CN111453619B - Self-adaptive bank bridge running mechanism intelligent monitoring and state evaluation system - Google Patents
Self-adaptive bank bridge running mechanism intelligent monitoring and state evaluation system Download PDFInfo
- Publication number
- CN111453619B CN111453619B CN201910047980.3A CN201910047980A CN111453619B CN 111453619 B CN111453619 B CN 111453619B CN 201910047980 A CN201910047980 A CN 201910047980A CN 111453619 B CN111453619 B CN 111453619B
- Authority
- CN
- China
- Prior art keywords
- state
- vibration
- module
- data
- shore bridge
- 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.)
- Active
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66C—CRANES; LOAD-ENGAGING ELEMENTS OR DEVICES FOR CRANES, CAPSTANS, WINCHES, OR TACKLES
- B66C15/00—Safety gear
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66C—CRANES; LOAD-ENGAGING ELEMENTS OR DEVICES FOR CRANES, CAPSTANS, WINCHES, OR TACKLES
- B66C13/00—Other constructional features or details
- B66C13/16—Applications of indicating, registering, or weighing devices
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66C—CRANES; LOAD-ENGAGING ELEMENTS OR DEVICES FOR CRANES, CAPSTANS, WINCHES, OR TACKLES
- B66C13/00—Other constructional features or details
- B66C13/18—Control systems or devices
- B66C13/22—Control systems or devices for electric drives
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66C—CRANES; LOAD-ENGAGING ELEMENTS OR DEVICES FOR CRANES, CAPSTANS, WINCHES, OR TACKLES
- B66C15/00—Safety gear
- B66C15/06—Arrangements or use of warning devices
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H17/00—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves, not provided for in the preceding groups
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention discloses a self-adaptive intelligent monitoring and state evaluation system for a shore bridge running mechanism, which comprises the following parts: the system comprises a data acquisition subsystem, a data processing subsystem and a cloud center; the data acquisition subsystem comprises a microphone, a vibration acceleration sensor and a data acquisition card; the data processing subsystem comprises a data compression module, a vibration multi-statistical feature extraction module, a noise multi-statistical feature extraction module, a deep learning module and a self-adaptive shore bridge state evaluation module. The system can acquire rich state information from vibration and noise signals of the shore bridge, and can adaptively perform intelligent evaluation on the state of the shore bridge according to the characteristics of the shore bridge, and then send a shore bridge state report to a mobile terminal of a user through the cloud center. The system has strong self-adaptive capacity, can meet the requirements of intellectualization and customization of an equipment monitoring system, and is beneficial to the development of port mechanical equipment towards the direction of automation, intellectualization and unmanned.
Description
Technical Field
The invention belongs to the technical field of state monitoring and fault diagnosis of mechanical equipment, and particularly relates to an intelligent monitoring and state evaluation system for a self-adaptive shore bridge operating mechanism.
Background
In recent years, domestic automatic docks are developed rapidly, and quayside container cranes (quay cranes) are used in large quantities, but the quay cranes have complex structures, various load action forms and complex and severe operating conditions, various faults can easily occur in the operating process, and once the faults occur, huge economic loss can be caused. At present, the shore bridge is mainly regularly checked by a manual inspection mode, potential safety hazards cannot be found in time at many times due to the complex structure of the shore bridge and inconsistent level of inspection personnel, and in addition, the shore bridge ensures the state health of the shore bridge by adopting a mode of regularly replacing parts for operating equipment, but the operating equipment such as a reduction gearbox and a lifting motor usually work under continuous large load, large impact and frequent start-stop working conditions, the parts are accelerated to fail under severe working conditions, and the mode of regular maintenance cannot adapt to the situation. Therefore, it is necessary to perform targeted intelligent monitoring and state evaluation on the operating mechanism of the quayside container crane.
After retrieval and consultation, no intelligent monitoring and state evaluation system for the shore bridge running mechanism exists, and the following problems exist in the existing mechanical equipment state monitoring and fault diagnosis system:
1. the traditional signal processing method is adopted as a tool for fault diagnosis of mechanical equipment, such as wavelet and empirical mode decomposition, and the design of characteristic indexes of the traditional methods needs the knowledge of human experts, the design process is time-consuming and labor-consuming, and the self-adaptive capacity is poor;
2. the vibration signal is mainly used as a data source for evaluating the state information of the equipment, the data mode is single, the multi-type sensor can obtain data in various modes, and the data in various modes is more favorable for obtaining different mode information of the state of the mechanical equipment.
The intelligent monitoring and state evaluation system for the self-adaptive shore bridge running mechanism mainly solves the following problems: 1. a deep learning method is adopted as a tool for intelligent state evaluation of the shore bridge, and the intelligent state monitoring method based on deep learning abandons manual experience intervention in signal feature extraction, so that self-adaptive intelligent evaluation of the state features of the shore bridge mechanism is realized; 2. the system adopts two modes of data, namely vibration signals and noise signals, and adopts different characteristic indexes suitable for signal characteristics for the two modes of data respectively, so that richer state information of the shore bridge running mechanism can be obtained, and the system has better robustness and adaptivity; 3. the use of the self-adaptive quayside container crane state evaluation module can enable the system to adapt to the characteristics of the quayside container crane, improves the self-adaptive capacity of the system and meets the 'customization' requirement of the equipment system.
Disclosure of Invention
Aiming at the problems, the invention provides an intelligent monitoring and state evaluation system of a self-adaptive shore bridge running mechanism, which is realized by the following technical scheme:
an intelligent monitoring and state evaluation system for a self-adaptive shore bridge running mechanism comprises a data acquisition subsystem, a data processing subsystem and a cloud center; the data acquisition subsystem comprises a microphone, a vibration acceleration sensor and a data acquisition card; the data processing subsystem is also called as an upper computer and comprises a data compression module, a noise multi-statistical feature extraction module, a vibration multi-statistical feature extraction module, a deep learning module and a self-adaptive quayside crane state evaluation module; the cloud center comprises a human-computer interaction module, a cloud database and a user information management module; the data acquisition subsystem is connected with the upper computer through a network cable; the upper computer is connected with the cloud center through the Internet remote communication technology.
A microphone in the data acquisition subsystem is arranged in an original shore bridge cab and used for replacing a human ear to acquire sound field information of the cab, a vibration acceleration sensor is arranged on a running mechanism in a shore bridge machine room, and a reasonable measuring point position is selected according to the actual structure and working conditions of a specific mechanism to be measured and used for acquiring vibration signals of a shore bridge lifting reduction box and a lifting motor; the microphone and the vibration acceleration sensor are connected with a data acquisition card, and the data acquisition card converts acquired analog signals into digital signals, conditions the digital signals and sends the conditioned digital signals to an upper computer.
And a data compression module of the data processing subsystem is used for compressing the noise and vibration data acquired by the data acquisition card, the grade of the quay crane running mechanism state by the deep learning evaluation module and the state evaluation result sent by the self-adaptive quay crane state evaluation module and uploading the compressed state evaluation result to a cloud center for storage.
The noise multi-statistical feature extraction module and the vibration multi-statistical feature extraction module of the data processing subsystem are respectively used for carrying out feature extraction on the noise and vibration data sent by the data acquisition card and sending the extracted noise and vibration features to the deep learning module;
the vibration multi-statistical characteristic extraction module is used for extracting the characteristics of the vibration signals sent by the data acquisition card and sending the extracted characteristics to the deep learning module.
Preferably, the features extracted by the vibration multi-statistical feature extraction module include: effective value, mean value, standard deviation, vibration intensity, kurtosis, skewness, peak value, margin, pulse index and frequency spectrum.
The noise multi-statistical feature extraction module is used for extracting features of noise signals sent by the data acquisition card and sending the extracted features to the deep learning module.
Preferably, the features extracted by the noise multi-statistical feature extraction module include: equivalent continuous A sound level (LAeq), mean value of A weighted sound pressure level (LAmean), value of 1/3 octave characteristic band sound pressure level, cumulative percentage sound level L5.
A deep learning evaluation module in the data processing subsystem is divided into a noise evaluation submodule, a vibration evaluation submodule and a model memory; the noise evaluation submodule and the vibration evaluation submodule give out state scores of noise and vibration of the shore bridge according to the noise and vibration characteristic values sent by the noise multi-statistical characteristic extraction module and the vibration multi-statistical characteristic extraction module, and send the two state scores of the noise and the vibration to the self-adaptive shore bridge state evaluation module; the model memory is used for storing noise and vibration deep learning evaluation models, and the two models are obtained by using the large data training of a shore bridge of a certain port machinery enterprise.
The deep learning vibration evaluation mainly comprises two processes of model training and model application, wherein the model training is to train a deep learning model by using selected fault-free shore bridge reduction box and lifting motor data (score 100) and fault shore bridge reduction box and lifting motor data (score 0), and the deep learning model learns the data to obtain the score estimation capability of a shore bridge reducer and a lifting motor; the model application is to evaluate newly-measured data of the quay crane by using a trained model to obtain state scores of a quay crane reduction gearbox and a hoisting motor. The lower the score of the vibration signal characteristic value is, the lower the probability that the signal is in a normal state is, and if the score is stable, the machine is in a stable state.
The deep learning noise evaluation is similar to the vibration evaluation process and is also divided into a model training process and a model application process, the model training process is to train a deep learning model by using sound field data (score 100) of a cab when a shore bridge running mechanism is not in fault and sound field data (score 0) of the cab when the shore bridge running mechanism is in fault, the score estimation capability of the sound field of the shore bridge cab is obtained through model training, and then the trained model is used for evaluating new data of the shore bridge to obtain the state score of the sound field of the shore bridge cab. The lower the shore bridge noise signal characteristic value score is, the lower the probability that the signal is in a normal state is, and the working state of the shore bridge running mechanism is evaluated by the model from the perspective of a sound field.
The self-adaptive quayside crane state evaluation module in the data processing subsystem is used for self-adapting different quayside cranes, receiving vibration and noise scores sent by the deep learning module, and giving the 'customized' hierarchical state evaluation of the quayside crane where the system is located by using a self-adaptive state evaluation algorithm, wherein the evaluation content comprises the following steps: grading the overall state of the quay crane, grading the state of a lifting reducer and grading the state of a lifting motor; the module is used for adapting to different shore bridges, and provides total state scoring, reducer state scoring and lifting motor state scoring suitable for the shore bridge for each shore bridge. According to the state score obtained by the self-adaptive shore bridge state evaluation module, the system classifies the shore bridge state into 3 levels, which are respectively: equipment healthy (score 70-100), sub-healthy (score 60-70) and unsafe (score below 60).
The self-adaptive shore bridge state evaluation algorithm uses a shallow neural network (only one hidden layer), and in the training process, the model in the deep learning evaluation module can be locked so as not to participate in updating, and only the shallow neural network in the self-adaptive shore bridge state evaluation module can be updated.
The core idea of the self-adaptive quayside container crane state evaluation algorithm is to finely adjust the scores on the basis of deep learning evaluation, so that finely adjusted data can better accord with the characteristics of the quayside container crane.
The most important difference between deep learning evaluation and adaptive quayside crane state evaluation is that: the deep learning evaluation is a model obtained by training all data samples of the shore bridge, and the self-adaptive shore bridge state evaluation is a model obtained by training the data of the shore bridge where the system is located. For example, if there are M landbridges, deep learning evaluation is to train using the M landbridge data; the self-adaptive evaluation modules A to M are arranged on the M quayside container bridges, and the self-adaptive evaluation module of the quayside container bridge A only uses the data of the quayside container bridge A for training; the deep learning training is the evaluation capability of all the quay crane fault knowledge, and the self-adaptive quay crane evaluation module is adapted to the characteristics of the installed quay crane according to the quay crane state knowledge learned by the deep learning model and provides a quay crane evaluation model suitable for the characteristics of the self-adaptive quay crane; the adaptive evaluation module has no fault learning capacity and only has the grading adaptation capacity.
The cloud center is used for storing original vibration signal data, noise signal data and scores of the self-adaptive shore bridge state evaluation module, the original vibration signal data and the noise signal data are sent by the upper computer, the stored data are used as historical data, data streams in the cloud center are displayed on a mobile terminal or a PC (personal computer) end of a user through a human-computer interaction module in a visualization mode, and states of complex shore bridges are displayed for the user more visually.
The cloud database of the cloud center is used for storing noise data and vibration data sent by the upper computer data compression module and grading the state of a quay crane operation mechanism by the deep learning evaluation module, grading the overall state of the quay crane by the self-adaptive quay crane state evaluation module, grading a lifting reduction box and grading a lifting motor; each shore bridge has an own ID identification number, and decompressed data can be respectively stored in independent data blocks according to the ID identification numbers.
The human-computer interaction module of the cloud center can be used for inquiring and displaying all data streams of the cloud center, and comprises vibration and noise data sent by a data acquisition card and state scoring of a quay crane running mechanism by a deep learning evaluation module, total scoring of a quay crane by a self-adaptive quay crane state evaluation module, and state scoring of a lifting reduction box and a lifting motor. The cloud center generates a shore bridge health state evaluation report according to the grade of the self-adaptive shore bridge state evaluation module on the state of each shore bridge, and sends the shore bridge health state evaluation report to a mobile terminal or a PC (personal computer) end of an administrator and a user through the human-computer interaction module; when the analysis result prompts that the equipment state reaches sub-health, automatically issuing a maintenance report; when the analysis result prompts that the equipment state is unsafe, alarm information is automatically issued.
Preferably, the user information management module of the cloud center divides the authority into an administrator authority and a user authority, and personnel with the administrator authority can edit and modify the content of the shore bridge state and can provide an instructive maintenance suggestion for a user; personnel with the user authority can only view the real-time state of the shore bridge, and the user authority is distributed by the administrator authority.
The system of the invention has the beneficial effects that:
(1) The intelligent state monitoring method based on deep learning abandons manual experience intervention in signal feature extraction, and realizes self-adaptive intelligent evaluation of the state features of the shore bridge mechanism;
(2) According to the method, the microphone is arranged in the original driving cab of the quayside container crane to acquire the sound information, and the noise signal and the vibration signal are simultaneously evaluated, so that the defect that the mechanical equipment information is not sufficiently acquired by only using a single vibration signal in the conventional mechanical equipment state evaluation is overcome, and the evaluation result is more accurate;
(3) The invention designs a self-adaptive shore bridge state evaluation module and a self-adaptive shore bridge state evaluation method, which can be used for customizing different shore bridges, have the specific analysis capability of the specific shore bridge and improve the self-adaptive capability of the system.
(4) According to the method and the system, the noise and the vibration signal of the shore bridge running mechanism are acquired, the intelligent monitoring and the state evaluation are carried out on the shore bridge running mechanism, the cloud center sends a shore bridge health state evaluation report to a user according to an evaluation result, the user can know the running state of the shore bridge in time, when the shore bridge running state is unsafe, the system can automatically issue alarm information, the monitoring efficiency is improved, and the user experience degree is high.
Drawings
Fig. 1 is a schematic structural diagram of an intelligent monitoring and state evaluation system of a self-adaptive shore bridge operating mechanism.
Fig. 2 is a schematic diagram of data flow of an upper computer of an intelligent monitoring and state evaluation system of an adaptive shore crane operating mechanism.
Fig. 3 is a flow chart of an implementation of the intelligent monitoring and state evaluation system of the self-adaptive shore bridge operating mechanism.
Detailed Description
The present invention will be described in further detail with reference to specific embodiments in order to make the technical field better understand the scheme of the present invention. It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention.
The present invention is described in further detail below with reference to the attached drawing figures.
An intelligent monitoring and state evaluation system for an adaptive shore bridge operating mechanism comprises a vibration acceleration sensor 112, a microphone 122, a signal acquisition card 200, an upper computer 300, a cloud center 400 and a human-computer interaction module 410, wherein the schematic structural diagram of the system is shown in fig. 1; the upper computer 300 comprises a noise multi-statistical feature extraction module 310, a vibration multi-statistical feature extraction module 320, a deep learning evaluation module 330, an adaptive quayside crane state evaluation module 340 and a data compression module 350.
The vibration acceleration sensor 112 is respectively installed at the positions of the lifting motor, the input and output shafts of the reduction gearbox and the base, and the specific installation positions are shown in table 1.
The vibration acceleration sensor 112 selects 608a11 of PCB company, which belongs to a high-impact vibration sensor.
The microphone 122 is installed in the cab, is located near the driver seat of the cab and 1 meter away from the ground, is arranged in front and at the back to partially offset the influence of the directivity of the microphone, improves the reliability of data and evaluation, and utilizes the microphone to acquire the dynamic sound pressure signal of the cab to simulate the auditory sense of a human.
The microphone 122 is a PCB378B02 free-field microphone, belonging to a condenser microphone.
Table 1 description of vibration acceleration sensor arrangement
The signal acquisition card 200 is connected with the microphone 122 through the vibration acceleration sensor 112. In the embodiment, 4 NI 9234 data acquisition cards and cDAQ9184 are selected to form a 16-channel data acquisition system, and the sampling frequency of the signal acquisition card on the vibration acceleration sensor is set to be 8192Hz, and the sampling frequency of the signal acquisition card on the microphone is set to be 25.6KHz.
The signal acquisition card 200 is connected with an upper computer (industrial personal computer) 300 through a network cable.
The signal acquisition card 200 sends the vibration signal data to the vibration multi-statistical feature extraction module 320 in the upper computer 300, and sends the noise signal data to the noise multi-statistical feature extraction module 310 in the upper computer 300.
The upper computer 300 comprises a noise multi-statistical feature extraction module 310, a vibration multi-statistical feature extraction module 320, a deep learning evaluation module 330, a self-adaptive quayside crane state evaluation module 340 and a data compression module 350, wherein the modules are installed on an industrial personal computer in a software mode. The operation logic sequence of various modules is certain, and the operation logic sequence can be judged as shown in fig. 2 through the flow schematic of data streams in the upper computer 300.
The upper computer 300 receives the sent data, and respectively sends the data to the noise multi-statistical-feature extraction module 310, the vibration multi-statistical-feature extraction module 320 and the data compression module 350, and the three modules process the data in parallel, so that the data processing speed can be increased.
The data compression module 350 receives the noise and vibration original data collected by the data acquisition card, scores of the quayside crane running mechanism state by the deep learning evaluation module and the state evaluation result sent by the self-adaptive quayside crane state evaluation module, and compresses the data by adopting a lossless data compression method.
The vibration multi-statistical feature extraction module 320 receives vibration signal data, performs a plurality of different vibration statistical feature extraction algorithm processes on the data, and specifically calculates characteristic values as an effective value (RMS) characteristic value, a mean characteristic value, a standard deviation characteristic value, a vibration intensity characteristic value, a kurtosis characteristic value, a skewness characteristic value and a spectrum characteristic, wherein the characteristic values can reflect characteristics of the signal such as energy size, average energy level, pulse vibration degree and the like.
The vibration multi-statistic feature extraction module 320 calculates the feature value once per second and sends the calculation result to the vibration evaluation sub-module 332 in the deep learning evaluation module.
The noise multi-statistical feature extraction module 310 receives noise signal data, and performs various algorithm processes of noise feature extraction on the data, specifically, the process processes are as follows:
(1) Fast Fourier transform, converting the abnormal noise signal in time domain into its corresponding power spectrum;
(2) Calculating the LAeq value to be an equivalent continuous A sound level, and reflecting the integral sound energy in a period of time;
wherein: l is a radical of an alcohol PA And T respectively represent the instantaneous A sound level (dB) at a certain time T and the specified measuring time(s)
(3) Calculating the LAmean value as the mean value of the A weighting sound pressure level, and better reflecting the subjective feelings of sensitivity and insensitivity to high-frequency noise and low-frequency noise of a human;
wherein:
-measuring the surface average a sound level, dB (a) (reference value 20 μ Pa) N-total number of points; lpi — sound level a measured at point i, dB (a) (reference value 20 μ Pa); KLi — background noise correction value at ith point, dB (a); k2-environmental correction value, dB (A); k3-ambient temperature and pressure correction value, dB (A).
(4) Calculating the sound pressure level value of the 1/3 octave characteristic frequency band as the sound energy in the frequency band, and the key point is to select the characteristic frequency band, such as: 1KHz, 2.5KHz, 6.3KHz, etc.;
(5) Calculating the value of L5, which is another statistical embodiment of common statistical parameters, ln values and sound pressure level peak values, wherein L5 refers to the sound level A with the occurrence time or times of more than 5% in the whole measurement time or times;
the noise multi-statistic feature extraction module 310 calculates the feature value once per minute, and sends the calculation result to the noise evaluation sub-module 311 in the deep learning evaluation module.
The deep learning evaluation module 330 is divided into a vibration evaluation sub-module 332 and a noise evaluation sub-module 331.
The vibration estimation sub-module 332 is divided into two processes, model training and model application.
Model training, namely training a deep learning evaluation module 330 capable of evaluating the states of all shore bridges; in this example, the vibration data of the new bank bridge in service, the abnormal maintenance bank bridge of maintaining is selected from among the high in the clouds center, calculates through vibration many statistical feature extraction module 320 and obtains the eigenvalue, adds the training label to the eigenvalue that obtains calculating: the state of the new service shore bridge is rated as 100, and the state of the abnormal maintenance shore bridge is rated as 0; training a deep learning evaluation module 330 by using the calculated characteristic value with the training label to obtain a model with the state evaluation capability on the quay crane hoisting equipment, and storing the trained model into a model memory 333; the model training is independent of the whole system operation, and the system operation only calls the trained model stored in the model memory 333 to calculate the state of the shore bridge in real time; model training is performed before a set of system is installed on a shore bridge, the trained models are solidified in a deep learning module 330 database, and any deep learning evaluation module 330 installed on the shore bridge is the same.
The model application is mainly to extract the feature values sent by the vibration multi-statistical feature extraction module 320 every second, and the trained vibration evaluation deep learning model in the model memory 333 to evaluate the feature values to obtain the score of the vibration of the second; the lower the signal characteristic value score is, the smaller the probability that the signal is in a normal state is, so that the state of the shore bridge speed reducer and the lifting motor can be effectively evaluated by the model.
The operation of the noise evaluation sub-module 331 and the vibration evaluation sub-module 332 is the same, except that the noise evaluation sub-module 331 trains the model using the noise data, and the feature value sent by the noise multi-statistic feature extraction module is evaluated once per minute during the model application process.
The deep learning evaluation model uses a 1D convolutional neural network, the network has 8 layers, wherein the first 5 layers of convolutional layers (Conv 1-Conv 5) and the last three layers are full connection layers (FC 1, FC2 and FC 3); the first five layers are characteristic extraction parts, and the last three layers are regression models.
The deep learning evaluation module 330 uses the deep convolutional neural network model to calculate the vibration score and the noise score once, and then sends the vibration score and the noise score to the adaptive quayside crane state evaluation module 340.
The self-adaptive shore bridge state evaluation module 340 is adapted to the characteristics of different shore bridges and different types of shore bridges, and carries out 'customized' shore bridge overall scoring, hoisting mechanism scoring, speed reducer state scoring and hoisting motor state scoring on the different shore bridges; after receiving the state scores obtained by the self-adaptive shore bridge state evaluation module, the cloud center grades the states of the shore bridges according to the scores, and when the noise and vibration analysis result prompts that the equipment state reaches sub-health (the score is 60-70), a maintenance report is issued to a worker; when the analysis result of the noise and the vibration prompts that the equipment state is unsafe (the score is lower than 60), the system automatically issues alarm information.
The adaptive quayside crane state evaluation module 340 is divided into two processes of model training and model application. In the model training, the vibration score and the noise score sent by the deep learning evaluation module are used as training data in the first month after the equipment is installed in the shore bridge and starts to operate, the intervention of artificial knowledge is used as a data label, and the self-adaptive shore bridge state evaluation model is trained; the model application is that a trained self-adaptive shore bridge state evaluation model is used for evaluating the vibration score and the noise score sent by the deep learning evaluation module 330, and a total score suitable for the shore bridge where the model is located, a lifting reducer score and a lifting motor score are given.
The intervention of the artificial knowledge refers to that noise data are graded and labeled according to human drivers, and vibration data are labeled according to vibration data analysis experts.
The cloud center 400 is used for storing a data compression packet sent by the upper computer data compression module, decompressing the data compression packet to obtain noise data and vibration data, the deep learning evaluation module scores the state of the quay crane running mechanism, and the self-adaptive quay crane state evaluation module scores the total state of the quay crane, scores the lifting reduction box and scores the lifting motor; each shore bridge has an own ID identification number, and decompressed data can be respectively stored in independent data blocks according to the ID identification numbers; the cloud center generates a shore bridge health status evaluation report according to the score of each shore bridge, sends the report to a mobile terminal or a PC (personal computer) end of a user through the human-computer interaction module 410, and sends a maintenance order to workers when finding that the shore bridge is in a sub-health state or an unsafe state.
The human-computer interaction module 410 displays all data streams in the cloud center 400, including vibration and noise data sent by a data acquisition card, various feature values obtained by calculation of the vibration multi-statistic feature extraction module and the noise multi-statistic feature extraction module, the deep learning evaluation module scores the state of a shore bridge running mechanism, the self-adaptive shore bridge state evaluation module scores the total score of a shore bridge, the states of a lifting reduction box and a lifting motor, and the data and the score can be sent to a PC (personal computer) terminal 411 and a mobile device 412 of a user.
The embodiment shows that the system is an intelligent monitoring and state evaluation system of the self-adaptive shore bridge running mechanism with the noise and vibration diagnosis technology; firstly, collecting vibration and noise signals capable of effectively acquiring shore bridge information, and extracting a characteristic value capable of effectively reflecting the state information of the shore bridge by using a multi-statistical characteristic extraction algorithm; then, a deep learning algorithm is introduced to evaluate the extracted multiple characteristic values; finally, a self-adaptive shore bridge evaluation algorithm is used for specifically analyzing the specific shore bridge; through this system, can fully acquire abundant state information from the vibration and the noise signal of bank bridge to these state information of effective, intelligent study, thereby the concrete analysis's of realization of "customization" ability, the erroneous judgement rate that can effectively reduce the system like this, the state information of accurate grasp bank bridge is favorable to harbour mechanical equipment to develop towards automatic, intelligent, unmanned direction, has fine using value that promotes.
Claims (7)
1. The utility model provides a self-adaptation bank bridge running gear intelligent monitoring and state evaluation system which characterized in that: the system comprises a data acquisition subsystem, a data processing subsystem and a cloud center;
the data acquisition subsystem comprises a microphone, a vibration acceleration sensor and a data acquisition card, wherein the microphone is arranged near a driver seat of a cab of the trolley, the microphone is respectively arranged in the directions of a sea side and a land side, the mounting positions of the vibration acceleration sensor are horizontal, radial and axial directions of a left lifting motor, a right lifting motor, a lifting mechanism base and a lifting mechanism reduction box high-speed shaft, and the mounting positions of a low-speed shaft of the lifting mechanism reduction box are radial; the data processing subsystem is also called as an upper computer and comprises a data compression module, a noise multi-statistical feature extraction module, a vibration multi-statistical feature extraction module, a deep learning module and a self-adaptive shore bridge state evaluation module, wherein the weighted features of the noise multi-statistical feature extraction module comprise equivalent continuous A sound level, the mean value of A weighted sound pressure level, 1/3 octave feature frequency band sound pressure level and accumulated percentile sound pressure level L5, the feature effective value extracted by the vibration multi-statistical feature extraction module comprises the mean value, the standard deviation, the vibration intensity, the kurtosis, the skewness, the peak value, the margin, the pulse index and the frequency spectrum, and the deep learning module adopts a convolutional neural network; the cloud center comprises a human-computer interaction module, a cloud database and a user information management module.
2. The system for intelligently monitoring and evaluating the state of the self-adaptive shore bridge running mechanism according to claim 1, wherein:
a microphone and a vibration acceleration sensor in the data acquisition subsystem are connected with a data acquisition card, and the data acquisition card converts acquired analog signals into digital signals, conditions the digital signals and sends the conditioned digital signals to an upper computer; the data compression module in the data processing subsystem is used for performing lossless compression on the noise and vibration data acquired by the data acquisition card, the scores of the states of the quay crane running mechanism by the deep learning evaluation module and the state evaluation results sent by the self-adaptive quay crane state evaluation module, and uploading the state evaluation results to the cloud center for storage; the noise multi-statistical feature extraction module and the vibration multi-statistical feature extraction module in the data processing subsystem are respectively used for carrying out feature extraction on the noise and vibration data sent by the data acquisition card and sending the extracted noise and vibration features to the deep learning module; a deep learning evaluation module in the data processing subsystem is divided into a noise evaluation submodule, a vibration evaluation submodule and a model memory; the noise evaluation submodule and the vibration evaluation submodule give state scores of noise and vibration of the quayside container crane according to the noise and vibration characteristic values sent by the noise multi-statistical characteristic extraction module and the vibration multi-statistical characteristic extraction module, and send the two state scores of the noise and the vibration to the self-adaptive quayside container crane state evaluation module; the model memory is used for storing the noise and vibration deep learning evaluation model; the deep learning vibration evaluation is mainly divided into two processes of model training and model application, and the deep learning model learns the data to obtain the grading estimation capability of the opposite-bank bridge speed reducer and the lifting motor; the model application is that a trained model is used for evaluating newly measured data of the shore bridge to obtain state scores of a reduction gearbox and a lifting motor of the shore bridge; the deep learning noise evaluation is similar to the vibration evaluation process and is also divided into a model training process and a model application process, the evaluation estimation capacity of the sound field of the shore bridge cab is obtained through the model training, and the trained model is used for evaluating the new data of the shore bridge cab to obtain the state score of the sound field of the shore bridge cab.
3. The system for intelligently monitoring and evaluating the state of the self-adaptive shore bridge running mechanism according to claim 1, wherein:
and a self-adaptive quayside crane state evaluation module in the data processing subsystem is used for self-adapting different quayside cranes, receiving vibration and noise scores sent by the deep learning module, giving a hierarchical state evaluation of 'customization' of the quayside crane where the system is located by using a self-adaptive state evaluation algorithm, and giving a total state score, a reducer state score and a lifting motor state score which are suitable for the quayside crane for each quayside crane.
4. The system for intelligently monitoring and evaluating the state of the self-adaptive shore bridge running mechanism according to claim 1, wherein:
the cloud center is used for storing original vibration signal data, noise signal data and scores of the self-adaptive shore bridge state evaluation module, the original vibration signal data and the noise signal data are sent by the upper computer, the stored data are used as historical data, data streams in the cloud center are displayed on a mobile terminal or a PC (personal computer) end of a user in a visual mode through the man-machine interaction module, and states of complex shore bridges are displayed for the user more visually.
5. The system for intelligently monitoring and evaluating the state of the self-adaptive shore bridge running mechanism according to claim 1, wherein:
the cloud database of the cloud center is used for storing noise data and vibration data sent by the upper computer data compression module and grading the state of a shore bridge running mechanism by the deep learning evaluation module, grading the overall state of the shore bridge by the self-adaptive shore bridge state evaluation module, grading a lifting reduction box and grading a lifting motor; each shore bridge has an own ID identification number, and decompressed data can be respectively stored in independent data blocks according to the ID identification numbers.
6. The system for intelligently monitoring and evaluating the state of the self-adaptive shore bridge running mechanism according to claim 1, wherein:
the human-computer interaction module of the cloud center can be used for inquiring and displaying all data streams of the cloud center, the cloud center can generate a bank bridge health state assessment report according to the score of each bank bridge, and the bank bridge health state assessment report is sent to a mobile terminal or a PC (personal computer) end of a user through the human-computer interaction module.
7. The system for intelligently monitoring and evaluating the state of the self-adaptive shore bridge running mechanism according to claim 1, wherein:
the user information management module of the cloud center divides the authority into administrator authority and user authority, and personnel with the administrator authority can edit and modify the content of the shore bridge state; personnel with the user authority can only view the real-time state of the shore bridge, and the user authority is distributed by the administrator authority.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910047980.3A CN111453619B (en) | 2019-01-18 | 2019-01-18 | Self-adaptive bank bridge running mechanism intelligent monitoring and state evaluation system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910047980.3A CN111453619B (en) | 2019-01-18 | 2019-01-18 | Self-adaptive bank bridge running mechanism intelligent monitoring and state evaluation system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111453619A CN111453619A (en) | 2020-07-28 |
| CN111453619B true CN111453619B (en) | 2023-01-17 |
Family
ID=71674350
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910047980.3A Active CN111453619B (en) | 2019-01-18 | 2019-01-18 | Self-adaptive bank bridge running mechanism intelligent monitoring and state evaluation system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111453619B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114153889A (en) * | 2020-09-08 | 2022-03-08 | 大连理工大学 | Power station monitoring system and method, readable medium and electronic device |
| CN112061988B (en) * | 2020-09-22 | 2023-01-17 | 上海振华重工(集团)股份有限公司 | Crane distributed network monitoring and single machine monitoring management system and method |
| CN117519054A (en) * | 2023-12-11 | 2024-02-06 | 广州智业节能科技有限公司 | High-efficient cold station control system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH10218571A (en) * | 1997-02-07 | 1998-08-18 | Maeda Seisakusho:Kk | Overload preventing device for crane |
| JPH10258987A (en) * | 1997-03-17 | 1998-09-29 | Mitsubishi Heavy Ind Ltd | Anti-swinging device for slung load |
| CN105819341A (en) * | 2016-06-02 | 2016-08-03 | 北京国泰星云科技有限公司 | System for profile identification of ship under container terminal quay crane and anti-collision of lifting sling and method for profile identification of ship under container terminal quay crane |
| CN107324214A (en) * | 2017-06-29 | 2017-11-07 | 天津大学 | Ocean platform crane intelligent state monitoring method |
| CN108681747A (en) * | 2018-05-11 | 2018-10-19 | 武汉理工大学 | Rotary machinery fault diagnosis based on deep learning and condition monitoring system and method |
-
2019
- 2019-01-18 CN CN201910047980.3A patent/CN111453619B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH10218571A (en) * | 1997-02-07 | 1998-08-18 | Maeda Seisakusho:Kk | Overload preventing device for crane |
| JPH10258987A (en) * | 1997-03-17 | 1998-09-29 | Mitsubishi Heavy Ind Ltd | Anti-swinging device for slung load |
| CN105819341A (en) * | 2016-06-02 | 2016-08-03 | 北京国泰星云科技有限公司 | System for profile identification of ship under container terminal quay crane and anti-collision of lifting sling and method for profile identification of ship under container terminal quay crane |
| CN107324214A (en) * | 2017-06-29 | 2017-11-07 | 天津大学 | Ocean platform crane intelligent state monitoring method |
| CN108681747A (en) * | 2018-05-11 | 2018-10-19 | 武汉理工大学 | Rotary machinery fault diagnosis based on deep learning and condition monitoring system and method |
Non-Patent Citations (1)
| Title |
|---|
| 集装箱岸边起重机主机构在线监测与诊断;刘敏毅等;《港口科技·信息化技术》;20141031;内文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111453619A (en) | 2020-07-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111453619B (en) | Self-adaptive bank bridge running mechanism intelligent monitoring and state evaluation system | |
| CN111507489B (en) | Cloud-edge-coordinated amusement equipment fault prediction and health management system and method | |
| CN114063507B (en) | Remote equipment control system based on intelligent agriculture and control method thereof | |
| CN110469462B (en) | Wind turbine generator system intelligent state monitoring system based on multiple templates | |
| CN104503399B (en) | The regimental Wind turbines status monitoring of one kind collection and fault diagnosis platform | |
| CN104200396B (en) | A kind of wind turbine component fault early warning method | |
| CN105300692B (en) | A kind of bearing failure diagnosis and Forecasting Methodology based on expanded Kalman filtration algorithm | |
| CN102620807A (en) | System and method for monitoring state of wind generator | |
| CN202676203U (en) | Large-size pumping station state monitoring and diagnosing and energy efficiency management device | |
| CN101718634A (en) | Equipment state comprehensive dynamic alarming method based on multivariate probability model | |
| CN112061988B (en) | Crane distributed network monitoring and single machine monitoring management system and method | |
| CN201110799Y (en) | Strange sound detector of car engine | |
| CN204113701U (en) | A kind of mine fan on-line monitoring and fault diagnosis system | |
| CN106022653A (en) | Ship risk early warning method and device | |
| CN115118333B (en) | Antenna health management system and method for satellite ground station | |
| CN104564542B (en) | A kind of fault diagnosis system and its diagnostic method based on big data technology | |
| CN113298134B (en) | System and method for remotely and non-contact health monitoring of fan blade based on BPNN | |
| CN111814848A (en) | Self-adaptive early warning strategy design method for temperature fault of wind turbine generator | |
| CN111594395A (en) | Wind turbine generator tower resonance identification method and device and monitoring alarm system | |
| CN115014617A (en) | Cable force synchronous monitoring method for cable-stayed bridge inhaul cable based on ground-based radar | |
| CN208516744U (en) | Escalator lubricating status monitors system and the voice data collection device for it | |
| CN113516159B (en) | Rail vehicle running part pinion shaft crack fault diagnosis method and system | |
| CN103999316A (en) | Method for generating a signal indicating an oscillation in an electrical energy supply network | |
| CN109886538B (en) | Railway signal equipment quality evaluation method and device based on dynamic monitoring data | |
| CN112801033A (en) | AlexNet network-based construction disturbance and leakage identification method along long oil and gas pipeline |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |