CN101408951B - Method for obtaining equivalent load spectrum and estimating weariness residual longevity of bridge crane based on neural network - Google Patents
Method for obtaining equivalent load spectrum and estimating weariness residual longevity of bridge crane based on neural network Download PDFInfo
- Publication number
- CN101408951B CN101408951B CN 200810180622 CN200810180622A CN101408951B CN 101408951 B CN101408951 B CN 101408951B CN 200810180622 CN200810180622 CN 200810180622 CN 200810180622 A CN200810180622 A CN 200810180622A CN 101408951 B CN101408951 B CN 101408951B
- Authority
- CN
- China
- Prior art keywords
- crane
- neural network
- fatigue
- load spectrum
- load
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Landscapes
- Control And Safety Of Cranes (AREA)
Abstract
The invention relates to a method for acquiring valent weight load spectrum and estimating fatigue surplus service life of a bridge crane based on the nerve network, and the method is applied to acquire the valent weight load spectrum of a general bridge crane and estimate the fatigue surplus service life of a box-shaped main girder thereof. The invention aims at solving the technical difficult problem in the prior art that a great deal of actual load spectrum actual measurement is difficult to actualize and the random variation tendency of the valent weight load spectrum of the bridge crane is difficult to express by using a determinate mathematic analytic relational expression due to the limitation of field experimental condition. In order to solve the technical difficult problem, the adopted technical proposal includes that the method comprises the following steps: (1) a nerve network model is established; (2) a leaning sample is obtained; (3) an LMBP nerve network is trained; (4) the valent weight load spectrum of the bridge crane is obtained; and at last, a fatigue stress spectrum and the like are calculated by adopting the valent weight load spectrum of the bridge crane obtained by the nerve network, so the fatigue surplus service life of the main girder of the crane can be estimated.
Description
Technical field
The present invention relates to a kind of overhead travelling crane equivalent load spectrum based on neural network obtains and the fatigue surplus life evaluation method, it is applicable to obtaining of general-purpose overhead crane equivalent load spectrum and the estimation of box girder fatigue surplus life thereof, and it is obtaining and the fatigue surplus life estimation applicable to other type crane equivalent load spectrum also.
Background technology
Hoisting machinery plays indispensable vital role in economic construction, be the special equipment in the major technologies and equipment industry, in case have an accident, economic loss is heavy, social influence is abominable.Especially in recent years, the sudden fracture of hoisting machinery structural system and failure event occur in domestic and international many countries and regions in succession, are that fatigue break destroys and cause the one of the main reasons of this class accident.Therefore, national governments and inspection body unprecedentedly pay attention to the research that the hoisting machinery structural system carries out deciding the longevity, lengthening the life.
Solve that crane is decided the longevity, the condition precedent of lengthening the life is to obtain representative typical load---the time history that can reflect the true operating position of crane metal construction, i.e. loading spectrum.At present the Crane Load spectrum obtain the two kinds of methods that mainly contain: the one, field measurement, namely obtain the load of crane by direct test---time history (loading spectrum), this not only needs to consume a large amount of manpower and materials, and because the repeated load of variation is born by the crane structure system, the result of each actual measurement is all not identical, this randomness and uncertainty cause measured result being directly applied to theoretical analysis and engineering practice; The 2nd, use traditional regression method, traditional regression method can find the general relationship of lifted load and working cycle number of times basically, but it is to be based upon a certain amount of mathematical analysis to concern relatively accurate statement on the basis that tradition returns (match) mode, and the concrete actual condition randomness of crane is very strong, is difficult to represent with a definite mathematical analysis relational expression.
Summary of the invention
The objective of the invention is to solve that existing method exists limits, is difficult to carry out a large amount of real loads spectrums because of the field experiment condition and survey and be difficult to represent the technical barrier of the randomness variation tendency of crane equivalent load spectrum with a definite mathematical analysis relational expression, and provide a kind of overhead travelling crane equivalent load spectrum based on neural network of conveniently obtaining equivalent load spectrum, effectively estimating the crane girder fatigue surplus life to obtain and the fatigue surplus life evaluation method.
For solving above-mentioned technical barrier, the present invention proposes a kind of overhead travelling crane equivalent load spectrum acquisition methods based on neural network, this method is utilized the Nonlinear Mapping function of neural network, select the feedforward neural network of the single output of dual input, adopt the specified lifted load of crane (lifting capacity), lifted load and working cycle time logarithmic data is as learning sample, using the error back propagation learning algorithm of LM (Levenberg-Marquardts) BP trains network, obtain optimum weights, realization is obtained the overhead travelling crane equivalent load spectrum, and it comprises the following steps:
(1) sets up neural network model
Specified lifted load and the lifted load of getting overhead travelling crane are input quantity, the working cycle number of times of different lifted loads within a working hour of corresponding such crane one specified lifted load is output quantity, make up the nerve network system of the single output of dual input, described neural network structure is three layers, be input layer, hidden layer and output layer, the input layer number is 2, and the output layer nodes is 1, and the number of hidden nodes is between 13～17;
(2) obtain learning sample
Input take the specified lifted load of overhead travelling crane and lifted load as learning sample, the working cycle number of times of different lifted loads within a working hour of corresponding such crane one specified lifted load is the hope output of learning sample, obtain learning sample, and specified lifted load value is divided into 8 parts at least learning sample carried out classification;
(3) training LMBP neural network
On the basis of the learning sample that neural network model that step (1) obtains and step (2) obtain, training LMBP neural network, and obtain optimum model parameter;
(4) the crane equivalent load spectrum obtains
The specified lifted load lifted loads different from least 8 grades of needs being estimated the overhead travelling crane of fatigue surplus life are input to the neural network that step (3) trains, and can obtain the equivalent load spectrum of this overhead travelling crane.
Sample data in the described step (2) is that overhead travelling crane is in lifted load and the corresponding working cycle number of times thereof under the normal operating conditions, makes learning sample meet the actual applying working condition of crane.
A kind of method that adopts the equivalent load spectrum estimation girder fatigue surplus life of the overhead travelling crane that obtains by neural network, it comprises the following steps:
(1) adopts the equivalent load spectrum of the overhead travelling crane that obtains by the described neural network of above-mentioned (1) to (4) step, calculate the Simulating of Fatigue Stress Spectra of this crane girder span centre dangerouse cross-section;
(2) from Simulating of Fatigue Stress Spectra, extract again cycle index and the global cycle number of times of stress amplitudes at different levels, stress amplitudes at different levels with rain flow method;
(3) calculate the equivalent stress width of cloth σ of this Simulating of Fatigue Stress Spectra according to Miner fatigue damage stress amplitude equivalence formula
Re
(4) with equivalent stress amplitude σ
ReThe computing formula of hardware fatigue surplus life is calculated in substitution: the fatigue surplus life that can estimate crane girder.
Because the present invention has adopted technique scheme, utilize the LMBP neural network that trains, can obtain quickly and efficiently the equivalent load spectrum of overhead travelling crane and the fatigue surplus life of estimation overhead travelling crane key structure spare.Therefore, compare with background technology, the present invention has following substantial characteristics:
(1) based on the neural network that has trained, set up the securing mechanism of the equivalent load spectrum of the type crane, and needn't seek to set up specified lifted load and lifted load and the analytical relation that complexity between corresponding working cycle number of times embarrasses, have realization simply, characteristics effectively efficiently;
(2) utilize the network model that trains, but the equivalent load spectrum by computing machine quick obtaining the type overhead travelling crane and estimate the fatigue surplus life of its girder, thereby greatly save loaded down with trivial details process and a large amount of input of crane field measurement, realize the purpose of obtaining equivalent load spectrum conveniently and accurately estimating the crane girder fatigue surplus life.
Description of drawings
Fig. 1 is schematic flow sheet of the present invention;
Fig. 2 is LMBP neural network topology structure schematic diagram of the present invention;
Fig. 3 is the equivalent stress spectral curve of 75 tons of general-purpose overhead cranes and the measured stress line chart of setting a song to music.
Embodiment
The present invention is described in further detail below in conjunction with drawings and Examples.
As depicted in figs. 1 and 2, the overhead travelling crane equivalent load spectrum acquisition methods based on neural network in the present embodiment, it comprises the following steps:
(1) sets up neural network model
Must obtain first the equivalent load spectrum of crane for realizing the fatigue surplus life estimation of crane key structure spare (girder), specified lifted load (lifting capacity) and the lifted load of getting overhead travelling crane are input quantity, the working cycle number of times of different lifted loads within a working hour of corresponding such crane one specified lifted load (lifting capacity) is output quantity, makes up the single output nerve network system of dual input.Described neural network structure is three layers, i.e. input layer, hidden layer and output layer, and selecting the input layer number is 2, the output layer nodes is 1, the number of hidden nodes optional 15.
(2) obtain learning sample
Primary work with the artificial neural network solving practical problems is to collect the learning sample data.For making the neural network after the study have good performance, collected sample data should be to begin to lift by crane a weight with overhead travelling crane to rise, end to can begin to lift by crane next article the time, a complete procedure that comprises overhead travelling crane operation and rest normally be in lifted load and corresponding working cycle number of times thereof under the normal operating conditions, make learning sample meet the actual applying working condition of crane.
The input of the present embodiment take the specified lifted load (lifting capacity) of overhead travelling crane and lifted load as learning sample, the working cycle number of times of different lifted loads within a working hour of a corresponding specified lifted load (lifting capacity) is the hope output of learning sample, obtain learning sample, and specified lifted load value is divided into 8 parts learning sample carried out classification (the sample classification is The more the better), be about to 0 load value to specified lifted load interval and be divided at least 8 parts, then by equal score value lifted load and corresponding working cycle number of times thereof are carried out classification; Learning sample data in the present embodiment realize by data acquisition system (DAS) and data recording: the support level place that at first signal wire of pulling force sensor is installed in lifting mechanism top sheave assembly axis, then the lifted load that the record pulling force sensor shows in the process that crane works, and within a working hour working cycle number of times of corresponding different lifted loads.Table 1 be depicted as the specified lifted load of obtaining be 100 tons, 120 tons, 140 tons, 160 tons with the lifted load of 180 tons of general-purpose overhead cranes and the working cycle of corresponding different lifted loads within a working hour time fractional part sample data bunch:
Table Bridge 1 formula crane hoisting load working cycle time fractional part sample data bunch
(3) training LMBP neural network
Only have the neural network after the training could realize its function.The learning sample normalization that step (2) obtains, the LMBP neural network that training is determined by step (1) obtains its model parameter.The neural metwork training process is the weight coefficient w1 by continuous adjustment input layer and hidden layer, and the weight coefficient w2 of hidden layer and output layer makes neural network very little until the requirement of satisfied application to the error of the output of non-learning sample and desired value.
(4) the crane equivalent load spectrum obtains
The neural network that step (3) trains has namely been set up the different lifted loads of general-purpose overhead crane of different specified lifted loads and the randomness variation tendency mapping relations of working cycle number of times---and be the equivalent load spectrum of general class overhead travelling crane.The specified lifted load of needs being estimated the general-purpose overhead crane of fatigue surplus life is input to the neural network that step (3) trains with 8 grades of (same well-behaved progression) different lifted loads, can obtain the equivalent load spectrum of this crane.As being that the specified lifted load of the general-purpose overhead crane of 28.5 meters of 75 tons, span is input to the neural network that step (3) trains with 8 grades of different lifted loads with specified lifted load, can obtain the equivalent load spectrum of this crane, its result is as follows, can see that their working cycle number of times variation tendency is consistent with the input sample.
Specified lifted load is the equivalent load spectrum of 75 tons general-purpose overhead crane
(5) fatigue surplus life estimation
The equivalent load spectrum of the general-purpose overhead crane that employing is obtained by neural network, calculate the Simulating of Fatigue Stress Spectra of this crane girder span centre dangerouse cross-section, utilize rain flow method from Simulating of Fatigue Stress Spectra, to extract cycle index and the global cycle number of times of stress amplitudes at different levels, stress amplitudes at different levels by 8 grades (same well-behaved progression), and then calculate the equivalent stress width of cloth σ of this Simulating of Fatigue Stress Spectra according to Miner fatigue damage stress amplitude equivalence formula
Re, with equivalent stress amplitude σ
ReThe computing formula of hardware fatigue surplus life is calculated in substitution.Can estimate the fatigue surplus life of this crane girder.
Be the Simulating of Fatigue Stress Spectra that the equivalent load spectrum of the general-purpose overhead crane of 28.5 meters of 75 tons, span changes into the girder dangerouse cross-section such as the specified lifted load that step (4) is obtained, and the equivalent stress width of cloth, and substitution calculates the computing formula of crane fatigue surplus life, can calculate the fatigue surplus life of this crane girder.As shown in Figure 3, be the Simulating of Fatigue Stress Spectra of the dangerouse cross-section that changed into by equivalent load spectrum and the correlation curve of measured stress spectrum, can see that they are very approaching.
Claims (3)
1. the overhead travelling crane equivalent load spectrum acquisition methods based on neural network is characterized in that comprising the following steps:
(1) sets up neural network model
Specified lifted load and the lifted load of getting overhead travelling crane are input quantity, the working cycle number of times of different lifted loads within a working hour of corresponding such crane one specified lifted load is output quantity, make up the nerve network system of the single output of dual input, described neural network structure is three layers, be input layer, hidden layer and output layer, the input layer number is 2, and the output layer nodes is 1, and the number of hidden nodes is between 13～17;
(2) obtain learning sample
Input take the specified lifted load of overhead travelling crane and lifted load as learning sample, the working cycle number of times of different lifted loads within a working hour of corresponding such crane one specified lifted load is the hope output of learning sample, obtain learning sample, and specified lifted load value is divided into 8 parts at least learning sample carried out classification;
(3) training LMBP neural network
On the basis of the learning sample that neural network model that step (1) obtains and step (2) obtain, training LMBP neural network, and obtain optimum model parameter;
(4) the crane equivalent load spectrum obtains
The specified lifted load lifted loads different from least 8 grades of needs being estimated the overhead travelling crane of fatigue surplus life are input to the neural network that step (3) trains, and can obtain the equivalent load spectrum of this overhead travelling crane.
2. the overhead travelling crane equivalent load spectrum acquisition methods based on neural network according to claim 1, it is characterized in that: the sample data in the described step (2) is that overhead travelling crane is in lifted load and the corresponding working cycle number of times thereof under the normal operating conditions, makes learning sample meet the actual applying working condition of crane.
3. the bridging crane main beam fatigue surplus life evaluation method based on neural network is characterized in that comprising the following steps:
(1) adopts equivalent load spectrum by the overhead travelling crane that obtains such as the described neural network of claim 1 (1) to (4) step, calculate the Simulating of Fatigue Stress Spectra of this crane girder span centre dangerouse cross-section;
(2) from Simulating of Fatigue Stress Spectra, extract again cycle index and the global cycle number of times of stress amplitudes at different levels, stress amplitudes at different levels with rain flow method;
(3) calculate the equivalent stress width of cloth σ of this Simulating of Fatigue Stress Spectra according to Miner fatigue damage stress amplitude equivalence formula
Re
(4) with equivalent stress amplitude σ
ReThe computing formula of hardware fatigue surplus life is calculated in substitution, can estimate the fatigue surplus life of crane girder.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200810180622 CN101408951B (en) | 2008-11-18 | 2008-11-18 | Method for obtaining equivalent load spectrum and estimating weariness residual longevity of bridge crane based on neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200810180622 CN101408951B (en) | 2008-11-18 | 2008-11-18 | Method for obtaining equivalent load spectrum and estimating weariness residual longevity of bridge crane based on neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101408951A CN101408951A (en) | 2009-04-15 |
| CN101408951B true CN101408951B (en) | 2013-04-03 |
Family
ID=40571957
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200810180622 Expired - Fee Related CN101408951B (en) | 2008-11-18 | 2008-11-18 | Method for obtaining equivalent load spectrum and estimating weariness residual longevity of bridge crane based on neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101408951B (en) |
Families Citing this family (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101830397A (en) * | 2010-04-02 | 2010-09-15 | 太原科技大学 | System and method for acquiring loading spectrum parameter data of bridge crane |
| CN102261988B (en) * | 2011-04-25 | 2013-06-26 | 吴峰崎 | Method for testing residual life of crane |
| CN102767133B (en) * | 2011-05-06 | 2014-11-19 | 东南大学 | Arranging method of sensor for detecting fatigue stress of steel box girder bridge |
| CN102324040B (en) * | 2011-09-09 | 2013-06-05 | 太原科技大学 | Methods for acquiring equivalent load spectrum and estimating residual fatigue life of bridge crane |
| CN102567632B (en) * | 2011-12-22 | 2014-12-10 | 上海交通大学 | Shore bridge structure wind vibration fatigue life forecasting method based on accumulated damage of probability |
| CN103389220B (en) * | 2013-08-01 | 2015-09-16 | 太原科技大学 | Large bridge crane load shedding equivalent method for testing and detecting |
| CN104809317B (en) * | 2015-05-19 | 2018-01-23 | 武汉理工大学 | Based on cluster analysis crane span structure class crane complete machine loading spectrum preparation method |
| CN106096262B (en) * | 2016-06-08 | 2018-11-23 | 南京航空航天大学 | Aero-engine load spectral correlative coefficient calculation method based on rain-flow counting circulation |
| CN107356445B (en) * | 2017-06-22 | 2019-05-07 | 长安大学 | Method for sorting, loading method and device are composed in working device of loader fatigue test |
| CN108760371A (en) * | 2018-07-04 | 2018-11-06 | 南京泽楚科技有限公司 | A kind of in-service hoisting machinery structure residual life computational methods |
| CN109408998B (en) * | 2018-11-08 | 2023-01-03 | 太原科技大学 | Fatigue life evaluation method by rapidly acquiring stress spectrum based on sample continuous increment |
| CN109766637B (en) * | 2019-01-12 | 2022-06-10 | 太原科技大学 | Bridge crane structure reliability optimization method based on Krigng agent model |
| CN109882998B (en) * | 2019-02-14 | 2021-08-27 | 中国联合网络通信集团有限公司 | Prediction method and system for air conditioner refrigerant |
| CN110032555A (en) * | 2019-04-16 | 2019-07-19 | 上海建科工程咨询有限公司 | A kind of neural network tower crane Risk Forecast Method and system |
| CN111131237B (en) * | 2019-12-23 | 2020-12-29 | 深圳供电局有限公司 | Microgrid attack identification method based on BP neural network and grid-connected interface device |
| CN111591893A (en) * | 2020-05-27 | 2020-08-28 | 太原科技大学 | Method for measuring hoisting load of automobile crane based on neural network |
| CN113138005B (en) * | 2021-05-13 | 2023-07-11 | 华东理工大学 | Vehicle random load pattern recognition system based on feature extraction and neural network |
-
2008
- 2008-11-18 CN CN 200810180622 patent/CN101408951B/en not_active Expired - Fee Related
Non-Patent Citations (4)
| Title |
|---|
| 徐格宁、左斌.起重机结构疲劳剩余寿命评估方法研究.《中国安全科学学报》.2007,第17卷(第3期),126-130. * |
| 李红梅.桥吊的状态监测与安全评估的系统建模及计算机实现.《桥吊的状态监测与安全评估的系统建模及计算机实现》.上海海运学院,2003, * |
| 熊莉，李丹柯.起重机结构寿命的模拟估算.《西南交通大学学报》.2002,第37卷(第3期),265-268. * |
| 肖涵.桥式起重机的金属结构疲劳.《桥式起重机的金属结构疲劳》.武汉科技大学,2004, * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101408951A (en) | 2009-04-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101408951B (en) | Method for obtaining equivalent load spectrum and estimating weariness residual longevity of bridge crane based on neural network | |
| CN105043631B (en) | Vibratory drilling method Cable force measuring method based on linear model | |
| CN102324040B (en) | Methods for acquiring equivalent load spectrum and estimating residual fatigue life of bridge crane | |
| CN105241660B (en) | High-speed rail large bridge performance test methods based on health monitoring data | |
| CN101806668B (en) | Cable structure health monitoring method based on cable tension monitoring | |
| CN106932135B (en) | Flexible inhaul cable force testing method for identifying vibration frequency based on weighted narrow-band peak searching method | |
| CN110095274B (en) | Steel trestle static load test evaluation method considering inelastic deformation | |
| Deng et al. | Fatigue reliability assessment for bridge welded details using long-term monitoring data | |
| CN103020429A (en) | Comprehensive decision-making and evaluating method for health condition of tied-arch bridge | |
| CN102261988B (en) | Method for testing residual life of crane | |
| CN108846197A (en) | A kind of crane's major girder non-destructive tests and degree of injury quantitative analysis method | |
| CN109060497A (en) | It is a kind of to consider the Analysis of Fatigue method lower than fatigue limit load | |
| CN105300692A (en) | Bearing fault diagnosis and prediction method based on extended Kalman filtering algorithm | |
| CN102486833A (en) | Method for predicting efficiency and detecting fault of device | |
| CN109978374B (en) | Risk assessment method for oil and gas pipeline system | |
| CN101667214B (en) | Method for arranging displacement measuring points | |
| CN107545110A (en) | A kind of dynamic stress accelerated life test section preparation method | |
| CN107122914A (en) | A kind of crane safety method of evaluating performance and system | |
| CN104933254A (en) | Vibration-method cable force measurement method of improved linear model on the basis of frequency ratio offset coefficient | |
| CN108319767A (en) | The method of suspension bridge sunpender stress assessment based on traveling load | |
| CN107908879A (en) | A kind of concrete beam bridge fatigue behaviour appraisal procedure | |
| CN102564875A (en) | Steel wire rope fatigue assessment system based on five point bending test | |
| CN108549847B (en) | Beam type structure crack damage identification method under non-reference data condition | |
| CN106446384B (en) | A kind of damnification recognition method of bridging crane main beam structure | |
| CN107421672A (en) | A kind of weighting Suo Li computational methods that peak is searched based on vibration frequency universe |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20130403 Termination date: 20141118 |
|
| EXPY | Termination of patent right or utility model |