Method, device, system and storage medium for condition monitoring of a belt system

Abstract

The present disclosure relates to the technical field of industrial automation, and discloses a state monitoring method, device and system for a belt system and a storage medium. The state monitoring method comprises: acquiring state data of multiple dimensions of the belt system by using a sensor group; determining a comparison result of the state data of each dimension and a corresponding data threshold; determining that the working state of the belt system is abnormal when the comparison result of the state data of one or more dimensions does not meet a safety condition; inputting the state data of multiple dimensions to a pre-trained state evaluation model when the comparison result of each state data meets the safety condition, and determining the grade type of the working state of the belt system by using the state evaluation model. The above method can efficiently acquire the working state of the belt system, can timely find abnormal problems of the belt system, and can more accurately understand the current working state of the belt system.

Description

Technical Field

[0001] This disclosure relates to the technical field of industrial automation, and for example to a method, apparatus, system and storage medium for condition monitoring of belt systems. Background Technology

[0002] The safe operation of belt conveyor systems directly impacts a company's production efficiency. While belt conveyors offer advantages such as high load capacity, large conveying capacity, wide applicability, and suitability for long-distance transportation, in actual production, long transport distances, heavy belt loads, and poor operating conditions can lead to frequent failures, including belt misalignment, belt tearing, belt breakage, belt slippage, and motor overheating and vibration. Belt conveyor system malfunctions disrupt normal production; therefore, timely monitoring of the system's operating status is crucial.

[0003] In related technologies, most methods of obtaining the working status of belt conveyor systems are manual inspection, manual correction, and video monitoring. However, these methods of obtaining the working status of belt conveyor systems are heavily dependent on manual labor, resulting in low efficiency and poor accuracy. Summary of the Invention

[0004] To provide a basic understanding of some aspects of the disclosed embodiments, a brief summary is given below. This summary is not intended as a general commentary, nor is it intended to identify key / important components or describe the scope of protection of these embodiments, but rather as a prelude to the detailed description that follows.

[0005] This disclosure provides a method, apparatus, system, and storage medium for monitoring the status of a belt system, which can efficiently acquire the working status of the belt system.

[0006] According to a first aspect of this disclosure, a method for monitoring the condition of a belt conveyor system is provided, characterized in that it includes:

[0007] Utilize a sensor array to acquire multi-dimensional status data of the belt system;

[0008] Determine the comparison results between the state data of each dimension and the corresponding data threshold;

[0009] If the comparison results of status data in one or more dimensions do not meet the safety conditions, the working status of the belt system is determined to be abnormal.

[0010] When the comparison results of each state data meet the safety conditions, the state data of multiple dimensions are input into the pre-trained state evaluation model, and the state evaluation model is used to determine the level type of the working state of the belt system.

[0011] In some embodiments, the multi-dimensional state data includes drive motor temperature, and / or reducer temperature, and / or drive motor vibration, and / or reducer vibration, and / or belt running speed, and / or belt tearing degree, and / or belt position offset degree, and / or belt slippage degree.

[0012] In some embodiments, inputting multi-dimensional state data into a pre-trained state evaluation model includes:

[0013] The state data for each dimension is regularized and aligned.

[0014] The data processing results of state data from multiple dimensions are input into a pre-trained state evaluation model.

[0015] In some embodiments, regularization processing for state data in one dimension includes:

[0016] Determine the maximum and minimum values ​​in the state data for this dimension, and calculate the first difference between the maximum and minimum values;

[0017] For each state data in this dimension, calculate the second difference between that state data and the minimum data, and use the ratio of the second difference to the first difference as the regularization result for each state data.

[0018] In some embodiments, alignment processing for state data in one dimension includes:

[0019] Identify the channel with the highest sampling frequency in this dimension, and determine the maximum data in the state data of the channel with the highest sampling frequency;

[0020] The regularization result of the maximum data of the channel with the highest sampling frequency is used as the alignment standard, and the regularization results of the state data of other channels in this dimension are adjusted based on this alignment standard.

[0021] In some embodiments, the data processing results of multi-dimensional state data are input into a pre-trained state evaluation model, including:

[0022] The data processing results of the state data in each dimension are truncated into data units of a specified length. Each data unit includes the regularization results of the state data of a specified number of channels in the corresponding dimension.

[0023] Multiple data units are input into a pre-trained state evaluation model.

[0024] In some embodiments, the state evaluation model is trained based on raw data, a positive sample set, and a negative sample set; the raw data is multi-dimensional state data collected by the sensor group when the belt system is in normal working condition, the positive sample set is obtained by data augmentation of the raw data, and the negative sample set is multi-dimensional state data collected by the sensor group when the belt system is in abnormal working condition.

[0025] In some embodiments, the loss functions used in the training process of the state evaluation model include the contrastive loss function and the cross-entropy loss function.

[0026] According to a second aspect of this disclosure, a condition monitoring device for a belt system is provided, comprising:

[0027] The data acquisition module is configured to acquire multi-dimensional status data of the belt system using a sensor array;

[0028] The data comparison module is configured to determine the comparison results between the state data of each dimension and the corresponding data threshold;

[0029] The first state evaluation module is configured to determine that the working state of the belt system is abnormal when the comparison results of state data in one or more dimensions do not meet the safety conditions.

[0030] The second state evaluation module is configured to input state data from multiple dimensions into a pre-trained state evaluation model when the comparison results of each state data meet the safety conditions, and use the state evaluation model to determine the level type of the working state of the belt system.

[0031] According to a third aspect of this disclosure, a condition monitoring device for a belt system is provided, including a processor and a memory storing program instructions, wherein the processor is capable of executing the condition monitoring method for a belt system provided in the first aspect of this disclosure.

[0032] According to a fourth aspect of this disclosure, a condition monitoring system for a belt system is provided, including a sensor array and a condition monitoring device for a belt system provided in a second or third aspect of this disclosure.

[0033] The status monitoring device is communicatively connected to each sensor in the sensor group, which is used to collect status data of the belt system in multiple dimensions.

[0034] According to a fifth aspect of this disclosure, a storage medium is provided that stores computer program instructions, which, when executed by a processor, perform the status monitoring method for a belt system provided in the first aspect.

[0035] The method, apparatus, system, and storage medium for monitoring the status of belt systems provided in this disclosure can achieve the following technical effects:

[0036] The method for monitoring the status of a belt conveyor system provided in this disclosure utilizes a sensor array to acquire multi-dimensional status data of the belt conveyor system in real time. By comparing the real-time acquired multi-dimensional status data with corresponding data thresholds, it quickly determines whether the operating status of the belt conveyor system is abnormal, thus enabling timely detection of abnormal problems. If an abnormality is not determined, the multi-dimensional status data can be further input into a pre-trained status evaluation model. The model determines the level of the belt conveyor system's operating status, providing a more accurate understanding of the current operating status and facilitating proactive countermeasures. Therefore, this method can efficiently acquire the operating status of the belt conveyor system, promptly detect abnormal problems, and provide a more accurate understanding of its current operating status, achieving refined management of the belt conveyor system's status.

[0037] The above general description and the description below are exemplary and illustrative only and are not intended to limit this disclosure. Attached Figure Description

[0038] One or more embodiments are illustrated by way of example with reference to the accompanying drawings. These illustrations and drawings do not constitute a limitation on the embodiments. Elements having the same reference numerals in the drawings are shown as similar elements. The drawings are not to be scaled. And wherein:

[0039] Figure 1 This is a schematic diagram of the structure of a condition monitoring system for a belt conveyor system provided in an embodiment of this disclosure;

[0040] Figure 2 This is a schematic diagram of another condition monitoring system for a belt conveyor system provided in this embodiment of the present disclosure;

[0041] Figure 3 This is a schematic flowchart of a condition monitoring method for a belt system provided in an embodiment of this disclosure;

[0042] Figure 4 This is a schematic flowchart of a condition monitoring method for a belt system provided in an embodiment of this disclosure;

[0043] Figure 5 This is a flowchart illustrating a training state evaluation model provided in an embodiment of this disclosure;

[0044] Figure 6 This is a schematic flowchart of a status monitoring device for a belt system provided in an embodiment of this disclosure;

[0045] Figure 7 This is a schematic flowchart of another condition monitoring device for a belt system provided in an embodiment of this disclosure. Detailed Implementation

[0046] To provide a more detailed understanding of the features and technical content of the embodiments of this disclosure, the implementation of the embodiments of this disclosure will be described in detail below with reference to the accompanying drawings. The accompanying drawings are for illustrative purposes only and are not intended to limit the embodiments of this disclosure. In the following technical description, for ease of explanation, several details are used to provide a full understanding of the disclosed embodiments. However, one or more embodiments may still be implemented without these details. In other cases, well-known structures and devices may be simplified in their depiction to simplify the drawings.

[0047] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of this disclosure described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion.

[0048] Unless otherwise stated, the term "multiple" means two or more.

[0049] In this embodiment of the disclosure, the character " / " indicates that the objects before and after it are in an "or" relationship. For example, A / B means: A or B.

[0050] The term "and / or" describes an association between objects, indicating that three relationships can exist. For example, A and / or B means: A or B, or A and B.

[0051] The term "correspondence" can refer to an association or binding relationship. The correspondence between A and B means that there is an association or binding relationship between A and B.

[0052] The safe operation of belt conveyor systems directly impacts a company's production efficiency. While belt conveyors offer advantages such as high load capacity, large conveying capacity, wide applicability, and suitability for long-distance transportation, in actual production, long transport distances, heavy belt loads, and poor operating conditions can lead to frequent failures, including belt misalignment, belt tearing, belt breakage, belt slippage, and motor overheating and vibration. Belt conveyor system malfunctions disrupt normal production; therefore, timely monitoring of the system's operating status is crucial.

[0053] In related technologies, most methods for obtaining the working status of belt conveyor systems rely on manual inspection, manual correction, and video monitoring. Manual inspection is slow, labor-intensive, and lacks real-time accuracy; manual correction has low automation and is inherently dangerous; and video monitoring can easily miss crucial information. Therefore, these technologies heavily depend on manual labor for obtaining the working status of belt conveyor systems, resulting in low efficiency and poor accuracy.

[0054] like Figure 1 As shown, this embodiment of the disclosure provides a status monitoring system for a belt conveyor system. The status monitoring system includes a status monitoring device 600 (700) and a sensor group 100. The sensor group 100 includes multiple sensors, and the status monitoring device 600 (700) is communicatively connected to each sensor in the sensor group 100. The multiple sensors in the sensor group 100 can be arranged at corresponding positions in the belt conveyor system to collect status data of the belt conveyor system in multiple dimensions. The status monitoring device 600 (700) receives the status data of multiple dimensions collected by the sensor group 100 and determines the working status of the belt conveyor system based on the status data of multiple dimensions.

[0055] In this embodiment of the disclosure, the belt system includes a drive motor, a reducer, and a belt. The status data of multiple dimensions include two or more of the following: drive motor temperature, reducer temperature, drive motor vibration, reducer vibration, belt running speed, belt tearing degree, belt position offset degree, and belt slippage degree.

[0056] In some embodiments, such as Figure 2 As shown, the sensor group 100 includes temperature sensors 101. There can be multiple temperature sensors 101, which can be arranged on the drive motor and / or the reducer. The temperature sensor 101 located on the drive motor is used to collect the drive motor temperature, and the temperature sensor 101 located on the reducer is used to collect the reducer temperature.

[0057] In some embodiments, such as Figure 2 As shown, the sensor group 100 includes vibration sensors 102. There can be multiple vibration sensors 102, which can be arranged on the drive motor and / or the reducer. The vibration sensor 102 installed on the drive motor is used to collect the vibration amount of the drive motor, and the vibration sensor 102 installed on the reducer is used to collect the vibration amount of the reducer.

[0058] In some embodiments, such as Figure 2 As shown, the sensor group 100 includes a speed sensor 103. There can be multiple speed sensors 103. The speed sensors 103 are used to collect the belt running speed.

[0059] In some embodiments, such as Figure 2 As shown, the sensor group 100 includes a belt tear sensor 104. There can be multiple belt tear sensors 104. The belt tear sensor 104 is used to collect the belt tear degree value.

[0060] In some embodiments, such as Figure 2 As shown, the sensor group 100 includes a belt misalignment sensor 105. There can be multiple belt misalignment sensors 105. The belt misalignment sensor 105 is used to collect the value of the belt position deviation.

[0061] In some embodiments, such as Figure 2 As shown, the sensor group 100 includes a belt slippage sensor 106. There can be multiple belt slippage sensors 106. The belt slippage sensor 106 is used to collect the belt slippage degree value.

[0062] In conjunction with the condition monitoring device for belt systems provided in the embodiments of this disclosure, the embodiments of this disclosure also provide a condition monitoring method for belt systems. This condition monitoring method for belt systems can be executed by the condition monitoring device (hereinafter referred to as the device). Figure 3 As shown, the condition monitoring method for a belt conveyor system includes the following steps:

[0063] S301, the device uses a sensor array to acquire multi-dimensional status data of the belt system.

[0064] In this embodiment of the disclosure, the belt system includes a drive motor, a reducer, and a belt. The status data of multiple dimensions include two or more of the following: drive motor temperature, reducer temperature, drive motor vibration, reducer vibration, belt running speed, belt tearing degree, belt position offset degree, and belt slippage degree.

[0065] S302, the device determines the comparison result between the state data of each dimension and the corresponding data threshold.

[0066] In the embodiments of this disclosure, data thresholds can be set for the state data of each dimension. For example, a first temperature threshold is set for the drive motor temperature, a second temperature threshold is set for the reducer temperature, a first vibration threshold is set for the drive motor vibration, a second vibration threshold is set for the reducer vibration, a running speed threshold is set for the belt running speed, a tearing degree threshold is set for the belt tearing degree, an offset degree threshold is set for the belt position offset degree, and a slippage degree threshold is set for the belt slippage degree.

[0067] S303, when the comparison results of the status data in one or more dimensions do not meet the safety conditions, the working status of the belt system is determined to be abnormal.

[0068] In this embodiment of the disclosure, the comparison result of the state data for each dimension is used to indicate the relationship between the state data of that dimension and the corresponding data threshold. Here, when the comparison result of the state data for one dimension indicates that the state data of that dimension is greater than the corresponding data threshold, it is determined that the comparison result of the state data for that dimension does not meet the safety conditions. When the comparison results of the state data for one or more dimensions do not meet the safety conditions, it is determined that the working state of the belt system is abnormal.

[0069] In this embodiment of the disclosure, for each dimension of state data, an average value of the state data for that dimension can be calculated, and this average value can be compared with a corresponding data threshold. The comparison result of the average value of the state data for that dimension and the corresponding data threshold is taken as the comparison result of the state data for that dimension and the corresponding data threshold.

[0070] In this embodiment of the disclosure, for each dimension of state data, the maximum value of the state data in that dimension can be calculated, and this maximum value can be compared with the corresponding data threshold. The comparison result of the maximum value of the state data in that dimension with the corresponding data threshold is taken as the comparison result of the state data of that dimension with the corresponding data threshold.

[0071] S304, when the comparison results of each state data meet the safety conditions, the device inputs the state data of multiple dimensions into the pre-trained state evaluation model.

[0072] As mentioned earlier, the comparison results of the state data for each dimension are used to indicate the relationship between the state data of that dimension and the corresponding data threshold. Here, when the comparison results of the state data for a dimension indicate that the state data of that dimension is not greater than the corresponding data threshold, it is determined that the comparison results of the state data for that dimension meet the safety conditions. When the comparison results of all state data meet the safety conditions, the state data of multiple dimensions are input into the pre-trained state evaluation model.

[0073] S305, the device uses a condition evaluation model to determine the level type of the working condition of the belt system.

[0074] In this embodiment of the disclosure, multiple different level types can be preset, such as emergency state, abnormal state, and relatively normal state. Specifically, the state evaluation model can determine the classification probability based on state data from multiple dimensions, and determine the level type of the belt system's working state according to the classification probability. For example, when the classification probability is in the range of (0.9, 1), the working state of the belt system is determined to be an emergency state; when the classification probability is in the range of (0.7, 0.9), the working state of the belt system is determined to be an abnormal state; and when the classification probability is in the range of (0.5, 0.7), the working state of the belt system is determined to be a relatively normal state.

[0075] The method for monitoring the status of a belt conveyor system provided in this disclosure utilizes a sensor array to acquire multi-dimensional status data of the belt conveyor system in real time. By comparing the real-time acquired multi-dimensional status data with corresponding data thresholds, it quickly determines whether the operating status of the belt conveyor system is abnormal, thus enabling timely detection of abnormal problems. If an abnormality is not determined, the multi-dimensional status data can be further input into a pre-trained status evaluation model. The model determines the level of the belt conveyor system's operating status, providing a more accurate understanding of the current operating status and facilitating proactive countermeasures. Therefore, this method can efficiently acquire the operating status of the belt conveyor system, promptly detect abnormal problems, and provide a more accurate understanding of its current operating status, achieving refined management of the belt conveyor system's status.

[0076] In some embodiments, inputting state data of multiple dimensions into a pre-trained state evaluation model includes: performing regularization and alignment processing on the state data of each dimension; and inputting the data processing results of the state data of multiple dimensions into the pre-trained state evaluation model.

[0077] Combination Figure 4 As shown in the embodiments of this disclosure, another method for monitoring the condition of a belt system is provided, including the following steps:

[0078] S401, the device uses a sensor array to acquire multi-dimensional status data of the belt system.

[0079] S402, the device determines the comparison result between the state data of each dimension and the corresponding data threshold.

[0080] S403, if the comparison results of the status data in one or more dimensions do not meet the safety conditions, the working status of the belt system is determined to be abnormal.

[0081] S404, when the comparison results of each state data meet the safety conditions, the device performs regularization and alignment processing on the state data of each dimension.

[0082] As mentioned earlier, the comparison results of the state data for each dimension are used to indicate the relationship between the state data of that dimension and the corresponding data threshold. Here, when the comparison result of the state data for a dimension indicates that the state data of that dimension is not greater than the corresponding data threshold, it is determined that the comparison result of the state data for that dimension meets the safety conditions. When the comparison results of all state data meet the safety conditions, the state data of each dimension is subjected to regularization and alignment processing.

[0083] In this embodiment of the disclosure, the regularization process for state data in one dimension includes: determining the maximum and minimum data in the state data of that dimension, calculating a first difference between the maximum and minimum data; for each state data in that dimension, calculating a second difference between the state data and the minimum data, and using the ratio of the second difference to the first difference as the regularization result for each state data.

[0084] In this embodiment of the disclosure, the regularization processing for a state data in one dimension can be calculated using the following formula 1:

[0085]

[0086] In Formula 1, X normal X represents the regularization result of state data in one dimension. i X represents a state data point in one dimension. max X represents the maximum value in the state data of one dimension. min This represents the minimum value in the state data of one dimension. It's understandable that after regularization of the state data in each dimension, the regularization result for each state data point is between 0 and 1.

[0087] In some embodiments, the alignment process for state data in one dimension includes: identifying the channel with the highest sampling frequency in that dimension, determining the maximum data in the state data of the channel with the highest sampling frequency; using the regularization result of the maximum data of the channel with the highest sampling frequency as an alignment standard, and adjusting the regularization results of the state data of other channels in that dimension based on the alignment standard.

[0088] In some embodiments, after adjusting the regularization results of the state data of other channels in this dimension based on the alignment criterion, blank positions may be generated in other channels. In this case, the regularization results of the state data that were previously located in the blank positions in the channel can be used to fill the blank positions.

[0089] S405, the device inputs the data processing results of the state data in multiple dimensions into the pre-trained state evaluation model.

[0090] In some embodiments, inputting the data processing results of state data in multiple dimensions into a pre-trained state evaluation model includes: truncating the data processing results of state data in each dimension into data units of a specified length, each data unit including the regularization results of state data in a specified number of channels of the corresponding dimension; and inputting multiple data units into the pre-trained state evaluation model. For example, a data unit can be represented as I0(m0, n0), where m0 is a specified length and n0 is a specified number.

[0091] S406, the device uses a condition evaluation model to determine the level type of the working condition of the belt system.

[0092] In this embodiment of the disclosure, multiple different level types can be preset, such as emergency state, abnormal state, and relatively normal state. Specifically, the state evaluation model can determine the classification probability based on the data processing results of state data from multiple dimensions, and determine the level type of the belt system's working state according to the classification probability. For example, when the classification probability is in the range of (0.9, 1), the working state of the belt system is determined to be an emergency state; when the classification probability is in the range of (0.7, 0.9), the working state of the belt system is determined to be an abnormal state; and when the classification probability is in the range of (0.5, 0.7), the working state of the belt system is determined to be a relatively normal state.

[0093] In this embodiment, the state evaluation model can be a deep neural network, such as a Long Short-Term Memory (LSTM) network or a Convolutional Neural Network (CNN). The state evaluation model is trained based on raw data, a set of positive samples, and a set of negative samples. The raw data consists of multi-dimensional state data collected by the sensor group when the belt conveyor system is operating normally. The set of positive samples is obtained by data augmentation of the raw data. The set of negative samples consists of multi-dimensional state data collected by the sensor group when the belt conveyor system is operating abnormally.

[0094] In this embodiment of the disclosure, there are multiple raw data sets, which can be represented as I(m,n), where m is the length of the raw data and n is the number of channels in the raw data. Combined with... Figure 5 As shown, a positive sample set is formed by performing data augmentation on multiple original datasets (such as shuffling the original data along the time dimension and adding random noise). The positive samples in the positive sample set can be represented as I′1(m,n), I′2(m,n), ..., I′ k (m,n). Furthermore, the sensor array collects multi-dimensional state data of the belt system when its operating state is abnormal. The set of state data collected under these conditions is considered a negative sample set, where each negative sample can be represented as N1(m,n), N2(m,n), ..., N... K (m,n).

[0095] Combination Figure 5As shown, sample pairs are constructed using the original data, a set of positive samples, and a set of negative samples. Each sample pair consists of one original data point, one positive sample, and one negative sample. These sample pairs are then input into the state evaluation model for training. The training process for the state evaluation model aims to maximize the probability of positive sample pairs (sample pairs with similarity) and minimize the probability of negative sample pairs (sample pairs with different categories). Specifically, the loss functions used in the training process of the state evaluation model include the contrastive loss function (or information entropy loss function) and the cross-entropy loss function to achieve the above objectives.

[0096] In this embodiment of the disclosure, the expression for the comparison loss function is shown in Formula 2:

[0097]

[0098] In Equation 2, L is the function value of the contrastive loss function, and z i z is the feature representation of the i-th original data. i+ It is its positive sample, z i-j It is a negative sample, sim(z) i ,z i+ ) is z i and z i+ Similarity measures between them (e.g., cosine similarity), sim((z i ,z i-j ) is z i and z i-j The similarity measure between them (e.g., cosine similarity), N is the number of channels in the data, k is the number of sample pairs, and τ is a temperature parameter used to control the shape of the contrastive loss function.

[0099] In this embodiment of the disclosure, supervised training of the state evaluation model is achieved by constraining the difference between the data and the label values ​​using the cross-entropy loss function. The expression of the cross-entropy loss function is shown in Equation 3:

[0100]

[0101] In Formula 3, It is the value of the cross-entropy loss function. To predict the label, y i This is a real label.

[0102] In some embodiments, when the operating state of the belt system is determined to be an emergency state, the state data for each dimension is narrowed down to its corresponding data threshold, and then the comparison result between the state data for each dimension and the corresponding narrowed data threshold is determined. Here, the comparison result of the state data for each dimension is used to indicate the magnitude relationship between the state data for that dimension and the corresponding narrowed data threshold. Based on the comparison results of the state data for each dimension, the target dimension whose state data is greater than the corresponding narrowed data threshold is determined, and the maintenance items of the belt system are located based on the target dimension. For example, if the temperature of the drive motor is greater than the corresponding narrowed first temperature threshold, the drive motor can be identified as a maintenance item of the belt system.

[0103] In this embodiment of the disclosure, for each dimension of state data, an average value of the state data for that dimension can be calculated, and this average value is compared with the corresponding reduced data threshold. The comparison result of the average value of the state data for that dimension and the corresponding reduced data threshold is taken as the comparison result of the state data for that dimension and the corresponding reduced data threshold.

[0104] In this embodiment of the disclosure, for each dimension of state data, the maximum value of the state data in that dimension can be calculated, and this maximum value is compared with the corresponding reduced data threshold. The comparison result of the maximum value of the state data in that dimension and the corresponding reduced data threshold is taken as the comparison result of the state data of that dimension and the corresponding reduced data threshold.

[0105] In some embodiments, when the working state of the belt conveyor system is determined to be an emergency state, a reduction ratio is determined based on the classification probability output by the state evaluation model. The state data for each dimension and its corresponding data threshold are then reduced based on this reduction ratio. Here, the reduction ratio is a positive number less than 1. The reduction ratio is negatively correlated with the classification probability, and the reduced data threshold is positively correlated with the reduction ratio. In other words, the smaller the classification probability, the larger the reduction ratio, and the larger the reduced data threshold.

[0106] Combination Figure 6 As shown in the figure, this disclosure provides a status monitoring device 600 for a belt system. The status monitoring device 600 includes a data acquisition module 601, a data comparison module 602, a first status evaluation module 603, and a second status evaluation module 604.

[0107] The data acquisition module 601 is configured to acquire multi-dimensional status data of the belt system using a group of sensors.

[0108] The data comparison module 602 is configured to determine the comparison results between the state data of each dimension and the corresponding data threshold.

[0109] The first state evaluation module 603 is configured to determine that the working state of the belt system is abnormal when the comparison results of state data in one or more dimensions do not meet the safety conditions.

[0110] The second state evaluation module 604 is configured to input state data of multiple dimensions into a pre-trained state evaluation model when the comparison results of each state data meet the safety conditions, and use the state evaluation model to determine the level type of the working state of the belt system.

[0111] The condition monitoring device for belt conveyor systems provided in this disclosure utilizes a sensor array to acquire multi-dimensional condition data of the belt conveyor system in real time. By comparing the real-time acquired multi-dimensional condition data with corresponding data thresholds, it quickly determines whether the operating state of the belt conveyor system is abnormal, thus enabling timely detection of abnormal problems. If an abnormality is not determined, the multi-dimensional condition data can be further input into a pre-trained condition evaluation model. The model determines the level of the belt conveyor system's operating state, providing a more accurate understanding of its current state and facilitating proactive countermeasures. Therefore, this method efficiently acquires the operating state of the belt conveyor system, promptly detects abnormal problems, and provides a more accurate understanding of its current operating state, achieving refined management of the belt conveyor system's condition.

[0112] In some embodiments, the multi-dimensional state data includes drive motor temperature, and / or reducer temperature, and / or drive motor vibration, and / or reducer vibration, and / or belt running speed, and / or belt tearing degree, and / or belt position offset degree, and / or belt slippage degree.

[0113] In some embodiments, the second state evaluation module 604 is configured to:

[0114] The state data for each dimension is regularized and aligned.

[0115] The data processing results of state data from multiple dimensions are input into a pre-trained state evaluation model.

[0116] In some embodiments, regularization processing for state data in one dimension includes:

[0117] Determine the maximum and minimum values ​​in the state data for this dimension, and calculate the first difference between the maximum and minimum values;

[0118] For each state data in this dimension, calculate the second difference between that state data and the minimum data, and use the ratio of the second difference to the first difference as the regularization result for each state data.

[0119] In some embodiments, alignment processing for state data in one dimension includes:

[0120] Identify the channel with the highest sampling frequency in this dimension, and determine the maximum data in the state data of the channel with the highest sampling frequency;

[0121] The regularization result of the maximum data of the channel with the highest sampling frequency is used as the alignment standard, and the regularization results of the state data of other channels in this dimension are adjusted based on this alignment standard.

[0122] In some embodiments, the second state evaluation module 604 is configured to:

[0123] The data processing results of the state data in each dimension are truncated into data units of a specified length. Each data unit includes the regularization results of the state data of a specified number of channels in the corresponding dimension.

[0124] Multiple data units are input into a pre-trained state evaluation model.

[0125] In some embodiments, the state evaluation model is trained based on raw data, a positive sample set, and a negative sample set; the raw data is multi-dimensional state data collected by the sensor group when the belt system is in normal working condition, the positive sample set is obtained by data augmentation of the raw data, and the negative sample set is multi-dimensional state data collected by the sensor group when the belt system is in abnormal working condition.

[0126] In some embodiments, the loss functions used in the training process of the state evaluation model include the contrastive loss function and the cross-entropy loss function.

[0127] Combination Figure 7 As shown, this disclosure provides a status monitoring device 700 for a belt conveyor system. The device 700 includes a processor 701 and a memory 702. Optionally, the device 700 may further include a communication interface 703 and a bus 704. The processor 701, communication interface 703, and memory 702 can communicate with each other via the bus 704. The communication interface 703 can be used for information transmission. The processor 701 can call logical instructions in the memory 702 to execute the status monitoring method for the belt conveyor system described in the above embodiment.

[0128] Furthermore, the logic instructions in the aforementioned memory 702 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium.

[0129] The memory 702, as a computer-readable storage medium, can be used to store software programs and computer-executable programs, such as program instructions / modules corresponding to the methods in the embodiments of this disclosure. The processor 701 executes functional applications and data processing by running the program instructions / modules stored in the memory 702, thereby implementing the status monitoring method for the belt system in the above embodiments.

[0130] The memory 702 may include a program storage area and a data storage area. The program storage area may store the operating system and application programs required for at least one function; the data storage area may store data created based on the use of the terminal device. Furthermore, the memory 702 may include high-speed random access memory and may also include non-volatile memory.

[0131] This disclosure provides a computer-readable storage medium storing computer-executable instructions configured to perform the above-described method for monitoring the status of a belt system.

[0132] The technical solutions of this disclosure can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes one or more instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in this disclosure. The aforementioned storage medium can be a non-transitory storage medium, such as a USB flash drive, external hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk, etc., and other media capable of storing program code.

[0133] The foregoing description and accompanying drawings fully illustrate embodiments of this disclosure to enable those skilled in the art to practice them. Other embodiments may include structural, logical, electrical, procedural, and other changes. The embodiments represent only possible variations. Individual components and functions are optional unless explicitly required, and the order of operation may vary. Parts and features of some embodiments may be included in or replace parts and features of other embodiments. Moreover, the terminology used in this application is for describing embodiments only and is not intended to limit the claims. As used in the description of embodiments and claims, the singular forms “a,” “an,” and “the” are intended to equally include the plural forms unless the context clearly indicates otherwise. Similarly, the term “and / or” as used in this application means including one or more of the associated listed items and all possible combinations thereof. Additionally, when used in this application, the term "comprise" and its variations "comprises" and / or "comprising" refer to the presence of stated features, integrals, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or groups thereof. Without further limitations, an element defined by the phrase "comprises a..." does not exclude the presence of other identical elements in the process, method, or apparatus that includes said element. In this document, each embodiment may focus on the differences from other embodiments, and similar or identical parts between embodiments can be referred to mutually. For methods, products, etc., disclosed in the embodiments, if they correspond to the method section disclosed in the embodiments, the relevant parts can be referred to the description of the method section.

[0134] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the embodiments of this disclosure. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0135] The methods and products (including but not limited to devices and equipment) disclosed in the embodiments herein can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of units may be merely a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the shown or discussed units may be through some interfaces, and the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the units may be selected to implement this embodiment according to actual needs. Furthermore, the functional units in the embodiments of this disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0136] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than that shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. In the descriptions corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different blocks may also occur in a different order than disclosed in the description, and sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. Each block in a block diagram and / or flowchart, and combinations of blocks in a block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

Claims

1. A method for monitoring the condition of a belt conveyor system, characterized in that, include: Utilize a sensor array to acquire multi-dimensional status data of the belt system; Determine the comparison results between the state data of each dimension and the corresponding data threshold; If the comparison results of status data in one or more dimensions do not meet the safety conditions, the working status of the belt system is determined to be abnormal. When the comparison results of each state data meet the safety conditions, the state data of each dimension are regularized and aligned. The data processing results of the state data of multiple dimensions are input into the pre-trained state evaluation model. The state evaluation model is used to determine the level type of the working state of the belt system. The alignment process for state data in one dimension includes: identifying the channel with the highest sampling frequency in that dimension, determining the maximum data in the state data of the channel with the highest sampling frequency; using the regularization result of the maximum data of the channel with the highest sampling frequency as the alignment standard, and adjusting the regularization results of the state data of other channels in that dimension based on the alignment standard.

2. The status monitoring method according to claim 1, characterized in that, The status data includes multiple dimensions such as drive motor temperature, and / or reducer temperature, and / or drive motor vibration, and / or reducer vibration, and / or belt running speed, and / or belt tearing degree, and / or belt position offset degree, and / or belt slippage degree.

3. The status monitoring method according to claim 1, characterized in that, Regularization processing for state data in one dimension includes: Determine the maximum and minimum values ​​in the state data for this dimension, and calculate the first difference between the maximum and minimum values; For each state data in this dimension, calculate the second difference between that state data and the minimum data, and use the ratio of the second difference to the first difference as the regularization result for each state data.

4. The status monitoring method according to claim 1, characterized in that, The data processing results of multi-dimensional state data are input into a pre-trained state evaluation model, including: The data processing results of the state data in each dimension are truncated into data units of a specified length. Each data unit includes the regularization results of the state data of a specified number of channels in the corresponding dimension. Multiple data units are input into a pre-trained state evaluation model.

5. The status monitoring method according to any one of claims 1 to 4, characterized in that, The state evaluation model is trained based on the original data, a set of positive samples, and a set of negative samples. The raw data consists of multi-dimensional state data collected by the sensor group when the belt system is operating normally. The positive sample set is obtained by data augmentation of the raw data, while the negative sample set consists of multi-dimensional state data collected by the sensor group when the belt system is operating abnormally.

6. The status monitoring method according to claim 5, characterized in that, The loss functions used in the training process of the state evaluation model include the contrastive loss function and the cross-entropy loss function.

7. A condition monitoring device for a belt conveyor system, characterized in that, include: The data acquisition module is configured to acquire multi-dimensional status data of the belt system using a sensor array; The data comparison module is configured to determine the comparison results between the state data of each dimension and the corresponding data threshold; The first state evaluation module is configured to determine that the working state of the belt system is abnormal when the comparison results of state data in one or more dimensions do not meet the safety conditions. The second state evaluation module is configured to perform regularization and alignment processing on the state data of each dimension when the comparison results of each state data meet the safety conditions, and input the data processing results of the state data of multiple dimensions into the pre-trained state evaluation model, and use the state evaluation model to determine the level type of the working state of the belt system. The alignment process for state data in one dimension includes: identifying the channel with the highest sampling frequency in that dimension, determining the maximum data in the state data of the channel with the highest sampling frequency; using the regularization result of the maximum data of the channel with the highest sampling frequency as the alignment standard, and adjusting the regularization results of the state data of other channels in that dimension based on the alignment standard.

8. A condition monitoring device for a belt conveyor system, comprising a processor and a memory storing program instructions, characterized in that, The processor is capable of executing the condition monitoring method for a belt system as described in any one of claims 1 to 6.

9. A condition monitoring system for a belt conveyor system, characterized in that, Includes a sensor array and a condition monitoring device for a belt system as described in claim 7 or 8; The status monitoring device is communicatively connected to each sensor in the sensor group, which is used to collect status data of the belt system in multiple dimensions.

10. A storage medium, characterized in that, The storage medium stores computer program instructions, which, when executed by a processor, perform the condition monitoring method for a belt system as described in any one of claims 1 to 6.

Images