.

If I ₀Drop on the I ' between Current Zone that slings _Uk± I _{λ umax}, then show the state of slinging.

If I ₀, work as I not in these two intervals ₀More close I ' _Dk, and | I ₀-I ' _Dk|/I ' _Dk≤ α, wherein α determines according to engineering is actual.Show down state, unison counter 2 adds such situation number of times of record, when counting machine 2/ counting machine 1＞m (wherein the value of m is chosen according to actual requirement of engineering), display alarm information; Otherwise work as | I ₀-I ' _Dk|/I ' _Dk＞α, demonstration is weighed or current sensor faults need be overhauled.

If I ₀More close I ' _Uk, and | I ₀-I ' _Uk|/I ' _Uk≤ α shows down state, and unison counter 2 adds such situation number of times of record, when counting machine 2/ counting machine 1＞m (wherein the value of m is chosen according to actual requirement of engineering), display alarm information; Otherwise work as | I ₀-I ' _Uk|/I ' _Uk＞α, demonstration is weighed or current sensor faults need be overhauled.

Thereby automatically judge it is the operation of slinging/put down, and weighing or current sensor in time reports an error when being out of order.

Now, Fig. 4 is described further in conjunction with actual.During 20 tons of materials of actual lifting, the main hook current sensor obtained 6 different values every 0.5 second be 15.10mA, 14.40mA, 14.38mA, 14.42mA, 14.43mA, 14.34mA.Remove 15.10mA and 14.34mA according to the median average filter, obtain aviation value 14.41mA, the actv. elimination accidental pulse current, weakened that examination is hung and waited the influence of moving.20 tons of cooresponding electric currents of slinging are 14.40 ± 0.06mA in weight-electric current prediction model, and putting down electric current is 13.23 ± 0.06mA.By find that relatively actual measurement electric current 14.41mA slings between Current Zone in model, then show the action of slinging.If the actual measurement electric current is 13.31mA, then it lift by crane electric current and is putting down between Current Zone, and its more close electric current of slinging is known in further judgement, and is less than 10% with the error of the electric current of slinging, and then is shown as the operation of slinging, and allows counting machine 2 add one.When counting machine 2 divided by the value of counting machine 1 (overall measurement number of times) during greater than m, just show and weigh or current sensor has fault.If the actual measurement electric current is 11.38mA, then it is further judged and knows its more close electric current that puts down at the lifting electric current with put down between Current Zone, and with model in put down electric current error greater than 10%, carry out then that alarm is weighed or current sensor has fault.

Claims (6)

1. a driving handling state detects and method for diagnosing faults automatically, it is characterized in that this method comprises that specifically step is as follows:

Step 1: under driving handling state automatic detection system off-line state, sling, put down operation,, obtain a plurality of weight of material W by LOAD CELLS and current sensor are gathered with various criterion weight material _kWith the cooresponding electric current I of slinging _Uk, put down electric current I _DkPerhaps steady current I _Sk, utilize method of least square to carry out weight-electric current prediction model that match obtains electric current under weight of material and the different handling states;

Step 2: in the online actual measurement continuous several times electric current and weight information are carried out filtering with median average filter method;

Step 3: by in the actual on-line operation, the intelligent comparative analysis in different handling state current interval under this weight under continual and steady electric current and weight-electric current prediction model during a certain material of handling, judge that accurately driving is sling, down state, and weighing or current sensor in time reports an error when being out of order;

Step 4: show weight information, handling state, or fault alarm.

2. driving handling state according to claim 1 detects and method for diagnosing faults automatically, it is characterized in that N the data of described median average filter method for continuous sampling, remove a maxim and a minimum value, calculate the center line average values of N-2 data then.

3. driving handling state according to claim 1 detects and method for diagnosing faults automatically, it is characterized in that: described weight of material W _kWith the electric current I of slinging _UkObtain I ' by the least square fitting formula fitting _Uk=a ₀W _k+ a ₁, a wherein ₀And a ₁Be fitting coefficient, the matched curve and the electric current I of slinging _UkMaximum error of measuring be I _{λ umax}, can estimate out at weight of material W _kThe time, the interval I ' of being of the current model of slinging _Uk± I _{λ umax}

4. driving handling state according to claim 1 detects and method for diagnosing faults automatically, it is characterized in that: described weight of material W _kWith put down electric current I _DkObtain I ' by the least square fitting formula fitting _Dk=a ₂W _k+ a ₃, a wherein ₂And a ₃Be fitting coefficient, matched curve with put down electric current I _DkMaximum error of measuring be I _{λ dmax}, can estimate out at weight of material W _kThe time, put down the interval I ' of being of current model _Dk± I _{λ dmax}

5. driving handling state according to claim 1 automatically detects and method for diagnosing faults, it is characterized in that: during a certain material of handling, when when vertical direction is static, electric current does not become I substantially _Sk, obtaining average current after taking multiple measurements is I ' _Sk, average current and vertical direction quiescent current I _SkMaximum error of measuring be I _{λ smax}, can estimate out vertical direction when static, between its Current Zone I ' _Sk± I _{λ smax}

6. driving handling state according to claim 1 detects and method for diagnosing faults automatically, it is characterized in that: in the described online actual measurement process, obtain weight of material information W ₀With cooresponding current information I ₀, in weight-electric current prediction model that match obtains with I ₀With W in the model ₀Compare between cooresponding Current Zone,

If I ₀∈ I ' _Sk± I _{λ smax}, then do not deal with, wait for observed reading next time,

If I ₀∈ I ' _Dk± I _{λ dmax}, then show down state,

If I ₀Drop on the I ' between Current Zone that slings _Us± I _{λ umax}, then show the state of slinging,

If I ₀, work as I not in these two intervals ₀More close I ' _Dk, and | I ₀-I ' _Dk|/I ' _Dk≤ α, wherein α determines according to engineering is actual, shows down state, writes down such situation number of times simultaneously, display alarm information when reaching certain number of times; Otherwise work as | I ₀-I ' _Dk|/I ' _Dk＞α, demonstration is weighed or current sensor faults need be overhauled,

If I ₀More close I ' _Uk, and | I ₀-I ' _Uk|/I ' _Uk≤ α shows down state, writes down such situation number of times simultaneously, display alarm information when reaching certain number of times; Otherwise work as | I ₀-I ' _Uk|/I ' _Uk＞α, demonstration is weighed or current sensor faults need be overhauled,

Thereby automatically judge it is the operation of slinging/put down, and weighing or current sensor in time reports an error when being out of order.

Images