Comprehensive protection system and method for CPD-I conveying equipment

Abstract

The invention discloses a comprehensive protection system and method for CPD-I conveying equipment, and relates to the technical field of automatic control. The CPD-I conveying equipment comprehensive protection method comprises the steps that driven wheel rotating speed, motor current and bearing temperature time sequence data of the CPD-I conveying equipment in the running process are obtained and input into a pre-trained multi-dimensional time sequence analysis model, and driven wheel rotating speed fluctuation, motor current heating and bearing temperature change characteristics are extracted; analyzing the rotating speed abnormity of the driven wheel, the current load of the motor and the temperature over-limit index of the bearing; whether the CPD-I conveying equipment is abnormal or not is judged; according to the method, the running trend of the equipment is dynamically tracked by utilizing the characteristics of the rotating speed change rate, the current thermal slope, the temperature slope and the like, and the abnormal trend in the state change can be identified and early warning can be timely given out before the serious physical abnormality of the equipment occurs.

Description

Technical Field

[0001] This invention relates to the field of automation control technology, specifically to a CPD-I integrated protection system and method for conveyor equipment. Background Technology

[0002] With the continuous improvement of industrial automation, conveying equipment plays a crucial role in material handling and processing. To ensure its long-term stable operation, continuous monitoring and protective control of its operating status are typically required. During operation, the motors, electrical components, and mechanical structures of conveying equipment can be affected by load fluctuations, environmental changes, or wear and tear, leading to operational abnormalities or even shutdowns. Therefore, data acquisition and operational status analysis of key parameters of conveying equipment have gradually become an important part of field control systems. Based on this, how to more effectively comprehensively evaluate key operating indicators and achieve early fault identification and response control has become a pressing issue in industrial settings.

[0003] Traditional methods for monitoring the operation of conveyor equipment mostly rely on static judgments of data at a single point in time or a single parameter (such as temperature, current, or speed), using threshold values ​​for fault identification. This method has two significant limitations:

[0004] First, there is a lack of in-depth analysis of the dynamic changes of parameters over time, and the timing characteristics of early faults are ignored. For example, in actual operation, short-term jumps in the driven wheel speed, the continuous rate of increase in current, or the slow cumulative increase in temperature are often early signs of equipment failure. However, these changes may be masked in single-point sampling or static averaging, making it difficult to identify anomalies in a timely manner.

[0005] Secondly, existing methods fail to achieve fusion modeling of multi-source data. The operating status of CPD-I conveying equipment is often affected by multiple factors such as mechanical, thermal, and electrical factors. Relying on only one indicator is prone to misjudgment or omission, and lacks the ability to identify anomalies in an overall manner. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a CPD-I integrated protection system and method for conveying equipment, which solves the problem that existing technologies are unable to identify potential abnormal trends during the operation of conveying equipment in a timely and accurate manner.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a comprehensive protection method for CPD-I conveyor equipment, comprising the following steps: acquiring time-series data of driven wheel speed, motor current, and bearing temperature during the operation of the CPD-I conveyor equipment, and preprocessing them; inputting the preprocessed time-series data of driven wheel speed, motor current, and bearing temperature into a pre-trained multi-dimensional time-series analysis model to extract the characteristics of driven wheel speed fluctuation, motor current heating, and bearing temperature change of the CPD-I conveyor equipment; based on the characteristics of driven wheel speed fluctuation, motor current heating, and bearing temperature change, analyzing the abnormal index of driven wheel speed, the motor current load index, and the bearing temperature over-limit index of the CPD-I conveyor equipment; determining whether the CPD-I conveyor equipment has experienced an abnormality based on the abnormal index of driven wheel speed, the motor current load index, and the bearing temperature over-limit index; and, upon determining a fault state, sending a fault alarm and performing a power-off protection operation on the CPD-I conveyor equipment.

[0008] Furthermore, the driven wheel speed timing data includes the driven wheel speed values ​​at several consecutive time points, the motor current timing data includes the motor current values ​​at several consecutive time points, and the bearing temperature timing data includes the bearing temperature values ​​at several consecutive time points.

[0009] Furthermore, the multidimensional time series analysis model includes a data window segmentation layer, a bidirectional gated loop layer, a self-attention layer, a feature fusion layer, and an output mapping layer.

[0010] Furthermore, the driven wheel speed fluctuation characteristics include the driven wheel speed range, driven wheel speed change rate, driven wheel speed fluctuation amplitude, and driven wheel speed jump count; the motor current heating characteristics include the motor current average value, motor current change rate, motor current integral value, and motor current thermal slope; and the bearing temperature change characteristics include the bearing temperature slope, bearing heating rate, and bearing temperature range.

[0011] Further, the specific steps for extracting the characteristics of driven wheel speed fluctuation, motor current heating, and bearing temperature change in the CPD-I conveyor are as follows: In the data window segmentation layer of the multidimensional time series analysis model, the driven wheel speed time series data, motor current time series data, and bearing temperature time series data are divided into multiple corresponding fixed-length time period sequences by sliding window processing; in the bidirectional gated loop layer of the multidimensional time series analysis model, forward and reverse state recursion processing is performed on the above fixed-length time period sequences respectively to obtain the bidirectional state output vectors of each type of data at different times; in the self-attention layer of the multidimensional time series analysis model, attention weights are assigned to each type of bidirectional state output vector in the time dimension; in the feature fusion layer of the multidimensional time series analysis model, the driven wheel speed state vector, motor current state vector, and bearing temperature state vector after attention processing are aligned and jointly mapped by channel to generate a fused time series joint feature representation; in the output mapping layer of the multidimensional time series analysis model, the joint feature representation is decoupled by feature channels and numerically restored to output the characteristics of driven wheel speed fluctuation, motor current heating, and bearing temperature change respectively.

[0012] Further, the specific steps for analyzing the abnormal driven wheel speed index of the CPD-I conveyor are as follows: Read the driven wheel speed range, driven wheel speed change rate, and driven wheel speed fluctuation amplitude of the CPD-I conveyor, and perform standardization processing; combine the standardized driven wheel speed range, driven wheel speed change rate, and driven wheel speed fluctuation amplitude with the corresponding driven wheel speed jump count for comprehensive analysis to obtain the abnormal driven wheel speed index of the CPD-I conveyor.

[0013] Further, the specific steps for analyzing the motor current load index of CPD-I conveying equipment are as follows: Read the mean motor current, motor current change rate, motor current integral value, and motor current thermal slope of the CPD-I conveying equipment, and perform standardization processing; Perform comprehensive analysis on the standardized mean motor current, motor current change rate, motor current integral value, and motor current thermal slope of the CPD-I conveying equipment to obtain the motor current load index of the CPD-I conveying equipment.

[0014] Further, the specific steps for analyzing the bearing temperature exceedance index of the CPD-I conveyor are as follows: read the bearing temperature slope, bearing heating rate, and bearing temperature range of the CPD-I conveyor and perform standardization processing; conduct a comprehensive analysis of the standardized bearing temperature slope, bearing heating rate, and bearing temperature range of the CPD-I conveyor to obtain the bearing temperature exceedance index of the CPD-I conveyor.

[0015] Furthermore, based on the abnormal driven wheel speed index, motor current load index, and bearing temperature over-limit index, the specific steps to determine whether the CPD-I conveyor equipment has malfunctioned are as follows: compare the abnormal driven wheel speed index, motor current load index, and bearing temperature over-limit index with the preset risk ranges for driven wheel speed, motor current, and bearing temperature, respectively; when any index is within its corresponding risk range, the CPD-I conveyor equipment is determined to have malfunctioned.

[0016] A comprehensive protection system for CPD-I conveyor equipment includes: a time-series data acquisition unit for acquiring and preprocessing time-series data of driven wheel speed, motor current, and bearing temperature during the operation of the CPD-I conveyor equipment; a multi-dimensional feature generation unit for inputting the preprocessed time-series data of driven wheel speed, motor current, and bearing temperature into a pre-trained multi-dimensional time-series analysis model to extract the characteristics of driven wheel speed fluctuation, motor current heating, and bearing temperature change of the CPD-I conveyor equipment; a state index analysis unit for analyzing the abnormal driven wheel speed index, motor current load index, and bearing temperature over-limit index of the CPD-I conveyor equipment based on the characteristics of driven wheel speed fluctuation, motor current heating, and bearing temperature change; an operational risk identification unit for determining whether the CPD-I conveyor equipment has experienced an abnormality based on the abnormal driven wheel speed index, motor current load index, and bearing temperature over-limit index; and a fault power-off protection unit for sending a fault alarm and performing power-off protection operations on the CPD-I conveyor equipment after determining a fault state.

[0017] The present invention has the following beneficial effects:

[0018] (1) The comprehensive protection method for CPD-I conveying equipment continuously collects and preprocesses the time-series data of driven wheel speed, motor current and bearing temperature, and combines it with a multi-dimensional time-series analysis model. It can effectively extract the fluctuation, heat generation and temperature rise characteristics that reflect the evolution of equipment status. During equipment operation, traditional solutions usually perform single-point monitoring based on fixed thresholds, which makes it difficult to capture the initial fluctuations in the abnormal development process in time, resulting in alarm lag and slow response. However, this method uses features such as speed change rate, current thermal slope and temperature slope to dynamically track the equipment operation trend. It can identify abnormal trends in status changes and issue early warnings in time before serious physical abnormalities occur in the equipment. At the same time, the system directly controls the power-off operation after identifying the fault status, which has a closed-loop capability from monitoring to response. It effectively improves the efficiency of fault handling and reduces the risk of equipment damage. It is suitable for CPD-I conveying equipment scenarios with high operational stability requirements.

[0019] (2) The CPD-I conveying equipment integrated protection method synchronously collects the driven wheel speed, motor current and bearing temperature data, and extracts their fluctuation, heating and temperature rise characteristics through a multi-dimensional time series analysis model. This avoids the problem of missing anomalies caused by dependence on a single data channel. In the feature extraction process, the model achieves full preservation and focus of various time series features through sliding window division, bidirectional recursive processing and self-attention mechanism, ensuring that key change information is not weakened. In the fusion stage, the feature vectors of different channels are aligned and jointly mapped to output a unified comprehensive feature representation, which can comprehensively evaluate the abnormal state with asynchronous performance. In the actual operating environment, the speed, current and temperature signals are affected by different interference conditions. A single signal may not be enough to reflect the complete equipment status. However, this method, through the multi-channel fusion mechanism, can still stably complete the anomaly identification through the other channels even if some signals are affected by noise, thus enhancing the adaptability to abnormal patterns and the fault tolerance of false alarms.

[0020] (3) The CPD-I comprehensive protection method for conveying equipment constructs an abnormal index, a load index, and an over-limit index for the driven wheel speed, motor current, and bearing temperature, respectively. In the data analysis stage, a calculation model is established through standardized feature values ​​to output a unified quantitative judgment index. This index structure has good adaptability and can adjust the risk range according to different equipment models or operating scenarios without modifying the model structure or feature extraction method. It is suitable for rapid deployment in diverse industrial environments. In the judgment stage, various indices are compared with the preset risk range to realize a flexible fault judgment mechanism, which can effectively cope with the benchmark drift problem caused by changes in external variables such as ambient temperature and power grid load.

[0021] (4) The CPD-I integrated protection system for conveying equipment divides data acquisition, feature generation, status analysis, risk identification and power outage protection into functional parts to form a clear logical structure. Each unit undertakes a specific processing task and can be deployed independently or linked in a unified manner, which improves the overall system's adaptability in the engineering environment. Especially in the later maintenance and functional upgrade, if it is necessary to change the analysis model or adjust the power outage logic, only the corresponding functional unit needs to be replaced. The original system architecture does not need to be significantly modified. In the actual scenario where the conveying equipment operates in a complex environment and the electrical and mechanical systems are highly coupled, this structure can significantly reduce the integration difficulty. In addition, after fault identification, the system directly drives the protection execution unit to complete the power outage control without relying on external intervention, realizing a closed-loop response mechanism, reducing the impact of human judgment or external communication delay on safety, and ensuring that the equipment can quickly stop operating after an anomaly occurs, avoiding further expansion of the damage range.

[0022] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0023] Figure 1 This is a flowchart of a comprehensive protection method for CPD-I conveying equipment according to the present invention.

[0024] Figure 2 This is a flowchart illustrating the specific steps involved in analyzing the abnormal rotational speed index of the driven wheel of a CPD-I conveyor in a comprehensive protection method for CPD-I conveyor equipment according to the present invention.

[0025] Figure 3 This is a block diagram of a CPD-I integrated protection system for conveying equipment according to the present invention. Detailed Implementation

[0026] Please see Figure 1 This invention provides a technical solution: a comprehensive protection method for CPD-I conveyor equipment, comprising the following steps: acquiring time-series data of driven wheel speed, motor current, and bearing temperature during the operation of the CPD-I conveyor equipment (measured by an external speed sensor, current transformer, and temperature sensor, respectively), and preprocessing the data; inputting the preprocessed time-series data of driven wheel speed, motor current, and bearing temperature into a pre-trained multi-dimensional time-series analysis model to extract the characteristics of driven wheel speed fluctuation, motor current heating, and bearing temperature change of the CPD-I conveyor equipment; analyzing the abnormal index of driven wheel speed, motor current load index, and bearing temperature over-limit index of the CPD-I conveyor equipment based on the characteristics of driven wheel speed fluctuation, motor current heating, and bearing temperature change; determining whether the CPD-I conveyor equipment has malfunctioned based on the abnormal index of driven wheel speed, motor current load index, and bearing temperature over-limit index; and, upon determining a fault state, sending a fault alarm and performing a power-off protection operation on the CPD-I conveyor equipment.

[0027] Preprocessing includes:

[0028] First, when removing outliers, the historical average value and fluctuation range are calculated for each type of time series data, and a reasonable threshold range is set as a reference standard. When the value of a data point deviates significantly from the threshold range, it is determined to be an outlier and removed from the original sequence to eliminate extreme value interference caused by occasional sensor errors or sudden environmental changes.

[0029] Subsequently, the data sequence remaining after removing extreme values ​​is checked for integrity. If missing data is found at certain time points, a missing data completion operation is performed. For missing data points between two normal sampling points, linear interpolation is used to estimate the missing value based on the trend between the two known data points. For missing points appearing at the beginning or end of the data, the value of the closest valid data point is used to fill in the missing data.

[0030] Meanwhile, the material level of the CPD-I conveyor can be monitored in real time by connecting an external blockage switch. When material accumulates in the equipment, the blockage switch is triggered, a blockage fault signal is output, and the conveyor stops, thus realizing the blockage protection function of the conveyor.

[0031] The deviation of the CPD-I conveyor can also be monitored in real time by connecting an external deviation switch. When the equipment deviates, a deviation fault signal is output to stop the conveyor and realize the deviation protection function of the conveyor.

[0032] It can also monitor the tearing condition of the conveyor belt of the CPD-I conveyor in real time by connecting an external tear switch. When the conveyor belt tears, a tear fault signal is output to stop the conveyor and realize the tear protection function of the conveyor.

[0033] If other types of switches or sensors are connected, they can be customized within the protection device.

[0034] Specifically, the driven wheel speed timing data includes the driven wheel speed values ​​at several consecutive time points, the motor current timing data includes the motor current values ​​at several consecutive time points, and the bearing temperature timing data includes the bearing temperature values ​​at several consecutive time points.

[0035] Specifically, by continuously collecting data at 20 time points with a sampling interval of 2 seconds, 40-second time-series data of driven wheel speed, motor current, and bearing temperature were constructed.

[0036] The multidimensional time series analysis model includes a data window segmentation layer, a bidirectional gated loop layer, a self-attention layer, a feature fusion layer, and an output mapping layer.

[0037] The model pre-training steps are as follows:

[0038] The operation data of several CPD-I conveying devices under normal and fault conditions were collected. The data included continuous time-series data of driven wheel speed, motor current and bearing temperature obtained at a sampling interval of 2 seconds during the operation of the device. Each set of data contained 20 consecutive time points, corresponding to a time span of 40 seconds.

[0039] In this set, normal state data and typical fault state data of known types are labeled as "normal" or "fault" respectively, serving as the label set required for supervised learning.

[0040] The collected raw time-series data were normalized, and multiple sets of fixed-length sample sequences were generated using the sliding window method. Each set of samples included the joint time series of three types of physical data. Each sample was accompanied by a corresponding label (e.g., normal, minor fault, severe fault) as the training target for the model.

[0041] A deep network structure is constructed, consisting of a data window segmentation layer, a bidirectional gated recurrent layer (Bi-GRU or Bi-LSTM), a self-attention layer, a feature fusion layer, and an output mapping layer. The parameters of each layer (such as the number of hidden units, the number of attention heads, and the mapping dimension) are initialized according to the experimental settings. For example, the bidirectional recurrent layer is set to 128 hidden units, the self-attention layer adopts a multi-head mechanism, and the feature fusion layer adopts a fully connected mapping to a 64-dimensional joint feature space.

[0042] The weighted cross-entropy loss function is used as the optimization objective, and the Adam optimizer is used for gradient updates. The model is trained iteratively for several rounds (e.g., 100 rounds) on the training set, and the convergence effect is monitored on the validation set. During training, a Dropout strategy is introduced to prevent overfitting, and an early stopping mechanism is used to avoid overtraining.

[0043] After model training, the output structure is changed from a classification output structure to a feature extraction structure, retaining the intermediate hidden layer and output mapping layer for subsequent extraction of three types of features: rotational speed fluctuation, current heating, and temperature change. The model no longer outputs fault labels, but instead outputs numerical representation vectors for each type of feature, serving as the basis for subsequent analysis.

[0044] The characteristics of driven wheel speed fluctuation include the driven wheel speed range, driven wheel speed change rate, driven wheel speed fluctuation amplitude, and driven wheel speed jump count. The characteristics of motor current heating include the motor current mean, motor current change rate, motor current integral value, and motor current thermal slope. The characteristics of bearing temperature change include the bearing temperature slope, bearing heating rate, and bearing temperature range.

[0045] The specific steps for extracting the characteristics of driven wheel speed fluctuation, motor current heating, and bearing temperature change of CPD-I conveyor equipment are as follows: In the data window segmentation layer of the multidimensional time series analysis model, the driven wheel speed time series data, motor current time series data, and bearing temperature time series data are divided into multiple corresponding fixed-length time series. Specifically, the sliding window length is set to 10 time points (i.e., 20 seconds) and the sliding step size is 5 time points (i.e., 10 seconds). Sliding window processing is performed on the driven wheel speed, motor current, and bearing temperature time series data respectively to obtain multiple non-overlapping or partially overlapping fixed-length time series. Each subsequence serves as the input unit for subsequent time series modeling.

[0046] In the bidirectional gated recurrent layer of the multidimensional time series analysis model, forward and backward state recursive processing is performed on the above fixed-length time series respectively to obtain the bidirectional state output vector of each type of data at different times. Specifically, each sliding window sequence is input into the bidirectional gated recurrent unit (BiGRU) for recursive modeling. The forward GRU extracts the state evolution trend from the start point to the end point of the sequence, and the backward GRU extracts the reverse logic of state change from the end point to the start point. The forward output vector and the backward output vector at each time point are concatenated to form the bidirectional state representation at that time point, reflecting the dynamic change characteristics of the input data in that time period.

[0047] In the self-attention layer of the multidimensional temporal analysis model, attention weights are assigned in the time dimension to various bidirectional state output vectors. Specifically, based on the bidirectional state vector at each time point, a scaling dot product attention mechanism is used to calculate the correlation between the current time point and other time points, generating a time-dimensional attention weight matrix. The state vectors at each time point are weighted and aggregated to form an attention convergence vector representing the global temporal features within the sliding window, thereby highlighting the contribution of key moments to the overall state evolution and improving the sensitivity of anomaly detection.

[0048] In the feature fusion layer of the multidimensional time series analysis model, the state vectors of the driven wheel speed, motor current, and bearing temperature after attention processing are aligned and jointly mapped to generate a fused time series joint feature representation. Specifically, the three types of attention convergence vectors are first expanded to make their dimensions consistent (e.g., unified to 128 dimensions); then the state representations of the three types of data are integrated by vector concatenation and input into a multilayer perceptron (MLP) for nonlinear mapping to extract the interaction relationship between features; finally, the fused comprehensive time series vector is obtained as a representation vector reflecting the multidimensional evolution characteristics of the equipment's operating state.

[0049] In the output mapping layer of the multidimensional time series analysis model, the joint feature representation is decoupled from the feature channels and numerically restored to output the features of driven wheel speed fluctuation, motor current heating and bearing temperature change respectively. Specifically, the fused integrated time series vector is split according to the physical category to obtain three feature sets of driven wheel speed fluctuation, motor current heating and bearing temperature change respectively.

[0050] For the characteristics of the driven wheel speed fluctuation, firstly calculate the difference between the maximum and minimum speed values ​​within each sliding window to obtain the speed range value; then calculate the difference between the speed at the end of the sequence and the speed at the beginning of the sequence divided by the window time to obtain the speed change rate; at the same time, average the adjacent speed differences within the window to obtain the speed fluctuation amplitude; finally, count the number of times the adjacent speed differences exceed the set threshold as the speed jump count.

[0051] To determine the heating characteristics of the motor current, the average value of the current sequence is first calculated to obtain the mean current. Then, the difference between the end current and the starting current is calculated and divided by the window time to obtain the current change rate. At the same time, the current values ​​are accumulated and summed within the window, taking into account the sampling interval, to obtain the current integral value. Finally, the trend of the square value of the current sequence changing with time is linearly fitted to obtain the motor current heating slope.

[0052] To determine the bearing temperature change characteristics, firstly, a linear fit is performed on time and temperature points to obtain the slope of the fitted line, which is used as the bearing temperature slope. Then, the difference between the end temperature and the starting temperature is divided by the window time to obtain the bearing heating rate. Finally, the difference between the highest temperature and the lowest temperature within the window is calculated to obtain the bearing temperature range.

[0053] In this implementation scheme, the continuous observation data of driven wheel speed, motor current, and bearing temperature are divided by a sliding window method. This divides the long-term series into multiple well-structured local segments, fully preserving the phased changes in the operating state. With the introduction of a bidirectional modeling mechanism, the forward evolution trend and backward backtracking characteristics of changes can be captured simultaneously within the same time period, enhancing the ability to identify asymmetric fluctuation processes. The self-attention module can perform weighted aggregation based on the response relationship between various time points within the time segment, amplifying and expressing weak but continuous trend changes, and avoiding the obscuring of key information by overall averaging. On this basis, the collected data are aligned and jointly encoded through a unified channel mapping method, enabling the expression of the coupled fluctuation relationship between multiple monitoring channels. Finally, the extracted numerical indicators, such as rate of change, range, and integral value, can accurately reflect the fluctuation trend and amplitude characteristics within a short period of time, thereby capturing abnormal signs before the equipment reaches the physical threshold.

[0054] Specifically, such as Figure 2 As shown, the specific steps for analyzing the abnormal drive wheel speed index of the CPD-I conveyor are as follows: Read the drive wheel speed range, drive wheel speed change rate, and drive wheel speed fluctuation amplitude of the CPD-I conveyor, and perform standardization processing (i.e., unit removal); Combine the standardized drive wheel speed range, drive wheel speed change rate, and drive wheel speed fluctuation amplitude with the corresponding drive wheel speed jump count for comprehensive analysis to obtain the abnormal drive wheel speed index of the CPD-I conveyor.

[0055] The specific formula for calculating the abnormal speed index of the driven wheel of the CPD-I conveyor is as follows: ;in, , , , , The following are, in order: driven wheel speed abnormality index, driven wheel speed fluctuation amplitude, driven wheel speed range, driven wheel speed jump count, and driven wheel speed change rate of the CPD-I conveyor equipment.

[0056] In this implementation scheme, by jointly analyzing data items such as the range of driven wheel speed, rate of change, fluctuation amplitude, and jump count, the dynamic fluctuation state during equipment operation can be effectively characterized. The index is constructed after standardization processing, which does not depend on specific equipment parameter ranges and helps to maintain consistent judgment under different operating conditions or models. This feature design can express the sudden changes, deviations, or unstable trends hidden in the original data with a unified index, thereby avoiding misjudgment caused by single-point numerical fluctuations. It also makes up for the insufficiency of relying solely on threshold triggering mechanisms to identify gradual risks. The overall calculation logic is closely based on the temporal fluctuation characteristics, which can reflect the abnormal performance of the driven wheel state earlier and more accurately.

[0057] Specifically, the steps for analyzing the motor current load index of CPD-I conveying equipment are as follows: Read the mean motor current, motor current change rate, motor current integral value, and motor current thermal slope of the CPD-I conveying equipment, and perform standardization processing (i.e., unit removal); Perform comprehensive analysis on the standardized mean motor current, motor current change rate, motor current integral value, and motor current thermal slope of the CPD-I conveying equipment to obtain the motor current load index of the CPD-I conveying equipment.

[0058] The specific formula for calculating the motor current load index of CPD-I transmission equipment is as follows: ;in, , , , , The values ​​are, in order, the motor current load index, the average motor current, the integral motor current, the thermal slope of the motor current, and the rate of change of the motor current for the CPD-I conveyor equipment.

[0059] In this implementation plan, by jointly analyzing the characteristics of motor current such as mean, rate of change, integral value, and thermal slope, the load bearing of the motor at different operating stages can be comprehensively reflected. The mean index reveals the overall distribution of the current level, the integral value reflects the cumulative trend of energy consumption, and the rate of change and thermal slope are sensitive to the fluctuation rate and temperature rise trend of the current load. After standardization, these data are used to construct a unified index to avoid interference caused by differences in dimensions or inconsistent numerical distribution. This helps to maintain the adaptability and comparability of the index under different scenarios. The overall analysis mechanism can identify abnormal behaviors such as long-term mild overload or frequent fluctuations before the current exceeds the safety limit, thereby providing a reliable basis for judging the equipment operating status and effectively improving the system's monitoring accuracy and response capability to operating stress.

[0060] Specifically, the steps for analyzing the bearing temperature exceedance index of CPD-I conveyor equipment are as follows: Read the bearing temperature slope, bearing heating rate, and bearing temperature range of CPD-I conveyor equipment, and perform standardization processing (i.e., unit removal); perform comprehensive analysis on the standardized bearing temperature slope, bearing heating rate, and bearing temperature range of CPD-I conveyor equipment to obtain the bearing temperature exceedance index of CPD-I conveyor equipment.

[0061] The specific formula for calculating the bearing temperature exceedance index of CPD-I conveyor equipment is as follows: ;in, , , , The parameters are, in order, the bearing temperature exceedance index, bearing heating rate, bearing temperature slope, and bearing temperature range of the CPD-I conveyor equipment.

[0062] In this implementation scheme, by analyzing the heating rate, slope, and range of bearing temperature, the characteristics of equipment state changes during the heating process can be revealed from multiple perspectives, including temperature growth rate, trend, and fluctuation amplitude. The heating rate reflects whether the temperature accumulation is too fast, the slope captures the continuous upward trend in the time series, and the range reflects the maximum fluctuation range within the observation period. By standardizing the above data and then comprehensively modeling it, the influence of the original numerical units and dimensions is effectively avoided, making it universal across time periods and different equipment. In actual operation, bearing failures often manifest as a slow temperature rise or local fluctuations in the early stages, before reaching the alarm threshold. This analysis method can identify these unstable trends in advance, issue timely risk warnings, and help reduce the risk of damage caused by temperature rise.

[0063] Specifically, based on the abnormal driven wheel speed index, the motor current load index, and the bearing temperature over-limit index, the specific steps to determine whether the CPD-I conveyor equipment has malfunctioned are as follows: The abnormal driven wheel speed index, the motor current load index, and the bearing temperature over-limit index are compared and analyzed with the preset risk ranges for driven wheel speed, motor current, and bearing temperature, respectively; when any index is within its corresponding risk range, the CPD-I conveyor equipment is determined to have malfunctioned.

[0064] In this implementation scheme, by mapping the driven wheel speed abnormality index, motor current load index, and bearing temperature over-limit index to their independently set risk ranges, it is possible to achieve unified identification of abnormal situations while maintaining the independence of the indicators. Each index is derived from a previously constructed time-series feature set, which can reflect the dynamic performance of speed fluctuations, current load, and temperature changes respectively. By comparing each index with the risk range item by item, the abnormal judgment mechanism can be triggered without relying on a single index threshold. When any index enters its risk range, potential faults can be identified immediately, effectively improving the identification coverage. This method avoids the dependence on all indicators exceeding the limits at the same time, enhances the system's response capability to atypical abnormal operating conditions, and is suitable for real-time status monitoring and protection needs in various operating scenarios.

[0065] Please see Figure 3 This invention provides a technical solution: a comprehensive protection system for CPD-I conveyor equipment, comprising: a time-series data acquisition unit for acquiring and preprocessing time-series data of driven wheel speed, motor current, and bearing temperature during the operation of the CPD-I conveyor; a multi-dimensional feature generation unit for inputting the preprocessed time-series data of driven wheel speed, motor current, and bearing temperature into a pre-trained multi-dimensional time-series analysis model to extract the characteristics of driven wheel speed fluctuation, motor current heating, and bearing temperature change of the CPD-I conveyor; a state index analysis unit for analyzing the abnormal driven wheel speed index, motor current load index, and bearing temperature over-limit index of the CPD-I conveyor based on the characteristics of driven wheel speed fluctuation, motor current heating, and bearing temperature change; an operational risk identification unit for determining whether the CPD-I conveyor has experienced an abnormality based on the abnormal driven wheel speed index, motor current load index, and bearing temperature over-limit index; and a fault power-off protection unit for sending a fault alarm and performing power-off protection operation on the CPD-I conveyor after determining a fault state.

[0066] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the invention.

[0067] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A comprehensive protection method for CPD-I delivery equipment, characterized in that, The method comprises the following steps: Obtain the time series data of the driven wheel speed, motor current and bearing temperature of the CPD-I conveying device during operation, and preprocess them; Input the preprocessed time series data of the driven wheel speed, motor current and bearing temperature into the pre-trained multi-dimensional time series analysis model respectively, and extract the driven wheel speed fluctuation, motor current heating and bearing temperature change characteristics of the CPD-I conveying device; Based on the driven wheel speed fluctuation, motor current heating and bearing temperature change characteristics, analyze the driven wheel speed anomaly index, motor current load index and bearing temperature overrun index of the CPD-I conveying device respectively; Based on the driven wheel speed anomaly index, motor current load index and bearing temperature overrun index, determine whether the CPD-I conveying device is abnormal; After determining the fault state, send a fault alarm and perform power-off protection operation on the CPD-I conveying device.

2. The CPD-I delivery apparatus integrated protection method of claim 1, wherein, The driven wheel speed time series data includes driven wheel speed values at a plurality of continuous time points, the motor current time series data includes motor current values at a plurality of continuous time points, and the bearing temperature time series data includes bearing temperature values at a plurality of continuous time points.

3. The CPD-I delivery apparatus integrated protection method of claim 1, wherein, The multi-dimensional time series analysis model includes a data window segmentation layer, a bidirectional gate recurrent layer, a self-attention layer, a feature fusion layer and an output mapping layer.

4. The CPD-I delivery apparatus integrated protection method of claim 1, wherein, The driven wheel speed fluctuation characteristics include driven wheel speed range value, driven wheel speed change rate, driven wheel speed fluctuation amplitude, driven wheel speed jump count, the motor current heating characteristics include motor current mean value, motor current change rate, motor current integral value, motor current heat slope, and the bearing temperature change characteristics include bearing temperature slope, bearing temperature rise rate and bearing temperature range value.

5. The CPD-I delivery apparatus integrated protection method of claim 3, wherein, The specific steps of extracting the driven wheel speed fluctuation, motor current heating and bearing temperature change characteristics of the CPD-I conveying device are as follows: In the data window segmentation layer of the multi-dimensional time series analysis model, perform sliding window division processing on the driven wheel speed time series data, motor current time series data and bearing temperature time series data to obtain a plurality of corresponding fixed-length time period sequences; In the bidirectional gate recurrent layer of the multi-dimensional time series analysis model, perform forward and reverse state recursion processing on the above fixed-length time period sequences respectively to obtain bidirectional state output vectors of each type of data at different time points; In the self-attention layer of the multi-dimensional time series analysis model, perform attention weight distribution on the time dimension of each bidirectional state output vector; In the feature fusion layer of the multi-dimensional time series analysis model, align and jointly map the driven wheel speed state vector, motor current state vector and bearing temperature state vector after attention processing to generate a fused time series joint feature representation; In the output mapping layer of the multi-dimensional time series analysis model, decouple the feature channels and restore the numerical values of the joint feature representation, and output the driven wheel speed fluctuation, motor current heating and bearing temperature change characteristics respectively.

6. The CPD-I delivery apparatus integrated protection method of claim 4, wherein, The specific steps of analyzing the driven wheel speed anomaly index of the CPD-I conveying device are as follows: Read the driven wheel speed range value, driven wheel speed change rate and driven wheel speed fluctuation amplitude of the CPD-I conveying device, and perform standardization processing; The driven wheel rotation speed abnormality index of the CPD-I conveying equipment is obtained by comprehensively analyzing the driven wheel rotation speed range difference, the driven wheel rotation speed change rate, the driven wheel rotation speed fluctuation amplitude of the CPD-I conveying equipment after standardization processing, and the corresponding driven wheel rotation speed jump count.

7. The CPD-I delivery apparatus integrated protection method of claim 4, wherein, The specific steps for analyzing the motor current load index of the CPD-I conveying equipment are as follows: Read the motor current mean value, the motor current change rate, the motor current integral value, and the motor current thermal slope of the CPD-I conveying equipment, and perform standardization processing; Comprehensively analyze the motor current mean value, the motor current change rate, the motor current integral value, and the motor current thermal slope of the CPD-I conveying equipment after standardization processing, and obtain the motor current load index of the CPD-I conveying equipment.

8. The CPD-I delivery apparatus integrated protection method of claim 4, wherein, The specific steps for analyzing the bearing temperature overrun index of the CPD-I conveying equipment are as follows: Read the bearing temperature slope, the bearing temperature rise rate, and the bearing temperature range difference of the CPD-I conveying equipment, and perform standardization processing; Comprehensively analyze the bearing temperature slope, the bearing temperature rise rate, and the bearing temperature range difference of the CPD-I conveying equipment after standardization processing, and obtain the bearing temperature overrun index of the CPD-I conveying equipment.

9. The CPD-I delivery apparatus integrated protection method of claim 1, wherein, Based on the driven wheel rotation speed abnormality index, the motor current load index, and the bearing temperature overrun index, the specific steps for determining whether the CPD-I conveying equipment is abnormal are as follows: Compare and analyze the driven wheel rotation speed abnormality index, the motor current load index, and the bearing temperature overrun index with the preset driven wheel rotation speed risk interval, the motor current risk interval, and the bearing temperature risk interval, respectively; When any index is in its corresponding risk interval, it is determined that the CPD-I conveying equipment has failed.

10. A CPD-I delivery device integrated protection system, applying the CPD-I delivery device integrated protection method of any one of claims 1-9, characterized in that, It includes: A time series data acquisition unit for acquiring the driven wheel rotation speed, motor current, and bearing temperature time series data of the CPD-I conveying equipment during operation and performing preprocessing; A multi-dimensional feature generation unit for inputting the preprocessed driven wheel rotation speed, motor current, and bearing temperature time series data into a pre-trained multi-dimensional time series analysis model, respectively, to extract the driven wheel rotation speed fluctuation, motor current heating, and bearing temperature change characteristics of the CPD-I conveying equipment; A state index analysis unit for analyzing the driven wheel rotation speed abnormality index, the motor current load index, and the bearing temperature overrun index of the CPD-I conveying equipment based on the driven wheel rotation speed fluctuation, the motor current heating, and the bearing temperature change characteristics; An operation risk identification unit for determining whether the CPD-I conveying equipment is abnormal based on the driven wheel rotation speed abnormality index, the motor current load index, and the bearing temperature overrun index; A fault power-off protection unit for sending a fault alarm and performing power-off protection operation on the CPD-I conveying equipment after determining the fault state.

Images