AGV fault diagnosis and prediction system and method based on data analysis

Abstract

The invention discloses an AGV fault diagnosis and prediction system and method based on data analysis, and relates to the technical field of automated guided vehicle management, and the AGV fault diagnosis and prediction system based on data analysis comprises a data management module, a fault diagnosis module, a fault prediction module and an adjustment control module. The method can quickly respond to possible operation problems by recognizing fault data in time, reduce downtime and predict possible faults in the future, so that an enterprise can turn from reactive fault processing to active fault prevention, thereby greatly improving the continuity and stability of production and reducing the production cost. By analyzing the key factors influencing the operation of the automatic guided vehicle, the operation parameters can be adjusted and optimized to deal with the predicted potential faults, so that the overall maintenance cost is reduced, and the service life of the automatic guided vehicle and related equipment can be effectively prolonged.

Description

Technical Field

[0001] The present invention relates to the technical field of automatic guided vehicle management, and in particular to an AGV fault diagnosis and prediction system and method based on data analysis. Background Art

[0002] Automated guided vehicles (AGVs) play a vital role in smart manufacturing companies. They are not only a key tool for realizing digitalization, automation, networking and intelligence of production, but also an important indicator of the level of industrial automation. AGVs are mainly responsible for material handling within the factory. Their efficiency directly affects the smoothness of the production process and the operational efficiency of the enterprise. Therefore, maintaining the efficient operation of the AGV system is crucial to ensuring production safety and improving production efficiency. In daily production, if the AGV system fails and is not detected and processed in time, it will not only cause the production process to be interrupted, increase the pressure of material accumulation and pending work, but also may cause safety accidents due to improper operation. In this case, fault detection and timely maintenance of the AGV system become particularly important. System failures may be caused by a variety of factors, including hardware failures, software program errors, or improper operation.

[0003] In the prior art, it is not convenient to effectively screen and extract key characteristic data that affect the operation of the automatic guided vehicle, thereby reducing the accuracy of fault detection, failing to identify fault data in a timely manner, increasing downtime, and it is not convenient to establish a fault prediction model to predict possible future faults, thereby increasing the fault incidence rate. At the same time, it is not convenient to analyze the key factors affecting the operation of the automatic guided vehicle, and it is not convenient to adjust and optimize the operating parameters to cope with the predicted potential faults, thereby reducing the operating efficiency of the automatic guided vehicle and increasing the risk of failures caused by inappropriate parameter settings, reducing the service life of the automatic guided vehicle and related equipment.

[0004] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention proposes an AGV fault diagnosis and prediction system and method based on data analysis, which solves the problem mentioned in the above background technology that the existing technology is not convenient for effectively screening and refining key feature data that affect the operation of the automatic guided vehicle, thereby reducing the accuracy of fault detection, failing to identify fault data in a timely manner, increasing downtime, and being inconvenient to establish a fault prediction model to predict possible future faults, thereby increasing the fault incidence rate. At the same time, it is not convenient to analyze the key factors affecting the operation of the automatic guided vehicle, and it is not convenient to adjust and optimize the operating parameters to cope with the predicted potential faults, thereby reducing the operating efficiency of the automatic guided vehicle and increasing the risk of failures caused by inappropriate parameter settings, thereby reducing the service life of the automatic guided vehicle and related equipment.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0007] According to one aspect of the present invention, there is provided an AGV fault diagnosis and prediction system based on data analysis, the AGV fault diagnosis and prediction system based on data analysis comprising: a data management module, a fault diagnosis module, a fault prediction module and an adjustment control module;

[0008] A data management module is used to obtain real-time operation status data of the automatic guided vehicle and pre-process the real-time operation status data of the automatic guided vehicle to obtain operation status characteristic data of the automatic guided vehicle;

[0009] A fault diagnosis module is used to analyze the obtained running state characteristic data of the automatic guided vehicle using a fault diagnosis algorithm to identify fault data of the current running state of the automatic guided vehicle;

[0010] A fault prediction module is used to establish a fault prediction model based on the fault data of the operating state of the automatic guided vehicle identified, and use the fault prediction model to predict the operating faults of the automatic guided vehicle at a future moment, and obtain key factors affecting the operating faults of the automatic guided vehicle;

[0011] An adjustment control module, used for adjusting the operation parameters of the automatic guided vehicle based on the obtained key factors affecting the operation failure of the automatic guided vehicle;

[0012] The fault diagnosis algorithm is used to analyze the obtained running state characteristic data of the automatic guided vehicle, and the fault data of the current running state of the automatic guided vehicle is identified to include:

[0013] Collect historical operating status data of automated guided vehicles;

[0014] Calculate the number of failures at each detection point within a preset time period, and divide the number of failures within the preset time period by the total operating time of each detection point to obtain the failure frequency distribution of each detection point;

[0015] Analyze the fault frequency distribution of each detection point, draw the fault frequency distribution curve of each detection point by statistical analysis method, and identify the area with higher fault frequency as the prone fault point of the automatic guided vehicle;

[0016] According to the fault frequency distribution and fault occurrence law of each detection point, a fault monitoring model is constructed and deployed on the automatic guided vehicle;

[0017] When a fault is detected in a fault-prone area, a fault diagnosis algorithm is used to detect abnormal values ​​of the running status characteristic data of the automatic guided vehicle, and identify potential fault data of the current automatic guided vehicle status.

[0018] Furthermore, the data management module includes: a data acquisition module and a data processing module;

[0019] A data acquisition module, used to acquire real-time operation status data of the automatic guided vehicle through sensing technology;

[0020] The data processing module is used to pre-process the acquired real-time running status data of the automatic guided vehicle to obtain the running status characteristic data of the automatic guided vehicle.

[0021] Furthermore, according to the fault frequency distribution and fault occurrence law of each detection point, a fault monitoring model is constructed, and the fault monitoring model is deployed on the automatic guided vehicle, including:

[0022] Collect historical fault data of the AGV and use the historical fault data as training samples for the fault monitoring model;

[0023] The fault frequency distribution and fault occurrence law of each detection point are used as input parameters of the fault monitoring model;

[0024] Use estimation algorithm to analyze outliers on fault data in training samples, divide fault data into normal operation area and abnormal operation area, and output the optimal grouping center matrix;

[0025] Normalize the training samples, calculate the pattern correlation of the normalized training samples, use the optimal grouping center matrix and performance threshold as the input of the fault monitoring model, and build fault monitoring models of different modes in combination with the pattern correlation;

[0026] The trained fault monitoring model is deployed in the fault-prone areas of the automated guided vehicle, and the performance deviation of each monitored detection point is compared with the preset performance threshold to determine whether the automated guided vehicle is in a fault state.

[0027] Furthermore, the trained fault monitoring model is deployed in the fault-prone area of ​​the automated guided vehicle, and the performance deviation of each detection point monitored is compared with the preset performance threshold to determine whether the automated guided vehicle is in a fault state, including:

[0028] Deploy the trained fault monitoring model in the fault-prone areas of the AGV, use the fault monitoring model to read the operating parameters of each detection point of the AGV, and calculate the performance indicators;

[0029] If the calculated performance index is not greater than the preset performance threshold, the automated guided vehicle is judged to be in a normal operating state, and the fault monitoring model is used to continue to read the operating parameters of each detection point of the automated guided vehicle and recalculate the performance index;

[0030] If the performance index is greater than a preset threshold, the automated guided vehicle is judged to be in an abnormal operating state, and a backup fault monitoring model is started for re-monitoring;

[0031] If the monitoring result of the standby fault monitoring model is normal, the judgment process mode is changed, and the fault monitoring model is used to continue to read the operating parameters of each detection point of the automatic guided vehicle and recalculate the performance index;

[0032] If the monitoring result of the backup fault monitoring model also indicates an abnormality, it is determined that a fault has occurred in the fault-prone area, and it is confirmed that the automatic guided vehicle is in a faulty state.

[0033] Furthermore, when a fault is detected in a fault-prone area, the fault diagnosis algorithm is used to detect abnormal values ​​of the running state characteristic data of the automatic guided vehicle, and the potential fault data of the current automatic guided vehicle state is identified, including:

[0034] Set the parameters of the fault diagnosis algorithm;

[0035] Generate the initial population of fault diagnosis models using chaotic sequences, and set the initial position and parameters for each fault diagnosis model;

[0036] All fault diagnosis models are evaluated and sorted according to fitness, and all fault diagnosis models are classified into different performance levels according to performance, and the optimal fault diagnosis model and parameters are saved;

[0037] Perform Gaussian mutation on the fault diagnosis model with the best performance, compare the fitness values ​​before and after the mutation, and save the fault diagnosis model with the highest fitness value;

[0038] Update the probability threshold of the fault diagnosis model with the highest fitness value, calculate the new position of the fault diagnosis model at different performance levels based on the new parameters, and finally update the position and parameters of all fault diagnosis models;

[0039] Update the environment variables, and if the preset conditions are met, relocate the fault diagnosis model with poor performance to improve it;

[0040] All fault diagnosis models are repeatedly evaluated and sorted according to their fitness, and random reverse learning is performed on the global optimal fault diagnosis model to generate a new optimal fault diagnosis model. If the fitness value of the new optimal fault diagnosis model is greater than the fitness value of the current optimal fault diagnosis model, the new optimal fault diagnosis model is output as the final fault diagnosis model, and the fault data of the current automatic guided vehicle status is output.

[0041] Furthermore, the formula for generating a new optimal fault diagnosis model by performing random reverse learning on the global optimal fault diagnosis model is:

[0042]

[0043] In the formula, represents the position of the new optimal fault diagnosis model generated by random reverse learning;

[0044] Represents the position of the current optimal fault diagnosis model;

[0045] lb represents the lower bound of the search space of fault diagnosis model parameters;

[0046] ub represents the upper bound of the search space of fault diagnosis model parameters;

[0047] rand() represents a generated random number.

[0048] Furthermore, the fault prediction module includes: a model building module and a model prediction module;

[0049] A model building module, for building a fault prediction model based on the identified fault data of the running state of the automatic guided vehicle;

[0050] The model prediction module is used to use the fault prediction model to predict the operation failure of the automatic guided vehicle at a future moment, and obtain the key factors affecting the operation failure of the automatic guided vehicle.

[0051] Furthermore, the fault prediction model is used to predict the operation faults of the automated guided vehicle at future moments, and the key factors affecting the operation faults of the automated guided vehicle include:

[0052] Based on the performance indicators, the model optimization algorithm is used to optimize the parameters of the fault prediction model and find the best fault prediction model parameter configuration;

[0053] The parameter configuration is applied to the constructed fault prediction model to make predictions and obtain the key factors affecting the operation failure of the automated guided vehicle.

[0054] Furthermore, based on the performance indicators, the parameters of the fault prediction model are optimized using the model optimization algorithm to find the best fault prediction model parameter configuration including:

[0055] Initialize the model optimization algorithm;

[0056] The initial positions of the fault prediction model parameters are determined using the spherical coordinate method, and the first several parameter configurations are selected as the initial solutions according to their fitness values.

[0057] Substituting the obtained fault prediction model parameter values ​​and adjustment factors into the update formula to obtain new fault prediction model parameter positions;

[0058] Check whether the stop condition is met. If so, output the best solution as the best fault prediction model parameter configuration. If not, continue to execute the parameter update loop.

[0059] According to another aspect of the present invention, there is also provided an AGV fault diagnosis prediction method based on data analysis, the AGV fault diagnosis prediction method based on data analysis comprises the following steps:

[0060] S1. Acquire real-time operation status data of the automatic guided vehicle, and pre-process the real-time operation status data of the automatic guided vehicle to obtain operation status characteristic data of the automatic guided vehicle;

[0061] S2. Analyze the obtained running state characteristic data of the automatic guided vehicle using a fault diagnosis algorithm to identify fault data of the current running state of the automatic guided vehicle;

[0062] S3. Based on the identified 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 a future moment is predicted by using the fault prediction model to obtain key factors affecting the running fault of the automatic guided vehicle;

[0063] S4. Adjust the operating parameters of the automatic guided vehicle based on the obtained key factors affecting the automatic guided vehicle operating failure.

[0064] The beneficial effects of the present invention are:

[0065] 1. The present invention can effectively screen and extract key feature data that affect the operation of the automatic guided vehicle by acquiring the operating status data of the automatic guided vehicle in real time and performing preprocessing, thereby improving the accuracy of fault detection. Timely identification of fault data can quickly respond to possible operating problems and reduce downtime. The fault prediction model established based on the identified fault data can further analyze these data and predict possible future faults, allowing enterprises to shift from reactive fault handling to proactive fault prevention, greatly improving the continuity and stability of production. By analyzing the key factors affecting the operation of the automatic guided vehicle, the operating parameters can be adjusted and optimized to deal with predicted potential faults. This parameter adjustment helps to improve the operating efficiency of the automatic guided vehicle and reduce the risk of failure caused by inappropriate parameter settings. Systematic fault diagnosis and prediction can reduce unnecessary maintenance and excessive repairs, thereby reducing overall maintenance costs. At the same time, through preventive maintenance, the service life of the automatic guided vehicle and related equipment can be effectively extended.

[0066] 2. Through real-time monitoring and data-driven fault diagnosis, the present invention can quickly identify and respond to faults, reduce system downtime and maintenance costs, and through in-depth analysis and prediction of fault data, it can achieve the transition from post-processing to preventive maintenance, discover problems in advance and intervene, avoid greater equipment damage and production interruptions. Through accurate fault diagnosis models, enterprises can more effectively allocate maintenance resources, optimize maintenance plans and budgets, and improve overall operational efficiency. Continuous monitoring and timely maintenance can avoid significant losses caused by faults, extend the service life of equipment, and improve the return on investment of assets, thereby reducing safety accidents caused by equipment failures, protecting the safety of operators and related facilities, and maintaining the stability of the production environment.

[0067] 3. The present invention builds a fault prediction model based on the identified fault data, so that the prediction is based on the actual observed data, thereby improving the accuracy of the prediction. The optimized model can more accurately identify possible faults in the future, so that preventive measures are more targeted and effective. The fault prediction model can analyze and predict future fault conditions in real time, and promptly discover the trends and influencing factors of fault development. This enables enterprises to respond quickly, reduce the possibility and scope of fault expansion, reduce maintenance time and cost, and use model optimization algorithms to dynamically adjust the parameters of the fault prediction model according to performance indicators to ensure that the model always operates in the best state. This dynamic adjustment helps the model adapt to changes in the operating environment, improves the generalization ability of the model and the flexibility of application. Through timely fault prediction and corresponding maintenance, intervention can be made before the fault causes serious damage, thereby extending the service life of the equipment and improving the overall efficiency of the equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0069] Figure 1 is a principle block diagram of an AGV fault diagnosis and prediction system based on data analysis according to an embodiment of the present invention;

[0070] Figure 2 It is a flow chart of an AGV fault diagnosis and prediction method based on data analysis according to an embodiment of the present invention.

[0071] In the figure:

[0072] 1. Data management module; 2. Fault diagnosis module; 3. Fault prediction module; 4. Regulation and control module. DETAILED DESCRIPTION

[0073] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0074] In the description of the present invention, unless otherwise specified, the meaning of "plurality" is two or more. In addition, the terms "first", "second", "third", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0075] According to an embodiment of the present invention, a system and method for AGV fault diagnosis and prediction based on data analysis are provided.

[0076] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to an AGV fault diagnosis and prediction system based on data analysis according to an embodiment of the present invention, the AGV fault diagnosis and prediction system based on data analysis includes: a data management module 1, a fault diagnosis module 2, a fault prediction module 3 and an adjustment control module 4;

[0077] The data management module 1 is used to obtain the real-time running status data of the automatic guided vehicle and pre-process the real-time running status data of the automatic guided vehicle to obtain the running status characteristic data of the automatic guided vehicle;

[0078] Specifically, the real-time operating status data of the automatic guided vehicle includes:

[0079] 1) Location information:

[0080] Coordinates: The specific location coordinates of the automated guided vehicle in a factory or warehouse.

[0081] Navigation path: The current driving path and the intended destination of the AGV.

[0082] 2) Speed ​​and acceleration data:

[0083] Speed: The current speed of the AGV.

[0084] Acceleration: Real-time data on whether the AGV is accelerating or decelerating.

[0085] 3) Battery and energy status:

[0086] Battery power: The remaining power of the AGV battery.

[0087] Energy consumption rate: energy consumption of the automated guided vehicle during operation.

[0088] 4) Load status:

[0089] Load capacity: The weight currently carried by the AGV.

[0090] Load state change: increase or decrease of the AGV load.

[0091] 5) Sensor data:

[0092] Environmental sensors: Data collected from the AGV’s surroundings (e.g., temperature, humidity, obstacle distance, etc.).

[0093] Mechanical sensors: Operational data about the mechanical components of the AGV (e.g. motor status, steering system, etc.).

[0094] 6) Operation status:

[0095] Fault Indication: Real-time indication of any system faults or warnings.

[0096] Operation Mode: Whether the AGV is in automatic mode, under manual control, or shut down.

[0097] 7) Communication status:

[0098] Network connection status: The network communication status of the automated guided vehicle, including the quality of the connection with the central control system.

[0099] 8) Software system status:

[0100] System log: A log record of the AGV software's operations that may include details of any exceptions or errors.

[0101] A fault diagnosis module 2 is used to analyze the obtained running state characteristic data of the automatic guided vehicle using a fault diagnosis algorithm to identify fault data of the current running state of the automatic guided vehicle;

[0102] Specifically, the fault data of the operating status of the automatic guided vehicle includes:

[0103] 1) Mechanical failure:

[0104] Drive system problems: such as motor failure, gearbox problem or drive chain damage.

[0105] Failure of the navigation system: such as unresponsive steering mechanism or insufficient positioning accuracy.

[0106] Load bearing mechanism problems: such as pallet lifting mechanism failure or stability issues.

[0107] 2) Electronic failure:

[0108] Sensor failure: The sensor outputs abnormal data, such as inaccurate data from distance sensors, weight sensors, or environmental sensors.

[0109] Control system failure: Failure of the control unit or microprocessor, resulting in errors in data processing or command execution.

[0110] Battery problem: The battery is insufficiently charged, is discharging at an abnormal rate, or has a battery management system malfunction.

[0111] 3) Software and communication failure:

[0112] Software error: An operational problem caused by a software logic error or programming defect.

[0113] Communication interruption: Delayed or interrupted communication with the control system, which may be caused by network problems or interface failures.

[0114] 4) Operation efficiency issues:

[0115] Speed ​​reduction: The automatic guided vehicle's running speed slows down abnormally or cannot reach the set speed.

[0116] Path deviation: The AGV does not follow the planned route and navigation deviation may occur.

[0117] 5) Safety and monitoring warning:

[0118] Safety system alarms: such as emergency stop being triggered, or safety boundaries being violated.

[0119] Maintenance Reminder: Maintenance warning based on pre-set maintenance intervals or detected component wear.

[0120] A fault prediction module 3 is used to establish a fault prediction model based on the fault data of the operating state of the automatic guided vehicle identified, and use the fault prediction model to predict the operating faults of the automatic guided vehicle at a future moment, and obtain key factors affecting the operating faults of the automatic guided vehicle;

[0121] Specifically, the key factors affecting the operation failure of automatic guided vehicles include:

[0122] Hardware wear and aging: Long-term wear of mechanical components such as tires, gears, and bearings is a common cause of failure.

[0123] Battery performance degradation: The reduction in battery life and charging efficiency may lead to unstable energy supply and affect the continuous operation ability of the automated guided vehicle.

[0124] Sensor accuracy and reliability: The accuracy and reliability of sensors directly affect the performance of the AGV’s navigation and positioning systems. Distorted or delayed sensor data can directly lead to navigation errors or accidents.

[0125] Software and algorithm errors: Software logic errors or algorithm defects may lead to incorrect decisions and command execution, such as path planning errors or response delays.

[0126] Environmental factors: Changes in the AGV’s operating environment, such as ground conditions, temperature fluctuations, humidity, etc., can also affect the AGV’s operating status.

[0127] Network and communication systems: The quality of network connections and the stability of communication protocols are key to ensuring smooth communication between the AGV and the control center and other AGVs. Network delays or interruptions may cause command transmission failures.

[0128] Operational errors: Operator errors or program setting errors may also cause automatic guided vehicle operation failures.

[0129] Maintenance and overhaul strategy: Improper or insufficient maintenance may lead to early failure. Timely maintenance and overhaul are important measures to prevent failure.

[0130] An adjustment control module 4, used for adjusting the operation parameters of the automatic guided vehicle based on the obtained key factors affecting the operation failure of the automatic guided vehicle;

[0131] Specifically, adjusting the operating parameters of the automatic guided vehicle includes:

[0132] Adjust speed and acceleration settings: Appropriately adjust the AGV’s driving speed and acceleration based on the operating environment and load conditions. For example, reduce speed in crowded or complex environments to reduce the risk of collisions and failures.

[0133] Optimize path selection and navigation parameters: Use advanced path planning algorithms to optimize the AGV's route to avoid areas prone to failure, such as uneven ground or signal interference. At the same time, adjust the navigation system's sensitivity and response threshold to improve navigation accuracy.

[0134] Battery management and energy optimization: Adjust the battery charging cycle and energy usage strategy based on battery performance degradation and energy consumption patterns. For example, increase the charging frequency or adjust the charging strategy to ensure that the battery works in the best condition.

[0135] Sensor calibration and parameter adjustment: Calibrate sensors regularly to ensure data accuracy and reliability. According to the specific characteristics of the sensor, adjust the data collection frequency and processing algorithm to improve the sensor's ability to adapt to environmental changes.

[0136] Update software and firmware: Regularly update the software and firmware of the automated guided vehicle to correct known defects and vulnerabilities. At the same time, optimize the control algorithm and decision logic based on operating data and fault history.

[0137] Adjust load and weight limits: Adjust the maximum load allowed based on the design and performance specifications of the AGV. Properly adjust the load distribution to prevent mechanical failures caused by overload or unbalanced loads.

[0138] Network and communication optimization: Strengthen the network connection and communication protocol settings of the AGV to improve the stability and security of data transmission. Adjust the communication frequency and retransmission strategy to cope with possible network fluctuations.

[0139] Fault response and recovery strategy: Set specific fault response processes and parameters, such as automatic speed reduction or stop in case of a fault, and restart parameters after fault recovery.

[0140] The fault diagnosis algorithm is used to analyze the obtained running state characteristic data of the automatic guided vehicle, and the fault data of the current running state of the automatic guided vehicle is identified to include:

[0141] Collect historical operating status data of automated guided vehicles;

[0142] Calculate the number of failures at each detection point within a preset time period, and divide the number of failures within the preset time period by the total operating time of each detection point to obtain the failure frequency distribution of each detection point;

[0143] Analyze the fault frequency distribution of each detection point, draw the fault frequency distribution curve of each detection point by statistical analysis method, and identify the area with higher fault frequency as the prone fault point of the automatic guided vehicle;

[0144] According to the fault frequency distribution and fault occurrence law of each detection point, a fault monitoring model is constructed and deployed on the automatic guided vehicle;

[0145] When a fault is detected in a fault-prone area, a fault diagnosis algorithm is used to detect abnormal values ​​of the running status characteristic data of the automatic guided vehicle, and identify potential fault data of the current automatic guided vehicle status.

[0146] It should be explained that, taking the fault diagnosis of automatic guided vehicles in warehouses as an example, in a large warehouse management system, all automatic guided vehicles are equipped with data recorders, which can record the position, speed, load status, battery power and other information of the automatic guided vehicles in real time. The historical data is stored in the central database for analysis and fault diagnosis. Using the historical data, the system administrator sets a preset time period of six months to count the number of failures of each detection point (such as battery, motor, sensor) of the automatic guided vehicle. For example, if the battery of an automatic guided vehicle has a sudden drop in power 12 times in six months, and the total running time during this period is 1200 hours, the failure frequency is 0.01 times / hour. Using statistical software such as R language (a programming language and environment designed for statistical analysis and graphical representation. It is widely used in statistical computing, data analysis, and drawing high-quality charts) or Python (a high-level programming language known for its clear syntax and easy reading and writing. It supports multiple programming paradigms such as object-oriented, imperative, functional and procedural programming), the failure frequency distribution curve of each detection point is drawn. These curves help identify that the failure frequency of certain specific detection points is abnormally high. For example, the detection point of the above-mentioned battery problem has a frequency of up to 0.05 times / hour in a specific area, which is much higher than other areas. Using machine learning-based classification algorithms (such as random forests or support vector machines), a fault monitoring model is constructed using the identified fault frequency and other operating parameters (such as load weight and speed). This model can predict which operating conditions will cause failures. When the fault monitoring model issues a fault alarm at a battery detection point of an automated guided vehicle, the system immediately executes a fault diagnosis algorithm, such as a box plot or Z-score analysis, to determine whether a fault actually exists. If the fault is confirmed to exist, the system will notify the maintenance team to intervene in time to avoid more serious failures or system downtime.

[0147] Preferably, the data management module 1 includes: a data acquisition module and a data processing module;

[0148] A data acquisition module, used to acquire real-time operation status data of the automatic guided vehicle through sensing technology;

[0149] The data processing module is used to pre-process the acquired real-time running status data of the automatic guided vehicle to obtain the running status characteristic data of the automatic guided vehicle.

[0150] Preferably, constructing a fault monitoring model according to the fault frequency distribution and fault occurrence law of each detection point, and deploying the fault monitoring model on the automatic guided vehicle includes:

[0151] Collect historical fault data of the AGV and use the historical fault data as training samples for the fault monitoring model;

[0152] The fault frequency distribution and fault occurrence law of each detection point are used as input parameters of the fault monitoring model;

[0153] An estimation algorithm (i.e., kernel density estimation algorithm) is used to analyze the outliers (i.e., singular values) of the fault data in the training samples, and the fault data is divided into normal operation areas and abnormal operation areas, and the optimal grouping (i.e., optimal clustering) center matrix is ​​output;

[0154] It should be explained that the kernel density estimation algorithm is used to calculate the probability density of each sample data point. Commonly used kernel functions include Gaussian kernel function and Epanechnikov kernel function. For each sample data point, the ratio of its probability density to the surrounding data points is calculated. This ratio can be used as a measure of the change in singular values.

[0155] It should be explained that the outliers (i.e., singular values) of the training samples are calculated by a kernel density estimation algorithm (such as the density-based clustering algorithm DBSCAN, a kernel density estimation algorithm), and the training samples are clustered by the FCM algorithm to distinguish normal and abnormal data points and determine the cluster centers.

[0156] It should be explained that the FCM algorithm (Fuzzy C-means) is a clustering algorithm based on fuzzy logic, which is used to divide data points into different categories (that is, to divide fault data into normal operating areas and abnormal operating areas). The FCM algorithm allows data points to belong to multiple categories, and the output is the membership value of each data point belonging to each category.

[0157] Normalize the training samples, calculate the mode correlation (i.e., modal membership) of the normalized training samples, use the optimal grouping center matrix and performance threshold as the input of the fault monitoring model, and build fault monitoring models of different modes (i.e., different modalities) in combination with the mode correlation;

[0158] It should be explained that the training data is normalized to eliminate the influence of different measurement scales. The clustering results and the calculated pattern correlation are used to build an anomaly monitoring model and set the performance threshold.

[0159] The trained fault monitoring model is deployed in the fault-prone areas of the automated guided vehicle, and the performance deviation of each monitored detection point is compared with the preset performance threshold to determine whether the automated guided vehicle is in a fault state.

[0160] It should be explained that, taking the fault monitoring of automatic guided vehicles in warehouses as an example, in a warehouse, all automatic guided vehicles are equipped with sensors and data recording systems, which can record various fault data including battery status, motor function, sensor accuracy, etc. The historical fault data is collected and stored in the central database of the factory, and used as subsequent fault monitoring model training samples to calculate the fault frequency distribution of each detection point. For example, if the navigation system of an automatic guided vehicle has 15 faults in six months, and the total running time of the automatic guided vehicle is 1500 hours, then its fault frequency is 0.01 times / hour. The fault data of the automatic guided vehicle is analyzed using the kernel density estimation algorithm to determine the distribution of the data and find the best clustering centers, which represent the most typical fault modes. The training samples are normalized to ensure that all input features are at the same order of magnitude to avoid the algorithm being biased towards features with larger values. The pattern correlation of the normalized data is calculated to determine the similarity of each sample to the best clustering center. The trained fault monitoring model is deployed on the automatic guided vehicle, and new sensor data is received in real time when the model is running. Whenever the data shows that the performance deviation of a certain detection point exceeds the preset performance threshold, the model will trigger an alarm to indicate a possible failure.

[0161] Preferably, deploying a trained fault monitoring model in a fault-prone area of ​​the automated guided vehicle, and comparing the performance deviation of each monitored detection point with a preset performance threshold to determine whether the automated guided vehicle is in a fault state includes:

[0162] Deploy the trained fault monitoring model in the fault-prone areas of the AGV, use the fault monitoring model to read the operating parameters of each detection point of the AGV, and calculate the performance indicators;

[0163] If the calculated performance index is not greater than the preset performance threshold, the automated guided vehicle is judged to be in a normal operating state, and the fault monitoring model is used to continue to read the operating parameters of each detection point of the automated guided vehicle and recalculate the performance index;

[0164] If the performance index is greater than a preset threshold, the automated guided vehicle is judged to be in an abnormal operating state, and a backup fault monitoring model is started for re-monitoring;

[0165] If the monitoring result of the standby fault monitoring model is normal, the judgment process mode is changed, and the fault monitoring model is used to continue to read the operating parameters of each detection point of the automatic guided vehicle and recalculate the performance index;

[0166] If the monitoring result of the backup fault monitoring model also indicates an abnormality, it is determined that a fault has occurred in the fault-prone area, and it is confirmed that the automatic guided vehicle is in a faulty state.

[0167] It should be explained that, taking the fault monitoring of the automatic guided vehicle in the warehouse as an example, in a warehouse, the key operating parameters of the automatic guided vehicle, such as battery status, motor efficiency, sensor accuracy, etc., are regarded as fault-prone areas. In order to monitor these areas, a trained fault monitoring model is deployed on each automatic guided vehicle. This model calculates performance indicators by analyzing these parameters in real time: the model reads the operating parameters of each detection point in real time, and compares the calculated performance indicators with the preset performance threshold. If the performance indicator does not exceed the threshold, the system determines that the automatic guided vehicle is in normal operation, and the model continues to monitor the data. If the performance indicator exceeds the threshold, the system first starts the backup fault monitoring model to perform a secondary monitoring of the automatic guided vehicle to confirm whether there is really an abnormality. If the results of the backup model show that the automatic guided vehicle is operating normally, it means that the operating environment or conditions of the automatic guided vehicle have changed, resulting in inaccurate initial model predictions. The system will readjust the model parameters or retrain the model to adapt to the new operating conditions. If the backup model also indicates an abnormality, the system confirms that the fault-prone area of ​​the automatic guided vehicle has indeed failed. At this time, the system will automatically send a fault report to the maintenance team and may automatically shut down according to the configuration to avoid further damage.

[0168] Preferably, when a fault is detected in a fault-prone area, the fault diagnosis algorithm is used to detect abnormal values ​​of the running state characteristic data of the automatic guided vehicle, and the potential fault data of the current automatic guided vehicle state is identified, including:

[0169] Set the parameters of the fault diagnosis algorithm;

[0170] The initial population of fault diagnosis models is generated by using chaotic sequences (i.e., Tent chaotic mapping), and the initial position and parameters are set for each fault diagnosis model.

[0171] All fault diagnosis models are evaluated and sorted according to fitness, and all fault diagnosis models are classified into different performance levels according to performance, and the optimal fault diagnosis model and parameters are saved;

[0172] Perform Gaussian mutation on the fault diagnosis model with the best performance, compare the fitness values ​​before and after the mutation, and save the fault diagnosis model with the highest fitness value;

[0173] Update the probability threshold of the fault diagnosis model with the highest fitness value, calculate the new position of the fault diagnosis model at different performance levels based on the new parameters, and finally update the position and parameters of all fault diagnosis models;

[0174] Update the environment variables, and if the preset conditions are met, relocate the fault diagnosis model with poor performance to improve it;

[0175] All fault diagnosis models are repeatedly evaluated and sorted according to their fitness, and random reverse learning is performed on the global optimal fault diagnosis model to generate a new optimal fault diagnosis model. If the fitness value of the new optimal fault diagnosis model is greater than the fitness value of the current optimal fault diagnosis model, the new optimal fault diagnosis model is output as the final fault diagnosis model, and the fault data of the current automatic guided vehicle status is output.

[0176] It should be explained that, taking the fault diagnosis of automatic guided vehicles in warehouses as an example, in a warehouse, the parameters of the fault diagnosis algorithm, such as search depth, population size, mutation rate, etc., are set to initialize the population of fault diagnosis models. The selection of parameters is based on historical fault data and expected diagnostic efficiency. The initial population is generated using chaotic sequences, and the initial position and parameters are set for each model. Due to its random nature, chaotic sequences help to cover a wide range of possible solutions. All fault diagnosis models are evaluated and sorted by fitness and classified into different performance levels. The best performing model and its parameters are retained as the benchmark, and Gaussian mutation is performed on the best performing model to explore new possible solutions by adjusting its parameters. The fitness values ​​before and after the mutation are compared, and the model with higher fitness is retained. The probability threshold of the model with the highest fitness is updated, and the new position of the model at different performance levels is calculated based on the new parameters. Finally, the position and parameters of all models are updated, and the changes in environmental variables are monitored. If the preset conditions are met, the repositioning and optimization of the poorly performing model is triggered. Repeat the evaluation and sorting to generate a new optimal model through random reverse learning.

[0177] Specifically, the fault diagnosis algorithm is a chaotic squirrel search algorithm that mixes random reverse learning and Gaussian mutation. The algorithm generates a chaotic initial population through the Tent chaotic mapping initialization strategy, enhances the uniformity of the initial population distribution, and achieves a more efficient search of the solution space. It adopts a nonlinear decreasing predator probability strategy to balance the global search and local development capabilities of the squirrel search algorithm.

[0178] Specifically, the squirrel search algorithm simulates the dynamic foraging behavior of a southern squirrel: in autumn, squirrels glide between different trees in the forest (hickory, oak, and ordinary trees) to find food sources, with hickory trees being the best food source; in winter, due to the reduced leaf coverage and the increased probability of being preyed on, squirrels become less active until the end of winter when they become active again. This process is repeated continuously, forming a squirrel search algorithm that uses a position-greedy selection strategy to continuously retain dominant individuals in the population during the iteration process to improve the algorithm's convergence speed; introducing random reverse learning and Gaussian mutation strategies to increase population diversity while improving the algorithm's ability to escape local optimality.

[0179] Preferably, the formula for generating a new optimal fault diagnosis model by performing random reverse learning on the global optimal fault diagnosis model is:

[0180]

[0181] In the formula, It represents the position of the new optimal fault diagnosis model generated by random reverse learning (i.e., the position of the randomly reversed squirrel in the chaotic squirrel search algorithm);

[0182] It is represented as the position of the current optimal fault diagnosis model (i.e. the current optimal squirrel position);

[0183] lb represents the lower bound of the search space of fault diagnosis model parameters;

[0184] ub represents the upper bound of the search space of fault diagnosis model parameters;

[0185] rand() represents a generated random number.

[0186] Specifically, a random number between [0, 1] is generated.

[0187] Preferably, the fault prediction module 3 includes: a model building module and a model prediction module;

[0188] A model building module, for building a fault prediction model based on the identified fault data of the running state of the automatic guided vehicle;

[0189] The model prediction module is used to use the fault prediction model to predict the operation failure of the automatic guided vehicle at a future moment, and obtain the key factors affecting the operation failure of the automatic guided vehicle.

[0190] Preferably, the fault prediction model is used to predict the operation fault of the automatic guided vehicle at a future time, and the key factors affecting the operation fault of the automatic guided vehicle include:

[0191] Based on the performance indicators, the model optimization algorithm is used to optimize the parameters of the fault prediction model and find the best fault prediction model parameter configuration;

[0192] The parameter configuration is applied to the constructed fault prediction model to make predictions and obtain the key factors affecting the operation failure of the automated guided vehicle.

[0193] Preferably, based on the performance indicators, the parameters of the fault prediction model are optimized using a model optimization algorithm, and finding the best fault prediction model parameter configuration includes:

[0194] Initialize the model optimization algorithm;

[0195] The initial positions of the fault prediction model parameters are determined using the spherical coordinate method, and the first several parameter configurations are selected as the initial solutions according to their fitness values.

[0196] Substituting the obtained fault prediction model parameter values ​​and adjustment factors into the update formula to obtain new fault prediction model parameter positions;

[0197] Check whether the stop condition is met. If so, output the best solution as the best fault prediction model parameter configuration. If not, continue to execute the parameter update loop.

[0198] It needs to be explained that, taking the fault prediction of the automatic guided vehicle in the warehouse as an example, in a warehouse, the model parameters are initialized using optimization algorithms such as genetic algorithms and salp sea squirt algorithms. These parameters may include the learning rate, number of iterations, regularization strength, etc. of the model, and the initial position of the fault prediction model parameters is set using the spherical coordinate method. This method ensures the diversity and extensiveness of the initial solution by mapping the parameter position onto the sphere. According to the fitness evaluation of each set of parameters, the first few sets of parameters with the highest fitness are selected as the initial solution. Then, using the update formula of the model optimization algorithm, the new parameter position is calculated according to the fitness and adjustment factor, and the parameter update cycle is repeated until the stop condition is met (such as reaching the maximum number of iterations or the fitness improvement is no longer significant). Once the stop condition is met, the current best parameter configuration is output as the optimal parameter setting of the fault prediction model, and the optimized parameter configuration is applied to the constructed fault prediction model. The model predicts future faults by analyzing the historical fault data and current operating status of the automatic guided vehicle. The model can identify key factors such as low battery, sensor failure, and navigation system errors, which cause the automatic guided vehicle to fail in the future.

[0199] Specifically, the model optimization algorithm is an adaptive salp algorithm. The traditional salp algorithm randomly generates the initial position of the population, which makes the algorithm easy to fall into the local optimum and has poor convergence speed and accuracy. Therefore, the quantum bit Bloch spherical coordinate encoding method is used to optimize the initial population position, and the inertia weight is used to balance the global and local search capabilities of the traditional salp algorithm to improve the algorithm performance so that it can converge faster.

[0200] The traditional salp algorithm is a new optimization algorithm proposed based on the hunting behavior of marine organisms salps. It often moves in the form of a chain. The entire population consists of leaders and followers. The leader is at the front of the chain, and the rest are followers. During the entire foraging process, the leader determines the foraging direction, and the followers only move following the leader.

[0201] According to another embodiment of the present invention, Figure 2 As shown, a method for AGV fault diagnosis and prediction based on data analysis is also provided, and the method for AGV fault diagnosis and prediction based on data analysis includes the following steps:

[0202] S1. Acquire real-time operation status data of the automatic guided vehicle, and pre-process the real-time operation status data of the automatic guided vehicle to obtain operation status characteristic data of the automatic guided vehicle;

[0203] S2. Analyze the obtained running state characteristic data of the automatic guided vehicle using a fault diagnosis algorithm to identify fault data of the current running state of the automatic guided vehicle;

[0204] S3. Based on the identified 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 a future moment is predicted by using the fault prediction model to obtain key factors affecting the running fault of the automatic guided vehicle;

[0205] S4. Adjust the operating parameters of the automatic guided vehicle based on the obtained key factors affecting the automatic guided vehicle operating failure.

[0206] In summary, with the help of the above technical solutions of the present invention, the present invention can quickly identify and respond to faults through real-time monitoring and data-driven fault diagnosis, reduce system downtime and maintenance costs, and realize the transformation from post-processing to preventive maintenance through in-depth analysis and prediction of fault data, discover problems in advance and intervene, avoid greater equipment damage and production interruption, and through accurate fault diagnosis models, enterprises can more effectively allocate maintenance resources, optimize maintenance plans and budgets, and improve overall operational efficiency. Continuous monitoring and timely maintenance can avoid major losses caused by faults, extend the service life of equipment, and improve the return on investment of assets, thereby reducing safety accidents caused by equipment failures, protecting the safety of operators and related facilities, and maintaining the stability of the production environment. The present invention constructs a fault prediction model based on the identified fault data, so that the prediction is based on actual observation data, thereby improving the accuracy of the prediction. The optimized model can more accurately identify possible faults in the future, so that preventive measures are more targeted and effective. The fault prediction model can analyze and predict future fault conditions in real time, and timely discover the trend and influencing factors of fault development. This enables enterprises to respond quickly, reduce the possibility and scope of fault expansion, reduce maintenance time and cost, and use model optimization algorithms to dynamically adjust the parameters of the fault prediction model according to performance indicators to ensure that the model always operates in the best state. This dynamic adjustment helps the model adapt to changes in the operating environment, improves the generalization ability of the model and the flexibility of application. Through timely fault prediction and corresponding maintenance, intervention can be carried out before the fault causes serious damage, thereby extending the service life of the equipment and improving the overall efficiency of the equipment.

[0207] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An AGV fault diagnosis and prediction system based on data analysis, characterized in that: The AGV fault diagnosis and prediction system based on data analysis includes: a data management module, a fault diagnosis module, a fault prediction module and an adjustment control module; A data management module is used to obtain real-time operation status data of the automatic guided vehicle and pre-process the real-time operation status data of the automatic guided vehicle to obtain operation status characteristic data of the automatic guided vehicle; A fault diagnosis module is used to analyze the obtained running state characteristic data of the automatic guided vehicle using a fault diagnosis algorithm to identify fault data of the current running state of the automatic guided vehicle; A fault prediction module is used to establish a fault prediction model based on the fault data of the operating state of the automatic guided vehicle identified, and use the fault prediction model to predict the operating faults of the automatic guided vehicle at a future moment, and obtain key factors affecting the operating faults of the automatic guided vehicle; An adjustment control module, used for adjusting the operation parameters of the automatic guided vehicle based on the obtained key factors affecting the operation failure of the automatic guided vehicle; The method of analyzing the obtained running state characteristic data of the automatic guided vehicle by using a fault diagnosis algorithm to identify the fault data of the current running state of the automatic guided vehicle includes: Collect historical operating status data of automated guided vehicles; Calculate the number of failures at each detection point within a preset time period, and divide the number of failures within the preset time period by the total operating time of each detection point to obtain the failure frequency distribution of each detection point; Analyze the fault frequency distribution of each detection point, draw the fault frequency distribution curve of each detection point by statistical analysis method, and identify the area with higher fault frequency as the prone fault point of the automatic guided vehicle; According to the fault frequency distribution and fault occurrence law of each detection point, a fault monitoring model is constructed and deployed on the automatic guided vehicle; When a fault is detected in a fault-prone area, a fault diagnosis algorithm is used to detect abnormal values ​​of the running status characteristic data of the automatic guided vehicle, and identify potential fault data of the current automatic guided vehicle status.

2. The AGV fault diagnosis and prediction system based on data analysis according to claim 1 is characterized in that: The data management module includes: a data acquisition module and a data processing module; The data acquisition module is used to acquire the real-time operation status data of the automatic guided vehicle through sensing technology; The data processing module is used to pre-process the acquired real-time running status data of the automatic guided vehicle to obtain the running status characteristic data of the automatic guided vehicle.

3. The AGV fault diagnosis and prediction system based on data analysis according to claim 1 is characterized in that: The method of constructing a fault monitoring model according to the fault frequency distribution and fault occurrence rule of each detection point, and deploying the fault monitoring model on the automatic guided vehicle includes: Collect historical fault data of the AGV and use the historical fault data as training samples for the fault monitoring model; The fault frequency distribution and fault occurrence law of each detection point are used as input parameters of the fault monitoring model; Use the estimation algorithm to perform outlier analysis on the fault data in the training samples, divide the fault data into normal operation area and abnormal operation area, and output the optimal grouping center matrix; Normalize the training samples, calculate the pattern correlation of the normalized training samples, use the optimal grouping center matrix and performance threshold as the input of the fault monitoring model, and build fault monitoring models of different modes in combination with the pattern correlation; The trained fault monitoring model is deployed in the fault-prone areas of the automated guided vehicle, and the performance deviation of each monitored detection point is compared with the preset performance threshold to determine whether the automated guided vehicle is in a fault state.

4. The AGV fault diagnosis and prediction system based on data analysis according to claim 3 is characterized in that: The method of deploying a trained fault monitoring model in a fault-prone area of ​​the automatic guided vehicle and comparing the performance deviation of each detected detection point with a preset performance threshold to determine whether the automatic guided vehicle is in a fault state includes: Deploy the trained fault monitoring model in the fault-prone areas of the AGV, use the fault monitoring model to read the operating parameters of each detection point of the AGV, and calculate the performance indicators; If the calculated performance index is not greater than the preset performance threshold, the automated guided vehicle is judged to be in a normal operating state, and the fault monitoring model is used to continue to read the operating parameters of each detection point of the automated guided vehicle and recalculate the performance index; If the performance index is greater than a preset threshold, the automated guided vehicle is judged to be in an abnormal operating state, and a backup fault monitoring model is started for re-monitoring; If the monitoring result of the standby fault monitoring model is normal, the judgment process mode is changed, and the fault monitoring model is used to continue to read the operating parameters of each detection point of the automatic guided vehicle and recalculate the performance index; If the monitoring result of the backup fault monitoring model also indicates an abnormality, it is determined that a fault has occurred in the fault-prone area, and it is confirmed that the automatic guided vehicle is in a faulty state.

5. The AGV fault diagnosis and prediction system based on data analysis according to claim 4 is characterized in that: When a fault is detected in a fault-prone area, the fault diagnosis algorithm is used to detect abnormal values ​​of the running state characteristic data of the automatic guided vehicle, and the potential fault data of the current automatic guided vehicle state is identified, including: Set the parameters of the fault diagnosis algorithm; Generate the initial population of fault diagnosis models using chaotic sequences, and set the initial position and parameters for each fault diagnosis model; All fault diagnosis models are evaluated and sorted according to fitness, and all fault diagnosis models are classified into different performance levels according to performance, and the optimal fault diagnosis model and parameters are saved; Perform Gaussian mutation on the fault diagnosis model with the best performance, compare the fitness values ​​before and after the mutation, and save the fault diagnosis model with the highest fitness value; Update the probability threshold of the fault diagnosis model with the highest fitness value, calculate the new position of the fault diagnosis model at different performance levels based on the new parameters, and finally update the position and parameters of all fault diagnosis models; Update the environment variables, and if the preset conditions are met, relocate the fault diagnosis model with poor performance to improve it; All fault diagnosis models are repeatedly evaluated and sorted according to their fitness, and random reverse learning is performed on the global optimal fault diagnosis model to generate a new optimal fault diagnosis model. If the fitness value of the new optimal fault diagnosis model is greater than the fitness value of the current optimal fault diagnosis model, the new optimal fault diagnosis model is output as the final fault diagnosis model, and the fault data of the current automatic guided vehicle status is output.

6. The AGV fault diagnosis and prediction system based on data analysis according to claim 5 is characterized in that: The formula for generating a new optimal fault diagnosis model by performing random reverse learning on the global optimal fault diagnosis model is: In the formula, represents the position of the new optimal fault diagnosis model generated by random reverse learning; Represents the position of the current optimal fault diagnosis model; lb represents the lower bound of the search space of fault diagnosis model parameters; ub represents the upper bound of the search space of fault diagnosis model parameters; rand() represents a generated random number.

7. The AGV fault diagnosis and prediction system based on data analysis according to claim 1 is characterized in that: The fault prediction module includes: a model building module and a model prediction module; The model building module is used to build a fault prediction model based on the identified fault data of the running state of the automatic guided vehicle; The model prediction module is used to predict the operation failure of the automatic guided vehicle at a future moment by using the fault prediction model, and obtain the key factors affecting the operation failure of the automatic guided vehicle.

8. The AGV fault diagnosis and prediction system based on data analysis according to claim 7 is characterized in that: The fault prediction model is used to predict the operation faults of the automatic guided vehicle at a future time, and the key factors affecting the operation faults of the automatic guided vehicle include: Based on the performance indicators, the model optimization algorithm is used to optimize the parameters of the fault prediction model and find the best fault prediction model parameter configuration; The parameter configuration is applied to the constructed fault prediction model to make predictions and obtain the key factors affecting the operation failure of the automated guided vehicle.

9. The AGV fault diagnosis and prediction system based on data analysis according to claim 8, characterized in that: The method of optimizing the parameters of the fault prediction model based on the performance index and finding the best fault prediction model parameter configuration includes: Initialize the model optimization algorithm; The initial positions of the fault prediction model parameters are determined using the spherical coordinate method, and the first several parameter configurations are selected as the initial solutions according to their fitness values. Substituting the obtained fault prediction model parameter values ​​and adjustment factors into the update formula to obtain new fault prediction model parameter positions; Check whether the stop condition is met. If so, output the best solution as the best fault prediction model parameter configuration. If not, continue to execute the parameter update loop.

10. An AGV fault diagnosis and prediction method based on data analysis, used to implement the AGV fault diagnosis and prediction system based on data analysis as described in any one of claims 1 to 9, characterized in that: The AGV fault diagnosis and prediction method based on data analysis includes the following steps: S1. Acquire real-time operation status data of the automatic guided vehicle, and pre-process the real-time operation status data of the automatic guided vehicle to obtain operation status characteristic data of the automatic guided vehicle; S2. Analyze the obtained running state characteristic data of the automatic guided vehicle using a fault diagnosis algorithm to identify fault data of the current running state of the automatic guided vehicle; S3. Based on the identified 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 a future moment is predicted by using the fault prediction model to obtain key factors affecting the running fault of the automatic guided vehicle; S4. Adjust the operating parameters of the automatic guided vehicle based on the obtained key factors affecting the automatic guided vehicle operating failure.

Images