A real-time fault monitoring and early warning system for container stackers
By arranging sensor network and central processing units on the container stacker to analyze data, identify faults and generate early warning instructions, the problem of inaccurate fault monitoring in the existing technology is solved, real-time fault monitoring and early warning are achieved, and the reliability of the equipment is improved. and security.
Patent Information
- Application Number
- CN202411258887.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-10
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-09-10
AI Technical Summary
The existing container recharger fault monitoring lacks comprehensive monitoring scope and data support, resulting in the inability to accurately judge the type, location and severity of the fault, resulting in false alarm information or missed reports, affecting the normal operation and maintenance of the equipment.
By laying a sensor network on the container stacker, real-time collection of operation data, and using the central processing unit to perform data analysis and prediction, generating an operation trend curve chart, identifying mismatched points sets, backtracking to the sensor network for fault location, and generating early warning instructions. Implement intelligent fault warning.
Real-time fault monitoring and early warning are realized, the false alarm rate and missed rate are reduced, the equipment is handled in a timely manner when there is a fault, and the equipment operation reliability and safety is improved.
Smart Images

Figure CN118797293B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to data processing related fields, and in particular to a real-time fault monitoring and early warning system for container stackers. Background Art
[0002] Container stackers are indispensable loading and unloading equipment in large logistics facilities such as ports, docks and logistics centers. They are mainly used to stack, transport and move containers. With the rapid development of the logistics industry, the frequency of use and load capacity of container stackers are increasing, making the safety and reliability of their operation particularly important. Traditional container stacker maintenance and troubleshooting mainly rely on manual inspections and regular repairs, which is not only inefficient, but also difficult to detect potential faults in a timely manner, which may lead to equipment downtime, production delays and even safety accidents.
[0003] In the current relevant technologies, container forklift fault monitoring is limited to the monitoring of specific components or systems. The lack of comprehensive monitoring scope and data support leads to the inability to accurately determine the type, location and severity of the fault, resulting in false alarms or missed alarms, affecting the normal operation and maintenance of the equipment. Summary of the invention
[0004] The present application provides a real-time fault monitoring and early warning system for container stackers, which collects and analyzes data of container stackers in real time, promptly discovers potential faults or abnormal conditions, processes and analyzes potential faults or abnormal conditions using algorithms, and identifies fault modes or abnormal conditions, thereby achieving the technical effect of realizing real-time fault monitoring and early warning, reducing false alarm and missed alarm rates, ensuring that equipment is handled in a timely manner when a fault occurs, avoiding equipment downtime or production delays, and improving the reliability and safety of equipment operation.
[0005] The present application provides a real-time fault monitoring method for a container stacker, comprising:
[0006] The structural data of the container stacker are traversed to identify key points, a sensor network is deployed according to the identified points, and the container stacker is sensed in real time through the sensor network to obtain a real-time operation data set, and the real-time operation data set corresponds to the sensor network; the real-time operation data set is transmitted to a central processing unit, and the operation prediction of the container stacker is performed based on the central processing unit, and an operation trend curve diagram of the container stacker is drawn; the operation trend curve diagram is covered and traversed in combination with the expected operation trend curve diagram to generate a mismatched curvature coordinate point set; the mismatched curvature coordinate point set is compared with the real-time operation data set, and the sensor network is traced back to locate the fault according to the comparison result to obtain multiple fault location data; continuous monitoring is performed based on the multiple fault location data, and the monitoring results are synchronized to the early warning unit to generate an early warning instruction, and the early warning instruction is executed to perform intelligent fault early warning on the container stacker.
[0007] In a possible implementation, the structural data of the container stacker is traversed to identify key points, a sensor network is deployed according to the identified points, the container stacker is sensed in real time through the sensor network, a real-time operation data set is obtained, and the following processing is performed:
[0008] Acquire monitoring requirement information of identification points on a container stacker; determine a sensor information set based on the monitoring requirement information, wherein the sensor information set includes sensor type information and sensor quantity information; deploy multiple sensors based on the identification points according to the sensor type information and the sensor quantity information, and configure the sensor network, wherein the sensor network includes sampling frequencies of multiple sensors; perform data sensing on the container stacker through the sensor network according to the sampling frequency, and determine the real-time operation data set.
[0009] In a possible implementation, the real-time operation data set is transmitted to a central processing unit, an operation prediction of the container stacker is performed based on the central processing unit, an operation trend curve of the container stacker is drawn, and the following processing is performed:
[0010] Feature extraction is performed based on the real-time operation data set to obtain multiple operation features; data is divided into the real-time operation data set according to the multiple operation features to generate a training set, a supervision set, and a test set; a loss function is introduced, and an operation prediction model is trained using the training set and the supervision set; the operation prediction model is verified and evaluated based on the test set, and the operation prediction model is output according to the evaluation result; the operation prediction model is added to the central processing unit, and the operation prediction of the container stacker is performed through the operation prediction model to generate an operation prediction data set; the operation prediction data set is synchronized to the operation trend coordinate system, and the operation trend curve chart is drawn.
[0011] In a possible implementation, the operation trend curve graph is covered and traversed in combination with the expected operation trend curve graph to generate a set of unmatched curvature coordinate points, and the following processing is performed:
[0012] Retrieve a historical operation data set of a container stacker from big data to determine a plurality of expected operation data; generate an expected operation trend curve graph based on the plurality of expected operation data; set a traversal time step, align the expected operation trend curve graph with the operation trend curve graph, and perform an overlay traversal according to the traversal time step based on the alignment result to generate a traversal curvature coordinate point set, wherein the traversal curvature coordinate point set includes a plurality of expected coordinate points and a plurality of actual coordinate points, and each expected coordinate point in the plurality of expected coordinate points corresponds to an actual coordinate point in the plurality of actual coordinate points; determine whether a difference value between a first expected coordinate point and a first actual coordinate point is greater than a preset difference threshold based on the traversal curvature coordinate point set; when the difference value between the first expected coordinate point and the first actual coordinate point is greater than the preset difference threshold, it is deemed that the curvature at the time point corresponding to the first expected coordinate point and the first real-time coordinate point does not match, and add the unmatched time point, the first expected coordinate point, and the first real-time coordinate point to the unmatched curvature coordinate point set.
[0013] In a possible implementation, based on comparing the unmatched curvature coordinate point set with the real-time running data set, the following processing is performed:
[0014] The unmatched curvature coordinate point set is used as index data to match the real-time operation data set, and multiple real-time operation data are extracted according to the matching result, and there is a one-to-one correspondence between the multiple real-time operation data and the unmatched curvature coordinate point set; whether each of the multiple real-time operation data is in a preset operation threshold range is determined in turn; the real-time operation data set that is not in the preset operation threshold range is extracted, and abnormal labels are marked for the real-time operation data set to generate abnormal label data; and the abnormal label data is added to the comparison result.
[0015] In a possible implementation, the sensor network is traced back to locate the fault according to the comparison result, a plurality of fault location data are obtained, and the following processing is performed:
[0016] Based on the abnormal label data, the comparison result is traced back to the sensor network to determine multiple abnormal data sources, and the multiple abnormal data sources correspond to N sensors, where N is an integer greater than or equal to 1; the N sensors are analyzed in combination with the abnormal data of the multiple abnormal data sources to generate a sensor abnormality information set, and the sensor abnormality information set includes abnormal sensor type information and abnormal sensor location information; multiple fault areas are delineated according to the abnormal sensor type information and the abnormal sensor location information; fault points are identified based on the multiple fault areas to generate the multiple fault location data.
[0017] In a possible implementation, continuous monitoring is performed based on the multiple fault location data, the monitoring results are synchronized to the early warning unit to generate an early warning instruction, and the following processing is performed:
[0018] A monitoring cycle is set based on the operating time of the container stacker; based on the multiple fault location data, the container stacker is continuously monitored according to the monitoring cycle to generate the monitoring result; the monitoring result is analyzed to obtain fault risk information, and multiple levels of warning thresholds are set based on the fault risk information; the warning unit is triggered based on the monitoring result, and the fault risk information is matched with the multiple levels of warning thresholds by the warning unit to generate warning information of a target level; and the warning information of the target level is added to the warning instruction.
[0019] The present application also provides a real-time fault warning system for a container stacker, comprising:
[0020] A real-time operation data set acquisition module, the real-time operation data set acquisition module is used to traverse the structural data of the container stacker to identify key points, deploy a sensor network according to the identified points, and perform real-time sensing of the container stacker through the sensor network to obtain a real-time operation data set, and there is a corresponding relationship between the real-time operation data set and the sensor network; an operation prediction module, the operation prediction module is used to transmit the real-time operation data set to a central processing unit, perform operation prediction of the container stacker based on the central processing unit, and draw an operation trend curve of the container stacker; a mismatched curvature coordinate point set generation module, the mismatched curvature coordinate point set generation module The matching curvature coordinate point set generation module is used to cover and traverse the operation trend curve diagram in combination with the expected operation trend curve diagram to generate an unmatched curvature coordinate point set; the fault location module is used to compare the unmatched curvature coordinate point set with the real-time operation data set, and trace back to the sensor network to locate the fault according to the comparison result to obtain multiple fault location data; the intelligent fault warning module is used to perform continuous monitoring based on the multiple fault location data, synchronize the monitoring results to the warning unit to generate a warning instruction, and execute the warning instruction to perform intelligent fault warning on the container stacker.
[0021] A real-time fault monitoring and early warning system for container stackers proposed in this application first traverses the structural data of the container stacker to identify key points, deploys a sensor network according to the identified points, performs real-time sensing of the container stacker through the sensor network, obtains a real-time operation data set, and there is a corresponding relationship between the real-time operation data set and the sensor network. Then, the real-time operation data set is transmitted to the central processing unit, and the operation of the container stacker is predicted based on the central processing unit. The operation trend curve of the container stacker is drawn, and then the operation trend curve is covered and traversed in combination with the expected operation trend curve. The mismatched curvature coordinate point set is formed, and then compared with the real-time operation data set based on the mismatched curvature coordinate point set. According to the comparison result, the fault is located back to the sensor network to obtain multiple fault location data. Finally, continuous monitoring is performed based on the multiple fault location data, and the monitoring results are synchronized to the early warning unit to generate early warning instructions. The early warning instructions are executed to perform intelligent fault early warning for the container stacker, thereby achieving the technical effect of realizing real-time fault monitoring and early warning, reducing false alarm rate and missed alarm rate, ensuring that the equipment is handled in time when a fault occurs, avoiding equipment shutdown or production delays, and improving the reliability and safety of equipment operation. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.
[0023] Figure 1 A flowchart of a real-time fault monitoring method for a container stacker provided in an embodiment of the present application.
[0024] Figure 2 A structural schematic diagram of a real-time fault warning system for a container stacker provided in an embodiment of the present application.
[0025] Explanation of the reference numerals: real-time operation data set acquisition module 10 , operation prediction module 20 , mismatched curvature coordinate point set generation module 30 , fault location module 40 , intelligent fault warning module 50 . DETAILED DESCRIPTION
[0026] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.
[0027] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.
[0028] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.
[0029] The present application embodiment provides a real-time fault monitoring method for a container stacker, and the method is applied to a real-time fault warning system for a container stacker, such as Figure 1 As shown, the method includes:
[0030] Step S100, traverse the structural data of the container stacker to identify key points, deploy a sensor network according to the identified points, perform real-time sensing on the container stacker through the sensor network, and obtain a real-time operation data set, and the real-time operation data set has a corresponding relationship with the sensor network. Specifically, collect detailed structural data of the container stacker, including design drawings, mechanical parts list, working principle description, etc., and analyze which parts of the container stacker are prone to failure or wear during operation based on the collected structural data. These parts are key points, such as the engine, hydraulic system, lifting mechanism, steering system, etc., and use specific marks or numbers to identify the key points. According to the characteristics of the key points and monitoring requirements, determine the appropriate sensor type. For example, for temperature monitoring, a temperature sensor can be selected; for pressure monitoring, a pressure sensor can be selected; for vibration monitoring, a vibration sensor can be selected. Install sensors at the identified key points, connect all sensors into a network through wired or wireless methods, ensure smooth communication between sensors, and be able to transmit data to the central processing unit in real time. After the sensor network is deployed, the container forklift is sensed in real time through the sensor network. The sensors continuously collect real-time data of key points to form a real-time operation data set. The real-time operation data set contains the real-time data of each key point and has a clear corresponding relationship with the sensor network.
[0031] In a possible implementation, step S100 further includes step S110, obtaining monitoring demand information of identification points on the container stacker. Specifically, identify key points on the container stacker, such as cargo lifting height, fork position, driving speed, etc., communicate with the stacker operator or maintenance personnel to determine the specific parameters and performance indicators that need to be monitored, collect and organize these monitoring demand information, and form a demand list. Step S120, determine a sensor information set in combination with the monitoring demand information, and the sensor information set includes sensor type information and sensor quantity information. Specifically, analyze the monitoring demand information, determine the most suitable sensor type for each parameter or performance indicator, determine the number of each sensor required according to the structure and operating environment of the container stacker, integrate this information, and form a sensor information set containing sensor type information and sensor quantity information. Step S130, based on the identification point, multiple sensors are deployed according to the sensor type information and the sensor quantity information, and the sensor network is configured, and the sensor network includes the sampling frequency of multiple sensors. Specifically, according to the sensor information set, determine the appropriate sensor, install the sensor at the identification point of the container stacker, ensure that the installation position is accurate and stable, and configure the sensor network, including setting parameters such as sampling frequency and data transmission method. Step S140, perform data sensing on the container stacker according to the sampling frequency through the sensor network to determine the real-time operation data set. Specifically, start the sensor network to start data collection according to the set sampling frequency, process and analyze the collected data, and form a real-time operation data set. This implementation method determines the most suitable sensor type and quantity according to actual monitoring needs, ensures the accuracy and effectiveness of the monitoring data, and achieves the technical effect of ensuring that the operating status of the container stacker is monitored comprehensively and accurately.
[0032] Step S200, the real-time operation data set is transmitted to the central processing unit, and the operation prediction of the container stacker is performed based on the central processing unit, and the operation trend curve of the container stacker is drawn. Specifically, the real-time operation data set collected by the sensor network is transmitted to the central processing unit according to a certain communication protocol (such as TCP / IP, Zigbee, etc.), wherein the central processing unit is a core unit responsible for receiving, processing and storing the real-time operation data set transmitted by the sensor network, and performing operation prediction and data analysis. The central processing unit preprocesses the received real-time operation data set, including data cleaning (removing noise, outliers, etc.), data standardization (unifying data units, ranges, etc.) and feature extraction (extracting key features related to operation prediction) and other steps. According to the operation characteristics and prediction requirements of the container stacker, the preprocessed real-time operation data set is analyzed using a data analysis algorithm (such as machine learning, neural network, etc.), the operation trend of the container stacker is predicted, the prediction result is output, and the prediction result is displayed in a graphical manner, that is, the operation trend curve of the container stacker is drawn, and the operation trend curve shows the change trend and abnormal situation of the container stacker during operation.
[0033] In a possible implementation, step S200 further includes step S210, performing feature extraction based on the real-time operation data set to obtain multiple operation features. Specifically, analyzing each data item in the real-time operation data set, determining key features related to the operation of the container stacker (various parameters or indicators describing the operation status of the container stacker), using feature extraction algorithms (such as PCA, LDA, etc.) or manual selection methods, extracting key features from the real-time operation data set, and sorting the extracted features to form feature vectors or feature matrices. Step S220, dividing the real-time operation data set according to the multiple operation features to generate a training set, a supervision set, and a test set. Specifically, according to the importance and relevance of the features, the real-time operation data set is divided into different subsets, most of the data is used as a training set for model training; a part of the data is used as a supervision set for model tuning during the training process; and the remaining data is used as a test set for evaluating the performance of the model. During the division process, the data distribution of each subset is made as similar as possible to the original data set to maintain the generalization ability of the model. Step S230, introducing a loss function, and using the training set and the supervision set to train the operation prediction model. Specifically, determine the machine learning algorithm (such as neural network, support vector machine, etc.) to build the operation prediction model, introduce the loss function (such as mean square error, cross entropy, etc.) to measure the difference between the predicted value of the operation prediction model and the true value, use the training set to train the operation prediction model, and adjust the operation prediction model parameters through iterative optimization algorithms (such as gradient descent) to minimize the loss function. During the training process, use the supervision set to verify the operation prediction model to avoid overfitting or underfitting.
[0034] Step S240, verify and evaluate the operation prediction model based on the test set, and output the operation prediction model according to the evaluation result. Specifically, use the test set to evaluate the trained operation prediction model, calculate the accuracy, recall rate, F1 value and other indicators of the operation prediction model, and judge whether the performance of the operation prediction model meets the requirements according to the evaluation results. If not, adjust the operation prediction model structure or parameters and retrain. If the performance of the operation prediction model meets the requirements, output it as an available operation prediction model. Step S250, add the operation prediction model to the central processing unit, use the operation prediction model to perform operation prediction on the container stacker, and generate an operation prediction data set. Specifically, deploy the trained operation prediction model to the central processing unit, and when there is new real-time operation data, input it into the operation prediction model for prediction, output the prediction result, and form an operation prediction data set. Step S260, synchronize the operation prediction data set to the operation trend coordinate system and draw the operation trend curve chart. Specifically, a two-dimensional coordinate system is constructed to display the operation trend of the container stacker, for example, an operation trend coordinate system with operation data as the vertical axis and operation time as the horizontal axis, and the data points in the operation prediction data set are mapped according to the definition of the operation trend coordinate system, and a graphics drawing tool or library (such as matplotlib, plotly, etc.) is used to draw an operation trend curve chart. This implementation method predicts and analyzes the operation of the container stacker through machine learning technology, achieving the technical effect of improving the accuracy of the prediction.
[0035] Step S300, the operation trend curve is covered and traversed in combination with the expected operation trend curve to generate a set of mismatched curvature coordinate points. Specifically, historical data of the container stacker under normal operating conditions are collected. These data are accumulated through long-term stable operation. Trend analysis is performed on the collected historical data using data analysis tools to find out the operating trend under normal operating conditions. Based on the normal operating trend, an expected operation trend curve is drawn. The expected operation trend curve represents the operating trend of the container stacker under ideal or standard conditions. The real-time drawn operation trend curve is aligned with the expected operation trend curve on the time axis, the operation trend curve is covered with the expected operation trend curve, and the curvature of the two is compared point by point or section by section. During the comparison process, areas with significant differences or anomalies between the operation trend curve and the expected operation trend curve are identified. These differences or anomalies are manifested as sudden changes in curvature, deviations from the normal range, etc. The identified differences or abnormal areas are converted into specific coordinate point sets, which represent the mismatched parts between the operating trend curve and the expected operating trend curve. The extracted coordinate point sets are sorted, including removing duplicate points, merging similar points, etc., to generate mismatched curvature coordinate point sets.
[0036] In a possible implementation, step S300 further includes step S310, retrieving the historical operation data set of the container stacker in the big data, and determining a plurality of expected operation data. Specifically, the past operation data of the container stacker is retrieved from the big data storage system, including various indicators such as speed, acceleration, position, load, energy consumption, etc., and a plurality of expected operation data points are determined by analyzing these data. These expected data points represent the operating state that the container stacker should reach under normal circumstances. Step S320, generating an expected operation trend curve chart based on the plurality of expected operation data. Specifically, using a data analysis tool or visualization software, the expected operation data is plotted into a trend curve chart, and the expected operation trend curve chart shows the operating state of the container stacker that changes over time under expected conditions. Step S330, set the traversal time step, align the expected operation trend curve graph with the operation trend curve graph, perform overlay traversal according to the alignment result and the traversal time step, generate a traversal curvature coordinate point set, the traversal curvature coordinate point set includes multiple expected coordinate points and multiple actual coordinate points, and each expected coordinate point in the multiple expected coordinate points corresponds to an actual coordinate point in the multiple actual coordinate points. Specifically, set a traversal time step, that is, during the traversal process, at each forward time interval, align the expected operation trend curve graph with the current or actual operation trend curve graph on the time axis, take corresponding coordinate points on the two curves according to the set traversal time step, and form a traversal curvature coordinate point set including multiple expected coordinate points and multiple actual coordinate points, that is, the traversal curvature coordinate point set is a set of coordinate points including corresponding time points on the expected operation trend curve graph and the actual operation trend curve graph.
[0037] Step S340, based on the traversed curvature coordinate point set, determine whether the difference between the first expected coordinate point and the first actual coordinate point is greater than a preset difference threshold. Specifically, take out a first pair of expected coordinate points and actual coordinate points from the traversed curvature coordinate point set, the first pair of expected coordinate points and actual coordinate points is any pair in the traversed curvature coordinate point set, calculate the difference between the two coordinate points (such as the distance, speed difference, acceleration difference, etc. of the coordinate points), and determine whether the difference exceeds the preset difference threshold. The preset difference threshold is a pre-set standard value used to determine whether the difference between the expected value and the actual value reaches a level that requires attention. Step S350, when the difference between the first expected coordinate point and the first actual coordinate point is greater than the preset difference threshold, it is considered that the curvature at the time point corresponding to the first expected coordinate point and the first real-time coordinate point does not match, and the unmatched time point, the first expected coordinate point, and the first real-time coordinate point are added to the unmatched curvature coordinate point set. Specifically, if the judged difference value exceeds the preset difference threshold, it is considered that at this time point, the actual operating state of the container stacker does not match the expected state. At this time, the information of this time point, the first expected coordinate point and the first actual coordinate point are added to the set of mismatched curvature coordinate point sets. The mismatched curvature coordinate point set is a set that includes all coordinate points and their corresponding time points that are found in the comparison process that do not match the actual and expected operating states. This implementation method accurately identifies the deviation between the actual operating state and the expected state of the container stacker by calculating the difference value between the first expected coordinate point and the first actual coordinate point, thereby achieving the technical effect of improving the accuracy of generating mismatched curvature coordinate point sets.
[0038] Step S400, based on the comparison between the unmatched curvature coordinate point set and the real-time operation data set, the sensor network is traced back to locate the fault according to the comparison result, and multiple fault location data are obtained. Specifically, the unmatched curvature coordinate point set is synchronized with the real-time operation data set in time series, the unmatched curvature coordinate point set is mapped to the corresponding time point of the real-time operation data set, the specific operation data corresponding to these coordinate points are found, and the operation data corresponding to the unmatched curvature coordinate points are deeply analyzed to find the parameters or data patterns that cause the curvature abnormality. According to the parameters or data patterns associated with the unmatched curvature coordinate points in the real-time operation data set, the corresponding sensor data is located, the sensor data is detected for abnormality, abnormal values, mutation points or data beyond the normal range are found, and the fault source or fault type is identified according to the abnormal situation of the sensor data and the structural characteristics of the container stacker. The sensor data related to the fault source is collected to form multiple fault location data, including information such as sensor location, fault type, and fault time.
[0039] In a possible implementation, based on the comparison between the unmatched curvature coordinate point set and the real-time operation data set, step S400 further includes step S410, using the unmatched curvature coordinate point set as index data, matching with the real-time operation data set, extracting multiple real-time operation data according to the matching result, and the multiple real-time operation data have a one-to-one correspondence with the unmatched curvature coordinate point set. Specifically, using the unmatched curvature coordinate point set as index data, searching the real-time operation data set for the real-time operation data set for the index data, and extracting multiple real-time operation data corresponding to the unmatched curvature coordinate point set. Step S420, sequentially judging whether each of the multiple real-time operation data is within a preset operation threshold interval. Specifically, traversing the extracted multiple real-time operation data, judging whether each real-time operation data is within a preset operation threshold interval one by one, wherein the preset operation threshold interval is a preset range set according to the normal operation parameters of the equipment, and is used to judge whether the real-time operation data is normal. Step S430, extract the real-time operation data set that is not in the preset operation threshold interval, and mark the abnormal labels of the real-time operation data set to generate abnormal label data. Specifically, the real-time operation data set that is not in the preset operation threshold interval is screened out, the abnormal labels of these real-time operation data sets are marked, and abnormal label data is generated. Step S440, add the abnormal label data to the comparison result. Specifically, the generated abnormal label data is added to the comparison result to form a complete comparison result report. This implementation method quantitatively judges the normality of the data by presetting the operation threshold interval, and clearly identifies the abnormal data by marking the abnormal labels, thereby achieving the technical effect of improving the accuracy and efficiency of problem identification.
[0040] In a possible implementation, the sensor network is traced back to locate the fault according to the comparison result to obtain multiple fault location data. Step S400 further includes step S450, tracing back the comparison result to the sensor network based on the abnormal label data to determine multiple abnormal data sources, and the multiple abnormal data sources correspond to N sensors, where N is an integer greater than or equal to 1. Specifically, based on the generated abnormal label data, the comparison result is traced back to the sensor network, and multiple abnormal data sources (specific sensors or sensor groups that cause abnormal data) are determined by comparing the abnormal label data with the real-time data of the sensor network. These abnormal data sources correspond to N sensors. Step S460, the N sensors are analyzed in combination with the abnormal data of the multiple abnormal data sources to generate a sensor abnormal information set, and the sensor abnormal information set includes abnormal sensor type information and abnormal sensor location information. Specifically, the N sensors are analyzed one by one in combination with the abnormal data of the multiple abnormal data sources, including but not limited to the mode, change trend, range, etc. of the sensor data, and a sensor abnormal information set including abnormal sensor type information and abnormal sensor location information is generated according to the analysis results. Step S470, delineate multiple fault areas according to the abnormal sensor type information and the abnormal sensor location information. Specifically, the fault areas are divided according to the abnormal sensor type information and the abnormal sensor location information. These fault areas may be specific physical locations, equipment components or sensor groups. Step S480, identify the fault points based on the multiple fault areas, and generate the multiple fault location data. Specifically, based on the multiple fault areas delineated, identify the specific fault points. The identification information may include the name of the fault point, location description, possible causes, etc., and generate multiple fault location data. This implementation method achieves the technical effect of quickly and accurately finding the source of the fault, thereby improving the efficiency and accuracy of fault handling by accurately identifying the source of abnormal data, analyzing sensor abnormal information, delineating fault areas and identifying fault points.
[0041] Step S500, based on the multiple fault location data, continuous monitoring is performed, the monitoring results are synchronized to the early warning unit to generate an early warning instruction, and the early warning instruction is executed to perform intelligent fault early warning on the container stacker. Specifically, the multiple fault location data obtained are continuously collected, and the fault status information of the container stacker is dynamically updated according to the fault location data collected in real time, and the updated fault status information is transmitted to the early warning unit in real time. After receiving the data, the early warning unit performs data verification, and judges the received fault status information according to the preset early warning conditions and rules to determine whether it is necessary to generate an early warning instruction. When the early warning conditions are met, the early warning unit automatically generates an early warning instruction, which includes detailed information such as the fault type, fault location, and fault degree. The generated early warning instruction is sent to the corresponding execution unit or operator, and the early warning information is displayed through sound and light alarms, display screens, etc. The execution unit or operator takes corresponding processing measures according to the received early warning instructions, such as shutdown inspection, emergency maintenance, etc., to ensure the safe operation of the container stacker. The embodiment of the present application adopts technical means such as real-time data collection and analysis of container stackers, timely discovery of potential faults or abnormal conditions, use of algorithms to process and analyze potential faults or abnormal conditions, and identification of fault modes or abnormal conditions, thereby achieving the technical effect of realizing real-time fault monitoring and early warning, reducing false alarm rates and missed alarm rates, ensuring that equipment is handled in a timely manner when a fault occurs, avoiding equipment downtime or production delays, and improving the reliability and safety of equipment operation.
[0042] In a possible implementation, continuous monitoring is performed based on the multiple fault location data, and the monitoring results are synchronized to the early warning unit to generate an early warning instruction. Step S500 further includes step S510, setting a monitoring cycle based on the operating time of the container stacker. Specifically, according to factors such as the operating time, workload, and maintenance history of the container stacker, a monitoring cycle (the time interval for continuous monitoring of the container stacker) is set. The monitoring cycle can be a fixed time interval or a dynamic time interval, and is adjusted according to the real-time status of the container stacker. Step S520, based on the multiple fault location data, the container stacker is continuously monitored according to the monitoring cycle to generate the monitoring results. Specifically, according to the generated multiple fault location data, the container stacker is continuously monitored, and in each monitoring cycle, the operating data of the container stacker is collected to generate monitoring results. Step S530, the monitoring results are analyzed to obtain fault risk information, and multiple levels of early warning thresholds are set based on the fault risk information. Specifically, the monitoring results are analyzed to identify potential fault risk information, and multiple levels of warning thresholds are set based on historical data, expert experience or machine learning algorithms according to factors such as the severity and impact range of the fault risk information. Step S540: trigger the warning unit based on the monitoring results, and match the fault risk information with the warning thresholds of the multiple levels through the warning unit to generate warning information of the target level. Specifically, when the monitoring results trigger the warning unit, the warning unit matches the fault risk information with the warning thresholds of the multiple levels according to the preset algorithms and rules, and generates warning information of the corresponding level according to the matching results. Step S550: add the warning information of the target level to the warning instruction. Specifically, add the generated warning information of the target level to the warning instruction, and the warning instruction may include information such as the warning level, fault risk description, and recommended treatment measures. This implementation method more accurately judges the severity and urgency of the fault by setting multiple levels of warning thresholds, and achieves the technical effect of improving the accuracy of fault warning.
[0043] In the above, refer to Figure 1 A method for real-time fault monitoring of a container stacker according to an embodiment of the present invention is described in detail. Figure 2 A real-time fault warning system for a container stacker according to an embodiment of the present invention is described.
[0044] A real-time fault warning system for container stackers according to an embodiment of the present invention is used to solve the technical problem that the existing fault monitoring of container stackers is limited to monitoring of specific components or systems, lacks a comprehensive monitoring scope and data support, and cannot accurately determine the type, location and severity of the fault, resulting in false alarms or missed alarms of warning information, affecting the normal operation and maintenance of the equipment, so as to achieve the technical effect of realizing real-time fault monitoring and early warning, reducing the false alarm rate and missed alarm rate, ensuring that the equipment is handled in time when a fault occurs, avoiding equipment downtime or production delays, and improving the reliability and safety of equipment operation. A real-time fault warning system for container stackers includes: a real-time operation data set acquisition module 10, an operation prediction module 20, a mismatched curvature coordinate point set generation module 30, a fault location module 40, and an intelligent fault warning module 50.
[0045] The real-time operation data set acquisition module 10 is used to traverse the structural data of the container stacker to identify key points, deploy a sensor network according to the identified points, and perform real-time sensing of the container stacker through the sensor network to obtain a real-time operation data set, and there is a corresponding relationship between the real-time operation data set and the sensor network; the operation prediction module 20 is used to transmit the real-time operation data set to the central processing unit, perform operation prediction of the container stacker based on the central processing unit, and draw an operation trend curve of the container stacker; the mismatched curvature coordinate point set generation module 30 is used to cover and traverse the operation trend curve in combination with the expected operation trend curve to generate a mismatched curvature coordinate point set; the fault location module 40 is used to compare the mismatched curvature coordinate point set with the real-time operation data set, trace back to the sensor network for fault location according to the comparison result, and obtain multiple fault location data; the intelligent fault warning module 50 is used to perform continuous monitoring based on the multiple fault location data, synchronize the monitoring results to the warning unit to generate a warning instruction, and execute the warning instruction to perform intelligent fault warning on the container stacker.
[0046] The specific configuration of the real-time operation data set acquisition module 10 will be described in detail below. As described above, the structural data of the container stacker is traversed to identify key points, a sensor network is deployed according to the identified points, and the container stacker is sensed in real time through the sensor network to obtain the real-time operation data set. The real-time operation data set acquisition module 10 may further include: a monitoring requirement information acquisition unit for acquiring monitoring requirement information of the identified points on the container stacker; a sensor information set determination unit for determining a sensor information set in combination with the monitoring requirement information, wherein the sensor information set includes sensor type information and sensor quantity information; a sensor network configuration unit for deploying multiple sensors based on the identified points according to the sensor type information and the sensor quantity information, and configuring the sensor network, wherein the sensor network includes sampling frequencies of multiple sensors; and a data sensing unit for sensing data of the container stacker according to the sampling frequency through the sensor network to determine the real-time operation data set.
[0047] The specific configuration of the operation prediction module 20 will be described in detail below. As described above, the real-time operation data set is transmitted to the central processing unit, the operation prediction of the container stacker is performed based on the central processing unit, and the operation trend curve of the container stacker is drawn. The operation prediction module 20 may further include: a feature extraction unit for extracting features based on the real-time operation data set to obtain multiple operation features; a data division unit for dividing the real-time operation data set according to the multiple operation features to generate a training set, a supervision set, and a test set; an operation prediction model training unit for introducing a loss function and training the operation prediction model using the training set and the supervision set; a verification and evaluation unit for verifying and evaluating the operation prediction model based on the test set, and outputting the operation prediction model according to the evaluation result; an operation prediction data set generation unit for adding the operation prediction model to the central processing unit, performing operation prediction on the container stacker through the operation prediction model, and generating an operation prediction data set; and a data set synchronization module, wherein the data set synchronization module is used to synchronize the operation prediction data set to the operation trend coordinate system and draw the operation trend curve.
[0048] The specific configuration of the mismatched curvature coordinate point set generation module 30 will be described in detail below. As described above, the operation trend curve graph is covered and traversed in combination with the expected operation trend curve graph to generate a mismatched curvature coordinate point set. The mismatched curvature coordinate point set generation module 30 may further include: an expected operation data determination unit is used to retrieve a historical operation data set of a container stacker in big data to determine a plurality of expected operation data; an expected operation trend curve graph generation unit is used to generate an expected operation trend curve graph based on the plurality of expected operation data; a covering traversal unit is used to set a traversal time step, align the expected operation trend curve graph with the operation trend curve graph, and cover traverse according to the alignment result according to the traversal time step to generate a traversed curvature coordinate point set, wherein the traversed curvature coordinate point set is generated. The set includes multiple expected coordinate points and multiple actual coordinate points, and each expected coordinate point in the multiple expected coordinate points corresponds to an actual coordinate point in the multiple actual coordinate points; the judgment unit is used to judge whether the difference value between the first expected coordinate point and the first actual coordinate point is greater than a preset difference threshold based on the traversed curvature coordinate point set; the mismatched curvature coordinate point set acquisition unit is used to regard the curvature at the time point corresponding to the first expected coordinate point and the first real-time coordinate point as mismatching when the difference value between the first expected coordinate point and the first actual coordinate point is greater than the preset difference threshold, and add the mismatched time point, the first expected coordinate point, and the first real-time coordinate point to the mismatched curvature coordinate point set.
[0049] The specific configuration of the fault location module 40 will be described in detail below. As described above, based on the comparison between the unmatched curvature coordinate point set and the real-time operation data set, the fault location module 40 may further include: a matching unit for matching the unmatched curvature coordinate point set as index data with the real-time operation data set, and extracting a plurality of real-time operation data according to the matching result, wherein the plurality of real-time operation data and the unmatched curvature coordinate point set have a one-to-one correspondence; a judgment unit for sequentially judging whether each of the plurality of real-time operation data is within a preset operation threshold interval; an abnormal identification unit for extracting a real-time operation data set that is not within the preset operation threshold interval, and identifying an abnormal label for the real-time operation data set to generate abnormal label data; and a comparison result generation unit for adding the abnormal label data to the comparison result.
[0050] Among them, according to the comparison result, the fault is traced back to the sensor network to locate the fault and multiple fault location data are obtained. The fault location module 40 may further include: an abnormal data source determination unit is used to trace the comparison result back to the sensor network based on the abnormal label data, and determine multiple abnormal data sources, and the multiple abnormal data sources have a corresponding relationship with N sensors, N is an integer greater than or equal to 1; a sensor abnormal information set generation unit is used to analyze the N sensors in combination with the abnormal data of the multiple abnormal data sources to generate a sensor abnormal information set, and the sensor abnormal information set includes abnormal sensor type information and abnormal sensor location information; a fault area division unit is used to delineate multiple fault areas according to the abnormal sensor type information and the abnormal sensor location information; a fault point identification unit is used to identify the fault point based on the multiple fault areas and generate the multiple fault location data.
[0051] The specific configuration of the intelligent fault warning module 50 will be described in detail below. As described above, based on the multiple fault location data, continuous monitoring is performed, and the monitoring results are synchronized to the warning unit to generate a warning instruction. The intelligent fault warning module 50 may further include: a monitoring cycle setting unit for setting a monitoring cycle based on the operating time of the container stacker; a continuous monitoring unit for continuously monitoring the container stacker according to the monitoring cycle based on the multiple fault location data to generate the monitoring result; a fault risk information acquisition unit for parsing the monitoring result to obtain fault risk information, and setting multiple levels of warning thresholds based on the fault risk information; a warning information generation unit for triggering the warning unit based on the monitoring result, and matching the fault risk information with the multiple levels of warning thresholds through the warning unit to generate a target level of warning information; and a warning instruction acquisition unit for adding the target level of warning information to the warning instruction.
[0052] A real-time fault warning system for a container stacker provided in an embodiment of the present invention can execute a real-time fault monitoring method for a container stacker provided in any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method.
[0053] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.
[0054] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
Claims
1. A real-time fault monitoring method for a container stacker. It is characterized in that The method is applied to a real-time fault warning system for a container stacker, and the method comprises: Traversing the structural data of the container stacker to identify key points, deploying a sensor network according to the identified points, performing real-time sensing on the container stacker through the sensor network, and obtaining a real-time operation data set, wherein the real-time operation data set has a corresponding relationship with the sensor network; The real-time operation data set is transmitted to a central processing unit, and an operation trend curve graph of the container stacker is drawn based on the operation prediction of the container stacker by the central processing unit; In combination with the expected operation trend curve graph, the operation trend curve graph is covered and traversed to generate a set of unmatched curvature coordinate points; Based on the comparison between the mismatched curvature coordinate point set and the real-time operation data set, the sensor network is traced back to locate the fault according to the comparison result to obtain a plurality of fault location data; Continuously monitoring the multiple fault location data, synchronizing the monitoring results to the early warning unit to generate an early warning instruction, and executing the early warning instruction to perform intelligent fault early warning on the container stacker; The operation trend curve graph is covered and traversed in combination with the expected operation trend curve graph to generate a set of mismatched curvature coordinate points, and the method includes: Retrieve the historical operation data set of the container forklift from the big data and determine multiple expected operation data; generating an expected operation trend curve graph based on the plurality of expected operation data; Setting a traversal time step, aligning the expected operation trend curve graph with the operation trend curve graph, performing overlay traversal according to the traversal time step based on the alignment result, and generating a traversal curvature coordinate point set, wherein the traversal curvature coordinate point set includes a plurality of expected coordinate points and a plurality of actual coordinate points, and each expected coordinate point in the plurality of expected coordinate points corresponds to an actual coordinate point in the plurality of actual coordinate points; Determining whether a difference between a first expected coordinate point and a first actual coordinate point is greater than a preset difference threshold based on the traversed curvature coordinate point set; When the difference between the first expected coordinate point and the first actual coordinate point is greater than a preset difference threshold, it is considered that the curvature at the time point corresponding to the first expected coordinate point and the first real-time coordinate point does not match, and the unmatched time point, the first expected coordinate point, and the first real-time coordinate point are added to the unmatched curvature coordinate point set.
2. A method for real-time fault monitoring of a container stacker according to claim 1, It is characterized in that Traversing the structural data of the container stacker to identify key points, deploying a sensor network according to the identified points, and performing real-time sensing on the container stacker through the sensor network to obtain a real-time operation data set, the method includes: Obtain monitoring demand information of identification points on container stackers; Determine a sensor information set in combination with the monitoring requirement information, wherein the sensor information set includes sensor type information and sensor quantity information; Arrange a plurality of sensors based on the identification points according to the sensor type information and the sensor quantity information, and configure the sensor network, wherein the sensor network includes sampling frequencies of the plurality of sensors; The sensor network performs data sensing on the container stacker according to the sampling frequency to determine the real-time operation data set.
3. A method for real-time fault monitoring of a container stacker according to claim 1, It is characterized in that The real-time operation data set is transmitted to a central processing unit, and an operation prediction of the container stacker is performed based on the central processing unit, and an operation trend curve graph of the container stacker is drawn. The method includes: Perform feature extraction based on the real-time operation data set to obtain multiple operation features; Dividing the real-time operation data set according to the multiple operation characteristics to generate a training set, a supervision set, and a test set; A loss function is introduced, and the running prediction model is trained using the training set and the supervision set; Performing verification and evaluation on the operation prediction model based on the test set, and outputting the operation prediction model according to the evaluation result; Adding the operation prediction model to the central processing unit, performing operation prediction on the container stacker through the operation prediction model, and generating an operation prediction data set; The operation prediction data set is synchronized to the operation trend coordinate system, and the operation trend curve graph is drawn.
4. A method for real-time fault monitoring of a container stacker according to claim 1, It is characterized in that Based on comparing the mismatched curvature coordinate point set with the real-time running data set, the method includes: Using the unmatched curvature coordinate point set as index data to match the real-time operation data set, and extracting a plurality of real-time operation data according to the matching result, wherein the plurality of real-time operation data have a one-to-one correspondence with the unmatched curvature coordinate point set; Determine in sequence whether each of the plurality of real-time operation data is within a preset operation threshold interval; Extracting a real-time operation data set that is not within the preset operation threshold interval, and marking an abnormal label for the real-time operation data set to generate abnormal label data; The abnormal label data is added to the comparison result.
5. A method for real-time fault monitoring of a container stacker according to claim 4, It is characterized in that According to the comparison result, the sensor network is traced back to locate the fault and a plurality of fault location data are obtained. The method includes: Based on the abnormal label data, the comparison result is traced back to the sensor network to determine multiple abnormal data sources, where the multiple abnormal data sources correspond to N sensors, where N is an integer greater than or equal to 1; Analyze the N sensors in combination with the abnormal data from the multiple abnormal data sources to generate a sensor abnormality information set, wherein the sensor abnormality information set includes abnormal sensor type information and abnormal sensor location information; Delineating a plurality of fault areas according to the abnormal sensor type information and the abnormal sensor location information; Fault point identification is performed based on the multiple fault areas to generate the multiple fault location data.
6. A method for real-time fault monitoring of a container stacker according to claim 1, It is characterized in that Continuously monitoring the multiple fault location data and synchronizing the monitoring results to the early warning unit to generate an early warning instruction, the method includes: Set the monitoring cycle based on the operating time of the container handler; Based on the multiple fault location data, the container stacker is continuously monitored according to the monitoring period to generate the monitoring result; Analyzing the monitoring results to obtain fault risk information, and setting multiple levels of warning thresholds based on the fault risk information; The early warning unit is triggered based on the monitoring result, and the early warning unit matches the fault risk information with the early warning thresholds of the multiple levels to generate early warning information of the target level; Add the warning information of the target level to the warning instruction.
7. A real-time fault warning system for container stackers, It is characterized in that The system is used to implement a real-time fault monitoring method for a container stacker according to any one of claims 1 to 6, and the system comprises: A real-time operation data set acquisition module, which is used to traverse the structural data of the container stacker to identify key points, deploy a sensor network according to the identified points, and perform real-time sensing on the container stacker through the sensor network to obtain a real-time operation data set, and there is a corresponding relationship between the real-time operation data set and the sensor network; An operation prediction module, the operation prediction module is used to transmit the real-time operation data set to a central processing unit, perform operation prediction on the container stacker based on the central processing unit, and draw an operation trend curve graph of the container stacker; A mismatched curvature coordinate point set generation module, the mismatched curvature coordinate point set generation module is used to cover and traverse the operation trend curve graph in combination with the expected operation trend curve graph to generate a mismatched curvature coordinate point set; A fault location module, the fault location module is used to compare the unmatched curvature coordinate point set with the real-time operation data set, trace back to the sensor network to locate the fault according to the comparison result, and obtain multiple fault location data; An intelligent fault warning module is used to perform continuous monitoring based on the multiple fault location data, synchronize the monitoring results to the warning unit to generate a warning instruction, and execute the warning instruction to perform intelligent fault warning on the container stacker.
Citation Information
Patent Citations
Article management method and system and storage medium
CN110245887A
Intelligent equipment management system and method based on Internet of Things
CN116823227A
Unmanned automatic empty box stacking machine detection and fault diagnosis system and method
CN117516975A
Intelligent equipment monitoring and early warning system
CN117826715A
Underground high-voltage cable joint fault vibration monitoring Internet of Things system
CN118310626A
