CN119105554A - An AGV fault diagnosis and prediction system and method based on data analysis - Google Patents
An AGV fault diagnosis and prediction system and method based on data analysis Download PDFInfo
- Publication number
- CN119105554A CN119105554A CN202411217801.3A CN202411217801A CN119105554A CN 119105554 A CN119105554 A CN 119105554A CN 202411217801 A CN202411217801 A CN 202411217801A CN 119105554 A CN119105554 A CN 119105554A
- Authority
- CN
- China
- Prior art keywords
- fault
- guided vehicle
- automatic guided
- model
- data
- 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.)
- Pending
Links
- 238000003745 diagnosis Methods 0.000 title claims abstract description 126
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000007405 data analysis Methods 0.000 title claims abstract description 19
- 238000012545 processing Methods 0.000 claims abstract description 16
- 238000012544 monitoring process Methods 0.000 claims description 73
- 238000004422 calculation algorithm Methods 0.000 claims description 69
- 238000001514 detection method Methods 0.000 claims description 53
- 238000009826 distribution Methods 0.000 claims description 23
- 238000005457 optimization Methods 0.000 claims description 18
- 238000012549 training Methods 0.000 claims description 17
- 230000002159 abnormal effect Effects 0.000 claims description 16
- 230000002441 reversible effect Effects 0.000 claims description 13
- 230000000739 chaotic effect Effects 0.000 claims description 10
- 238000013523 data management Methods 0.000 claims description 10
- 238000007781 pre-processing Methods 0.000 claims description 9
- 230000035772 mutation Effects 0.000 claims description 8
- 238000004458 analytical method Methods 0.000 claims description 7
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 5
- 230000005856 abnormality Effects 0.000 claims description 4
- 238000007619 statistical method Methods 0.000 claims description 4
- 238000005516 engineering process Methods 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 238000012423 maintenance Methods 0.000 abstract description 26
- 238000004519 manufacturing process Methods 0.000 abstract description 13
- 230000002829 reductive effect Effects 0.000 abstract description 10
- 238000007726 management method Methods 0.000 abstract description 5
- 230000002265 prevention Effects 0.000 abstract description 2
- 238000004891 communication Methods 0.000 description 11
- 241000555745 Sciuridae Species 0.000 description 10
- 230000006378 damage Effects 0.000 description 6
- 241000251557 Ascidiacea Species 0.000 description 5
- 230000001133 acceleration Effects 0.000 description 5
- 230000007613 environmental effect Effects 0.000 description 5
- 230000003449 preventive effect Effects 0.000 description 5
- 238000010845 search algorithm Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 238000005265 energy consumption Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000013507 mapping Methods 0.000 description 3
- 230000007246 mechanism Effects 0.000 description 3
- 230000002035 prolonged effect Effects 0.000 description 3
- 230000004044 response Effects 0.000 description 3
- 241000723418 Carya Species 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 230000015556 catabolic process Effects 0.000 description 2
- 230000003247 decreasing effect Effects 0.000 description 2
- 238000006731 degradation reaction Methods 0.000 description 2
- 230000001934 delay Effects 0.000 description 2
- 230000002431 foraging effect Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000013439 planning Methods 0.000 description 2
- 238000012805 post-processing Methods 0.000 description 2
- 230000008439 repair process Effects 0.000 description 2
- 230000007704 transition Effects 0.000 description 2
- 241000196324 Embryophyta Species 0.000 description 1
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 241000251555 Tunicata Species 0.000 description 1
- 238000005299 abrasion Methods 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000004634 feeding behavior Effects 0.000 description 1
- 230000019637 foraging behavior Effects 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 230000000670 limiting effect Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000007257 malfunction Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000007659 motor function Effects 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 244000062645 predators Species 0.000 description 1
- 239000002096 quantum dot Substances 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000011895 specific detection Methods 0.000 description 1
- 238000005728 strengthening Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/80—Arrangements for reacting to or preventing system or operator failure
- G05D1/86—Monitoring the performance of the system, e.g. alarm or diagnosis modules
Landscapes
- Engineering & Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention discloses an AGV fault diagnosis prediction system and method based on data analysis, and relates to the technical field of automatic guided vehicle management. According to the invention, possible operation problems can be responded quickly by timely identifying fault data, the downtime is reduced, and possible faults in the future are predicted, so that enterprises can switch from reactive fault processing to active fault prevention, thereby greatly improving the continuity and stability of production, and the operation parameters can be adjusted and optimized by analyzing key factors influencing the operation of the automatic guided vehicle so as to cope with the predicted potential faults, thereby reducing the overall maintenance cost, and further effectively prolonging the service lives of the automatic guided vehicle and related equipment.
Description
Technical Field
The invention relates to the technical field of automatic guided vehicle management, in particular to an AGV fault diagnosis prediction system and method based on data analysis.
Background
Automatic Guided Vehicles (AGVs) play a critical role in intelligent manufacturing enterprises, and are not only key tools for realizing production digitization, automation, networking and intellectualization, but also important marks for measuring the industrial automation level. The AGV is mainly responsible for carrying materials in factories, and whether the AGV is efficient or not directly influences the smoothness of a production flow and the operation efficiency of enterprises. Therefore, the maintenance of the efficient operation of the AGV system is important for ensuring production safety and improving production efficiency. In daily production, if an AGV system fails and cannot be timely detected and processed, not only can the production flow be interrupted and the pressure of material accumulation and work to be processed be increased, but also safety accidents can be caused due to improper operation. In this case, fault detection and timely maintenance of the AGV system becomes particularly important. System failures may be caused by a variety of factors including hardware failures, software program errors, or misuse of operations.
In the prior art, key characteristic data influencing the operation of the automatic guided vehicle is inconvenient to effectively screen and extract, so that the accuracy of fault detection is reduced, fault data cannot be identified in time, the downtime is increased, a fault prediction model is inconvenient to build to predict faults which possibly occur in the future, so that the fault occurrence rate is improved, meanwhile, key factors influencing the operation of the automatic guided vehicle are inconvenient to analyze, and operation parameters are inconvenient to adjust and optimize, so that the predicted potential faults are dealt with, the operation efficiency of the automatic guided vehicle is further reduced, the fault risk caused by improper parameter setting is improved, and the service lives of the automatic guided vehicle and related equipment are reduced.
For the problems in the related art, no effective solution has been proposed at present.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides an AGV fault diagnosis prediction system and method based on data analysis, which solve the problems that the prior art is inconvenient to effectively screen and extract key characteristic data influencing the operation of an automatic guided vehicle, so that the accuracy of fault detection is reduced, fault data cannot be recognized in time, the downtime is increased, a fault prediction model is inconvenient to be built to predict faults possibly occurring in the future, the fault occurrence rate is improved, meanwhile, the key factors influencing the operation of the automatic guided vehicle are inconvenient to analyze, and the operation parameters are inconvenient to adjust and optimize so as to cope with the predicted potential faults, thereby reducing the operation efficiency of the automatic guided vehicle, improving the fault risk caused by improper parameter setting, and reducing the service life of the automatic guided vehicle and related equipment.
In order to achieve the above purpose, the invention is realized by the following technical scheme:
According to one aspect of the invention, an AGV fault diagnosis and prediction system based on data analysis is provided, and comprises a data management module, a fault diagnosis module, a fault prediction module and an adjustment control module;
The data management module is used for acquiring the real-time running state data of the automatic guided vehicle, and preprocessing the real-time running state data of the automatic guided vehicle to obtain the running state characteristic data of the automatic guided vehicle;
the fault diagnosis module is used for analyzing the obtained running state characteristic data of the automatic guided vehicle by using a fault diagnosis algorithm and identifying the fault data of the running state of the current automatic guided vehicle;
the fault prediction module is used for establishing a fault prediction model based on the recognized fault data of the running state of the automatic guided vehicle, and predicting the running fault of the automatic guided vehicle at the future moment by using the fault prediction model to obtain key factors influencing the running fault of the automatic guided vehicle;
the adjusting control module is used for adjusting the operation parameters of the automatic guided vehicle based on the obtained key factors influencing the operation faults of the automatic guided vehicle;
Analyzing the obtained running state characteristic data of the automatic guided vehicle by using a fault diagnosis algorithm, wherein the identification of the fault data of the running state of the current automatic guided vehicle comprises the following steps:
Collecting historical running state data of the automatic guided vehicle;
Calculating the occurrence times of faults of each detection point in a preset time period, dividing the occurrence times of faults in the preset time period by the total running time of each detection point, and obtaining the fault frequency distribution of each detection point;
Analyzing the obtained fault frequency distribution of each detection point, drawing a fault frequency distribution curve of each detection point by a statistical analysis method, and identifying an area with higher fault frequency as a fault point easy to occur of the automatic guided vehicle;
constructing a fault monitoring model according to the fault frequency distribution and the fault occurrence rule of each detection point, and deploying the fault monitoring model on the automatic guide vehicle;
when the fault of the region with the fault point is detected, the abnormal value detection is carried out on the running state characteristic data of the automatic guided vehicle by using a fault diagnosis algorithm, and the potential fault data of the current state of the automatic guided vehicle is identified.
Further, the data management module comprises a data acquisition module and a data processing module;
the data acquisition module is used for acquiring real-time running state data of the automatic guided vehicle through a sensing technology;
the data processing module is used for preprocessing the acquired real-time running state data of the automatic guided vehicle to obtain running state characteristic data of the automatic guided vehicle.
Further, constructing a fault monitoring model according to the fault frequency distribution and the fault occurrence rule of each detection point, and deploying the fault monitoring model on the automatic guided vehicle comprises:
collecting historical fault data of the automatic guide vehicle, and taking the historical fault data as a training sample of a fault monitoring model;
using the fault frequency distribution and the fault occurrence rule of each detection point as input parameters of a fault monitoring model;
Performing outlier analysis on fault data in the training samples by using an estimation algorithm, dividing the fault data into a normal operation area and an abnormal operation area, and outputting an optimal grouping center matrix;
carrying out normalization processing on the training samples, calculating the mode association degree of the training samples after the normalization processing, using the optimal grouping center matrix and the performance threshold as the input of the fault monitoring model, and constructing the fault monitoring models of different modes by combining the mode association degree;
and deploying a trained fault monitoring model in a fault-prone area of the automatic guided vehicle, and comparing the performance deviation of each monitored detection point with a preset performance threshold value to judge whether the automatic guided vehicle is in a fault state.
Further, deploying a trained fault monitoring model in a fault prone region of the automatic guided vehicle, and comparing the performance deviation of each monitored detection point with a preset performance threshold value to determine whether the automatic guided vehicle is in a fault state, including:
deploying a trained fault monitoring model in a fault-prone area of the automatic guided vehicle, reading operation parameters of each detection point of the automatic guided vehicle by using the fault monitoring model, and calculating performance indexes;
if the calculated performance index is not greater than the preset performance threshold, judging that the automatic guided vehicle is in a normal running state, and continuously reading the running parameters of each detection point of the automatic guided vehicle by using a fault monitoring model and recalculating the performance index;
If the performance index is greater than a preset threshold, judging that the automatic guided vehicle is in an abnormal running state, and starting a standby fault monitoring model for re-monitoring;
if the monitoring result of the standby fault monitoring model is normal, judging that the process mode is changed, and continuously reading the operation parameters of each detection point of the automatic guided vehicle by using the fault monitoring model and recalculating the performance index;
if the monitoring result of the standby fault monitoring model also indicates abnormality, judging that the fault-prone area is faulty, and confirming that the automatic guided vehicle is in a fault state.
Further, when the fault is detected in the fault-prone point area, detecting abnormal values of the running state characteristic data of the automatic guided vehicle by using a fault diagnosis algorithm, and identifying potential fault data of the current automatic guided vehicle state comprises:
setting parameters of a fault diagnosis algorithm;
Generating an initial fault diagnosis model population by using a chaotic sequence, and setting an initial position and parameters for each fault diagnosis model;
evaluating and sorting all fault diagnosis models according to the fitness, classifying all fault diagnosis models into different performance levels according to the performance, and storing the optimal fault diagnosis models and parameters;
performing Gaussian mutation on the fault diagnosis model with the optimal representation, comparing the fitness values before and after mutation, and storing the fault diagnosis model with the highest fitness value;
Updating the probability threshold value of the fault diagnosis model with the highest fitness value, calculating the new positions of the fault diagnosis models on different performance levels according to the new parameters, and finally updating the positions and parameters of all the fault diagnosis models;
Updating environment variables, and if the environment variables reach preset conditions, improving a fault diagnosis model with poor repositioning performance;
and repeatedly evaluating and sequencing all the fault diagnosis models according to the fitness, carrying out random reverse learning on the global optimal fault diagnosis model to generate a new optimal fault diagnosis model, and outputting the new optimal fault diagnosis model as a final fault diagnosis model and outputting fault data of the current automatic guided vehicle state if the fitness value of the new optimal fault diagnosis model is larger than that of the current optimal fault diagnosis model.
Further, the formula for generating a new optimal fault diagnosis model by carrying out random reverse learning on the global optimal fault diagnosis model is as follows:
in the formula, Representing the position of a new optimal fault diagnosis model generated by random reverse learning;
a location expressed as a current optimal fault diagnosis model;
lb represents the lower bound of the search space for the fault diagnosis model parameters;
ub is denoted as the upper bound of the search space for the fault diagnosis model parameters;
rand () is represented as a generated random number.
Further, the fault prediction module comprises a model building module and a model prediction module;
the model building module is used for building a fault prediction model based on the recognized fault data of the running state of the automatic guided vehicle;
And the model prediction module is used for predicting the operation faults of the automatic guided vehicle at the future moment by utilizing the fault prediction model to obtain key factors influencing the operation faults of the automatic guided vehicle.
Further, predicting the operation fault of the automatic guided vehicle at the future moment by using the fault prediction model, and obtaining key factors affecting the operation fault of the automatic guided vehicle includes:
optimizing parameters of a fault prediction model by using a model optimization algorithm based on the performance index, and finding out the optimal parameter configuration of the fault prediction model;
and applying the parameter configuration to the constructed fault prediction model to predict so as to obtain key factors influencing the operation fault of the automatic guided vehicle.
Further, optimizing parameters of the fault prediction model based on the performance index by using a model optimization algorithm, and finding the optimal parameter configuration of the fault prediction model comprises:
initializing a model optimization algorithm;
determining the initial position of the fault prediction model parameters by using a spherical coordinate method, sorting according to the fitness value of the initial position, and selecting a plurality of parameter configurations as initial solutions;
the obtained parameter values of the fault prediction model and the adjustment factors are brought into an update formula to obtain new parameter positions of the fault prediction model;
And checking whether a stopping condition is met, if so, outputting the optimal solution as the optimal fault prediction model parameter configuration, and if not, continuing to execute the parameter updating cycle.
According to another aspect of the present invention, there is also provided an AGV fault diagnosis prediction method based on data analysis, including the steps of:
s1, acquiring real-time running state data of an automatic guiding vehicle, and preprocessing the real-time running state data of the automatic guiding vehicle to obtain running state characteristic data of the automatic guiding vehicle;
S2, analyzing the obtained running state characteristic data of the automatic guided vehicle by using a fault diagnosis algorithm, and identifying fault data of the running state of the current automatic guided vehicle;
S3, based on the recognized fault data of the running state of the automatic guided vehicle, a fault prediction model is established, and the running fault of the automatic guided vehicle at the future moment is predicted by using the fault prediction model, so that key factors influencing the running fault of the automatic guided vehicle are obtained;
s4, adjusting the operation parameters of the automatic guided vehicle based on the obtained key factors influencing the operation faults of the automatic guided vehicle.
The beneficial effects of the invention are as follows:
1. According to the invention, the running state data of the automatic guided vehicle is obtained in real time and preprocessed, so that the key characteristic data affecting the running of the automatic guided vehicle can be effectively screened and extracted, and the accuracy of fault detection is improved. The fault data can be identified in time to quickly respond to possible operation problems, the downtime is reduced, the data can be further analyzed based on a fault prediction model established by the identified fault data, the possible faults in the future can be predicted, the enterprise can be enabled to switch from reactive fault processing to active fault prevention, the continuity and stability of production are greatly improved, and the operation parameters can be adjusted and optimized to cope with the predicted potential faults by analyzing key factors influencing the operation of the automatic guided vehicle. Such parameter adjustment helps to improve the operating efficiency of the automated guided vehicle and reduce the risk of failure due to improper parameter settings, and systematic failure diagnosis prediction can reduce unnecessary maintenance and excessive repair, thereby reducing overall maintenance costs. Meanwhile, through preventive maintenance, the service lives of the automatic guide vehicle and related equipment can be effectively prolonged.
2. The invention can rapidly identify and respond to faults through real-time monitoring and data-driven fault diagnosis, reduce the downtime and maintenance cost of the system, realize the transition from post-processing to preventive maintenance through deep analysis and prediction of fault data, discover problems in advance and intervene, avoid larger equipment damage and production interruption, enable enterprises to more effectively allocate maintenance resources, optimize maintenance plans and budgets, improve the overall operation efficiency, continuously monitor and timely maintain, avoid serious loss caused by faults, prolong the service life of equipment, improve the return rate of investment of the assets, further reduce safety accidents caused by equipment faults, protect operators and related facilities, and maintain the stability of production environment.
3. According to the invention, the fault prediction model is constructed based on the identified fault data, so that the prediction is based on actual observation data, and the accuracy of the prediction is improved. The optimized model can accurately identify possible faults in the future, so that preventive measures are more targeted and effective, the fault prediction model can analyze and predict the fault condition in the future in real time, and the trend and influencing factors of the fault development can be found in time. The method enables enterprises to quickly respond, reduces the possibility and range of fault expansion, reduces maintenance time and cost, dynamically adjusts parameters of a fault prediction model according to performance indexes by using a model optimization algorithm, and ensures that the model always operates in an optimal state. The dynamic adjustment helps the model to adapt to the change of the running environment, improves the generalization capability and the application flexibility of the model, and can intervene before serious damage is caused by faults through timely fault prediction and corresponding maintenance, so that the service life of the equipment is prolonged, and the overall use efficiency of the equipment is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a functional block diagram of an AGV fault diagnosis prediction system based on data analysis according to an embodiment of the present invention;
FIG. 2 is a flow chart of an AGV fault diagnosis prediction method based on data analysis according to an embodiment of the invention.
In the figure:
1. The system comprises a data management module, a fault diagnosis module, a fault prediction module and an adjustment control module.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments.
In the description of the present invention, unless otherwise indicated, the meaning of "a plurality" is two or more. Furthermore, the terms "first," "second," "third," and the like are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
According to the embodiment of the invention, an AGV fault diagnosis prediction system and method based on data analysis are provided.
Referring to the drawings and the specific embodiments, as shown in fig. 1, the data analysis-based AGV fault diagnosis and prediction system according to the embodiment of the present invention includes a data management module 1, a fault diagnosis module 2, a fault prediction module 3, and an adjustment control module 4;
The data management module 1 is used for acquiring real-time running state data of the automatic guided vehicle, and preprocessing the real-time running state data of the automatic guided vehicle to obtain running state characteristic data of the automatic guided vehicle;
specifically, the real-time running state data of the automatic guided vehicle includes:
1) Position information:
Coordinates-specific location coordinates of the automated guided vehicle in the factory or warehouse.
Navigation path, automatic guiding vehicle current running path and preset destination.
2) Velocity and acceleration data:
speed is the current running speed of the automatic guided vehicle.
Acceleration, namely, automatically guiding real-time data of acceleration or deceleration of the vehicle.
3) Battery and energy status:
and the battery power is the residual power of the automatic guiding vehicle battery.
The energy consumption rate is the energy consumption condition of the automatic guiding vehicle in the running process.
4) Load state:
Load weight-automatically guiding the current weight of the vehicle.
Load state change, namely increasing or decreasing the load of the automatic guided vehicle.
5) Sensor data:
Environmental sensors-data collected from the environment surrounding the automated guided vehicle (e.g., temperature, humidity, obstacle distance, etc.).
Mechanical sensors-operational data regarding mechanical components of the automated guided vehicle (e.g., motor status, steering system, etc.).
6) Operating state:
Fault indication-real-time indication of any system fault or warning.
And the operation mode is whether the automatic guiding vehicle is in an automatic mode, a manual control state or a stop state.
7) Communication state:
The network connection condition is the network communication state of the automatic guiding vehicle, including the connection quality with the central control system.
8) Software system state:
System logging-logging of the operation of the automated guided vehicle software, may include detailed information of any anomalies or errors.
The fault diagnosis module 2 is used for analyzing the obtained running state characteristic data of the automatic guided vehicle by using a fault diagnosis algorithm and identifying the fault data of the running state of the current automatic guided vehicle;
Specifically, the fault data of the operation state of the automatic guided vehicle includes:
1) Mechanical failure:
Drive system problems such as motor failure, gearbox problems or drive chain damage.
Failure of the navigation system, such as unresponsiveness of the steering mechanism or insufficient positioning accuracy.
Load bearing mechanism problems such as tray lifting mechanism failure or stability problems.
2) Electronic failure:
sensor failure-sensor output anomaly data such as inaccurate distance sensor, weight sensor, or environmental sensor data.
Control system failure-control unit or microprocessor failure, resulting in data processing or command execution errors.
Battery problems, such as battery undercharge, abnormal discharge speed, or battery management system malfunction.
3) Software and communication failures:
software errors-operational problems caused by software logic errors or programming defects.
Communication disruption-communication delay or disruption with the control system may be caused by network problems or interface failures.
4) Operational efficiency problems:
The running speed is reduced, and the running speed of the automatic guided vehicle is abnormally reduced or can not reach the set speed.
Path deviation-the automatic guided vehicle does not travel along the predetermined route, and navigation deviation may occur.
5) Safety and monitoring warnings:
security system alarms, such as emergency stops, are triggered or security boundaries are violated.
And (3) maintenance reminding, namely providing maintenance warning according to a preset maintenance period or detected component abrasion condition.
The fault prediction module 3 is used for establishing a fault prediction model based on the recognized fault data of the running state of the automatic guided vehicle, and predicting the running fault of the automatic guided vehicle at the future moment by using the fault prediction model to obtain key factors influencing the running fault of the automatic guided vehicle;
Specifically, key factors affecting the operation failure of the automatic guided vehicle include:
hardware wear and degradation, mechanical component wear in long-term operation, such as tires, gears, and bearings, are common causes of failure.
Battery performance degradation, a decrease in battery life and charging efficiency, may result in an unstable energy supply, affecting the continuous operation capability of the automatic guided vehicle.
Sensor accuracy and reliability the accuracy and reliability of the sensor directly affects the effectiveness of the navigation and positioning system of the automated guided vehicle. Sensor data distortion or delay can directly lead to navigation errors or accidents.
Software and algorithm errors software logic errors or algorithm flaws may lead to erroneous decisions and command execution, such as path planning errors or response delays.
Environmental factors such as ground conditions, temperature fluctuations, humidity, etc. may also affect the operational status of the automated guided vehicle.
Network and communication system, network connection quality and stability of communication protocol are key to ensure smooth communication between the automatic guided vehicle and the control center and other automatic guided vehicles. Network delays or interrupts may cause command transmission to fail.
Operator operation errors or program setting errors can also lead to failure of the automatic guided vehicle.
Maintenance and overhaul strategies improper or insufficient maintenance may lead to early occurrence of faults. Timely maintenance and repair is an important measure for preventing faults.
The adjusting control module 4 is used for adjusting the operation parameters of the automatic guided vehicle based on the obtained key factors influencing the operation faults of the automatic guided vehicle;
specifically, adjusting the operating parameters of the automated guided vehicle includes:
And adjusting the speed and the acceleration, namely properly adjusting the running speed and the acceleration of the automatic guided vehicle according to the working environment and the load condition. For example, speed is reduced in crowded or complex environments to reduce the risk of collisions and failures.
Optimizing the path selection and navigation parameters, namely optimizing the driving route of the automatic guided vehicle by utilizing an advanced path planning algorithm, and avoiding the area prone to faults, such as uneven ground or signal interference areas. And simultaneously, the sensitivity and the response threshold of the navigation system are adjusted to improve the navigation accuracy.
Battery management and energy optimization, namely adjusting the charging period and the energy use strategy of the battery according to the battery performance attenuation and the energy consumption mode. For example, increasing the charging frequency or adjusting the charging strategy ensures that the battery is operating in an optimal state.
Sensor calibration and parameter adjustment, namely, calibrating the sensor periodically, and ensuring the accuracy and reliability of data. Aiming at the specific characteristics of the sensor, the data acquisition frequency and the processing algorithm are adjusted, and the adaptability of the sensor to environmental changes is improved.
Updating software and firmware, namely updating the software and firmware of the automatic guided vehicle periodically, and correcting known defects and vulnerabilities. Meanwhile, according to the operation data and the fault history, a control algorithm and decision logic are optimized.
And (3) adjusting the load and weight limit, namely adjusting the allowable maximum load according to the design and performance indexes of the automatic guide vehicle. The load distribution is suitably adjusted to prevent mechanical failure due to overload or unbalanced loads.
Network and communication optimization, namely strengthening network connection and communication protocol setting of the automatic guided vehicle and improving stability and safety of data transmission. The communication frequency and retransmission policy are adjusted to cope with possible network fluctuations.
And setting specific fault response flow and parameters, such as automatic speed reduction or stopping when a fault occurs, and restarting parameters after fault recovery.
Analyzing the obtained running state characteristic data of the automatic guided vehicle by using a fault diagnosis algorithm, wherein the identification of the fault data of the running state of the current automatic guided vehicle comprises the following steps:
Collecting historical running state data of the automatic guided vehicle;
Calculating the occurrence times of faults of each detection point in a preset time period, dividing the occurrence times of faults in the preset time period by the total running time of each detection point, and obtaining the fault frequency distribution of each detection point;
Analyzing the obtained fault frequency distribution of each detection point, drawing a fault frequency distribution curve of each detection point by a statistical analysis method, and identifying an area with higher fault frequency as a fault point easy to occur of the automatic guided vehicle;
constructing a fault monitoring model according to the fault frequency distribution and the fault occurrence rule of each detection point, and deploying the fault monitoring model on the automatic guide vehicle;
when the fault of the region with the fault point is detected, the abnormal value detection is carried out on the running state characteristic data of the automatic guided vehicle by using a fault diagnosis algorithm, and the potential fault data of the current state of the automatic guided vehicle is identified.
In the case of fault diagnosis of automatic guided vehicles in a warehouse, in a large warehouse management system, all automatic guided vehicles are equipped with data recorders, and the data recorders can record information such as the position, speed, load state, battery level and the like of the automatic guided vehicle in real time. The historical data is stored in a central database for analysis and fault diagnosis, and using the historical data, a system administrator sets a preset time period of six months, and counts the number of faults for each automatic guided vehicle detection point (such as battery, motor, sensor). For example, if a battery of an automatic guided vehicle has a problem of a sudden drop in power of 12 times in six months, during which the total running time is 1200 hours, the failure frequency is 0.01 times/hour, and a failure frequency distribution curve of each detection point is drawn using statistical software such as R language (a programming language and environment designed for statistical analysis and graphic representation, which is widely used for statistical calculation, data analysis, and drawing of high-quality charts) or Python (a high-level programming language known for its clear syntax and easiness of reading and writing). These curves help identify that the failure frequency of certain specific detection points is abnormally high, such as the detection points of the battery problems described above are as high as 0.05 times/hour in a specific area, much higher than in other areas, and a failure monitoring model is constructed using machine learning based classification algorithms (such as random forests or support vector machines) using the identified failure frequency and other operating parameters (such as load weight and speed). This model can predict which operating conditions will lead to failure. When the fault monitoring model gives out fault alarms at the battery detection points of a certain automatic guided vehicle, the system immediately executes a fault diagnosis algorithm, such as a box graph or Z score analysis, to determine whether a fault exists. If a fault is confirmed, the system will notify the maintenance team to intervene in time to avoid a more serious fault or system shutdown.
Preferably, the data management module 1 comprises a data acquisition module and a data processing module;
the data acquisition module is used for acquiring real-time running state data of the automatic guided vehicle through a sensing technology;
the data processing module is used for preprocessing the acquired real-time running state data of the automatic guided vehicle to obtain running state characteristic data of the automatic guided vehicle.
Preferably, constructing a fault monitoring model according to the fault frequency distribution and the fault occurrence rule of each detection point, and deploying the fault monitoring model on the automatic guided vehicle comprises:
collecting historical fault data of the automatic guide vehicle, and taking the historical fault data as a training sample of a fault monitoring model;
using the fault frequency distribution and the fault occurrence rule of each detection point as input parameters of a fault monitoring model;
Performing outlier (i.e., singular value) analysis on fault data in the training samples by using an estimation algorithm (i.e., a kernel density estimation algorithm), dividing the fault data into a normal operation area and an abnormal operation area, and outputting an optimal grouping (i.e., optimal clustering) center matrix;
It should be noted that the probability density for each sample data point is calculated using a kernel density estimation algorithm. Common kernel functions include gaussian kernel function and Epanechnikov kernel function, etc., and for each sample data point, the ratio of its probability density to the surrounding data points is calculated, which can be a measure of the amount of singular value variation.
It should be explained that, the abnormal value (i.e. singular value) of the training sample is calculated by a kernel density estimation algorithm (such as a density-based clustering algorithm DBSCAN, kernel density estimation algorithm), and the training sample is clustered by an FCM algorithm to distinguish normal and abnormal data points and determine a clustering center.
It should be explained that, the FCM algorithm (Fuzzy C-means) is a Fuzzy logic based clustering algorithm, which is used to divide data points into different categories (i.e. fault data is divided into a normal operation area and an abnormal operation area), the FCM algorithm allows the data points to belong to membership degrees of multiple categories, and outputs membership degree values of each data point belonging to each category.
Carrying out normalization processing on the training samples, calculating the mode association degree (namely the mode membership degree) of the training samples after the normalization processing, using the optimal grouping center matrix and the performance threshold as the input of the fault monitoring model, and constructing the fault monitoring models of different modes (namely different modes) by combining the mode association degree;
it should be explained that the training data is normalized to eliminate the influence of different measurement scales. And constructing an anomaly monitoring model by using the clustering result and the calculated mode association degree, and setting a performance threshold.
And deploying a trained fault monitoring model in a fault-prone area of the automatic guided vehicle, and comparing the performance deviation of each monitored detection point with a preset performance threshold value to judge whether the automatic guided vehicle is in a fault state.
It should be noted that, taking fault monitoring of automatic guided vehicles in a warehouse as an example, in one warehouse, all automatic guided vehicles are equipped with sensors and data recording systems capable of recording various fault data including battery status, motor function, sensor accuracy, etc. Historical fault data is collected and stored in a central database of the plant for use as a subsequent fault monitoring model training sample to calculate the fault frequency distribution for each of the detection points. For example, if a navigation system of an automated guided vehicle fails 15 times in six months and the total running time of the automated guided vehicle is 1500 hours, the failure frequency is 0.01 times/hour, the failure data of the automated guided vehicle is analyzed by using a kernel density estimation algorithm, the distribution of the data is determined, the best clustering centers are found, the centers represent the most typical failure modes, the training samples are normalized, and all input features are ensured to be in the same magnitude, so that the algorithm is prevented from biasing the features with larger values. Calculating the mode association degree of the normalized data, determining the similarity degree of each sample and the optimal clustering center, deploying a trained fault monitoring model on the automatic guided vehicle, and receiving new sensor data in real time when the model runs. Whenever the data shows that the performance deviation of a certain detection point exceeds a preset performance threshold, the model triggers an alarm, prompting a possible fault.
Preferably, deploying a trained fault monitoring model in a fault prone region of the automatic guided vehicle, and comparing the performance deviation of each monitored detection point with a preset performance threshold value to determine whether the automatic guided vehicle is in a fault state comprises:
deploying a trained fault monitoring model in a fault-prone area of the automatic guided vehicle, reading operation parameters of each detection point of the automatic guided vehicle by using the fault monitoring model, and calculating performance indexes;
if the calculated performance index is not greater than the preset performance threshold, judging that the automatic guided vehicle is in a normal running state, and continuously reading the running parameters of each detection point of the automatic guided vehicle by using a fault monitoring model and recalculating the performance index;
If the performance index is greater than a preset threshold, judging that the automatic guided vehicle is in an abnormal running state, and starting a standby fault monitoring model for re-monitoring;
if the monitoring result of the standby fault monitoring model is normal, judging that the process mode is changed, and continuously reading the operation parameters of each detection point of the automatic guided vehicle by using the fault monitoring model and recalculating the performance index;
if the monitoring result of the standby fault monitoring model also indicates abnormality, judging that the fault-prone area is faulty, and confirming that the automatic guided vehicle is in a fault state.
It should be noted that, taking fault monitoring of an automatic guided vehicle in a warehouse as an example, in one warehouse, key operation parameters of the automatic guided vehicle, such as battery status, motor efficiency, sensor accuracy, etc., are regarded as a fault-prone area. To monitor these areas, a trained fault monitoring model is deployed on each automated guided vehicle. The model calculates performance indexes by analyzing the parameters in real time, wherein the model reads the operation parameters of each detection point in real time, and the calculated performance indexes are compared with a preset performance threshold. If the performance index does not exceed the threshold, the system judges that the automatic guided vehicle is in a normal running state, the model continues to monitor data, if the performance index exceeds the threshold, the system starts a standby fault monitoring model to monitor the automatic guided vehicle for the second time so as to confirm whether the automatic guided vehicle really has abnormality, and if the result of the standby model shows that the automatic guided vehicle runs normally, the running environment or condition of the automatic guided vehicle changes, so that the primary model prediction is inaccurate. The system will readjust the model parameters or retrain the model to accommodate the new operating conditions and if the standby model also indicates an anomaly, the system confirms that the failure-prone region of the automated guided vehicle did fail. At this point, the system will automatically send a fault report to the maintenance team and possibly shut down automatically according to the configuration to avoid further damage.
Preferably, when the fault is detected in the fault-prone point area, detecting an abnormal value of the running state characteristic data of the automatic guided vehicle by using a fault diagnosis algorithm, and identifying the fault data of the potential current automatic guided vehicle state includes:
setting parameters of a fault diagnosis algorithm;
generating an initial fault diagnosis model population by using a chaotic sequence (namely Tent chaotic mapping), and setting initial positions and parameters for each fault diagnosis model;
evaluating and sorting all fault diagnosis models according to the fitness, classifying all fault diagnosis models into different performance levels according to the performance, and storing the optimal fault diagnosis models and parameters;
performing Gaussian mutation on the fault diagnosis model with the optimal representation, comparing the fitness values before and after mutation, and storing the fault diagnosis model with the highest fitness value;
Updating the probability threshold value of the fault diagnosis model with the highest fitness value, calculating the new positions of the fault diagnosis models on different performance levels according to the new parameters, and finally updating the positions and parameters of all the fault diagnosis models;
Updating environment variables, and if the environment variables reach preset conditions, improving a fault diagnosis model with poor repositioning performance;
and repeatedly evaluating and sequencing all the fault diagnosis models according to the fitness, carrying out random reverse learning on the global optimal fault diagnosis model to generate a new optimal fault diagnosis model, and outputting the new optimal fault diagnosis model as a final fault diagnosis model and outputting fault data of the current automatic guided vehicle state if the fitness value of the new optimal fault diagnosis model is larger than that of the current optimal fault diagnosis model.
It should be noted that, in the case of fault diagnosis of an automatic guided vehicle in a warehouse, parameters of a fault diagnosis algorithm, such as a search depth, a population size, a mutation rate, etc., are set in one warehouse to initialize a population of a fault diagnosis model. The selection of parameters is based on historical fault data and expected diagnostic efficiency, using a chaotic sequence to generate an initial population, and setting initial positions and parameters for each model. The chaotic sequence is helpful to cover a wide range of possible solutions due to its random nature, evaluate and rank all fault diagnosis models by fitness, and classify them into different performance levels. And (3) keeping the model with the optimal performance and parameters thereof as a benchmark, carrying out Gaussian variation on the model with the optimal performance, and searching a new possible solution by adjusting the parameters thereof. Comparing fitness values before and after mutation, reserving a model with higher fitness, updating a probability threshold of the model with highest fitness, calculating new positions of the models on different performance levels according to new parameters, finally updating positions and parameters of all the models, monitoring changes of environmental variables, and triggering repositioning and optimizing of the model with poor performance if preset conditions are met. The ranking is repeatedly evaluated, and a new optimal model is generated through random reverse learning.
Specifically, the fault diagnosis algorithm is a chaotic squirrel search algorithm with mixed random reverse learning and Gaussian variation, the algorithm generates a chaotic initial population through a Tent chaotic mapping initialization strategy, the uniformity of initial population distribution is enhanced, the more efficient search of a solution space is realized, and the global search and local development capacity of the squirrel search algorithm are balanced by adopting a nonlinear decreasing predator probability strategy.
Specifically, the squirrel search algorithm simulates the dynamic foraging behavior of southern squirrel by the squirrel searching for food sources by gliding between different trees in the forest (hickory, oak, common tree) where hickory is the best food source, and by the increased probability of predation due to reduced winter leaf coverage, the squirrel becomes inactive again until the end of winter. The method is repeated continuously, a squirrel search algorithm is formed, dominant individuals in the population are kept continuously in an iterative process by using a position greedy selection strategy, so that the algorithm convergence speed is improved, a random reverse learning and Gaussian variation strategy is introduced, the population diversity is increased, and meanwhile the capability of the algorithm to jump out of local optimum is improved.
Preferably, the formula for generating a new optimal fault diagnosis model by carrying out random reverse learning on the global optimal fault diagnosis model is as follows:
in the formula, The position of a new optimal fault diagnosis model generated by random reverse learning (namely, the position of a random reverse squirrel in a chaotic squirrel searching algorithm) is represented;
the position expressed as the current optimal fault diagnosis model (namely the current optimal squirrel position);
lb represents the lower bound of the search space for the fault diagnosis model parameters;
ub is denoted as the upper bound of the search space for the fault diagnosis model parameters;
rand () is represented as a generated random number.
Specifically, a random number between [0,1] is generated.
Preferably, the fault prediction module 3 comprises a model building module and a model prediction module;
the model building module is used for building a fault prediction model based on the recognized fault data of the running state of the automatic guided vehicle;
And the model prediction module is used for predicting the operation faults of the automatic guided vehicle at the future moment by utilizing the fault prediction model to obtain key factors influencing the operation faults of the automatic guided vehicle.
Preferably, the fault prediction model is used for predicting the operation fault of the automatic guided vehicle at the future moment, and the key factors affecting the operation fault of the automatic guided vehicle include:
optimizing parameters of a fault prediction model by using a model optimization algorithm based on the performance index, and finding out the optimal parameter configuration of the fault prediction model;
and applying the parameter configuration to the constructed fault prediction model to predict so as to obtain key factors influencing the operation fault of the automatic guided vehicle.
Preferably, optimizing parameters of the fault prediction model based on the performance index by using a model optimization algorithm, and finding the optimal fault prediction model parameter configuration includes:
initializing a model optimization algorithm;
determining the initial position of the fault prediction model parameters by using a spherical coordinate method, sorting according to the fitness value of the initial position, and selecting a plurality of parameter configurations as initial solutions;
the obtained parameter values of the fault prediction model and the adjustment factors are brought into an update formula to obtain new parameter positions of the fault prediction model;
And checking whether a stopping condition is met, if so, outputting the optimal solution as the optimal fault prediction model parameter configuration, and if not, continuing to execute the parameter updating cycle.
In the case of failure prediction of an automatic guided vehicle in a warehouse, the model parameters are initialized in one warehouse by using an optimization algorithm such as a genetic algorithm and a goblet sea squirt algorithm. These parameters may include the learning rate, number of iterations, regularization strength, etc. of the model, with spherical coordinates setting the initial position for the failure prediction model parameters. The method ensures the diversity and the universality of the initial solution by mapping the parameter positions on the spherical surface, and selects the first several groups of parameters with the highest fitness as the initial solution according to the fitness evaluation of each group of parameters. Then, using the update formula of the model optimization algorithm, new parameter positions are calculated according to the fitness and adjustment factors, and the parameter update loop is repeatedly performed until a stopping condition is met (e.g., the maximum number of iterations is reached or the fitness improvement is no longer significant). Once the stopping condition is met, outputting the current optimal parameter configuration as the optimal parameter setting of the fault prediction model, applying the optimized parameter configuration to the constructed fault prediction model, predicting future faults by analyzing historical fault data and current running state of the automatic guided vehicle, and identifying key factors such as insufficient battery power, sensor failure, navigation system errors and the like, which cause the future faults of the automatic guided vehicle.
Specifically, the model optimization algorithm is a self-adaptive ascidian algorithm, and the traditional ascidian algorithm randomly generates a population initial position, so that the algorithm is easy to sink into local optimum and has poor convergence speed and precision, therefore, the initial population position is optimized by using a method of quantum bit Bloch spherical coordinate coding, the global and local searching capacity of the traditional ascidian algorithm is balanced by using an inertia weight, and the algorithm performance is improved so that the algorithm can converge more quickly.
The traditional ascidian algorithm is a new optimization algorithm proposed according to the feeding behavior of the marine organism ascidian, and the new optimization algorithm is often moved in a chain form, the whole population consists of a leader and a follower, the leader is at the forefront end of the chain, and the rest is the follower. The foraging direction is determined by the leader in the whole foraging process, and the follower only moves along with the leader.
According to another embodiment of the present invention, as shown in fig. 2, there is also provided an AGV fault diagnosis prediction method based on data analysis, including the steps of:
s1, acquiring real-time running state data of an automatic guiding vehicle, and preprocessing the real-time running state data of the automatic guiding vehicle to obtain running state characteristic data of the automatic guiding vehicle;
S2, analyzing the obtained running state characteristic data of the automatic guided vehicle by using a fault diagnosis algorithm, and identifying fault data of the running state of the current automatic guided vehicle;
S3, based on the recognized fault data of the running state of the automatic guided vehicle, a fault prediction model is established, and the running fault of the automatic guided vehicle at the future moment is predicted by using the fault prediction model, so that key factors influencing the running fault of the automatic guided vehicle are obtained;
s4, adjusting the operation parameters of the automatic guided vehicle based on the obtained key factors influencing the operation faults of the automatic guided vehicle.
In summary, by means of the technical scheme, the invention can rapidly identify and respond to faults through real-time monitoring and data-driven fault diagnosis, reduce the downtime and maintenance cost of the system, realize the transition from post-processing to preventive maintenance through deep analysis and prediction of fault data, discover problems in advance and intervene, avoid larger equipment damage and production interruption, enable enterprises to more effectively allocate maintenance resources, optimize maintenance plans and budgets, improve the overall operation efficiency, continuously monitor and timely maintain, avoid serious losses caused by faults, prolong the service life of equipment, improve the investment return rate of assets, further reduce safety accidents caused by equipment faults, protect the safety of operators and related facilities, and maintain the stability of production environment. According to the invention, the fault prediction model is constructed based on the identified fault data, so that the prediction is based on actual observation data, and the accuracy of the prediction is improved. The optimized model can accurately identify possible faults in the future, so that preventive measures are more targeted and effective, the fault prediction model can analyze and predict the fault condition in the future in real time, and the trend and influencing factors of the fault development can be found in time. The method enables enterprises to quickly respond, reduces the possibility and range of fault expansion, reduces maintenance time and cost, dynamically adjusts parameters of a fault prediction model according to performance indexes by using a model optimization algorithm, and ensures that the model always operates in an optimal state. The dynamic adjustment helps the model to adapt to the change of the running environment, improves the generalization capability and the application flexibility of the model, and can intervene before serious damage is caused by faults through timely fault prediction and corresponding maintenance, so that the service life of the equipment is prolonged, and the overall use efficiency of the equipment is improved.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (10)
1. The AGV fault diagnosis and prediction system based on the data analysis is characterized by comprising a data management module, a fault diagnosis module, a fault prediction module and an adjustment control module;
The data management module is used for acquiring the real-time running state data of the automatic guided vehicle, and preprocessing the real-time running state data of the automatic guided vehicle to obtain the running state characteristic data of the automatic guided vehicle;
the fault diagnosis module is used for analyzing the obtained running state characteristic data of the automatic guided vehicle by using a fault diagnosis algorithm and identifying the fault data of the running state of the current automatic guided vehicle;
the fault prediction module is used for establishing a fault prediction model based on the recognized fault data of the running state of the automatic guided vehicle, and predicting the running fault of the automatic guided vehicle at the future moment by using the fault prediction model to obtain key factors influencing the running fault of the automatic guided vehicle;
the adjusting control module is used for adjusting the operation parameters of the automatic guided vehicle based on the obtained key factors influencing the operation faults of the automatic guided vehicle;
The analyzing the obtained running state characteristic data of the automatic guided vehicle by using the fault diagnosis algorithm, and the identifying the fault data of the running state of the current automatic guided vehicle comprises the following steps:
Collecting historical running state data of the automatic guided vehicle;
Calculating the occurrence times of faults of each detection point in a preset time period, dividing the occurrence times of faults in the preset time period by the total running time of each detection point, and obtaining the fault frequency distribution of each detection point;
Analyzing the obtained fault frequency distribution of each detection point, drawing a fault frequency distribution curve of each detection point by a statistical analysis method, and identifying an area with higher fault frequency as a fault point easy to occur of the automatic guided vehicle;
constructing a fault monitoring model according to the fault frequency distribution and the fault occurrence rule of each detection point, and deploying the fault monitoring model on the automatic guide vehicle;
when the fault of the region with the fault point is detected, the abnormal value detection is carried out on the running state characteristic data of the automatic guided vehicle by using a fault diagnosis algorithm, and the potential fault data of the current state of the automatic guided vehicle is identified.
2. The AGV fault diagnosis and prediction system based on data analysis according to claim 1, wherein the data management module comprises a data acquisition module and a data processing module;
the data acquisition module is used for acquiring real-time running state data of the automatic guided vehicle through a sensing technology;
the data processing module is used for preprocessing the acquired real-time running state data of the automatic guided vehicle to obtain running state characteristic data of the automatic guided vehicle.
3. The system of claim 1, wherein the constructing a fault monitoring model based on the fault frequency distribution and the fault occurrence law of each detection point and deploying the fault monitoring model on the automatic guided vehicle comprises:
collecting historical fault data of the automatic guide vehicle, and taking the historical fault data as a training sample of a fault monitoring model;
using the fault frequency distribution and the fault occurrence rule of each detection point as input parameters of a fault monitoring model;
Performing outlier analysis on fault data in the training samples by using an estimation algorithm, dividing the fault data into a normal operation area and an abnormal operation area, and outputting an optimal grouping center matrix;
carrying out normalization processing on the training samples, calculating the mode association degree of the training samples after the normalization processing, using the optimal grouping center matrix and the performance threshold as the input of the fault monitoring model, and constructing the fault monitoring models of different modes by combining the mode association degree;
and deploying a trained fault monitoring model in a fault-prone area of the automatic guided vehicle, and comparing the performance deviation of each monitored detection point with a preset performance threshold value to judge whether the automatic guided vehicle is in a fault state.
4. The system of claim 3, wherein deploying the trained fault monitoring model in the fault prone region of the automated guided vehicle and comparing the monitored performance deviation of each detection point with a preset performance threshold to determine whether the automated guided vehicle is in a fault state comprises:
deploying a trained fault monitoring model in a fault-prone area of the automatic guided vehicle, reading operation parameters of each detection point of the automatic guided vehicle by using the fault monitoring model, and calculating performance indexes;
if the calculated performance index is not greater than the preset performance threshold, judging that the automatic guided vehicle is in a normal running state, and continuously reading the running parameters of each detection point of the automatic guided vehicle by using a fault monitoring model and recalculating the performance index;
If the performance index is greater than a preset threshold, judging that the automatic guided vehicle is in an abnormal running state, and starting a standby fault monitoring model for re-monitoring;
if the monitoring result of the standby fault monitoring model is normal, judging that the process mode is changed, and continuously reading the operation parameters of each detection point of the automatic guided vehicle by using the fault monitoring model and recalculating the performance index;
if the monitoring result of the standby fault monitoring model also indicates abnormality, judging that the fault-prone area is faulty, and confirming that the automatic guided vehicle is in a fault state.
5. The system of claim 4, wherein the detecting the abnormal value of the characteristic data of the operation state of the automatic guided vehicle by using the fault diagnosis algorithm when the fault occurs in the fault-prone point area, and identifying the fault data of the potential current state of the automatic guided vehicle comprises:
setting parameters of a fault diagnosis algorithm;
Generating an initial fault diagnosis model population by using a chaotic sequence, and setting an initial position and parameters for each fault diagnosis model;
evaluating and sorting all fault diagnosis models according to the fitness, classifying all fault diagnosis models into different performance levels according to the performance, and storing the optimal fault diagnosis models and parameters;
performing Gaussian mutation on the fault diagnosis model with the optimal representation, comparing the fitness values before and after mutation, and storing the fault diagnosis model with the highest fitness value;
Updating the probability threshold value of the fault diagnosis model with the highest fitness value, calculating the new positions of the fault diagnosis models on different performance levels according to the new parameters, and finally updating the positions and parameters of all the fault diagnosis models;
Updating environment variables, and if the environment variables reach preset conditions, improving a fault diagnosis model with poor repositioning performance;
and repeatedly evaluating and sequencing all the fault diagnosis models according to the fitness, carrying out random reverse learning on the global optimal fault diagnosis model to generate a new optimal fault diagnosis model, and outputting the new optimal fault diagnosis model as a final fault diagnosis model and outputting fault data of the current automatic guided vehicle state if the fitness value of the new optimal fault diagnosis model is larger than that of the current optimal fault diagnosis model.
6. The system of claim 5, wherein the formula for generating a new optimal fault diagnosis model by performing random reverse learning on a global optimal fault diagnosis model is as follows:
in the formula, Representing the position of a new optimal fault diagnosis model generated by random reverse learning;
a location expressed as a current optimal fault diagnosis model;
lb represents the lower bound of the search space for the fault diagnosis model parameters;
ub is denoted as the upper bound of the search space for the fault diagnosis model parameters;
rand () is represented as a generated random number.
7. The AGV fault diagnosis and prediction system based on data analysis according to claim 1, wherein the fault prediction module comprises a model building module and a model prediction module;
The model building module is used for building a fault prediction model based on the recognized fault data of the running state of the automatic guided vehicle;
The model prediction module is used for predicting the operation faults of the automatic guided vehicle at the future moment by using the fault prediction model to obtain key factors influencing the operation faults of the automatic guided vehicle.
8. The system of claim 7, wherein the predicting the operation failure of the automatic guided vehicle at the future time using the failure prediction model to obtain the key factors affecting the operation failure of the automatic guided vehicle comprises:
optimizing parameters of a fault prediction model by using a model optimization algorithm based on the performance index, and finding out the optimal parameter configuration of the fault prediction model;
and applying the parameter configuration to the constructed fault prediction model to predict so as to obtain key factors influencing the operation fault of the automatic guided vehicle.
9. The system of claim 8, wherein optimizing parameters of the fault prediction model based on the performance metrics using a model optimization algorithm to find an optimal fault prediction model parameter configuration comprises:
initializing a model optimization algorithm;
determining the initial position of the fault prediction model parameters by using a spherical coordinate method, sorting according to the fitness value of the initial position, and selecting a plurality of parameter configurations as initial solutions;
the obtained parameter values of the fault prediction model and the adjustment factors are brought into an update formula to obtain new parameter positions of the fault prediction model;
And checking whether a stopping condition is met, if so, outputting the optimal solution as the optimal fault prediction model parameter configuration, and if not, continuing to execute the parameter updating cycle.
10. An AGV fault diagnosis and prediction method based on data analysis for implementing the AGV fault diagnosis and prediction system based on data analysis according to any one of claims 1 to 9, characterized in that the AGV fault diagnosis and prediction method based on data analysis includes the steps of:
s1, acquiring real-time running state data of an automatic guiding vehicle, and preprocessing the real-time running state data of the automatic guiding vehicle to obtain running state characteristic data of the automatic guiding vehicle;
S2, analyzing the obtained running state characteristic data of the automatic guided vehicle by using a fault diagnosis algorithm, and identifying fault data of the running state of the current automatic guided vehicle;
S3, based on the recognized fault data of the running state of the automatic guided vehicle, a fault prediction model is established, and the running fault of the automatic guided vehicle at the future moment is predicted by using the fault prediction model, so that key factors influencing the running fault of the automatic guided vehicle are obtained;
s4, adjusting the operation parameters of the automatic guided vehicle based on the obtained key factors influencing the operation faults of the automatic guided vehicle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411217801.3A CN119105554A (en) | 2024-09-02 | 2024-09-02 | An AGV fault diagnosis and prediction system and method based on data analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411217801.3A CN119105554A (en) | 2024-09-02 | 2024-09-02 | An AGV fault diagnosis and prediction system and method based on data analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN119105554A true CN119105554A (en) | 2024-12-10 |
Family
ID=93713404
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411217801.3A Pending CN119105554A (en) | 2024-09-02 | 2024-09-02 | An AGV fault diagnosis and prediction system and method based on data analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119105554A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119646713A (en) * | 2024-12-25 | 2025-03-18 | 南通泰达物联网科技有限公司 | A charging service platform data analysis method and system based on the Internet of Things |
-
2024
- 2024-09-02 CN CN202411217801.3A patent/CN119105554A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119646713A (en) * | 2024-12-25 | 2025-03-18 | 南通泰达物联网科技有限公司 | A charging service platform data analysis method and system based on the Internet of Things |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117789422A (en) | Combustible gas alarm control system and method | |
| CN108460144B (en) | Coal equipment fault early warning system and method based on machine learning | |
| US11521105B2 (en) | Machine learning device and machine learning method for learning fault prediction of main shaft or motor which drives main shaft, and fault prediction device and fault prediction system including machine learning device | |
| CN117786373A (en) | Equipment operation diagnosis system based on big data corrugated paper processing | |
| CN116862199A (en) | Building construction optimizing system based on big data and cloud computing | |
| CN117911012B (en) | Equipment health management system based on equipment ecological detection and running state evaluation | |
| KR102644388B1 (en) | Dynamic vehicle performance analyzer with smoothing filter | |
| CN119539769A (en) | An intelligent operation and maintenance method and system for a solar thermal power generation system based on the Internet of Things | |
| CN119312148B (en) | Equipment failure prediction method and related equipment based on industrial Internet of Things | |
| CN119441777B (en) | A gantry crane intelligent fault diagnosis method and system | |
| zu Wickern | Challenges and reliability of predictive maintenance | |
| CN112390127B (en) | Health degree model-based preventive maintenance strategy generation method for escalator | |
| CN119105554A (en) | An AGV fault diagnosis and prediction system and method based on data analysis | |
| KR102474332B1 (en) | Prediction method of machine health stability in smart factory system, and recording medium thereof | |
| CN119847066A (en) | A method and system for intelligently controlling the operation status of a precision CNC machine tool | |
| CN120278705A (en) | Method and system for analyzing equipment state data applied to automatic electric | |
| CN118364411A (en) | A system and method for locating stacking faults of unmanned forklifts based on machine learning | |
| CN120145300A (en) | Evaluation method of relay protection for smart substation based on multi-source data fusion | |
| CN118797538B (en) | Industrial energy storage system management method and system based on multi-source data | |
| Gandhi et al. | Machine maintenance decision support system: a systematic literature review | |
| KR102334590B1 (en) | Fault detection prediction system for manufacturing machines using artificial neural networks | |
| Raj et al. | Predictive Maintenance in Autonomous Vehicles Using Machine Learning Techniques | |
| CN119313263B (en) | Small warehouse cargo management method and system based on artificial intelligence | |
| CN120736377A (en) | Elevator control reminder and fault self-diagnosis method and system | |
| CN118861944B (en) | A system construction method combining Internet of Things and virtual reality technology |
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 |