A method and system for monitoring abnormal operation of a hose reel

Abstract

This application relates to the field of industrial equipment monitoring technology, and more particularly to a method and system for monitoring the abnormal operation of a hose reel. The method comprises: acquiring multi-dimensional operating data during the operation of the reel, including motor speed data, current data, and hose temperature data; constructing a detection correction factor; using a LOF anomaly detection algorithm based on the operating data to detect the operating status of the equipment, and using the detection correction factor to correct the LOF detection result to determine the equipment operating status. This application improves the accuracy of reel operation monitoring.

Description

Technical Field

[0001] The present application relates to the technical field of industrial equipment monitoring, and in particular to a method and system for monitoring abnormal operation of a hose reel. Background Art

[0002] A hose is a rubber product used to transport gases and liquids. It is formed by extrusion through an extruder and then wound and packaged using a hose reel. The reel's operational stability during the hose reeling process directly impacts the final hose packaging quality. For example, insufficient tension on the hose during reeling can result in a loose hose after reeling. Excessive tension can cause the motor to overheat, and choppy hose movement can lead to uneven tension after reeling. Therefore, it is often necessary to monitor the reel's operating status during operation to promptly detect any anomalies and adjust the reel. The LOF (Local Outlier Factor) algorithm is a common anomaly detection algorithm used in medical diagnostics, network security, and industrial equipment monitoring. Its main principle is to determine whether a data point is an outlier by calculating the density of its neighboring data points. If a data point is located in a low-density area and its neighboring data points are located in a high-density area, then this data point may be an outlier. If the density of a data point is similar to the density of its neighboring data points, then this data point may be a normal data point.

[0003] However, hose reels are complex process equipment, involving the coordinated motion of multiple mechanical components, such as the motor drive, reel rotation, and reciprocating motion of the wire guide. The complexity of these mechanical motions can lead to nonlinear relationships and coupling effects in sensor data, making it difficult for traditional LOF anomaly detection algorithms to accurately distinguish between normal fluctuations and true anomalies. Consequently, traditional LOF anomaly detection algorithms lack accuracy when applied to complex hose reels. Summary of the Invention

[0004] In order to solve the problem of insufficient accuracy of traditional LOF detection algorithms for monitoring the operation of hose reels, the present application provides a method and system for monitoring the abnormal operation of hose reels.

[0005] In a first aspect, the present application provides a method for monitoring abnormal operation of a hose reel, which adopts the following technical solution:

[0006] A method for monitoring abnormal operation of a hose reel includes the following steps: obtaining multi-dimensional operating data during the operation of the reel, the operating data including motor speed data, current data, and hose temperature data; constructing a detection correction factor; using a LOF abnormality detection algorithm based on the operating data to detect the operating status of the equipment, and using the detection correction factor to correct the LOF detection result to determine the operating status of the equipment;

[0007] The abnormal winding degree of the hose winding at each moment is calculated based on the motor speed and current at each moment; an observation sequence of the hose temperature data at each moment is constructed, and the temperature anomaly index at the corresponding moment is obtained based on the ratio of the hose temperature data in the observation sequence to the limit temperature of the hose deformation; for any moment, the adjacent winding sequence at that moment is constructed based on the abnormal winding degree, and the adjacent temperature sequence is constructed based on the temperature anomaly index; the abnormal amplitude of the period is determined based on the size of the data in the adjacent winding sequence and the adjacent temperature sequence, and the abnormal fluctuation of the period is determined based on the fluctuation of the data in the adjacent winding sequence and the adjacent temperature sequence, and the product of the abnormal amplitude of the period and the abnormal fluctuation of the period is used as the detection correction factor.

[0008] The beneficial effects are as follows: the speed data and current data during the operation of the coiling machine are analyzed to construct the abnormal winding degree. The adjacent winding sequence at a certain moment is constructed based on the abnormal winding degree. The temperature anomaly index at each temperature collection moment is constructed based on the temperature of the hose and the maximum temperature that the hose can withstand, and the adjacent temperature sequence at a certain moment is constructed based on the temperature anomaly index. The abnormal winding degree can reflect whether there is a jamming phenomenon during the hose winding process. The jamming phenomenon is mainly reflected in the change of motor speed and current. The temperature anomaly index reflects the temperature of the hose during the winding process. The generation of the hose temperature mainly includes the tension it is subjected to and the contact friction with its wire guide. Therefore, the temperature change can reflect the abnormal tension on the hose itself or whether the wire guide is abnormal. Combined with the size of the data in the adjacent temperature sequence and the adjacent winding sequence, the jamming situation in the adjacent time period at a certain moment and the temperature of the hose, and based on this, a detection correction factor is constructed. The detection correction factor is used to correct the detection results of the traditional LOF detection algorithm, thereby improving the accuracy of the final equipment detection. Compared with the traditional LOF detection algorithm, the detection results in this application take into account the changing relationship between motor current, motor speed and hose temperature more comprehensively based on the current data, and are more sensitive to drastic and large-amplitude sudden changes in data, making it easier to distinguish normal fluctuations from abnormal fluctuations in operating data, and therefore have higher accuracy.

[0009] Optionally, the step of calculating the abnormal winding degree at any moment includes: acquiring current data and rotational speed data at that moment, and taking the ratio of the current data to the rotational speed data at that moment as the abnormal winding degree at that moment.

[0010] The beneficial effect is: during the hose winding process, when the hose winding is stuck, the load on the motor increases, the current increases, and at the same time the motor speed decreases. Therefore, combining the motor current and the motor speed can determine whether the equipment is stuck during operation, and then the operating status of the equipment can be determined based on this.

[0011] Optionally, the calculation step for the abnormal winding degree at any moment includes: obtaining the abnormal winding degree at that moment and before that moment Current data and speed data are obtained; the current mean value of the current data and the speed mean value of the speed data are obtained, and the ratio of the current mean value to the speed mean value is used as the abnormal winding degree at that moment.

[0012] The beneficial effects are: in the process of calculating the abnormal winding degree, the calculation is performed based on the average of multiple data, which improves the accuracy and robustness of the abnormal winding degree calculation and reduces the influence of noise signals on the abnormal winding degree calculation.

[0013] Optionally, the collection frequency of the hose temperature data is lower than the collection frequency of the speed data and the current data, and the collection frequency of the speed data is equal to that of the current data; It is the difference between the current data acquisition frequency and the hose temperature data acquisition frequency.

[0014] The beneficial effect is that during the operation of the equipment, the temperature data changes relatively slowly, thereby reducing the frequency of temperature collection.

[0015] Optionally, the step of constructing the observation sequence of each moment in the hose temperature data includes: determining the length of the observation sequence , for any moment, the moment and the moments before that moment The hose temperature data is taken as the observation sequence at that moment.

[0016] The beneficial effect is: a certain moment and the moment before The data is used as the observation sequence at that moment, which is convenient for analyzing the temperature changes in the recent time period and accurately reflecting the current temperature conditions.

[0017] Optionally, the step of obtaining the temperature anomaly index at the corresponding moment according to the ratio of the hose temperature data in the observation sequence to the limit temperature of the hose deformation includes: taking the sum of the ratios of multiple hose temperature data in the observation sequence to the limit temperature as the temperature anomaly index.

[0018] The beneficial effect is that the hose deformation limit temperature represents the maximum temperature the hose can withstand. During coiling machine operation, the closer the hose temperature approaches the limit temperature, the more dangerous the current operation state is, and the more likely the hose will deform. Therefore, the ratio of the hose temperature to the limit temperature is used as the temperature anomaly index. The larger the temperature anomaly index, the worse the current temperature conditions are and the more likely the hose will deform.

[0019] Optionally, the step of constructing the adjacent winding sequence and the adjacent temperature sequence includes: for any moment, the moment and the moment before the moment The abnormal winding degrees constitute the adjacent winding sequence, and the moment and the moments before this moment The temperature anomaly indices constitute a neighboring temperature sequence.

[0020] Optionally, the step of obtaining the abnormal amplitude of the time period includes: obtaining the mean of the data in the adjacent winding sequence and the mean of the data in the adjacent temperature sequence, and taking the sum of the two means as the abnormal amplitude of the time period;

[0021] Obtain the standard deviation of the data in the adjacent winding series and the standard deviation of the data in the adjacent temperature series, and take the sum of the two means as the abnormal fluctuation of the period.

[0022] The beneficial effect is that the mean values of adjacent winding sequences and adjacent temperature sequences indicate the magnitude of the coiler jam anomaly and the overall temperature anomaly index, reflecting the overall degree of anomaly within that time period. Therefore, the sum of the two means is used as the period anomaly magnitude. The standard deviation of the data in the two sequences reflects the fluctuation range of the data in the two sequences and can reflect the sudden change of data in the sequences. Therefore, the sum of the two standard deviations is used as the period anomaly, further improving the accuracy of anomaly detection.

[0023] Optionally, the step of correcting the LOF detection result based on the detection correction factor to determine the operating status of the equipment includes: constructing a monitoring vector corresponding to each temperature collection moment based on the abnormality score and the detection correction factor at each temperature collection moment, clustering each monitoring vector using a clustering method, and judging the operating status at each moment based on the clustering result.

[0024] The beneficial effects are as follows: the detection correction factor and the anomaly score obtained by the traditional LOF detection algorithm constitute a monitoring vector, the monitoring vectors at each moment are clustered through clustering, and the operating status at each moment is judged according to the clustering results, thereby realizing the detection of the equipment operating status.

[0025] In a second aspect, the present application provides a system for monitoring abnormal operation of a hose reel, which adopts the following technical solution:

[0026] A system for monitoring abnormal operation of a hose reel includes a processor and a memory. The memory stores computer program instructions. When the computer program instructions are executed by the processor, the system implements the above-mentioned method for monitoring abnormal operation of a hose reel.

[0027] The beneficial effect is that the above-mentioned method for monitoring the abnormal operation of the hose reel is generated into a computer program and stored in the memory so as to be loaded and executed by the processor, thereby making a system based on the memory and the processor for easy use.

[0028] This application has the following technical effects: In this application, the abnormal winding degree is extracted based on the current data and speed data during the operation of the coiler, and the temperature anomaly index is constructed based on the temperature data of the hose. Then, the two are combined to construct a detection correction factor to correct the traditional LOF detection results, thereby improving the accuracy of the final detection of the equipment operation status. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 This is a method flow chart of a method for monitoring abnormal operation of a hose reel according to an embodiment of the present application.

[0030] Figure 2 This is a method flow chart of step S2 in a method for monitoring abnormal operation of a hose reel according to an embodiment of the present application. DETAILED DESCRIPTION

[0031] The present application discloses a method and system for monitoring abnormal operation of a hose reel. The system acquires multi-dimensional operational data during the reel's operation, extracts the coupling relationship between the data in each dimension, constructs a detection correction factor, and uses the detection correction factor to optimize the detection results of a traditional LOF detection algorithm. Compared to the traditional LOF algorithm, which relies on neighboring data to determine the status of the current data point, the method in this application takes into account the relationship between multiple dimensional data, reducing the traditional detection algorithm's reliance on neighboring data, thereby improving the accuracy of the final reel operation detection.

[0032] Reference Figure 1 A method for monitoring abnormal operation of a hose reel includes steps S1 to S3.

[0033] S1: Obtain multi-dimensional operating data during the operation of the coiler, including motor speed data, current data, and hose temperature data.

[0034] The coiling machine primarily consists of a frame, a winding reel, a motor, and a wire guide. During operation, the wire guide guide activates the hose, while the motor drives the winding reel. This rotation allows the hose to be wound around the reel, completing the process. Coiling machines are conventional equipment in this field, so their detailed structure will not be detailed here.

[0035] During coiling machine operation, motor speed and current data, as well as hose temperature data on the winding reel, are collected. In this embodiment, the speed and current data are collected at a frequency of 100 Hz, meaning 100 data points are collected per second. Because temperature changes are relatively slow, the hose temperature data collection frequency is set to 1 Hz in this embodiment.

[0036] In the process of collecting the above data, the collected data may contain anomalies such as noise and missing values due to factors such as environmental interference and sensor errors. In order to improve the data quality, this application uses a median filtering algorithm to denoise the collected data. The regression filling method is used to fill in the missing values. At the same time, in order to avoid the influence of different dimensions on the final calculation results, this application uses the maximum normalization method to normalize the collected current data and speed data. Among them, the median filtering algorithm, the regression filling method, and the maximum normalization method are well-known technologies, and the specific process will not be repeated here.

[0037] S2: Construct detection correction factor.

[0038] The step of constructing the detection correction factor includes: step S21 to step S23.

[0039] S21: calculating the abnormal winding degree of the hose winding at each moment based on the motor speed and current at each moment;

[0040] During hose reel operation, if the hose is not winding smoothly or becomes stuck, the motor must overcome greater resistance to drive the reel. This sudden increase in motor load causes a sudden surge in current. Furthermore, as resistance increases, the motor's speed cannot be maintained at a normal level and may decrease to a certain extent. These changes in motor speed and the sudden change in winding resistance can affect proper hose winding. Furthermore, prolonged and frequent motor jams can cause the motor to overheat, resulting in significant damage to the equipment.

[0041] Therefore, in this application, the abnormal winding degree that occurs during the hose winding process is extracted by analyzing the current and motor speed data.

[0042] In one embodiment, the step of calculating the abnormal winding degree includes: for the abnormal winding degree at any moment, taking the ratio of the current data to the motor speed data at the moment as the abnormal winding degree at the moment.

[0043] When the motor current increases, it indicates that the motor load has increased, and the hose may be stuck during coiling. At the same time, when the motor speed decreases, it also indicates that the coiling process may be stuck. Combining current data and motor speed data can synergistically reflect the abnormal winding degree of the hose during coiling equipment.

[0044] In another embodiment, the calculation of abnormal winding degree can also be obtained by the following steps: first, obtain the data at any moment and before that moment. The system collects current and speed data, obtains the average current and speed, and uses the ratio of the average current to the average speed as the abnormal winding degree. When a jam occurs during hose winding, resistance disappears, but it takes time for the speed and current to recover. This means that during a certain period of time, the current is higher than the normal current level, while the motor speed is lower than the normal speed level. By calculating the average of multiple data points, the final calculated abnormal winding degree is less affected by individual extreme data points, improving the accuracy and robustness of the abnormal winding degree calculation.

[0045] In this embodiment = is the difference between the current data acquisition frequency and the hose temperature data acquisition frequency. In this embodiment, the hose temperature data acquisition frequency is used. For example, if the temperature data sampling frequency in this application is 1 Hz, then the current mean is the average value of the current within one second, which can reflect the overall current magnitude during that period.

[0046] S22: constructing an observation sequence of the hose temperature data at each moment, and obtaining a temperature anomaly index at the corresponding moment based on the ratio of the hose temperature data in the observation sequence to the limit temperature of the hose deformation;

[0047] During the hose coiling process, friction between the wire guide and the hose generates a certain amount of heat. The hose also generates some heat when subjected to tension. During normal coiling, the tension on the hose is relatively stable (this can also be understood as no jamming during the coiling process). Furthermore, there is rolling friction between the wire guide and the hose. Jams in the coiling equipment or malfunctions in the wire guide can cause the hose temperature to rise. Excessively high hose temperatures can cause the hose to crack or deform. Therefore, analyzing the hose temperature is crucial to further identify the coiling machine's operating status.

[0048] Determine the length of the observation sequence , for any moment, the moment and the moments before that moment The hose temperature data is taken as the observation sequence at that moment.

[0049] Since the temperature changes relatively slowly and accumulates continuously, an observation sequence is set for each moment and multiple hose temperature data in the observation sequence are analyzed to improve the accuracy of the final calculated temperature anomaly indication.

[0050] Calculate the ratio of each hose temperature data point in the observation sequence to the hose deformation limit temperature, and use the cumulative result of these ratios as the temperature anomaly index. The hose deformation limit temperature refers to the temperature at which the hose begins to deform under a certain tension. This temperature is a preset value, typically determined through multiple tests, and is set based on actual production conditions and hose material.

[0051] In this embodiment, the length of the observation sequence is 30. In combination with the fact that the collection frequency of the hose temperature data in this application is 1 Hz, it can also be understood that the observation sequence includes the hose temperature data of the latest 30 seconds at a certain moment.

[0052] The closer the hose temperature data in the observation sequence is to the limit temperature of the hose deformation, the more dangerous the hose's current state is, and the more likely it is that the abnormal heating of the hose is caused by an abnormality in the coiling machine.

[0053] S23: For any moment, construct the adjacent winding sequence of the moment based on the abnormal winding degree, and construct the adjacent temperature sequence based on the temperature anomaly index; determine the period abnormal amplitude based on the size of the data in the adjacent winding sequence and the adjacent temperature sequence, and determine the period abnormal fluctuation based on the fluctuation of the data in the adjacent winding sequence and the adjacent temperature sequence, and use the product of the period abnormal amplitude and the period abnormal fluctuation as the detection correction factor.

[0054] For any moment, the mixed anomaly degree of the moment is calculated by the temperature anomaly index and abnormal winding degree of the moment. In order to improve the accuracy and robustness of the temperature mixing degree calculation at the moment, the present embodiment obtains the temperature anomaly index and the abnormal winding degree of the moment. The present invention provides a plurality of temperature anomaly indices and a plurality of abnormal winding degrees corresponding to the temperature anomaly indices, wherein the plurality of temperature anomaly indices constitute an adjacent winding sequence, and the plurality of temperature anomaly indices constitute an adjacent temperature sequence.

[0055] The period abnormal amplitude and period abnormal fluctuation are obtained based on the adjacent winding sequence and the adjacent temperature sequence; the product of the period abnormal amplitude and the period abnormal fluctuation is used as the mixed abnormality degree; and the mixed abnormality degree is used as the detection correction factor.

[0056] Specifically, the mean of the abnormal winding degree data in the adjacent winding sequence and the mean of the temperature anomaly index in the adjacent temperature sequence are obtained, and the sum of the two means is used as the abnormal amplitude of the period. The standard deviation of the data in the adjacent winding sequence and the standard deviation of the data in the adjacent temperature sequence are obtained, and the sum of the two standard deviations is used as the abnormal fluctuation of the period.

[0057] The mean of the abnormal winding degree in the adjacent winding sequence describes the overall level of the data in the adjacent winding sequence. The standard deviation of the data in the adjacent winding sequence describes the degree of fluctuation in the data in the adjacent winding sequence. Corresponding to the abnormal jamming during the winding equipment operation, the mixed abnormality degree is calculated by combining the abnormal amplitude and abnormal fluctuation of the period, thereby improving the accuracy of the final detection of the equipment operation status.

[0058] S3: Detect the equipment operating status using the LOF anomaly detection algorithm based on the operating data, and use the detection correction factor to correct the LOF detection result to determine the equipment operating status.

[0059] The above steps yield a detection correction factor corresponding to each temperature collection moment, which is then used to correct the detection results of the traditional LOF algorithm. Therefore, the mean current and mean speed values corresponding to the temperature collection moment are first used as inputs to the LOF algorithm, which then outputs the LOF anomaly score for that moment. Each temperature collection moment now corresponds to a detection correction factor and an anomaly score. The detection correction factor and anomaly score for the same temperature collection moment constitute the monitoring vector for that temperature collection moment. Using a clustering method, the monitoring vectors from multiple moments are clustered to produce multiple clusters.

[0060] In this example, the K-means clustering algorithm is used for clustering, with the number of clusters set to 4. The mean modulus of the monitoring vectors within each cluster is calculated, and the clusters are sorted in ascending order from smallest to largest. The sorted clusters are then labeled as normal, slightly abnormal, moderately abnormal, and severely abnormal, respectively. The label of the severe cluster corresponding to each temperature collection moment is used to determine the device's operating status at that moment. Different signals are issued based on different operating states, providing a graded warning of the device's operating status.

[0061] An embodiment of the present application also discloses an operation abnormality monitoring system for a hose reel, comprising a processor and a memory, wherein the memory stores computer program instructions. When the computer program instructions are executed by the processor, an operation abnormality monitoring method for a hose reel according to the present application is implemented.

[0062] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface. The configuration and functions of these components are known in the art and will not be described in detail here.

[0063] The above are all preferred embodiments of the present application, and are not intended to limit the scope of protection of the present application. Therefore, any equivalent changes made based on the structure, shape, and principle of the present application should be included in the scope of protection of the present application.

Claims

1. A method for monitoring abnormal operation of a hose reel, characterized in that: The method comprises the following steps: obtaining multi-dimensional operation data of the coiling machine during operation, the operation data including motor speed data, current data and hose temperature data; constructing a detection correction factor; Based on the operating data, the LOF anomaly detection algorithm is used to detect the operating status of the equipment, and the detection correction factor is used to correct the LOF detection results to determine the operating status of the equipment; Calculate the abnormal winding degree of the hose winding at each moment based on the motor speed and current at each moment; An observation sequence for each moment in the hose temperature data is constructed, and the temperature anomaly index at the corresponding moment is obtained based on the ratio of the hose temperature data in the observation sequence to the extreme temperature of the hose deformation. For any moment, a neighboring winding sequence is constructed based on the abnormal winding degree, and a neighboring temperature sequence is constructed based on the temperature anomaly index. The abnormal amplitude of the time period is determined based on the size of the data in the neighboring winding sequence and the neighboring temperature sequence, and the abnormal fluctuation of the time period is determined based on the fluctuation of the data in the neighboring winding sequence and the neighboring temperature sequence. The product of the abnormal amplitude of the time period and the abnormal fluctuation of the time period is used as the detection correction factor. The step of calculating the abnormal winding degree at any moment includes: obtaining the current data and the rotational speed data at the moment, and taking the ratio of the current data to the rotational speed data at the moment as the abnormal winding degree at the moment.

2. The method for monitoring abnormal operation of a hose reel according to claim 1, characterized in that: The calculation steps for the abnormal winding degree at any moment include: obtaining the time at that moment and the time before that moment Current data and speed data are obtained; the current mean value of the current data and the speed mean value of the speed data are obtained, and the ratio of the current mean value to the speed mean value is used as the abnormal winding degree at that moment.

3. The method for monitoring abnormal operation of a hose reel according to claim 2, characterized in that: The collection frequency of hose temperature data is lower than that of speed data and current data, while the collection frequency of speed data and current data is equal; It is the difference between the current data acquisition frequency and the hose temperature data acquisition frequency.

4. The method for monitoring abnormal operation of a hose reel according to claim 1, characterized in that: The steps of constructing the observation sequence of each moment in the hose temperature data include: determining the length of the observation sequence , for any moment, the moment and the moments before that moment The hose temperature data is taken as the observation sequence at that moment.

5. The method for monitoring abnormal operation of a hose reel according to claim 4, characterized in that: The step of obtaining the temperature anomaly index at the corresponding moment according to the ratio of the hose temperature data in the observation sequence to the limit temperature of the hose deformation includes: taking the sum of the ratios of multiple hose temperature data in the observation sequence to the limit temperature as the temperature anomaly index.

6. The method for monitoring abnormal operation of a hose reel according to claim 1, characterized in that: The steps of constructing adjacent winding sequence and adjacent temperature sequence include: for any moment, the moment and the moments before the moment The abnormal winding degrees constitute the adjacent winding sequence, and the moment and the moments before this moment The temperature anomaly indices constitute a neighboring temperature sequence.

7. The method for monitoring abnormal operation of a hose reel according to claim 6, characterized in that: The step of obtaining the abnormal amplitude of the time period includes: obtaining the mean of the data in the adjacent winding sequence and the mean of the data in the adjacent temperature sequence, and taking the sum of the two means as the abnormal amplitude of the time period; The step of obtaining abnormal fluctuations during a period includes obtaining a standard deviation of data in adjacent winding sequences and a standard deviation of data in adjacent temperature sequences, and taking the sum of the two standard deviations as the abnormal fluctuations during the period.

8. The method for monitoring abnormal operation of a hose reel according to claim 1, characterized in that: The steps of correcting the LOF detection result based on the detection correction factor to determine the operating status of the equipment include: constructing a monitoring vector corresponding to each temperature collection moment based on the abnormality score and the detection correction factor at each temperature collection moment, clustering each monitoring vector using a clustering method, and determining the operating status at each moment based on the clustering result; The average current and speed corresponding to the temperature collection time are used as the input of the LOF algorithm, and the LOF anomaly score at the corresponding time is output.

9. A system for monitoring abnormal operation of a hose reel, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a method for monitoring abnormal operation of a hose reel according to any one of claims 1 to 8 is implemented.

Images