Scraper operation state self-diagnosis and early warning system

Abstract

The invention relates to a scraper operation state self-diagnosis and early warning system, and belongs to the technical field of scraper operation monitoring. The system comprises a state acquisition module used for acquiring multiple types of state data of the scraper within a preset time period before the current moment; the feature analysis module is used for performing time-frequency domain decomposition processing on the state data of each type to obtain state features of the state data of each type; the weight assignment module is used for matching each state feature with a preset abnormal weight table to obtain a matched abnormal parameter type corresponding to each type of state feature and an initial weight of the matched abnormal parameter type; the abnormality judgment module is used for adjusting the initial weight of each matched abnormal parameter type to obtain a final weight, calculating the state value of each potential operation abnormality, and screening all operation abnormality states of the scraper at the current moment; and the early warning module is used for carrying out early warning on each abnormal operation state. According to the invention, accurate diagnosis and early warning of the operation state of the scraper conveyor can be realized.

Description

Technical Field

[0001] This invention relates to the field of scraper conveyor operation monitoring technology, and in particular to a scraper conveyor operation status self-diagnosis and early warning system. Background Technology

[0002] As a key piece of equipment for material conveying in industries such as mining and metallurgy, the stability of scraper conveyors directly affects production efficiency and safety. In traditional scraper conveyor operation and maintenance, the condition of the scraper conveyor is mainly determined by manual periodic inspections or monitoring of a single parameter.

[0003] However, manual inspection has obvious shortcomings. Due to the complex structure of the scraper conveyor, potential faults in key parts such as transmission components, chute interfaces and sprocket areas are difficult to detect intuitively through manual inspection. Moreover, the inspection cycle is fixed and cannot capture sudden abnormalities in the operation of the scraper conveyor in real time, resulting in delayed fault detection. Often, the fault is only detected when it has already caused significant damage, which increases maintenance costs and downtime.

[0004] Single-parameter monitoring methods typically only detect a single indicator such as vibration or temperature, which cannot comprehensively reflect the overall operating status of the scraper conveyor. When a scraper conveyor malfunctions, the change in a single parameter may not be significant, easily leading to missed or false diagnoses. For example, early wear of transmission components may be accompanied by abnormal vibration and temperature rise. Monitoring only vibration may not be able to detect potential temperature changes in time, and vice versa. This approach cannot achieve real-time, comprehensive, and accurate diagnosis and early warning of the scraper conveyor's operating status, and it is difficult to meet the needs of modern industry for intelligent maintenance of scraper conveyors. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a self-diagnosis and early warning system for the operating status of a scraper conveyor. The technical solution of this invention is as follows: A self-diagnosis and early warning system for the operating status of a scraper conveyor, comprising: The status acquisition module is used to collect multiple types of status data of the scraper conveyor during a preset time period before the current moment through multiple types of sensors pre-arranged on the scraper conveyor. The feature parsing module is used to perform time-frequency domain decomposition on each type of state data to obtain the state features of each type of state data. The weight assignment module is used to match each type of state feature with a preset anomaly weight table to obtain the matching anomaly parameter type and its initial weight corresponding to each type of state feature; the preset anomaly weight table stores the correspondence between anomaly parameter types and initial weights. The anomaly detection module is used to adjust the initial weight of each matching anomaly parameter type according to the state characteristics of each matching anomaly parameter type to obtain the final weight, and to calculate the state value of each potential operating anomaly according to each matching anomaly parameter type, its final weight, and state characteristics. Based on the state value of each potential operating anomaly, it filters all operating anomaly states of the scraper conveyor at the current moment from all potential operating anomalies. The early warning module is used to issue early warnings for each abnormal operation state based on the status value of each abnormal operation state of the scraper conveyor at the current moment.

[0006] Preferably, the feature parsing module includes: The energy distribution acquisition unit is used to decompose each type of state data within a preset time period into multiple unit time windows through a sliding time window, and obtain the time-frequency energy distribution of each unit time window according to the Fourier transform. The feature extraction unit is used to obtain the frequency features and time-domain features of the state data of each type at each time node based on the time-frequency energy distribution of each unit time window of each type. The frequency features and time-domain features of each type constitute the state features of the state data of each type.

[0007] Preferably, the weight assignment module includes: The matching unit is used to calculate the matching degree between the parameter type name of each type of state feature and the abnormal type name of each abnormal parameter type in the preset abnormal weight table using text semantic matching technology. The assignment unit is used to determine whether the matching degree between each parameter type name and each exception type name is greater than the matching degree threshold. If the matching degree between any parameter type name and any exception type name is greater than the matching degree threshold, it is determined that the parameter type name and the exception type name are successfully matched, and the exception parameter type and its initial weight corresponding to the exception type name are used as the matching exception parameter type and its initial weight of the state feature corresponding to the parameter type name.

[0008] Preferably, the matching unit includes: The vector splitting subunit is used to convert parameter type names and exception type names into state sentence vectors and exception sentence vectors respectively according to the language vector model; The matching degree calculation subunit is used to calculate the matching degree between each state sentence vector and each exception sentence vector.

[0009] Preferably, the matching degree calculation subunit calculates the matching degree MD(a,b) between the state sentence vector a and the abnormal sentence vector b using formula (1): (1); In formula (1), Let N(a) represent the Euclidean distance, log() represent the logarithmic function, N() represent the total number of counts, N(a) represent the number of elements in the state vector a, and N(b) represent the number of elements in the abnormal vector b.

[0010] Preferably, the anomaly detection module includes: The weight adjustment unit is used to calculate the distribution index of each matched anomaly parameter type based on the state characteristics of each matched anomaly parameter type, and adjust its initial weight based on the distribution index of each matched anomaly parameter type to obtain the final weight of each matched anomaly parameter type. The status value judgment unit is used to calculate the status value of each potential operating abnormality based on each matched abnormal parameter type and its final weight, status characteristics and the preset risk level of each potential operating abnormality, and to filter potential operating abnormalities with status values ​​higher than the abnormal status threshold from all potential operating abnormalities, as the current operating abnormal status of the scraper conveyor.

[0011] Preferably, the weight adjustment unit includes: An anomaly matrix construction sub-unit is used to construct an anomaly matrix based on the state characteristics of each matching anomaly parameter type; The distribution index calculation subunit is used to input the anomaly matrix into the vector space, calculate the vector distance of the state features of each matching anomaly parameter type in the anomaly matrix based on the vector space, and perform probability transformation on the vector distance to obtain the distribution index of each matching anomaly parameter type. The final weight calculation subunit is used to superimpose the distribution index of each matched abnormal parameter type onto the initial weight of each matched abnormal parameter type to obtain the activation weight of each matched abnormal parameter type. The activation weights of all matched abnormal parameter types are normalized to obtain the final weight of each matched abnormal parameter type.

[0012] Preferably, the distribution index calculation subunit calculates the vector distance D of the state features of any matching anomaly parameter type in the anomaly matrix based on the vector space using formula (2): (2); In formula (2), R represents the feature vector corresponding to the state feature of the matched anomaly parameter type, and X represents the anomaly matrix. Let T denote the Euclidean distance, and let T denote the matrix transpose. The sum of the eigenvectors corresponding to each type of state feature in the anomaly matrix is ​​represented by the sum of the state feature vectors, and E represents the standard matrix.

[0013] Preferably, the early warning module includes: Anomaly level judgment unit is used to judge the anomaly level of each abnormal operating state based on the state value of each abnormal operating state of the scraper conveyor at the current moment; The early warning unit is used to issue early warnings for each operational anomaly based on the anomaly level of each anomaly.

[0014] All of the above-mentioned optional technical solutions can be combined arbitrarily, and the present invention will not provide a detailed description of the structure after each combination.

[0015] By means of the above solution, the beneficial effects of the present invention are as follows: The status acquisition module collects multiple types of status data of the scraper conveyor, and the feature analysis module performs time-frequency domain decomposition processing on each type of status data. This ensures that the subsequent analysis of all abnormal operating states of the scraper conveyor is based on multi-dimensional indicators, which can avoid missed or false judgments.

[0016] By using the weight assignment module to match the abnormal parameter type and its initial weight for each type of state feature, and then adjusting the initial weight to obtain the final weight through the anomaly judgment module, all abnormal operating states of the scraper conveyor are determined based on each matched abnormal parameter type, its final weight, and state feature. An early warning module then issues an alert for each abnormal operating state. This provides an abnormal state analysis mechanism that can be flexibly adjusted according to real-time status, effectively improving the accuracy of judging abnormal operating states of the scraper conveyor. It enables real-time, comprehensive, and accurate diagnosis and early warning of the scraper conveyor, meeting the needs of modern industry for intelligent maintenance of scraper conveyors.

[0017] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, the preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the structure of a scraper conveyor operation status self-diagnosis and early warning system provided in an embodiment of the present invention. Detailed Implementation

[0019] The following examples are used to illustrate the present invention, but are not intended to limit the scope of the invention.

[0020] like Figure 1 As shown, this embodiment of the invention provides a self-diagnosis and early warning system for the operating status of a scraper conveyor, which includes: The status acquisition module is used to collect multiple types of status data of the scraper conveyor during a preset time period before the current moment through multiple types of sensors pre-arranged on the scraper conveyor. The feature parsing module is used to perform time-frequency domain decomposition on each type of state data to obtain the state features of each type of state data. The weight assignment module is used to match each type of state feature with a preset anomaly weight table to obtain the matching anomaly parameter type and its initial weight corresponding to each type of state feature; the preset anomaly weight table stores the correspondence between anomaly parameter types and initial weights. The anomaly detection module is used to adjust the initial weight of each matching anomaly parameter type according to the state characteristics of each matching anomaly parameter type to obtain the final weight, and to calculate the state value of each potential operating anomaly according to each matching anomaly parameter type, its final weight, and state characteristics. Based on the state value of each potential operating anomaly, it filters all operating anomaly states of the scraper conveyor at the current moment from all potential operating anomalies. The early warning module is used to issue early warnings for each abnormal operation state based on the status value of each abnormal operation state of the scraper conveyor at the current moment.

[0021] Specifically, the status acquisition module includes various types of sensors, such as temperature sensors, vibration sensors, current sensors, voltage sensors, position sensors, and strain sensors. Correspondingly, the various types of status data include temperature, vibration, current, voltage, position, and strain. The specific value of the preset time period can be set as needed; in this embodiment, a preset time period of 30 minutes is preferred. It should be noted that after acquiring status data through various sensors, time alignment processing is performed on the various types of status data to facilitate analysis.

[0022] In the feature parsing module, the time-frequency domain decomposition process analyzes the state data simultaneously in both time and frequency dimensions. Specifically, it extracts features from each type of state data in both time and frequency dimensions, obtains time-domain features and frequency features, and then combines them to obtain the state features for each type.

[0023] In the weight assignment module, the preset abnormal weight table is a table used to store the abnormal parameter types and their corresponding initial weights, as shown in Table 1.

[0024] Table 1 Preset Anomaly Weight Table Exception parameter types Initial weights abnormal tension 0.25 abnormal high temperature 0.20 abnormal current 0.15 Vibration deformation 0.12 Location anomaly 0.10 Voltage overload abnormality 0.09 Speed ​​abnormality fault 0.09

[0025] In the anomaly detection module, potential operational anomalies of the scraper conveyor include motor failure, belt failure, scraper failure, track damage, and abnormal vibration. The status value is a numerical value used to quantify the severity of a potential operational anomaly. The operational anomaly status is an abnormal operating state selected from all potential operational anomalies that meets certain criteria.

[0026] In one specific embodiment, the feature parsing module includes: The energy distribution acquisition unit is used to decompose each type of state data within a preset time period into multiple unit time windows through a sliding time window, and obtain the time-frequency energy distribution of each unit time window according to the Fourier transform. The feature extraction unit is used to obtain the frequency features and time-domain features of the state data of each type at each time node based on the time-frequency energy distribution of each unit time window of each type. The frequency features and time-domain features of each type constitute the state features of the state data of each type.

[0027] Specifically, in the energy distribution acquisition unit, the step size of the sliding time window can be set as needed, for example, a step size of 3 seconds. After determining the step size of the sliding time window, the state data of each type within the preset time period is decomposed into multiple non-overlapping unit time windows according to the step size. Then, the data of each unit time window is processed by applying Fast Fourier Transform to obtain the frequency domain signal of each unit time window. Next, Short Time Fourier Transform is applied to each unit time window to obtain the time domain signal of each unit time window. Combining the distribution of the frequency domain signal and the time domain signal of each unit time window, the time-frequency energy distribution of each unit time window is obtained.

[0028] In the feature extraction unit, the frequency characteristics of each unit time window are determined based on the time-frequency energy distribution of each unit time window, including maximum and minimum values, bandwidth, variance, and energy. The frequency characteristics of a certain unit time window are used as the frequency characteristics of each time node in that unit time window. The time domain characteristics of each unit time window are determined by the degree of energy fluctuation in the time-frequency energy distribution of each unit time window, including kurtosis, skewness, and time zero crossover rate. The time domain characteristics of a certain unit time window are used as the time domain characteristics of each time node in that unit time window.

[0029] In one specific embodiment, the weight assignment module includes: The matching unit is used to calculate the matching degree between the parameter type name of each type of state feature and the abnormal type name of each abnormal parameter type in the preset abnormal weight table using text semantic matching technology. The assignment unit is used to determine whether the matching degree between each parameter type name and each exception type name is greater than the matching degree threshold. If the matching degree between any parameter type name and any exception type name is greater than the matching degree threshold, it is determined that the parameter type name and the exception type name are successfully matched, and the exception parameter type and its initial weight corresponding to the exception type name are used as the matching exception parameter type and its initial weight of the state feature corresponding to the parameter type name.

[0030] Specifically, in the matching unit, text semantic matching techniques include bag-of-words model, TF-IDF model, and cosine similarity, etc. In this embodiment of the invention, it is preferred to calculate the matching degree through cosine similarity. For example, taking Table 1 as an example, if a certain type of state feature is temperature type, that is, its parameter type name is "temperature", then the matching degree between "temperature" and each abnormal type name in the preset abnormal weight table, such as "tension abnormality", "high temperature abnormality", etc., is calculated.

[0031] In the assignment unit, the matching degree threshold is a threshold obtained based on empirical values. Taking the example above, if the matching degree between "temperature" and "high temperature anomaly" is greater than the matching degree threshold, then "temperature" and "high temperature anomaly" are determined to be a successful match, and the high temperature anomaly type corresponding to "high temperature anomaly" and its initial weight are used as the matching anomaly parameter type and its initial weight for the state feature corresponding to "temperature".

[0032] In one specific embodiment, the matching unit includes: The vector splitting subunit is used to convert parameter type names and exception type names into state sentence vectors and exception sentence vectors respectively according to the language vector model; The matching degree calculation subunit is used to calculate the matching degree between each state sentence vector and each exception sentence vector.

[0033] Specifically, in the vector splitting subunit, the language vector model is a model that converts text into vectors. It maps parameter type names and exception type names to a high-dimensional space, so that semantically similar texts also exhibit similarity in the vector space, thereby obtaining state sentence vectors and exception sentence vectors. Types of language vector models include Word2Vec, GloVe, and BERT.

[0034] In a specific embodiment, the matching degree calculation subunit calculates the matching degree MD(a,b) between the state sentence vector a and the abnormal sentence vector b using formula (1): (1); In formula (1), Let N(a) represent the Euclidean distance, log() represent the logarithmic function, N() represent the total number of counts, N(a) represent the number of elements in the state vector a, and N(b) represent the number of elements in the abnormal vector b.

[0035] Specifically, in formula (1) This represents the directional similarity between the state sentence vector a and the anomaly sentence vector b, calculated based on cosine similarity. The matching degree between state sentence vector a and abnormal sentence vector b in the vector dimension is represented by the above directional similarity and the matching degree in the vector dimension. The matching degree between state sentence vector a and abnormal sentence vector b is obtained by combining the above directional similarity and the matching degree in the vector dimension.

[0036] In one specific embodiment, the anomaly detection module includes: The weight adjustment unit is used to calculate the distribution index of each matched anomaly parameter type based on the state characteristics of each matched anomaly parameter type, and adjust its initial weight based on the distribution index of each matched anomaly parameter type to obtain the final weight of each matched anomaly parameter type. The status value judgment unit is used to calculate the status value of each potential operating abnormality based on each matched abnormal parameter type and its final weight, status characteristics and the preset risk level of each potential operating abnormality, and to filter potential operating abnormalities with status values ​​higher than the abnormal status threshold from all potential operating abnormalities, as the current operating abnormal status of the scraper conveyor.

[0037] Specifically, in the weight adjustment unit, the distribution index is used to represent the abnormal distribution trend of state feature values ​​that match the abnormal parameter type.

[0038] In the state value judgment unit, when calculating the state value SV of a potential operational anomaly based on each matched anomaly parameter type and its final weight, state characteristics, and the preset risk level of a potential operational anomaly, it is achieved through formula (3): (3); In formula (3), Wi represents the final weight of the i-th matching anomaly parameter type, and R represents the preset risk level of the potential operational anomaly. σi represents the maximum value of the frequency feature of the i-th matching anomaly parameter type, σi represents the variance of the frequency feature of the i-th matching anomaly parameter type, and n represents the number of matching anomaly parameter types.

[0039] The risk level preset for each potential operational anomaly is accurately defined based on expert experience, and the anomaly threshold is an operational anomaly threshold determined based on empirical values.

[0040] It should be noted that each potential operational anomaly has a pre-defined correspondence with multiple anomaly parameter types. For example, if a potential operational anomaly is a scraper failure, its corresponding anomaly parameter types are abnormal speed, abnormal load, and abnormal current, and are unrelated to temperature anomaly. Therefore, when calculating the state value of scraper failure, the final weight of temperature anomaly is set to 0.

[0041] In one specific embodiment, the weight adjustment unit includes: An anomaly matrix construction sub-unit is used to construct an anomaly matrix based on the state characteristics of each matching anomaly parameter type; The distribution index calculation subunit is used to input the anomaly matrix into the vector space, calculate the vector distance of the state features of each matching anomaly parameter type in the anomaly matrix based on the vector space, and perform probability transformation on the vector distance to obtain the distribution index of each matching anomaly parameter type. The final weight calculation subunit is used to superimpose the distribution index of each matched abnormal parameter type onto the initial weight of each matched abnormal parameter type to obtain the activation weight of each matched abnormal parameter type. The activation weights of all matched abnormal parameter types are normalized to obtain the final weight of each matched abnormal parameter type.

[0042] Specifically, in the anomaly matrix construction sub-unit, each row of the anomaly matrix represents a state feature that matches the anomaly parameter type, and each column represents the feature value of a state feature.

[0043] In the distribution index calculation subunit, the vector space refers to a multi-dimensional space to which the anomaly matrix is ​​mapped. In this multi-dimensional space, each element of the anomaly matrix is ​​represented by a vector. Non-linear processing in the vector space makes it easier to compare the matching anomaly parameter types within the anomaly matrix. Vector distance refers to the offset distance between the state features of each matching anomaly parameter type and the state features of other matching anomaly parameter types in the anomaly matrix. A larger vector distance between matching anomaly parameter types indicates a higher probability of the matching anomaly type being anomalous. During probability transformation, the Softmax function is typically used to probabilistically quantize all vector distances.

[0044] In the final weight calculation subunit, the normalization process adjusts the activation weights of all matching abnormal parameter types to a uniform range (usually between 0 and 1), and the sum of the final weights of all matching abnormal parameter types is 1.

[0045] In a specific embodiment, the distribution index calculation subunit calculates the vector distance D of a state feature of a certain matching anomaly parameter type in the anomaly matrix using formula (2): (2); In formula (2), R represents the feature vector corresponding to the state feature of the matched anomaly parameter type, and X represents the anomaly matrix. Let T denote the Euclidean distance, and let T denote the matrix transpose. The sum of the eigenvectors corresponding to each type of state feature in the anomaly matrix is ​​represented by the sum of the state feature vectors, and E represents the standard matrix.

[0046] Specifically, formula (2) calculates the vector distance by combining two parts, the first part... The second part is used to measure the energy distribution of the state features of this matched anomaly parameter type in the anomaly matrix. The standard matrix is ​​used to quantify the significance of the state features of the matched anomaly parameter type relative to the anomaly matrix. The standard matrix is ​​a matrix in which all elements are 1.

[0047] The feature vector corresponding to the state feature matching the abnormal parameter type is a vector obtained by concatenating the state features matching the abnormal parameter type according to the arrangement order of the state features in each column of the abnormal matrix. For example, if the frequency feature arrangement order of a certain state feature in the abnormal matrix is ​​0.5, 1.2, 0.8, and the time domain feature arrangement order is 0.7, 1.5, 1.0, then the feature vector corresponding to the state feature is [0.5, 1.2, 0.8, 0.7, 1.5, 1.0].

[0048] In one specific embodiment, the early warning module includes: Anomaly level judgment unit is used to judge the anomaly level of each abnormal operating state based on the state value of each abnormal operating state of the scraper conveyor at the current moment; The early warning unit is used to issue early warnings for each operational anomaly based on the anomaly level of each anomaly.

[0049] Specifically, in the anomaly level judgment unit, the anomaly level of the operation anomaly is usually divided into three types: severe, moderate, and mild. Each level has a state value range set according to experience. For example, the state value range of the severe anomaly level is [0.8, 1]. If the state value of a certain operation anomaly is 0.85, then the operation anomaly is determined to belong to the severe anomaly level.

[0050] In the early warning unit, the warning method varies depending on the level of abnormality. For example, the warning method for a mild abnormality is to display a blue warning sign on the display screen or the device control panel; the warning method for a moderate abnormality is to display an orange warning on the device screen, or to notify the operator by telephone, SMS, etc.; the warning method for a severe abnormality is for the system to issue a red alarm and notify all relevant personnel by emergency broadcast, SMS, telephone, etc.

[0051] Based on all the above embodiments, the self-diagnosis and early warning system for scraper conveyor operation status proposed in this invention has the following beneficial effects: First, the status acquisition module collects multiple types of status data of the scraper conveyor within a preset time period before the current moment through various sensors pre-deployed on the scraper conveyor. Then, the feature analysis module performs time-frequency domain decomposition processing on each type of status data to obtain the status features of each type of status data. This enables real-time acquisition of multi-dimensional status data of the scraper conveyor and the performance of time-frequency domain decomposition to obtain time-domain and frequency-domain status features, providing an accurate and real-time data foundation for subsequent abnormal operation status analysis of the scraper conveyor.

[0052] Next, the weight assignment module matches the state features of each type with the preset abnormal weight table to obtain the matching abnormal parameter type and its initial weight corresponding to each type of state feature. Then, the abnormal judgment module adjusts the initial weight of the state features of each matching abnormal parameter type to obtain the final weight. Based on each matching abnormal parameter type, its final weight, and the state features, the module calculates the state value of each potential operational abnormality. Based on the state value of each potential operational abnormality, the module filters all operational abnormalities of the scraper conveyor at the current moment from all potential operational abnormalities. This provides an abnormal state analysis mechanism that can be flexibly adjusted according to the real-time status, improving the accuracy and reliability of the abnormal analysis results of the scraper conveyor's operational status.

[0053] Finally, the early warning module provides early warnings for each abnormal operating state based on the current status value of each abnormal operating state of the scraper conveyor, enabling real-time, comprehensive, and accurate diagnosis and early warning of the scraper conveyor.

[0054] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. It should be noted that, for those skilled in the art, the specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and examples, without departing from the technical principles of the present invention. Furthermore, several improvements and modifications can be made under the given conditions, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A self-diagnosis and early warning system for the operating status of a scraper conveyor, characterized in that, include: The status acquisition module is used to collect multiple types of status data of the scraper conveyor during a preset time period before the current moment through multiple types of sensors pre-arranged on the scraper conveyor. The feature parsing module is used to perform time-frequency domain decomposition on each type of state data to obtain the state features of each type of state data. The weight assignment module is used to match each type of state feature with a preset anomaly weight table to obtain the matching anomaly parameter type and its initial weight corresponding to each type of state feature. The preset abnormal weight table stores the correspondence between abnormal parameter types and initial weights; The anomaly detection module is used to adjust the initial weight of each matching anomaly parameter type according to the state characteristics of each matching anomaly parameter type to obtain the final weight, and to calculate the state value of each potential operating anomaly according to each matching anomaly parameter type, its final weight, and state characteristics. Based on the state value of each potential operating anomaly, it filters all operating anomaly states of the scraper conveyor at the current moment from all potential operating anomalies. The early warning module is used to issue early warnings for each abnormal operation state based on the status value of each abnormal operation state of the scraper conveyor at the current moment.

2. The self-diagnosis and early warning system for the operating status of a scraper conveyor according to claim 1, characterized in that, The feature parsing module includes: The energy distribution acquisition unit is used to decompose each type of state data within a preset time period into multiple unit time windows through a sliding time window, and obtain the time-frequency energy distribution of each unit time window according to the Fourier transform. The feature extraction unit is used to obtain the frequency features and time-domain features of the state data of each type at each time node based on the time-frequency energy distribution of each unit time window of each type. The frequency features and time-domain features of each type constitute the state features of the state data of each type.

3. The self-diagnosis and early warning system for the operating status of a scraper conveyor according to claim 1, characterized in that, The weight assignment module includes: The matching unit is used to calculate the matching degree between the parameter type name of each type of state feature and the abnormal type name of each abnormal parameter type in the preset abnormal weight table using text semantic matching technology. The assignment unit is used to determine whether the matching degree between each parameter type name and each exception type name is greater than the matching degree threshold. If the matching degree between any parameter type name and any exception type name is greater than the matching degree threshold, it is determined that the parameter type name and the exception type name are successfully matched, and the exception parameter type and its initial weight corresponding to the exception type name are used as the matching exception parameter type and its initial weight of the state feature corresponding to the parameter type name.

4. The self-diagnosis and early warning system for the operating status of a scraper conveyor according to claim 3, characterized in that, The matching unit includes: The vector splitting subunit is used to convert parameter type names and exception type names into state sentence vectors and exception sentence vectors respectively according to the language vector model; The matching degree calculation subunit is used to calculate the matching degree between each state sentence vector and each exception sentence vector.

5. The self-diagnosis and early warning system for the operating status of a scraper conveyor according to claim 4, characterized in that, The matching degree calculation subunit calculates the matching degree MD(a,b) between the state sentence vector a and the abnormal sentence vector b using formula (1): （1）； In formula (1), Let N(a) represent the Euclidean distance, log() represent the logarithmic function, N() represent the total number of counts, N(a) represent the number of elements in the state vector a, and N(b) represent the number of elements in the abnormal vector b.

6. The self-diagnosis and early warning system for the operating status of a scraper conveyor according to claim 1, characterized in that, The anomaly detection module includes: The weight adjustment unit is used to calculate the distribution index of each matched anomaly parameter type based on the state characteristics of each matched anomaly parameter type, and adjust its initial weight based on the distribution index of each matched anomaly parameter type to obtain the final weight of each matched anomaly parameter type. The status value judgment unit is used to calculate the status value of each potential operating abnormality based on each matched abnormal parameter type and its final weight, status characteristics and the preset risk level of each potential operating abnormality, and to filter potential operating abnormalities with status values ​​higher than the abnormal status threshold from all potential operating abnormalities, as the current operating abnormal status of the scraper conveyor.

7. The self-diagnosis and early warning system for the operating status of a scraper conveyor according to claim 6, characterized in that, The weight adjustment unit includes: An anomaly matrix construction sub-unit is used to construct an anomaly matrix based on the state characteristics of each matching anomaly parameter type; The distribution index calculation subunit is used to input the anomaly matrix into the vector space, calculate the vector distance of the state features of each matching anomaly parameter type in the anomaly matrix based on the vector space, and perform probability transformation on the vector distance to obtain the distribution index of each matching anomaly parameter type. The final weight calculation subunit is used to superimpose the distribution index of each matched abnormal parameter type onto the initial weight of each matched abnormal parameter type to obtain the activation weight of each matched abnormal parameter type. The activation weights of all matched abnormal parameter types are normalized to obtain the final weight of each matched abnormal parameter type.

8. The self-diagnosis and early warning system for the operating status of a scraper conveyor according to claim 7, characterized in that, The distribution index calculation subunit calculates the vector distance D of the state features of any matching anomaly parameter type in the anomaly matrix based on the vector space using formula (2): （2）； In formula (2), R represents the feature vector corresponding to the state feature of the matched anomaly parameter type, and X represents the anomaly matrix. Let T denote the Euclidean distance, and let T denote the matrix transpose. The sum of the eigenvectors corresponding to each type of state feature in the anomaly matrix is ​​represented by the sum of the state feature vectors, and E represents the standard matrix.

9. The self-diagnosis and early warning system for the operating status of a scraper conveyor according to claim 1, characterized in that, The early warning module includes: Anomaly level judgment unit is used to judge the anomaly level of each abnormal operating state based on the state value of each abnormal operating state of the scraper conveyor at the current moment; The early warning unit is used to issue early warnings for each operational anomaly based on the anomaly level of each anomaly.

Images