Intelligent material conveying deviation remote monitoring system based on Internet of Things

Abstract

The invention discloses an intelligent material conveying deviation remote monitoring system based on Internet of Things, which relates to the technical field of material conveying deviation remote monitoring and comprises a signal monitoring module, an abnormal window calibration module, a distortion evaluation module, a data correction module and a remote optimization module. The distortion evaluation module is used for acquiring ultrasonic propagation distortion characteristic information in a signal propagation abnormal window in real time, analyzing and evaluating the echo signal distortion degree under the condition of ultrasonic signal propagation path distortion after acquisition, and dividing the echo signal distortion degree into three types, namely a light distortion degree, a medium distortion degree and a heavy distortion degree; and the data correction module is used for taking corresponding correction measures on the to-be-corrected measurement data under the condition of different echo signal distortion degrees based on the evaluation result. According to the method, the problem of measurement errors caused by propagation path distortion of ultrasonic signals in conveyor belt deviation monitoring is solved, accurate distortion evaluation and adaptive data correction are realized, and the monitoring precision and stability are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of remote monitoring of material conveying deviation, and particularly to an intelligent remote monitoring system for material conveying deviation based on the Internet of Things. Background Art

[0002] Material conveying refers to the process of transporting solid, liquid or gaseous materials from one place to another through conveyor belts, pipelines, track systems or other conveying equipment in the fields of industrial production, logistics transportation, mining, etc. This process plays a crucial role in modern manufacturing and automated production systems. However, during the material conveying process, due to mechanical wear, uneven loads, environmental changes or equipment failures, the conveying system often experiences problems such as material deviation, such as conveyor belt deviation, uneven material distribution in pipelines, and track transportation position deviation. These deviation phenomena not only reduce the conveying efficiency, resulting in material waste, but may also cause equipment damage and even production accidents. Therefore, how to accurately monitor and timely adjust the deviation problem of material conveying has become an important technical challenge. Traditional deviation monitoring methods usually rely on manual inspections or local detections by single sensors, which have problems such as lag, high cost, and limited coverage, and are difficult to meet the monitoring requirements of large-scale, long-distance, and high-precision. Therefore, an intelligent remote monitoring method for deviation based on the Internet of Things has emerged. The introduction of Internet of Things technology enables the conveying system to collect deviation data in real time through distributed sensors and remotely transmit it to the cloud or control center via wireless communication technology. Combining big data analysis and artificial intelligence algorithms, it can accurately identify the trend of conveying deviation, realize automatic alarm, intelligent adjustment and even predictive maintenance, ensure the efficient and stable operation of the conveying process, thereby reducing operation and maintenance costs and improving production safety and intelligent level.

[0003] The existing Internet of Things (IoT)-based intelligent material conveying deviation remote monitoring technology mainly relies on core technologies such as multi-sensor fusion, wireless communication, cloud computing, and artificial intelligence analysis to achieve efficient monitoring and intelligent management of the conveying system. Specifically, the system usually deploys high-precision sensors, such as laser displacement sensors, inertial measurement units (IMUs), photoelectric detection devices, or pressure sensors, at key positions on the conveyor belt, pipeline, or track to collect displacement data, vibration conditions, pressure distribution, and material movement trajectories of the conveying system in real time. These data are then transmitted to a remote cloud server or edge computing node through wireless communication technologies (such as Wi-Fi, LoRa, NB-IoT, or 5G). In the data center, the system uses artificial intelligence and big data analysis technologies to perform real-time calculations on the collected information to determine whether there is an abnormal deviation in the conveying system. Once a deviation exceeding the preset threshold is detected, the system can send an alarm notification (such as a text message, APP push, or email) to the operation and maintenance personnel through a remote control platform. Moreover, it can even adjust the operating parameters of the conveying equipment in combination with intelligent control algorithms, such as automatically adjusting the conveyor belt tension, changing the material feeding angle, or adjusting the track direction, to achieve self-correction. In addition, the system can also perform trend prediction based on historical data to detect potential faults in advance, prevent equipment damage or production interruption caused by conveying deviation, thereby improving the stability of the conveying system, reducing maintenance costs, and enhancing overall production efficiency.

[0004] The existing technology has the following deficiencies:

[0005] In the conveyor belt systems of large mines, steel plants, and cement plants, when the conveyor belt deviates, ultrasonic sensors installed at the edge are used to detect the deviation situation in real time. However, due to the irregular structures such as rubber skirts, metal baffles, and roller brackets at the edge of the conveyor belt, the propagation path of ultrasonic signals is distorted, resulting in scattering, multiple reflections, or occlusion, and thus causing the echo signal to be distorted. However, the existing IoT-based intelligent material conveying deviation remote monitoring technology cannot correct the measurement data according to the degree of echo distortion under the condition of ultrasonic signal propagation path distortion, but directly uses the data returned by the sensor for deviation calculation, failing to correct the measurement error caused by signal distortion. This may lead to misjudgment of the conveyor belt deviation state, trigger unnecessary alarms or adjustments, cause fluctuations in the conveyor belt tension, affect the operation of the equipment. At the same time, the ranging error may also mask the real deviation problem, increase the risk of conveyor belt wear, even cause tearing or breaking accidents, and affect the judgment of the operation and maintenance personnel on the conveyor belt state, resulting in delayed or excessive maintenance and increased operation and maintenance costs.

[0006] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present disclosure. Therefore, it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0007] The object of the present invention is to provide an intelligent material conveying deviation remote monitoring system based on the Internet of Things to solve the problems in the above-mentioned background technology.

[0008] To achieve the above object, the present invention provides the following technical solution: An intelligent material conveying deviation remote monitoring system based on the Internet of Things, including a signal monitoring module, an abnormal window calibration module, a distortion evaluation module, a data correction module, and a remote optimization module;

[0009] The signal monitoring module collects ultrasonic signals through ultrasonic sensors deployed on both sides of the conveyor belt and monitors the propagation characteristics of the ultrasonic signals in real time to analyze whether the propagation path of the ultrasonic signals is distorted;

[0010] The abnormal window calibration module, in the case where the analysis result is that the propagation path of the ultrasonic signal is distorted, determines the time period during which the propagation path of the ultrasonic signal is distorted, calibrates it as a signal propagation abnormal window, and calibrates all measurement data collected by the ultrasonic sensor within the signal propagation abnormal window as measurement data to be corrected;

[0011] The distortion evaluation module obtains the ultrasonic propagation distortion characteristic information within the signal propagation abnormal window in real time, analyzes it after obtaining, evaluates the echo signal distortion degree in the case of the distortion of the ultrasonic signal propagation path, and classifies it into three categories: mild distortion degree, moderate distortion degree, and severe distortion degree;

[0012] The data correction module takes corresponding correction measures for the measurement data to be corrected in the case of different echo signal distortion degrees based on the evaluation result;

[0013] The remote optimization module remotely transmits the corrected measurement data based on Internet of Things remote monitoring and adaptive optimization, and dynamically optimizes the calibration strategy of the signal propagation abnormal window and the measurement data correction method in combination with historical data.

[0014] Preferably, in the abnormal window calibration module, in the case where the analysis result is that the propagation path of the ultrasonic signal is distorted, by analyzing the continuous time series change rate of the ultrasonic measurement data and comparing it with the duration threshold of the abnormal signal, the time period during which the propagation path of the ultrasonic signal is distorted is determined, and this time period is calibrated as the signal propagation abnormal window;

[0015] Within the signal propagation abnormal window, by matching the timestamps of the measurement data and combining the relevance between the spatial position of the measurement data and the signal propagation path, all ultrasonic sensor measurement data within this time window are screened and calibrated as measurement data to be corrected.

[0016] Preferably, in the distortion evaluation module, the ultrasonic propagation distortion characteristic information within the signal propagation anomaly window is obtained in real time and preprocessed after acquisition; the spectral amplitude dynamic characteristic information and the energy amplitude distribution characteristic information are extracted from the preprocessed ultrasonic propagation distortion characteristic information, and analyzed to generate a spectral envelope distortion index and an amplitude distribution stability coefficient respectively; a distortion degree analysis model is constructed for the generated spectral envelope distortion index and amplitude distribution stability coefficient, and a distortion evaluation coefficient is generated through weighted summation; a preset distortion evaluation coefficient threshold interval is determined, and after determination, it is compared with the generated distortion evaluation coefficient. According to the comparison result, the distortion degree of the echo signal in the case of ultrasonic signal propagation path distortion is evaluated and classified into three categories: mild distortion degree, moderate distortion degree, and severe distortion degree.

[0017] Preferably, the acquisition logic of the spectral envelope distortion index is as follows:

[0018] The spectral amplitude dynamic characteristic information is extracted from the preprocessed ultrasonic propagation distortion characteristic information, specifically including the instantaneous amplitude of the echo signal received by the ultrasonic sensor at different times within the signal propagation anomaly window and the frequency offset of the ultrasonic signal at the receiving end, and are respectively calibrated as and , represents the instantaneous amplitude of the echo signal received by the ultrasonic sensor at time within the signal propagation anomaly window, represents the frequency offset of the ultrasonic signal at the receiving end at time within the signal propagation anomaly window, , is a positive integer;

[0019] The instantaneous amplitudes of the echo signals received by the ultrasonic sensor at different times within the signal propagation anomaly window are constructed into a set, and the maximum value within the set is calibrated as ;

[0020] The frequency offsets of the ultrasonic signals at the receiving end at different times within the signal propagation anomaly window are sorted according to the numerical size, and the median is calibrated as ;

[0021] Calculate the standard deviation of the frequency offsets of the ultrasonic signals at the receiving end at different times within the signal propagation anomaly window, according to the formula: ;

[0022] Calculate the spectral envelope distortion index, and the specific calculation formula is as follows:

[0023]

[0024] In the formula, is the spectral envelope distortion index.

[0025] Preferably, the acquisition logic of the amplitude distribution stability coefficient is as follows:

[0026] Extract the energy amplitude distribution characteristic information from the preprocessed ultrasonic propagation distortion characteristic information, specifically including the amplitude change rate of the echo signal received by the ultrasonic sensor at different times within the signal propagation abnormal window, the power of the ultrasonic signal, and the ratio of the maximum amplitude to the root mean square amplitude of the ultrasonic signal within the signal propagation abnormal window, and calibrate them as , and , represents the amplitude change rate of the echo signal received by the ultrasonic sensor at time within the signal propagation abnormal window, represents the power of the ultrasonic signal at time within the signal propagation abnormal window, represents the ratio of the maximum amplitude to the root mean square amplitude of the ultrasonic signal within the signal propagation abnormal window, , is a positive integer;

[0027] Construct a set of the powers of the ultrasonic signals at different times within the signal propagation abnormal window, and calibrate the maximum value within the set as ;

[0028] Calculate the amplitude distribution stability coefficient, and the specific calculation formula is as follows:

[0029]

[0030] In the formula, is the amplitude distribution stability coefficient.

[0031] Preferably, construct a distortion degree analysis model for the generated spectral envelope distortion index and the amplitude distribution stability coefficient , and generate a distortion evaluation coefficient through weighted summation. The specific calculation formula is as follows:

[0032]

[0033] In the formula, is the distortion evaluation coefficient, and are the non-zero weight coefficients of the spectral envelope distortion index and the amplitude distribution stability coefficient respectively, and .

[0034] Preferably, a threshold interval of a preset distortion evaluation coefficient is determined and, after determination, compared with the generated distortion evaluation coefficient to evaluate the distortion degree of the echo signal in the case of ultrasonic signal propagation path distortion according to the comparison result, and classify it into three categories: mild distortion degree, moderate distortion degree, and severe distortion degree. The specific comparison and analysis are as follows:

[0035] If , the distortion degree of the echo signal in the case of ultrasonic signal propagation path distortion is a mild distortion degree;

[0036] If , the distortion degree of the echo signal in the case of ultrasonic signal propagation path distortion is a moderate distortion degree;

[0037] If , the distortion degree of the echo signal in the case of ultrasonic signal propagation path distortion is a severe distortion degree.

[0038] Preferably, in the data correction module, based on the evaluation result, corresponding correction measures are taken for the measurement data to be corrected under different echo signal distortion degrees, specifically including:

[0039] If the evaluation result is a mild distortion degree, the correction measure taken for the measurement data to be corrected in the case of mild distortion degree is specifically: directly adopt the measurement data to be corrected and perform filtering processing on it to remove high-frequency noise while retaining the original characteristics of the measurement data;

[0040] If the evaluation result is a moderate distortion degree, the correction measure taken for the measurement data to be corrected in the case of moderate distortion degree is specifically: perform adaptive compensation correction on the measurement data to be corrected, specifically including using trend analysis of historical measurement data, combining time series interpolation and machine learning-based prediction methods for offset correction;

[0041] If the evaluation result is a severe distortion degree, the correction measure taken for the measurement data to be corrected in the case of severe distortion degree is specifically: perform abnormal rejection and redundant data replacement processing on the measurement data to be corrected, specifically including screening and removing high-distortion data points within the signal propagation abnormal window, and using data of adjacent ultrasonic sensors and multi-sensor data fusion methods for measurement data replacement.

[0042] In the above technical solution, the technical effects and advantages provided by the present invention:

[0043] 1. The present invention utilizes a signal monitoring module to collect ultrasonic signals in real time, and combines an abnormal window calibration module to accurately identify the time period of signal propagation distortion, enabling the system to avoid misjudging the deviation state of the conveyor belt due to signal abnormalities and improving the reliability of measurement data. In addition, by introducing a distortion evaluation module, the technical solution of the present invention can accurately quantify the distortion degree of ultrasonic signals based on the spectral envelope distortion index and the amplitude distribution stability coefficient, and classify the distortion situation into three categories: mild, moderate, and severe according to the distortion evaluation coefficient, ensuring that targeted treatment measures can be taken for different degrees of signal distortion, rather than applying a unified correction strategy in a one-size-fits-all manner, thereby effectively reducing the false alarm rate and the system misjudgment rate, and improving the accuracy and stability of the conveyor belt monitoring system.

[0044] 2. The data correction module of the present invention adopts methods such as filtering and noise reduction, adaptive compensation, abnormal elimination, and redundant replacement for different degrees of distortion situations, enabling the system to directly use the original data in the case of mild distortion, perform offset correction on the measurement data through time series interpolation or machine learning prediction in the case of moderate distortion, and eliminate the distorted data and perform data compensation by combining adjacent sensor data or multi-sensor fusion methods in the case of severe distortion, ensuring the credibility of the measurement data. This strategy of adaptively adjusting the correction method based on the signal distortion degree avoids the drawbacks of the traditional monitoring method of uniformly processing all measurement data, enabling the system to dynamically adjust the data correction strategy, ensuring both the integrity of the data and improving the accuracy of the conveyor belt deviation monitoring, effectively reducing unnecessary equipment adjustments caused by measurement data errors, extending the service life of the equipment, and reducing the maintenance cost. In addition, the system adopts a remote optimization module, which uploads the corrected measurement data to the cloud through Internet of Things remote data transmission, and performs dynamic optimization in combination with historical data, continuously adjusting the calibration strategy of the signal propagation abnormal window and the measurement data correction method, ensuring that the system can adapt to different working conditions changes during long-term operation, and improving the intelligent level and adaptive ability of the system.

[0045] 3. The advantages of the present invention are as follows. Firstly, through the dynamic calibration of the signal propagation anomaly window, it ensures that ultrasonic signals can still be effectively recognized under abnormal propagation conditions, reducing the impact of signal distortion on the measurement results. Secondly, through the distortion evaluation method based on mathematical modeling, it accurately quantifies the degree of signal distortion, enabling the system to take hierarchical correction measures for different distortion situations, rather than simply using data filtering or fixed threshold processing, enhancing the accuracy and robustness of the measurement data. Finally, the combination of remote optimization and historical data analysis endows the system with self-adaptive learning ability, enabling it to continuously optimize measurement parameters and correction strategies based on long-term operation data, thereby further improving the intelligence level of conveyor belt deviation monitoring. Compared with traditional monitoring technologies based on single ranging methods, this solution can not only reduce misjudgments and false alarms during the conveyor belt deviation detection process, but also significantly improve the long-term stability and maintenance efficiency of the conveyor belt monitoring system, having high engineering application value and industrial promotion prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0047] Figure 1 It is a module schematic diagram of the intelligent material conveying deviation remote monitoring system based on the Internet of Things of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0048] Now, the exemplary embodiments will be described more comprehensively with reference to the accompanying drawings. However, the exemplary embodiments can be implemented in various forms and should not be construed as limited to the examples set forth herein; rather, these exemplary embodiments are provided so that the present disclosure will be more comprehensive and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art.

[0049] The present invention provides an intelligent material conveying deviation remote monitoring system based on the Internet of Things as Figure 1 shown, including a signal monitoring module, an anomaly window calibration module, a distortion evaluation module, a data correction module, and a remote optimization module;

[0050] The signal monitoring module collects ultrasonic signals through ultrasonic sensors deployed on both sides of the conveyor belt and monitors the propagation characteristics of the ultrasonic signals in real time to analyze whether the propagation path of the ultrasonic signals is distorted;

[0051] In order to collect ultrasonic signals through ultrasonic sensors deployed on both sides of the conveyor belt and monitor the propagation characteristics of ultrasonic signals in real time to analyze whether the propagation path of ultrasonic signals is distorted, multi-sensor collaborative monitoring, signal feature analysis, and data fusion algorithms can be adopted. First, ultrasonic sensors are arranged at fixed intervals and synchronously trigger ultrasonic pulses at the same time interval, receive echo signals, and use digital signal processing (DSP) technology to extract characteristic parameters such as echo waveform consistency, signal propagation delay, echo energy attenuation rate, and incident angle change amplitude. Then, the system analyzes the signal characteristics in real time through an adaptive threshold detection method, calculates the current signal change rate based on historical baseline characteristics to determine whether abnormal waveform distortion occurs. When a signal propagation path distortion is detected, further combined with the multi-sensor data fusion algorithm, the ranging data of adjacent sensors are analyzed for consistency. If there are obvious deviations, it is further confirmed that the signal propagation path has been distorted and an abnormal mark is made, providing a basis for subsequent correction calculations. In addition, machine learning algorithms (such as support vector machine SVM or neural network) are used to train models to identify normal signal patterns and signal distortion patterns, improving the system's adaptability to different working conditions and enabling it to accurately distinguish environmental interference from actual misalignment situations.

[0052] The main purpose of this is to improve the accuracy of conveyor belt misalignment monitoring data, avoid ranging errors caused by signal propagation path distortion, and thus reduce false alarms and unnecessary corrective adjustments. Due to the complex operating environment of the conveyor belt, there may be irregular structures such as rubber skirtings, metal baffles, and roller brackets at its edges, resulting in problems such as multiple reflections, scattering, and energy attenuation of ultrasonic signals. Directly using the raw ranging data without analysis may lead to misjudgment of misalignment, thus affecting the stability of the conveying system. By monitoring the propagation characteristics of ultrasonic signals in real time, the abnormal area can be immediately calibrated when signal distortion occurs, providing a basis for subsequent data correction and ensuring that only reliable data is used for misalignment calculation. In addition, through data fusion and machine learning to optimize the monitoring algorithm, the system can continuously optimize the measurement accuracy during long-term operation, improve the ability to identify abnormal situations, enable the conveyor belt monitoring system to have adaptive learning ability, ultimately reduce maintenance costs, reduce false alarms, and improve the long-term stability and intelligent level of the conveying system.

[0053] The abnormal window calibration module, in the case where the analysis result is that the propagation path of the ultrasonic signal is distorted, determines the time period during which the propagation path of the ultrasonic signal is distorted, calibrates it as a signal propagation abnormal window, and calibrates all measurement data collected by the ultrasonic sensor within the signal propagation abnormal window as measurement data to be corrected;

[0054] In this embodiment, in the abnormal window calibration module, when the analysis result indicates that the propagation path of the ultrasonic signal is distorted, by analyzing the continuous time-series change rate of the ultrasonic measurement data and comparing it with the duration threshold of the abnormal signal, the time period during which the propagation path of the ultrasonic signal is distorted is determined, and this time period is calibrated as the signal propagation abnormal window;

[0055] To analyze the continuous time-series change rate of the ultrasonic measurement data, compare it with the duration threshold of the abnormal signal, and determine the time period during which the propagation path of the ultrasonic signal is distorted, a combination of time-series data analysis, sliding window detection, change rate calculation, and abnormal persistence analysis can be used for dynamic monitoring and determination. First, establish a time series model for the continuously collected ultrasonic measurement data, perform local change rate analysis on the propagation characteristics of the ultrasonic signal through the sliding window algorithm, calculate the change rates of multiple parameters such as the echo waveform consistency deviation, signal propagation time delay change rate, and echo energy attenuation rate, and compare their short-term change trends. If the change rate at a certain moment within the sliding window exceeds the set dynamic abnormal threshold, mark this data point as a potential abnormal point. To avoid misjudgment caused by accidental noise interference, further introduce the comparison of the duration threshold of the abnormal signal, that is, if the number of abnormal points within multiple consecutive sliding windows exceeds the set abnormal accumulation threshold, it is considered that the propagation path of the ultrasonic signal is distorted, and then the starting time of this abnormal state is determined. Within the time range where it persists, this time period is calibrated as the signal propagation abnormal window. In addition, combined with historical data for trend analysis, if the change pattern of the abnormal signal matches the previously confirmed distortion pattern in history, the credibility of the abnormal window is further improved, and the sliding window parameters are adjusted to adapt to the signal distortion characteristics under different working conditions.

[0056] The main purpose of adopting this method is to ensure high real-time performance and high accuracy in detecting the distortion of the ultrasonic signal propagation path, reduce misjudgment, and improve the robustness of the system. Due to the complex operating environment of the conveyor belt, its edge structure may transiently or periodically affect the ultrasonic signal propagation path. If the determination is based only on single-point anomalies, false alarms may be caused by instantaneous noise or short-term interference. Through continuous time-series change rate analysis, it is possible to effectively distinguish between short-term signal fluctuations and real signal propagation path distortions, improving the reliability of the determination. At the same time, the introduction of a sliding window detection enables the system to have an adaptive ability, capable of dynamically adjusting detection parameters under different signal fluctuation patterns to ensure accurate identification of the distortion time period. In addition, the duration threshold comparison mechanism for abnormal signals can further reduce misjudgment situations, enabling the system to trigger abnormal window calibration only in the case of real signal distortions, avoiding unnecessary interference with the subsequent data correction process. This method can not only improve the stability of conveyor belt deviation monitoring but also optimize data processing efficiency, reduce system resource consumption, and make the entire IoT deviation remote monitoring system have a higher level of intelligence.

[0057] Within the abnormal window of signal propagation, by matching the timestamps of the measurement data and combining the spatial position of the measurement data with the correlation of the signal propagation path, all ultrasonic sensor measurement data within this time window are screened and marked as measurement data to be corrected for subsequent echo signal distortion evaluation and measurement data correction.

[0058] Within the signal propagation anomaly window, by matching the timestamps of the measurement data and combining the relevance between the spatial location of the measurement data and the signal propagation path, all ultrasonic sensor measurement data within this time window are screened and calibrated as measurement data to be corrected, which can be achieved through methods such as timestamp synchronization matching, spatial location mapping, and signal propagation model analysis. First, the system performs timestamp matching on the measurement data of all ultrasonic sensors to ensure that all measurement data within the same time window are stored in the standard time series format, and uses a global clock synchronization mechanism (such as based on NTP or high-precision time synchronization protocol) to eliminate the time error between different sensors. Subsequently, combining the spatial location coordinate information of the sensors on both sides of the conveyor belt, by constructing a sensor spatial topology model, it is calculated whether the measurement points of each sensor within this window are within the range of the affected signal propagation path. If the spatial location of a certain measurement point matches the distorted signal propagation path, it is further determined whether it belongs to the affected data. In addition, using signal propagation path inversion analysis, according to the known conveyor belt deviation trend and ultrasonic signal propagation characteristics, the change pattern of the signal propagation path within this window is analyzed to ensure that only the data points truly affected by the distortion are screened out without mislabeling normal measurement points. Finally, all measurement data that meet the time window, spatial location, and signal propagation path matching are calibrated as measurement data to be corrected and stored in the subsequent echo signal distortion evaluation module for further processing.

[0059] The main purpose of doing this is to ensure that the selection of the measurement data to be corrected is accurate, reasonable, and has a high degree of credibility, and to avoid deviations in the subsequent measurement data correction process due to mis-screening or omission. The structure of the conveyor belt edge is complex, and some sensors may only be affected by partial signal propagation distortion. If data screening is only based on the time window, some unaffected measurement data may be calibrated as measurement data to be corrected, thus increasing the unnecessary computational burden and even affecting the correction accuracy. Therefore, by combining timestamp matching, spatial location analysis, and signal propagation path mapping, the affected data points can be accurately screened to ensure that the measurement data to be corrected only contains the data truly disturbed by signal propagation distortion, thereby improving the reliability of subsequent distortion evaluation and measurement data correction. In addition, using signal propagation path inversion analysis can help the system dynamically optimize the measurement data screening mechanism to adapt to different conveyor belt deviation modes, improve the system's adaptability, ensure the accuracy and reliability of the monitoring data during long-term operation, and ultimately improve the intelligence level of the entire conveyor belt deviation remote monitoring system.

[0060] The distortion evaluation module obtains in real time the ultrasonic propagation distortion characteristic information within the signal propagation abnormal window, analyzes it after acquisition, evaluates the echo signal distortion degree in the case of ultrasonic signal propagation path distortion, and classifies it into three categories: mild distortion degree, moderate distortion degree, and severe distortion degree;

[0061] In this embodiment, in the distortion evaluation module, the ultrasonic propagation distortion characteristic information within the signal propagation abnormal window is obtained in real time and preprocessed after acquisition; the spectral amplitude dynamic characteristic information and the energy amplitude distribution characteristic information are extracted from the preprocessed ultrasonic propagation distortion characteristic information, and analyzed to generate a spectral envelope distortion index and an amplitude distribution stability coefficient respectively; a distortion degree analysis model is constructed for the generated spectral envelope distortion index and amplitude distribution stability coefficient, and a distortion evaluation coefficient is generated by weighted summation; a preset distortion evaluation coefficient threshold interval is determined, and after determination, it is compared with the generated distortion evaluation coefficient, and the echo signal distortion degree in the case of ultrasonic signal propagation path distortion is evaluated according to the comparison result, and it is classified into three categories: mild distortion degree, moderate distortion degree, and severe distortion degree.

[0062] In the distortion evaluation module, in order to obtain in real time the ultrasonic propagation distortion characteristic information within the signal propagation abnormal window, methods such as time window data caching, event-triggered data acquisition, and parallel data stream processing can be used. First, based on the output of the abnormal window calibration module, the system dynamically updates the start time and end time of the abnormal window and establishes a timestamp index structure to ensure accurate positioning of the ultrasonic measurement data within this time range. Second, the system adopts an event-triggered mechanism. When the abnormal window is calibrated, the high-speed data buffer is immediately called to extract the original measurement data of all ultrasonic signals within this window and store them in chronological order. In addition, in order to improve the real-time performance and processing efficiency of the data, a parallel data stream processing architecture can be adopted, that is, while data is being acquired, the sliding window algorithm is used to perform dynamic streaming processing on the data to extract key feature data, including the instantaneous amplitude, instantaneous frequency offset, amplitude change rate, average power, and peak factor of the echo signal, etc., to ensure the complete continuity of these data within the abnormal window for subsequent analysis. At the same time, in order to avoid affecting the evaluation accuracy due to data loss or timing disorder, the system can use a time synchronization protocol (such as NTP or high-precision clock synchronization based on GPS) to ensure that data from different sensors is acquired under the same time reference benchmark to improve the data alignment accuracy.

[0063] The main purpose of preprocessing is to improve the quality of signal feature data, reduce noise interference, and enhance the accuracy of subsequent distortion assessment calculations. Since ultrasonic signals may be affected by factors such as mechanical vibration, electromagnetic interference, and environmental noise when propagating in a complex environment, directly using unprocessed data may lead to inaccurate signal distortion assessment. Therefore, after data acquisition is completed, noise suppression processing is first required. Adaptive filtering algorithms (such as Kalman filtering or wavelet denoising) can be used to filter out high-frequency noise while retaining the key features in the signal. Secondly, outlier detection and removal are carried out. Z-score or density-based outlier detection (DBSCAN) algorithms are used to identify and remove abnormal data points that may be caused by sudden interference. Then, normalization processing is performed on the amplitude and frequency data of the echo signal. Min-Max normalization or Z-score standardization methods can be used to make the data within the same magnitude range and improve the stability of subsequent calculations. In addition, the system also needs to perform time series alignment and interpolation completion. If the data at some measurement points is missing due to transmission delay or device jitter, linear interpolation or spline interpolation methods are used to fill the data gap to ensure the integrity of the time series. Finally, all preprocessed data will be stored in the feature data buffer for subsequent extraction of spectral amplitude dynamic feature information and energy amplitude distribution feature information to ensure the accuracy and stability of the distortion assessment process.

[0064] In order to extract the spectral amplitude dynamic feature information and the energy amplitude distribution feature information from the preprocessed ultrasonic propagation distortion feature information, methods such as time-frequency analysis, eigenvector calculation, statistical feature extraction, and adaptive signal decomposition can be adopted to ensure the high-precision extraction and stability of the data. First, for the instantaneous amplitude and instantaneous frequency data of the ultrasonic signal, the short-time Fourier transform (STFT) or the continuous wavelet transform (CWT) can be applied to analyze the energy distribution of the signal in different frequency components in the time-frequency domain, so as to extract the spectral amplitude dynamic feature information, which is mainly composed of the instantaneous amplitude and instantaneous frequency offset data of the echo signal. In specific implementation, the system first performs frame segmentation on the ultrasonic signal data, and performs Fourier transform within each frame to calculate the power spectral density within a specific frequency range. Subsequently, principal component analysis or independent component analysis is used to extract the spectral eigenvector with the largest variance contribution, and the distortion degree of the signal is evaluated by calculating the normalized power ratio and frequency concentration. In addition, in order to extract the energy amplitude distribution feature information, the system will use sliding window energy statistical analysis based on the amplitude change rate, average power, and peak factor of the echo signal to calculate the deviation degree of the signal energy distribution. First, the ultrasonic signal is segmented and calculated within multiple time windows, the root mean square energy, power mean, and peak factor of each segment of the signal are calculated, and the histogram statistical method is used to calculate the signal energy distribution within different amplitude ranges. In addition, in order to enhance the sensitivity to distorted signals, the system can apply higher-order statistical moments (such as skewness and kurtosis) to evaluate the distribution pattern of the signal energy, and use spectral entropy calculation to measure the complexity and stability of the signal energy. Finally, these extracted spectral amplitude dynamic feature information and energy amplitude distribution feature information will be standardized and stored in the eigenvector database for subsequent calculation of the spectral envelope distortion index and the amplitude distribution stability coefficient to ensure the accuracy and robustness of the distortion evaluation.

[0065] To determine the pre-set distortion evaluation coefficient threshold interval, methods such as statistical analysis, machine learning classification, dynamic adaptive optimization, and historical data regression analysis can be used to ensure the rationality and adaptability of the threshold setting. First, based on a large amount of ultrasonic signal distortion evaluation data under normal and abnormal working conditions, statistical analysis methods (such as mean-standard deviation analysis or kernel density estimation) are used to model the distribution of the distortion evaluation coefficient (EDEC) at different distortion levels, and quantile analysis (such as the P25-P75 quartile method) is used to preliminarily determine the critical values of mild distortion, moderate distortion, and severe distortion. Second, to improve the accuracy of classification, unsupervised clustering algorithms (such as K-means clustering or Gaussian mixture model GMM) can be used to perform clustering analysis on the distortion evaluation coefficients in historical data, automatically dividing a reasonable distortion evaluation interval from the data distribution to ensure that the threshold can adapt to different working conditions. Subsequently, based on a classification model (such as support vector machine SVM or decision tree classification), a large number of labeled ultrasonic signal distortion samples are trained so that the system can dynamically optimize the threshold and automatically update the threshold interval when new data arrives, improving the adaptability of the evaluation. In addition, combined with historical data regression analysis, exponential smoothing or Bayesian update methods can be used to perform time series prediction on the threshold, enabling the system to adaptively adjust as the state of the conveyor belt equipment changes, ensuring that the threshold can maintain stability and dynamically adapt to the actual usage environment. Finally, during the real-time operation of the system, the newly calculated distortion evaluation coefficient is compared with the current threshold interval, and after a certain time span, the threshold is fine-tuned according to the new data distribution to make the result of the distortion evaluation more accurate and avoid misjudgment or detection lag caused by a fixed threshold setting.

[0066] In this embodiment, the acquisition logic of the spectral envelope distortion index is as follows:

[0067] Extract the spectral amplitude dynamic feature information from the preprocessed ultrasonic propagation distortion feature information, specifically including the instantaneous amplitude of the echo signal received by the ultrasonic sensor at different times within the signal propagation abnormal window and the frequency offset of the ultrasonic signal at the receiving end, and respectively calibrate them as and , represents the instantaneous amplitude of the echo signal received by the ultrasonic sensor at time within the signal propagation abnormal window, represents the frequency offset of the ultrasonic signal at the receiving end at time within the signal propagation abnormal window, , is a positive integer;

[0068] To obtain the instantaneous amplitude of the echo signal received by the ultrasonic sensor at different times within the signal propagation anomaly window and the frequency offset of the ultrasonic signal at the receiving end in real time, technologies such as high-speed data acquisition, digital signal processing, time-frequency analysis, and filtering optimization can be adopted to ensure the accuracy and real-time nature of data acquisition. First, based on the output of the anomaly window calibration module, the system triggers the high-speed ADC (analog-to-digital converter) of the ultrasonic sensor to perform continuous signal sampling. Usually, the echo signal is digitized and stored at a high sampling rate (such as in the MHz range) to capture the complete signal characteristics. During the sampling process, the system uses a hardware trigger or a timed sampling mechanism to ensure that the data of all sensors are recorded under the same time reference, and a timestamp index is used to manage the data for subsequent processing. In the acquired digital signal, the instantaneous amplitude of the echo signal can be directly obtained by taking the absolute amplitude of the signal or calculating the signal envelope through the Hilbert Transform, thereby obtaining the instantaneous intensity of the signal. Since the echo signal may be disturbed by environmental noise, an adaptive filter (such as Kalman filter or low-pass filter) is also required to denoise the signal to extract the true instantaneous amplitude. In addition, the frequency offset of the ultrasonic signal at the receiving end can be calculated by the short-time Fourier transform (STFT) or the Hilbert Transform to calculate the instantaneous phase change rate of the signal, and the frequency offset is calculated based on the phase difference, that is, the frequency change trend is deduced through continuous phase changes to reflect whether the signal has shifted during propagation. Since different sensors may be affected by different environments, multi-sensor fusion technology can also be adopted to compare the frequency offset information of adjacent sensors to exclude data anomalies caused by sudden interference of a single sensor. Finally, all the data obtained in real time will be stored in a cache queue and matched with the anomaly window timestamp to ensure the time consistency of these data, providing high-quality input for subsequent signal analysis and distortion assessment.

[0069] Construct a set of the instantaneous amplitudes of the echo signals received by the ultrasonic sensor at different times within the signal propagation anomaly window, and calibrate the maximum value within the set as ;

[0070] Sort the frequency offsets of the ultrasonic signals at the receiving end at different times within the signal propagation anomaly window according to the numerical size, and calibrate the median as ;

[0071] Calculate the standard deviation of the frequency offsets of the ultrasonic signals at the receiving end at different times within the signal propagation anomaly window , according to the formula: ;

[0072] Calculate the spectral envelope distortion index, and the specific calculation formula is as follows:

[0073]

[0074] In the formula, is the spectral envelope distortion index.

[0075] To ensure that the spectral envelope distortion index can accurately reflect the dynamic characteristic changes of ultrasonic signals under the condition of propagation path distortion, the design of this calculation formula fully considers the statistical distribution characteristics of signal frequency offset and the normalization characteristics of echo signal amplitude, and uses exponential operation and logarithmic operation to improve the sensitivity and stability to signal distortion. First, before calculating , it is necessary to sort the instantaneous frequency offsets of ultrasonic signals at different times within the signal propagation abnormal window, and determine its median and standard deviation . Among them, the role of the median is to provide a robust reference value, reduce the influence of extreme outliers on the calculation, and make the distortion evaluation more stable; the role of the standard deviation is to measure the distribution range of frequency offset, making the influence of signals with larger abnormal amplitudes on the distortion index more significant. Then, calculate the term , which is used to measure the deviation degree of the frequency offset at a single moment from the overall signal offset distribution, and use the exponential operation , whose role is to amplify those signals far from the normal offset range, making the severely distorted frequency offset points contribute larger exponential values and improving the sensitivity of the calculation to abnormal signals.

[0076] On the other hand, for the instantaneous amplitude of the echo signal, in order to make the signals measured by different ultrasonic sensors comparable, it is necessary to first calculate its normalized value, that is, , where is the maximum amplitude within the signal propagation abnormal window. The role of this normalization process is to limit the signal amplitude within the range of [0,1], ensuring that signal data under different measurement environments can be standardized. Then, calculate , whose role is to reduce the influence of signals with larger amplitudes through logarithmic transformation, making pay more attention to signals with smaller amplitudes but still likely to cause distortion, thereby avoiding the dominance of the calculation result by individual ultra-high amplitude signals. Finally, take the average of the calculation results at all times, whose role is to ensure that can characterize the overall distortion degree of the signal within the entire abnormal window, rather than being affected only by the data at a single moment, improving stability. Through this calculation method, It can amplify the distortion characteristics when the signal distortion is severe and remain stable when the signal fluctuation is small, enabling the system to accurately evaluate the distortion degree of the ultrasonic signal under the condition of propagation path distortion and providing a reliable numerical basis for subsequent data correction.

[0077] Spectrum envelope distortion index The magnitude directly reflects the distortion degree of the echo signal of the ultrasonic signal under the condition of propagation path distortion. The larger its value, the more severe the signal distortion; conversely, it indicates that the signal has higher stability and less distortion. Specifically, It mainly quantifies the distortion degree of the signal through the distribution characteristics of the instantaneous frequency offset of the ultrasonic signal and the instantaneous amplitude of the echo signal. When the signal propagation path is distorted, the propagation characteristics of the ultrasonic signal will be affected by environmental interference, such as multiple reflections, attenuation, scattering, etc., which will lead to a larger or abnormally distributed instantaneous frequency offset , and then lead to an increase in . Among them, if significantly deviates from its standard distribution (that is, differs greatly from , and is also large), then the value of this item will rise rapidly, enhancing the ability to evaluate the distortion degree, indicating that the spectrum of the signal in this window has a large offset and the signal quality is poor. Similarly, if the instantaneous amplitude of the echo signal is low, that is, the energy of the signal decays severely during propagation, then the value of will be small, resulting in a decrease in , indicating insufficient signal energy and heavy distortion. To sum up, when is at a low level, it means that the signal propagation is relatively stable and the echo distortion is light; while when significantly increases, it indicates that the echo signal has severe distortion during propagation, such as excessive signal attenuation, nonlinear distortion or enhanced interference, etc. Therefore,

[0078] In this embodiment, the acquisition logic of the amplitude distribution stability coefficient is as follows:

[0079] Extract the energy amplitude distribution characteristic information from the preprocessed ultrasonic propagation distortion characteristic information, specifically including the amplitude change rate of the echo signal received by the ultrasonic sensor at different times within the signal propagation abnormal window, the power of the ultrasonic signal, and the ratio of the maximum amplitude to the root mean square amplitude of the ultrasonic signal within the signal propagation abnormal window, and calibrate them respectively as , and , represent the amplitude change rate of the echo signal received by the ultrasonic sensor at moments within the signal propagation anomaly window, represent the power of the ultrasonic signal at moments within the signal propagation anomaly window, represent the ratio of the maximum amplitude to the root mean square amplitude of the ultrasonic signal within the signal propagation anomaly window, , where \(n\) is a positive integer;

[0080] In order to obtain in real time the amplitude change rate of the echo signal received by the ultrasonic sensor at different moments within the signal propagation anomaly window, the power of the ultrasonic signal, and the ratio of the maximum amplitude to the root mean square amplitude of the ultrasonic signal, methods such as high-precision signal sampling, digital filtering, time series differential calculation, and statistical analysis can be adopted to ensure the accuracy and stability of the data. First, the system will perform high-frequency sampling on the echo signal of the ultrasonic sensor based on a high-speed ADC (analog-to-digital converter), usually obtaining the original time-domain waveform data of the echo signal at a sampling rate of the MHz level, and adopting adaptive filtering (such as Kalman filtering or Butterworth filtering) to remove environmental noise to ensure data quality. After obtaining the original signal, the system will calculate the amplitude change rate of the echo signal, that is, calculate the relative change rate of the amplitude between different sampling moments, and its specific implementation method is through a first-order difference operation, that is , this value can quantify the dynamic fluctuation characteristics of the signal and reflect the severity of the amplitude change. Secondly, the system will calculate the power of the ultrasonic signal , this value can be obtained through the root mean square (RMS) calculation formula of the signal , this value is used to characterize the distribution of signal energy on the time axis and can effectively identify the overall intensity change of the signal. Finally, in order to obtain the ratio of the maximum amplitude to the root mean square amplitude of the ultrasonic signal , the system will first calculate the maximum echo amplitude from the sampling data, and then calculate the root mean square amplitude , and use for normalization comparison to reflect whether the signal energy is concentrated at certain instantaneous points or is more evenly distributed. This calculation method can quickly identify whether there is abnormal instantaneous high-amplitude distortion in the signal, or whether the power is abnormally concentrated due to non-linear propagation. Finally, all the calculated data will be stored in the time-index database and matched with the signal propagation anomaly window to ensure the time consistency of these data and provide reliable input for subsequent signal distortion evaluation.

[0081] Construct a set of the powers of ultrasonic signals at different times within the signal propagation anomaly window, and calibrate the maximum value within the set as ;

[0082] Calculate the amplitude distribution stability coefficient. The specific calculation formula is as follows:

[0083]

[0084] In the formula, is the amplitude distribution stability coefficient.

[0085] The amplitude distribution stability coefficient 's calculation formula aims to comprehensively evaluate the amplitude distribution stability of ultrasonic signals under the condition of propagation path distortion, so as to reflect the abnormal degree of signal energy distribution. First of all, this calculation method performs a logarithmic transformation on the amplitude change rate of the echo signal, that is . Its function is to compress the data range with large amplitude changes, so that small amplitude change signals will not be ignored, and at the same time avoid the dominant effect of large amplitude change signals on the calculation, improving the calculation stability. Secondly, the exponential normalization factor is adopted in the formula. Its function is to enhance the influence of high-power signals on the overall stability evaluation, ensure more sensitive discrimination of amplitude stability in the case of strong signal power, and at the same time suppress the noise influence of low-power signals. In addition, in order to further supplement the stability evaluation of the amplitude distribution, this calculation formula introduces the ratio of the maximum amplitude to the root mean square amplitude of the ultrasonic signal, and calculates its contribution to the overall stability through . The main purpose is to measure whether the signal energy is evenly distributed and avoid misjudgment caused by the influence of transient high-amplitude signals. Finally, the averaging operation is adopted in the calculation process. Its function is to ensure that the evaluation results within the entire signal propagation anomaly window are more robust and will not cause calculation deviation due to abnormal values at a certain time point. Through this calculation method, can accurately reflect the energy stability of ultrasonic signals, ensure high calculation accuracy under different signal distortion degrees, and provide a reliable data basis for subsequent signal correction.

[0086] The amplitude distribution stability coefficient directly reflects the change of energy stability of ultrasonic signals under the condition of propagation path distortion, so as to be used to evaluate the distortion degree of echo signals. When is relatively small, it indicates that the amplitude change of the ultrasonic signal is small, the power distribution is relatively uniform, the ratio of the maximum amplitude to the root mean square amplitude is low, the signal is not significantly distorted during propagation, indicating that the signal distortion degree is light, the echo signal is relatively stable, and the distortion influence is small. And when When the value increases, it means the amplitude change rate of the echo signal is relatively large, with frequent mutations or violent fluctuations, and the energy and power are relatively unstable, and the ratio of the maximum amplitude to the root mean square amplitude is relatively high, indicating that the signal energy is concentrated in a few time points or spatial ranges, and the signal may suffer strong attenuation, interference or abnormal peaks due to propagation path distortion. At this time, the energy distribution pattern of the signal is abnormal, indicating an increase in the degree of distortion. Specifically, when is at a relatively low level, the signal distortion is relatively light and the measurement data is reliable; while when increases significantly, it indicates that the ultrasonic signal has suffered severe multiple reflections, interference or nonlinear distortion during propagation, resulting in a decrease in the accuracy of the measurement data. Therefore, The calculation result of can be used to classify the degree of distortion of the echo signal, dividing it into mild distortion, moderate distortion and severe distortion, providing a scientific basis for subsequent measurement data correction, and ensuring the accuracy and stability of the remote monitoring of the conveyor belt deviation.

[0087] In this embodiment, a distortion degree analysis model is constructed for the generated spectral envelope distortion index and the amplitude distribution stability coefficient to generate a distortion evaluation coefficient through weighted summation. The specific calculation formula is as follows:

[0088]

[0089] In the formula, is the distortion evaluation coefficient, and are the non-zero weight coefficients of the spectral envelope distortion index and the amplitude distribution stability coefficient respectively, and .

[0090] In order to calculate the distortion evaluation coefficient , when calculating the distortion evaluation coefficient , it is necessary to set non-zero weight coefficients ( and ) to adjust the influence weights of the two indicators on the final distortion evaluation result. represents 's contribution degree to the distortion evaluation, which is mainly used to measure the spectral characteristic changes of the ultrasonic signal under propagation path distortion, and is suitable for detecting the frequency shift and energy concentration change of the signal; represents The contribution degree to distortion evaluation mainly reflects the amplitude stability and power distribution characteristics of the signal, and is applicable to evaluating the uniformity and attenuation of signal energy. The values of these two weight coefficients can be determined through empirical setting, data-driven optimization methods, or machine learning model training. For example, based on historical data regression analysis, the optimal weight values can be calculated by fitting with the least squares method and the relationship with the actual signal distortion degree, and the optimal weight values can be calculated; or Bayesian optimization or genetic algorithms can be used to dynamically adjust and to minimize the misjudgment rate. In addition, the system can also adaptively adjust the weight coefficients according to different working conditions. For example, when the signal frequency drift is relatively serious, is increased, and when the amplitude fluctuation is relatively large, is increased to ensure the robustness and adaptability of distortion evaluation. Finally, after the calculation is completed, the system will compare it with the preset distortion evaluation threshold interval to determine the distortion degree of the echo signal, and classify it accordingly.

[0091] In this embodiment, a preset distortion evaluation coefficient threshold interval is determined, and after determination, it is compared with the generated distortion evaluation coefficient to evaluate the distortion degree of the echo signal in the case of ultrasonic signal propagation path distortion, and it is classified into three categories: mild distortion degree, moderate distortion degree, and severe distortion degree. The specific comparison and analysis are as follows:

[0092] If , the distortion degree of the echo signal in the case of ultrasonic signal propagation path distortion is a mild distortion degree;

[0093] This situation means that the distortion of the echo signal of the ultrasonic signal in the case of propagation path distortion is relatively light, the amplitude fluctuation of the signal is small, the spectral characteristics are relatively stable, and there is no obvious interference or energy loss. In this case, the echo signal still has a high credibility, and the measurement data is basically not affected by the propagation path distortion and can be directly used for conveyor belt deviation monitoring without additional correction or compensation. Due to the low distortion degree, the system can give priority to maintaining the originality of the measurement data in the subsequent processing process, and only perform basic data storage and trend analysis without triggering further signal correction or abnormal alarm.

[0094] If , the distortion degree of the echo signal in the case of ultrasonic signal propagation path distortion is a moderate distortion degree;

[0095] This situation indicates that the echo signal has been distorted to a certain extent under the condition of ultrasonic signal propagation path distortion. There may be signal attenuation, increased amplitude fluctuation, or partial spectrum drift. However, the overall signal can still be used for measurement, although the accuracy has decreased. In this case, the system needs to perform appropriate data correction, such as through interpolation compensation, filtering to reduce noise, or offset correction methods based on historical data, to improve the accuracy of the measurement data. At the same time, the system will record this status in the device maintenance database to evaluate the long-term changes in the conveyor belt operation status during subsequent trend analysis and ensure that long-term measurement errors do not accumulate due to moderate distortion.

[0096] If , the distortion degree of the echo signal under the condition of ultrasonic signal propagation path distortion is a severe distortion degree.

[0097] This situation means that the echo signal of the ultrasonic signal has been severely distorted under the condition of propagation path distortion, and may be affected by factors such as multiple reflections, severe attenuation, noise interference, or phase drift, resulting in the measurement data being no longer reliable. In this case, the system needs to trigger an abnormal alarm and perform mandatory data correction, such as eliminating the distorted signal, using redundant sensor data for compensation, or even taking temporary shutdown detection measures to ensure the safe operation of the conveyor belt. In addition, the system will adjust the working mode of the measurement device according to this status, such as increasing the sampling frequency or changing the signal processing algorithm, to cope with the signal distortion problem in a harsh environment, and generate an abnormal report for the operation and maintenance personnel to evaluate the status of the conveyor belt and take necessary maintenance measures.

[0098] The data correction module, based on the evaluation result, takes corresponding correction measures for the measurement data to be corrected under different degrees of echo signal distortion;

[0099] In this embodiment, in the data correction module, based on the evaluation result, corresponding correction measures are taken for the measurement data to be corrected under different degrees of echo signal distortion, specifically including:

[0100] If the evaluation result is a mild distortion degree, the correction measure taken for the measurement data to be corrected in the case of mild distortion degree is: directly adopt the measurement data to be corrected and perform filtering on it to remove high-frequency noise, while retaining the original characteristics of the measurement data, without further correction;

[0101] To implement this correction measure, signal processing methods such as adaptive filtering, band-pass filtering, or wavelet denoising can be adopted to remove high-frequency noise and environmental interference in the measurement data while ensuring that the original form of the data is not tampered with. The specific implementation method is as follows: First, detect the noise frequency distribution in the ultrasonic signal through spectral analysis. If the noise is mainly concentrated in the high-frequency region, low-pass filtering can be used to remove unnecessary high-frequency components; if there is low-frequency interference in the signal, band-pass filtering is adopted to retain the main frequency band of the ultrasonic signal and shield irrelevant frequency bands. In addition, Kalman filtering can also be used. This algorithm can smooth short-term sudden noise according to the statistical characteristics of the measurement data without affecting the key features of the signal. The reason for adopting this correction method is that in the case of mild distortion, the main information of the signal is still complete and only affected by a small amount of interference. Directly removing data may lead to information loss. Therefore, a moderate noise suppression method is adopted to make the data more stable while retaining its original features to ensure the accuracy of conveyor belt deviation monitoring.

[0102] If the evaluation result is a medium distortion degree, the correction measure taken for the measurement data to be corrected in the case of medium distortion degree is specifically as follows: Perform adaptive compensation correction on the measurement data to be corrected, specifically including using trend analysis of historical measurement data, combining time series interpolation and machine learning-based prediction methods for offset correction to improve the accuracy of the measurement data.

[0103] To implement this correction measure, techniques such as time series analysis, regression modeling, and adaptive signal compensation can be adopted to correct the measurement data of ultrasonic signals within the signal propagation abnormal window. The specific implementation method is as follows: First, the system will use methods such as moving average method, exponential smoothing, or autoregressive integrated moving average (ARIMA) model based on historical measurement data to predict the change trend of the signal under normal conditions, and compare the current measurement data with the predicted trend to detect the degree of data offset. Subsequently, interpolation algorithms (such as Lagrange interpolation, spline interpolation) are used to compensate the measurement data to make it return to the expected reasonable range. In addition, machine learning-based prediction models (such as LSTM, XGBoost) can also be adopted. Use a large amount of historical data to train the model so that the system can automatically learn the change pattern of ultrasonic signals and intelligently adjust the data compensation strategy when signal distortion occurs. The reason for adopting this correction method is that medium distortion usually means that the signal is partially interfered or distorted but still has recoverability. Directly discarding the data may lead to information loss, and through the compensation correction method based on historical trends, the normal state of the signal can be effectively restored, improving the availability and measurement accuracy of the data.

[0104] If the evaluation result is a severe distortion degree, the correction measures taken for the measurement data to be corrected in the case of severe distortion degree are as follows: perform abnormal rejection and redundant data replacement processing on the measurement data to be corrected, specifically including screening and rejecting the high-distortion data points within the abnormal signal propagation window, and using the data of adjacent ultrasonic sensors and the method of multi-sensor data fusion to replace the measurement data to ensure that the corrected data has a high credibility.

[0105] To implement this correction measure, methods such as outlier detection, data fusion, and redundant compensation can be used to eliminate and replace the severely distorted measurement data. The specific implementation method is as follows: First, the system will screen out the measurement data that exceeds the reasonable range through anomaly detection methods based on statistical analysis (such as Z-score, DBSCAN, isolation forest algorithm), and label it as an abnormal data point. Subsequently, use multi-sensor data fusion technology to obtain redundant data from adjacent ultrasonic sensors or other types of sensors (such as laser ranging, infrared detection, etc.), and perform data comparison to select the measurement value closest to the true state. For some sensors that may lack directly available redundant data, Kalman filtering or Bayesian inference can be used to estimate a reasonable alternative measurement value based on the system state. The reason for adopting this correction method is that in the case of severe distortion, the measurement data has been severely distorted. If the data is continued to be used, it may lead to misjudgment of the conveyor belt deviation monitoring. Therefore, it is necessary to eliminate the high-distortion data and generate reliable alternative data through redundant sensors or prediction algorithms to ensure the reliability of system monitoring.

[0106] The remote optimization module, based on Internet of Things remote monitoring and adaptive optimization, remotely transmits the corrected measurement data, and dynamically optimizes the calibration strategy of the signal propagation abnormal window and the measurement data correction method in combination with historical data to improve the intelligent level of remote monitoring of conveyor belt deviation.

[0107] To implement the functions of the remote optimization module, technologies such as Internet of Things remote data transmission, dynamic model optimization, and adaptive parameter adjustment can be adopted to continuously improve the intelligent level of remote monitoring of conveyor belt deviation. First, in terms of data transmission, the system can remotely transmit the corrected measurement data to the cloud data center through the edge computing gateway based on MQTT (Message Queuing Telemetry Transport Protocol), HTTP / HTTPS, or a real-time communication protocol based on WebSocket to ensure low latency and high reliability of data transmission. During the data transmission process, data compression technologies (such as LZ4, Snappy) can be used to reduce bandwidth occupancy, and at the same time, AES-256 or TLS encryption is used to ensure the security of remote data transmission. At the data receiving end, the cloud server will store all measurement data and adopt a distributed database (such as Apache Cassandra or InfluxDB) to efficiently manage historical data and support fast query of massive data. Secondly, to optimize the calibration strategy of the signal propagation anomaly window, the system will use methods such as dynamic time warping (DTW), sliding window statistical analysis, or adaptive threshold adjustment, combined with historical data analysis to analyze the law of ultrasonic signal distortion. For example, a Bayesian update model can be constructed based on historical distortion evaluation data to adaptively adjust the boundaries of the anomaly window under different working conditions, making the calibration of the signal propagation anomaly window more accurate and reducing false positives and missed detections. In addition, the system can adopt a reinforcement learning algorithm (such as deep Q learning DQN) to enable the model to adaptively adjust the set parameters of the anomaly window during the process of continuously accumulating data, making the optimization strategy more intelligent.

[0108] In terms of optimizing the measurement data correction method, the system can combine historical correction data and actual equipment operation feedback, and dynamically adjust the correction strategy through long-term data regression analysis or clustering algorithms. Specifically, the system can use K-means clustering or Gaussian Mixture Model (GMM) to cluster historical distortion evaluation data and corrected measurement data, analyze the optimal correction method under different working conditions, and adjust the current data correction method accordingly. In addition, the system can build a time series prediction model based on the LSTM (Long Short-Term Memory Network) or Transformer model to learn the long-term trend of the conveyor belt deviation, so as to adjust the correction parameters in advance and improve the accuracy of predictive correction. For example, if the system detects that the distortion degree has been increasing continuously for a certain period of time, it can dynamically increase the weight of the measurement data correction to enhance the compensation ability for signal anomalies. The reason for adopting this method is that during the long-term operation of the conveyor belt deviation monitoring system, environmental factors (such as temperature, humidity, dust) may affect the measurement accuracy. Therefore, it is necessary to continuously optimize the correction strategy to adapt to different working condition changes and improve the stability and accuracy of the monitoring system. Finally, based on these optimization methods, the remote optimization module can continuously optimize the calibration strategy of the signal propagation anomaly window and the measurement data correction method in the Internet of Things environment, enabling the conveyor belt deviation remote monitoring system to have stronger intelligent adaptive capabilities.

[0109] The above formulas are all dimensionless and take their numerical calculations. The formulas are obtained by collecting a large amount of data for software simulation to get a formula that is closest to the real situation. The preset parameters in the formulas are set by technicians in this field according to the actual situation.

[0110] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center in a wired or wireless (such as infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that contains one or more collections of available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, or magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0111] It should be understood that in various embodiments of the present application, the order of the above processes does not mean the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0112] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. A professional technician can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0113] In several embodiments provided in the present application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the above-described embodiments are merely illustrative. For example, the division of the units is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be indirect couplings or communication connections through some interfaces, devices, or units, and can be in an electrical, mechanical, or other form.

[0114] The unit described as a separation component may or may not be physically separated. The component shown as a unit may or may not be a physical unit, that is, it may be located in one place or may be distributed over multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0115] In addition, each functional unit in various embodiments of the present application may be integrated into a processing unit, may also exist separately as individual physical units, or two or more units may be integrated into one unit.

[0116] As described above, the above are only specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. An intelligent material conveying deviation remote monitoring system based on the Internet of Things, characterized in that It includes a signal monitoring module, an abnormal window calibration module, a distortion evaluation module, a data correction module, and a remote optimization module; The signal monitoring module collects ultrasonic signals through ultrasonic sensors deployed on both sides of the conveyor belt, and monitors the propagation characteristics of the ultrasonic signals in real time to analyze whether the propagation path of the ultrasonic signals is distorted; The abnormal window calibration module, in the case where the analysis result is that the propagation path of the ultrasonic signal is distorted, determines the time period during which the propagation path of the ultrasonic signal is distorted, calibrates it as a signal propagation abnormal window, and calibrates all measurement data collected by the ultrasonic sensor within the signal propagation abnormal window as measurement data to be corrected; The distortion evaluation module obtains the ultrasonic propagation distortion characteristic information within the signal propagation abnormal window in real time, analyzes it after obtaining, evaluates the echo signal distortion degree in the case of the distortion of the ultrasonic signal propagation path, and classifies it into three categories: mild distortion degree, moderate distortion degree, and severe distortion degree; The data correction module takes corresponding correction measures for the measurement data to be corrected under different echo signal distortion degrees based on the evaluation result; The remote optimization module remotely transmits the corrected measurement data based on Internet of Things remote monitoring and adaptive optimization, and dynamically optimizes the calibration strategy of the signal propagation abnormal window and the measurement data correction method in combination with historical data.

2. The intelligent material conveying deviation remote monitoring system based on the Internet of Things according to claim 1, characterized in that, In the abnormal window calibration module, in the case where the analysis result is that the propagation path of the ultrasonic signal is distorted, by analyzing the continuous time series change rate of the ultrasonic measurement data and comparing it with the duration threshold of the abnormal signal, the time period during which the propagation path of the ultrasonic signal is distorted is determined, and this time period is calibrated as the signal propagation abnormal window; Within the signal propagation abnormal window, by matching the timestamps of the measurement data and combining the relevance between the spatial position of the measurement data and the signal propagation path, all ultrasonic sensor measurement data within this time window are screened and calibrated as measurement data to be corrected.

3. The intelligent material conveying deviation remote monitoring system based on the Internet of Things according to claim 2, characterized in that, In the distortion evaluation module, the ultrasonic propagation distortion characteristic information within the signal propagation abnormal window is obtained in real time and preprocessed after obtaining; the spectral amplitude dynamic characteristic information and the energy amplitude distribution characteristic information are extracted from the preprocessed ultrasonic propagation distortion characteristic information, and they are analyzed to generate the spectral envelope distortion index and the amplitude distribution stability coefficient respectively; a distortion degree analysis model is constructed for the generated spectral envelope distortion index and amplitude distribution stability coefficient, and a distortion evaluation coefficient is generated through weighted summation; a preset distortion evaluation coefficient threshold interval is determined, and after determination, it is compared with the generated distortion evaluation coefficient, and the echo signal distortion degree in the case of the distortion of the ultrasonic signal propagation path is evaluated according to the comparison result, and it is classified into three categories: mild distortion degree, moderate distortion degree, and severe distortion degree.

4. The intelligent material conveying offset remote monitoring system based on the Internet of Things according to claim 3, characterized in that The acquisition logic of the spectral envelope distortion index is as follows: Extract the spectral amplitude dynamic characteristic information from the preprocessed ultrasonic propagation distortion characteristic information, specifically including the instantaneous amplitude of the echo signal received by the ultrasonic sensor at different times within the signal propagation abnormal window and the frequency offset of the ultrasonic signal at the receiving end, and calibrate them respectively as and , indicating the instantaneous amplitude of the echo signal received by the ultrasonic sensor at the time of within the signal propagation abnormal window, indicating the frequency offset of the ultrasonic signal at the receiving end at the time of within the signal propagation abnormal window, , being a positive integer; Construct a set of the instantaneous amplitudes of the echo signals received by the ultrasonic sensor at different moments within the signal propagation anomaly window, and calibrate the maximum value within the set as ; Sort the frequency offsets of the ultrasonic signals at different times within the signal propagation anomaly window at the receiving end according to the numerical values, and calibrate the median as ; Calculate the standard deviation of the frequency offset of the ultrasonic signal at the receiving end at different times within the signal propagation anomaly window , according to the formula: ; Calculate the spectral envelope distortion index, and the specific calculation formula is as follows: In the formula, is the spectral envelope distortion index.

5. The intelligent material conveying deviation remote monitoring system based on the Internet of Things according to claim 4, characterized in that The acquisition logic of the amplitude distribution stability coefficient is as follows: Extract the energy amplitude distribution feature information from the preprocessed ultrasonic propagation distortion feature information, specifically including the amplitude change rate of the echo signal received by the ultrasonic sensor at different times within the signal propagation abnormal window, the power of the ultrasonic signal, and the ratio of the maximum amplitude to the root mean square amplitude of the ultrasonic signal within the signal propagation abnormal window, and calibrate them respectively as 、 and , represents the amplitude change rate of the echo signal received by the ultrasonic sensor at the moment within the signal propagation abnormal window, represents the power of the ultrasonic signal at the moment within the signal propagation abnormal window, represents the ratio of the maximum amplitude to the root mean square amplitude of the ultrasonic signal within the signal propagation abnormal window, , is a positive integer; Construct a set of the powers of ultrasonic signals at different moments within the signal propagation anomaly window, and calibrate the maximum value within the set as ; Calculate the amplitude distribution stability coefficient, and the specific calculation formula is as follows: In the formula, is the amplitude distribution stability coefficient.

6. The intelligent material conveying deviation remote monitoring system based on the Internet of Things according to claim 5, characterized in that, For the generated spectral envelope distortion index and the amplitude distribution stability coefficient Construct a distortion degree analysis model, and generate a distortion evaluation coefficient through weighted summation. The specific calculation formula is as follows: In the formula, is the distortion evaluation coefficient, and are respectively the non - zero weight coefficients of the spectral envelope distortion index and the amplitude distribution stability coefficient , and .

7. The intelligent material conveying deviation remote monitoring system based on the Internet of Things according to claim 6, characterized in that, Determine the threshold interval of the preset distortion evaluation coefficient , and after determination, compare it with the generated distortion evaluation coefficient . According to the comparison result, evaluate the distortion degree of the echo signal under the condition of ultrasonic signal propagation path distortion, and classify it into three categories: mild distortion degree, moderate distortion degree and severe distortion degree. The specific comparison and analysis are as follows: If , the distortion degree of the echo signal in the case of ultrasonic signal propagation path distortion is a mild distortion degree; If , the degree of echo signal distortion in the case of ultrasonic signal propagation path distortion is medium distortion degree; If , the distortion degree of the echo signal in the case of ultrasonic signal propagation path distortion is a severe distortion degree.

8. The intelligent material conveying deviation remote monitoring system based on the Internet of Things according to claim 7, characterized in that, In the data correction module, based on the evaluation results, corresponding correction measures are taken for the measurement data to be corrected under different degrees of echo signal distortion, specifically including: If the evaluation result is a mild distortion degree, the correction measure taken for the measurement data to be corrected under the mild distortion degree is specifically: directly adopt the measurement data to be corrected, and perform filtering processing on it to remove high-frequency noise while retaining the original characteristics of the measurement data; If the evaluation result is a moderate distortion degree, the correction measure taken for the measurement data to be corrected under the moderate distortion degree is specifically: perform adaptive compensation correction on the measurement data to be corrected, specifically including using the trend analysis of historical measurement data, combining time series interpolation and machine learning-based prediction methods for offset correction; If the evaluation result is a severe distortion degree, the correction measure taken for the measurement data to be corrected under the severe distortion degree is specifically: perform abnormal rejection and redundant data replacement processing on the measurement data to be corrected, specifically including screening and removing the high-distortion data points within the abnormal signal propagation window, and using the data of adjacent ultrasonic sensors and the method of multi-sensor data fusion for measurement data replacement.

Images