A method and system for monitoring the operating efficiency of flexible equipment

Abstract

The present invention relates to the field of data processing technology, and discloses a method and system for monitoring the operating efficiency of flexible equipment. The method comprises obtaining the tensile stress, bending angle and temperature change of the flexible equipment under different working conditions, analyzing the change rate and frequency of the tensile stress to divide the tensile stress into several levels, analyzing the change rate and duration of the bending angle to divide the bending angle into several intervals, and extracting the rate, amplitude and frequency of temperature change to divide the temperature change into multiple modes; constructing a multi-dimensional feature vector and classifying it based on the tensile stress level, bending angle interval and temperature change mode, and establishing an efficiency monitoring database for the flexible equipment under all working conditions; obtaining the tensile stress, bending angle and temperature data of the flexible equipment in real time, matching them with the efficiency monitoring database, and optimizing the operating efficiency monitoring process of the flexible equipment according to the comprehensive monitoring scoring matrix obtained by matching.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and system for monitoring the operating efficiency of flexible equipment. Background Art

[0002] Since the automated control system used by flexible equipment can quickly respond to production changes and achieve refined processing, thereby improving product quality, it has been widely used in the production process, and the operating efficiency of flexible equipment has also received increasing attention. At present, the monitoring of the operating efficiency of flexible equipment mostly uses data analysis methods, but this method requires professional data analysts and technical means, which greatly reduces the monitoring efficiency.

[0003] It can be seen that how to solve the problem of low efficiency of monitoring the operating efficiency of flexible equipment by existing methods has become a technical problem that needs to be urgently solved by those skilled in the art. Summary of the invention

[0004] The present invention provides a method and system for monitoring the operating efficiency of flexible equipment, which solves the problem of how to improve the monitoring of the operating efficiency of flexible equipment.

[0005] In order to solve the above technical problems, the first aspect of the present invention provides a method for monitoring the operating efficiency of a flexible device, comprising:

[0006] Obtaining historical monitoring data of the flexible device under various working conditions; the historical monitoring data includes historical tensile stress data, historical bending angle data and historical temperature change data;

[0007] Classifying the historical tensile stress data according to a preset tensile stress threshold value to obtain tensile stress level data of the flexible device under corresponding working conditions;

[0008] quantifying the angle change rate of the flexible device under the corresponding working condition based on the historical bending angle data to perform interval division, and obtaining the bending angle interval data of the flexible device under the corresponding working condition;

[0009] Extracting key features from the historical temperature change data, and dividing the historical temperature change data into patterns according to a preset temperature change threshold and the key features, to obtain temperature change pattern data of the flexible device under corresponding working conditions;

[0010] Based on the tensile stress level data, the bending angle interval data and the temperature change mode data, a multi-dimensional feature vector is constructed and classified to obtain a feature classification result of the flexible device to construct a full-operating-condition efficiency monitoring database;

[0011] Real-time tensile stress data, real-time bending angle data and real-time temperature change data of the flexible device are obtained to match them with the full-operating efficiency monitoring database, and the operating efficiency monitoring process of the flexible device is optimized based on the comprehensive monitoring scoring matrix obtained by matching.

[0012] As one of the preferred solutions, the historical tensile stress data is graded according to a preset tensile stress threshold to obtain tensile stress grade data of the flexible device under corresponding working conditions, including:

[0013] constructing a stress reference data matrix through the historical tensile stress data, and quantifying stress change frequency data according to the stress reference data matrix;

[0014] Constructing a two-dimensional stress frequency matrix according to the stress change frequency data, and performing a primary level division on the two-dimensional stress frequency matrix by a clustering algorithm based on Euclidean distance to obtain a preliminary stress level;

[0015] The preliminary stress level is adjusted based on the preset tensile stress threshold to obtain tensile stress level data of the flexible device under corresponding working conditions.

[0016] As one of the preferred solutions, quantifying the angle change rate of the flexible device under the corresponding working condition based on the historical bending angle data to divide the intervals, and obtaining the bending angle interval data of the flexible device under the corresponding working condition, includes:

[0017] The historical bending angle data is processed in sections using a first fixed time window, and the angle change trend curve of each section is calculated by cubic polynomial least squares fitting to quantify the angle change rate of the deformation area;

[0018] The angle change rates of the deformation regions are grouped by a clustering algorithm, the slope of the angle change trend curve in each cluster group is quantified to divide the boundary interval, and an angle hierarchical quantization matrix is ​​constructed based on the boundary interval;

[0019] The angle graded quantization matrix is ​​subjected to Fourier transform to obtain a change frequency characteristic, and the distance between the change frequency characteristic and a preset change frequency threshold is quantized to adjust the constructed angle graded quantization matrix to obtain the bending angle interval data of the flexible device under the corresponding working condition.

[0020] As one of the preferred solutions, the extracting key features from the historical temperature change data includes:

[0021] The historical temperature change data is constructed into a historical temperature monitoring data matrix by a time series algorithm, and a first difference between adjacent data in the historical temperature monitoring data matrix is ​​quantified to obtain a historical temperature change rate based on the first difference and a time window;

[0022] Performing Fourier transformation on the historical temperature monitoring data matrix to obtain the historical temperature frequency component, and quantifying the second difference between the minimum temperature value and the maximum temperature value of the historical temperature monitoring data matrix within the time window to obtain the historical temperature change amplitude;

[0023] The historical temperature change rate, the historical temperature frequency component and the historical temperature change amplitude are combined to obtain the key feature.

[0024] As one of the preferred solutions, the historical temperature change data is divided into patterns by using the preset temperature change threshold and the key feature to obtain the temperature change pattern data of the flexible device under the corresponding working condition, including:

[0025] quantifying the similarity between the key features by a preset clustering algorithm, and classifying the key features based on the similarity to obtain a number of cluster centers;

[0026] Establishing an initial boundary of the historical temperature change data based on the cluster center, and calculating a temperature change quantization value of the historical temperature change data based on the initial boundary to obtain a temperature change pattern;

[0027] Performing five-layer wavelet decomposition on the historical temperature change data to extract local temperature features, reconstructing the local temperature features through temperature feature scale coefficients to obtain temperature change reconstructed features;

[0028] The variation deviation between the temperature variation reconstruction feature and the preset temperature variation threshold is quantified, and the temperature variation pattern is adjusted according to the variation deviation to obtain the temperature variation pattern data of the flexible device under the corresponding working condition.

[0029] As one of the preferred solutions, the multi-dimensional feature vector is constructed based on the tensile stress level data, the bending angle interval data and the temperature change mode data, including:

[0030] Aligning the tensile stress level data, the bending angle interval data, and the temperature change pattern data on a time axis, and performing data extraction through a second fixed time window to construct a multi-dimensional feature data matrix;

[0031] The multidimensional feature data matrix is ​​subjected to correlation calculation, and a first correlation coefficient between the tensile stress level data and the bending angle interval data, a second correlation coefficient between the tensile stress level data and the temperature change pattern data, and a third correlation coefficient between the bending angle interval data and the temperature change pattern data are weightedly summed to obtain the multidimensional feature vector; the multidimensional feature vector is used to reflect the working condition characteristics of the flexible device.

[0032] As one of the preferred solutions, the step of constructing and classifying a multi-dimensional feature vector to obtain a feature classification result of the flexible device includes:

[0033] Roughly classifying the multi-dimensional feature vectors by a support vector machine to obtain a number of data clusters; the data clusters are used to characterize the multi-dimensional feature vectors corresponding to different working conditions;

[0034] The data cluster is refined and classified by a preset clustering algorithm to obtain multiple feature clusters, and the cluster center within the feature cluster is calculated to output as the feature classification result of the flexible device; the cluster center is the mean of all data points in the feature cluster, which is used to characterize the typical multi-dimensional feature vector of the corresponding working condition.

[0035] As one of the preferred solutions, obtaining the feature classification result of the flexible device to construct a full-operating efficiency monitoring database includes:

[0036] A working condition parameter matrix is ​​constructed based on the cluster centers, a linear quantization mapping is performed on the working condition parameter matrix to construct a working condition comprehensive feature space through three-dimensional vector space transformation, and the working condition comprehensive feature space is grid-divided to determine a working condition characteristic value according to the number of data points in each grid;

[0037] The operating condition characteristic values ​​are hierarchically divided through a three-layer decision tree to obtain an operating condition segmentation standard, and the operating condition parameter matrix is ​​calibrated according to the operating condition segmentation standard. A time sliding window is used to extract the steady-state component of the operating condition parameter change of the calibrated operating condition parameter matrix to construct a dynamic monitoring index for the operating condition parameters;

[0038] The dynamic monitoring indicators of the operating parameters are combined, and a four-layer feedforward neural network is used to extract the combined features of the operating parameters in the combination results. The combined features of the operating parameters are feature mapped through a linear rectification function, and the mapping results are graded to obtain the full operating efficiency monitoring database.

[0039] As one of the preferred solutions, the real-time tensile stress data, real-time bending angle data and real-time temperature change data of the flexible device are obtained to match with the full-operating efficiency monitoring database, and the operating efficiency monitoring process of the flexible device is optimized according to the comprehensive monitoring scoring matrix obtained by matching, including:

[0040] Acquire real-time tensile stress data, real-time bending angle data, and real-time temperature change data of the flexible device, and calculate cosine similarity with the full-operating efficiency monitoring database, select corresponding real-time data based on the relationship between a preset similarity threshold and the calculation result, and determine a quantitative value of parameter matching degree based on the autocorrelation coefficient of the real-time data;

[0041] A comprehensive monitoring scoring matrix is ​​constructed according to the quantitative values ​​of the parameter matching degree, and the operation efficiency score of the comprehensive monitoring scoring matrix is ​​calculated by support vector regression of a Gaussian kernel function, so as to quantitatively calibrate the operation efficiency score through weighted combination;

[0042] A Kalman filter is used to smooth the quantified and calibrated operating efficiency score, the operating efficiency score of the current working condition is updated through recursive calculation, and the operating efficiency score of the current working condition is divided according to a preset efficiency division interval, and the division result is obtained as the output of the operating efficiency monitoring result of the flexible equipment under the current working condition.

[0043] A second aspect of the present invention provides a system for monitoring the operating efficiency of a flexible device, comprising:

[0044] A data acquisition module, used to acquire historical monitoring data of the flexible device under various working conditions; the historical monitoring data includes historical tensile stress data, historical bending angle data and historical temperature change data;

[0045] A level classification module, used to classify the historical tensile stress data according to a preset tensile stress threshold value, and obtain tensile stress level data of the flexible device under corresponding working conditions;

[0046] An interval division module, used for quantifying the angle change rate of the flexible device under the corresponding working condition based on the historical bending angle data to perform interval division, and obtaining the bending angle interval data of the flexible device under the corresponding working condition;

[0047] A mode division module is used to extract key features from the historical temperature change data, and divide the historical temperature change data into modes according to a preset temperature change threshold and the key features, so as to obtain temperature change mode data of the flexible device under corresponding working conditions;

[0048] A database construction module, used to construct and classify a multi-dimensional feature vector based on the tensile stress level data, the bending angle interval data and the temperature change mode data, to obtain a feature classification result of the flexible device, so as to construct a full-operating-condition efficiency monitoring database;

[0049] The efficiency monitoring module is used to obtain the real-time tensile stress data, real-time bending angle data and real-time temperature change data of the flexible device to match them with the full-operating efficiency monitoring database, and optimize the operating efficiency monitoring process of the flexible device according to the comprehensive monitoring scoring matrix obtained by matching.

[0050] Compared with the prior art, the embodiments of the present invention have the following advantages:

[0051] (1) By collecting and analyzing the historical monitoring data of flexible equipment under various working conditions, we can have a more comprehensive understanding of the operating status of flexible equipment and improve the accuracy and reliability of monitoring; by classifying the historical data into levels, intervals and modes, it is helpful to accurately identify the operating status of flexible equipment under different working conditions, providing strong support for subsequent analysis and monitoring;

[0052] (2) By constructing and classifying multi-dimensional feature vectors to form an efficiency monitoring database, rapid and efficient monitoring of the operating efficiency of flexible equipment can be achieved to improve monitoring efficiency. By collecting the operating data of flexible equipment in real time and matching it with the full-condition efficiency monitoring database, the operating efficiency of the equipment can be monitored in real time based on the matched comprehensive monitoring score matrix, potential problems can be discovered in a timely manner, and timely feedback can be provided for equipment maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solution of the present invention, the drawings required for use in the implementation mode will be briefly introduced below. Obviously, the drawings described below are only some implementation modes of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0054] Figure 1 is a flow chart of a method for monitoring operating efficiency of a flexible device provided by a certain embodiment of the present invention;

[0055] Figure 2 It is a structural diagram of a working condition operation efficiency monitoring system for flexible equipment provided by a certain embodiment of the present invention. DETAILED DESCRIPTION

[0056] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings and embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] In the description of this application, the terms "first", "second", "third", etc. are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", "third", etc. may explicitly or implicitly include one or more of the feature. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0058] In the description of the present application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be a connection between the two components. The terms "vertical", "horizontal", "left", "right", "upper", "lower" and similar expressions used herein are for illustrative purposes only, and do not indicate or imply that the system or component referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0059] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as those commonly understood by those skilled in the art. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood by specific circumstances.

[0060] In one embodiment, if Figure 1 As shown, the first aspect of the present invention provides a method for monitoring the operating efficiency of a flexible device, comprising:

[0061] S1. Obtaining historical monitoring data of the flexible device under various working conditions; the historical monitoring data includes historical tensile stress data, historical bending angle data and historical temperature change data;

[0062] Specifically, the present invention obtains historical tensile stress data, historical bending angle data and historical temperature change data of the flexible device under different working conditions within a preset period of time in the past (such as one year); wherein the historical tensile stress data include the change in the size and direction of the tensile stress of the flexible device during the continuous operation time of each working condition, the bending angle data include the change rate and duration of the bending angle of the flexible device during the continuous operation time of each working condition, and the historical temperature change data include the temperature change rate, amplitude, frequency, etc. of the flexible device during the continuous operation time of each working condition; then, these historical monitoring data are cleaned, including removing duplicate data, invalid data and abnormal data, etc., to ensure the integrity and consistency of the data, and the cleaned data are classified and sorted according to the data type (tensile stress, bending angle, temperature change) and working condition, so as to facilitate subsequent data calls, so as to have a more comprehensive understanding of the operating status of the flexible device, thereby improving the accuracy and reliability of monitoring.

[0063] S2. Classifying the historical tensile stress data according to a preset tensile stress threshold value to obtain tensile stress level data of the flexible device under corresponding working conditions;

[0064] In one embodiment, step S2 includes:

[0065] constructing a stress reference data matrix through the historical tensile stress data, and quantifying stress change frequency data according to the stress reference data matrix;

[0066] Constructing a two-dimensional stress frequency matrix according to the stress change frequency data, and performing a primary level division on the two-dimensional stress frequency matrix by a clustering algorithm based on Euclidean distance to obtain a preliminary stress level;

[0067] The preliminary stress level is adjusted based on the preset tensile stress threshold to obtain tensile stress level data of the flexible device under corresponding working conditions.

[0068] Specifically, the present invention divides the historical tensile stress data from 200 Pa to 1000 Pa into 40 quantization intervals of 20 Pa to form a stress reference data matrix, and divides the difference between the stress values ​​at every two adjacent sampling moments in the stress reference data matrix by the sampling time interval to obtain the stress change per unit time; because when the flexible device operates under stable conditions, the stress change fluctuates within the range of ±5 Pa per millisecond, then the stress change per unit time is spectrally analyzed by Fourier transform, and the main frequency component in the analysis result is extracted to obtain stress change frequency data, that is, the stress change frequency under stable conditions is concentrated in the range of 5 Hz to 20 Hz.

[0069] Then, a two-dimensional stress frequency matrix is ​​constructed according to the stress change frequency data. The matrix takes the stress change frequency data corresponding to the stress change per unit time as the horizontal axis and the stress direction offset angle as the vertical axis. The clustering algorithm based on Euclidean distance is used to calculate the similarity between the data points in the two-dimensional stress frequency matrix for initial grade division, so as to cluster the data points with high similarity into the same stress level and obtain the preliminary stress level.

[0070] Finally, the preliminary stress level is divided into intervals again in the form of upper and lower limit thresholds and center values. For example, if the center value of a stress level interval is 500 Pa, the upper and lower limit thresholds are ±50 Pa, and the data points with stress values ​​in the range of 450 Pa to 550 Pa belong to this level, then the preliminary stress level is adjusted according to the preset tensile stress threshold. That is, when it is detected that the preliminary stress level exceeds the preset threshold of ±5 Pa per millisecond, the stress data contained in the preliminary stress level is subjected to spectral analysis. If the stress change frequency exceeds the range of 5 Hz to 20 Hz, the preliminary stress level to which the data point belongs is adjusted to achieve fine-grained division of stress levels and obtain the tensile stress level data of the flexible device under the corresponding working conditions.

[0071] In practical applications, the stress change amount of flexible equipment is large and the change frequency is high during the startup phase. As the flexible equipment gradually enters a stable operating state, the stress change amount and change frequency gradually decrease and tend to be stable. When the flexible equipment has an abnormal working condition, the stress change characteristics will deviate from the normal range. By real-time monitoring of the stress change amount and change frequency, combined with the fine-grained division results of the stress level, the operating state of the flexible equipment can be judged. The present invention can more accurately reflect the stress state of the flexible equipment under different working conditions by constructing a stress reference data matrix and quantifying the stress change frequency, thereby improving the accuracy of the evaluation; using a clustering algorithm to grade the two-dimensional stress frequency matrix, the distribution of the flexible equipment under different stress states can be clearly displayed, which helps to better understand and analyze the stress characteristics of the flexible equipment; through accurate tensile stress level division, it can provide strong support for the management and maintenance of flexible equipment, and help to formulate more reasonable maintenance plans and decisions.

[0072] S3, quantifying the angle change rate of the flexible device under the corresponding working condition based on the historical bending angle data to perform interval division, and obtaining the bending angle interval data of the flexible device under the corresponding working condition;

[0073] In one embodiment, step S3 includes:

[0074] The historical bending angle data is processed in sections using a first fixed time window, and the angle change trend curve of each section is calculated by cubic polynomial least squares fitting to quantify the angle change rate of the deformation area;

[0075] The angle change rates of the deformation regions are grouped by a clustering algorithm, the slope of the angle change trend curve in each cluster group is quantified to divide the boundary interval, and an angle hierarchical quantization matrix is ​​constructed based on the boundary interval;

[0076] The angle graded quantization matrix is ​​subjected to Fourier transform to obtain a change frequency characteristic, and the distance between the change frequency characteristic and a preset change frequency threshold is quantized to adjust the constructed angle graded quantization matrix to obtain the bending angle interval data of the flexible device under the corresponding working condition.

[0077] Specifically, the present invention uses a first fixed time window (such as a 20-millisecond time window) to perform segmented processing on the historical bending angle data, and calculates the angle change trend curve of each segment through cubic polynomial least squares fitting. The curve coefficient reflects the fast and slow characteristics of the angle change. When the flexible device is in normal operation, the angle change trend curve is smooth and the coefficient value is small. When the flexible device is abnormally deformed, the angle changes dramatically and the coefficient value increases significantly. Based on the angle change trend curve, the angle change rate within it is averaged to obtain the angle change rate of the deformation area.

[0078] A clustering algorithm based on Euclidean distance is used to classify the angular change rate of the deformation area. The clustering radius is set to 5 degrees per second. The angle change data are initially grouped according to the cluster center value. The slope of the angle change trend curve in each cluster group is calculated, and the slope value is divided into boundary intervals of 10 degrees per second. An angle hierarchical quantization matrix is ​​constructed based on the boundary intervals. That is, the angle data with similar change characteristics are divided into the same group based on the boundary intervals to form an angle hierarchical quantization matrix.

[0079] The angle graded quantization matrix is ​​subjected to Fourier transform to obtain the change frequency characteristics, and the preset change frequency threshold is set to 0.5 Hz to 2 Hz. According to the distance between the actual frequency value corresponding to the angle graded quantization matrix data and the boundary of the threshold interval, fine-grained adjustment is performed within the range of plus or minus 2 degrees to obtain the bending angle interval data of the flexible device under the corresponding working conditions.

[0080] When the flexible device is operating normally, the frequency of angle changes is mainly concentrated in the range of 0.5 Hz to 2 Hz, which corresponds to the normal movement cycle of the flexible device. When the flexible device is abnormally deformed, the frequency of angle changes will deviate from this range. The frequency characteristics in the angle change sequence are extracted by Fourier transform, which can realize the identification of abnormal deformation. The deviation between the frequency characteristic value and the preset threshold interval determines the amplitude of fine-grained adjustment of the angle interval. The larger the deviation, the larger the adjustment amplitude. Generally speaking, in the actual operation of flexible equipment, there are significant differences in deformation characteristics under different working conditions. The angle changes are large and the frequency is high in the startup stage, the angle changes are gentle and the frequency is stable in the stable operation stage, and the angle changes gradually decrease until they stop in the shutdown stage.

[0081] The present invention can accurately capture the angle change trend and quantify the angle change rate of the deformation area through cubic polynomial least squares fitting, thereby providing an accurate data basis for subsequent clustering analysis and frequency feature extraction; the angle change rate of the deformation area is grouped through a clustering algorithm, and the boundary interval is divided by the quantization slope, thereby simplifying the processing process of complex data and improving the analysis efficiency; by extracting the change frequency characteristics in the angle graded quantization matrix, it is helpful to understand the operating status of the flexible device; by comparing with the preset change frequency threshold to adaptively adjust the angle graded quantization matrix, the obtained bending angle interval data is more in line with the actual situation, thereby improving the accuracy and practicality of the data.

[0082] S4, extracting key features from the historical temperature change data, and dividing the historical temperature change data into patterns according to a preset temperature change threshold and the key features, to obtain temperature change pattern data of the flexible device under corresponding working conditions;

[0083] In one embodiment, extracting key features from the historical temperature change data includes:

[0084] The historical temperature change data is constructed into a historical temperature monitoring data matrix by a time series algorithm, and a first difference between adjacent data in the historical temperature monitoring data matrix is ​​quantified to obtain a historical temperature change rate based on the first difference and a time window;

[0085] Performing Fourier transformation on the historical temperature monitoring data matrix to obtain the historical temperature frequency component, and quantifying the second difference between the minimum temperature value and the maximum temperature value of the historical temperature monitoring data matrix within the time window to obtain the historical temperature change amplitude;

[0086] The historical temperature change rate, the historical temperature frequency component and the historical temperature change amplitude are combined to obtain the key feature.

[0087] Specifically, the present invention arranges the historical temperature change data in chronological order to form time series data and performs standardization processing, and then extracts temperature-related features such as temperature amplitude and change rate, and converts the time series data into an array format to facilitate training of the autoregressive integrated moving average model with the temperature-related features to form a historical temperature monitoring data matrix, in which each row represents the temperature data at a time point, and each column represents a feature dimension; the difference between adjacent temperature data in the historical temperature monitoring data matrix is ​​calculated according to a fixed time window of 2 seconds as the first difference, so as to obtain the temperature change rate by the ratio of the first difference to the time window.

[0088] Perform Fourier transformation on the data in the historical temperature monitoring data matrix, extract the principal component in the transformation result, and obtain the historical temperature frequency component; calculate the second difference between the minimum temperature and the maximum temperature in the historical temperature monitoring data matrix within a fixed time window of 2 seconds as the historical temperature change amplitude; finally, combine the historical temperature change rate, the historical temperature frequency component and the historical temperature change amplitude to obtain the key features. The present invention calculates the temperature change rate, frequency component and change amplitude to comprehensively capture the dynamic characteristics of temperature data including change trend, periodic change and change range, and provides rich information for subsequent temperature monitoring and analysis.

[0089] In one embodiment, the historical temperature change data is divided into patterns by using the preset temperature change threshold and the key feature to obtain the temperature change pattern data of the flexible device under the corresponding working condition, including:

[0090] quantifying the similarity between the key features by a preset clustering algorithm, and classifying the key features based on the similarity to obtain a number of cluster centers;

[0091] Establishing an initial boundary of the historical temperature change data based on the cluster center, and calculating a temperature change quantization value of the historical temperature change data based on the initial boundary to obtain a temperature change pattern;

[0092] Performing five-layer wavelet decomposition on the historical temperature change data to extract local temperature features, reconstructing the local temperature features through temperature feature scale coefficients to obtain temperature change reconstructed features;

[0093] The variation deviation between the temperature variation reconstruction feature and the preset temperature variation threshold is quantified, and the temperature variation pattern is adjusted according to the variation deviation to obtain the temperature variation pattern data of the flexible device under the corresponding working condition.

[0094] Specifically, the present invention uses a mean clustering algorithm based on Euclidean distance to realize temperature feature vector classification, uses Euclidean distance to calculate the similarity between key features, sets the cluster radius to 5 degrees Celsius per minute, classifies key features based on the calculated similarity, obtains several cluster centers, and divides the initial boundary of historical temperature change data according to the cluster center, sets the temperature change rate threshold interval to 10 degrees Celsius per minute, adjusts and calculates the slope of the temperature change trend curve according to the initial boundary to obtain the temperature change quantification value, and establishes a fine-grained division interval for the quantification value every 5 degrees Celsius to obtain a temperature change pattern, which can characterize the temperature change interval of different working stages of the flexible device under different working conditions. Usually in the actual operation of flexible equipment, the temperature change shows obvious working condition characteristics, the temperature rises rapidly during the startup process and the change frequency is high, the temperature changes slowly and the frequency is stable during stable operation, and the temperature gradually decreases during the shutdown process until the ambient temperature is restored. By real-time monitoring of temperature change characteristics and combining the fine-grained pattern division results, the judgment and early warning of the equipment operation status can be realized.

[0095] Then, the historical temperature change data is subjected to five-layer wavelet decomposition to extract local temperature features. Wavelet decomposition can extract the features of temperature data at different time scales. The temperature change features from high frequency to low frequency are obtained through five-layer decomposition. The high-frequency component reflects the instantaneous fluctuation of temperature, and the low-frequency component reflects the long-term change trend of temperature. The local temperature features are reconstructed using the temperature characteristic scale coefficient to obtain the temperature change reconstruction features to identify abnormal temperature changes. The change deviation between the temperature change reconstruction features and the preset temperature change threshold is calculated, and the temperature change pattern is adjusted according to the size of the change deviation, that is, the temperature change pattern is fine-grained adjusted within the range of plus or minus 2 degrees Celsius to obtain the temperature change pattern data of the flexible equipment under the corresponding working conditions.

[0096] The present invention can quickly identify similarities and differences in temperature data by quantifying the similarities between key features and performing cluster analysis, thereby improving the efficiency of feature extraction, helping to reveal the inherent laws of temperature changes, and providing a basis for extracting temperature change patterns; the initial boundaries and temperature change quantization values ​​established based on cluster centers can accurately describe the overall trends and characteristics of historical temperature change data, helping to form accurate and reliable temperature change patterns; the five-layer wavelet decomposition can deeply analyze the detailed features of temperature data, extract local temperature features, and reconstruct features through temperature feature scale coefficients, which can further optimize temperature change features and improve the accuracy and reliability of temperature change pattern data; by quantifying the change deviation between the reconstructed features of temperature changes and a preset temperature change threshold, and adjusting the temperature change pattern according to the deviation, adaptive optimization can be achieved to ensure that the temperature change pattern data is more in line with actual conditions, thereby improving the practicality and application value of the data.

[0097] S5. constructing and classifying a multi-dimensional feature vector based on the tensile stress level data, the bending angle interval data, and the temperature change mode data to obtain a feature classification result of the flexible device to construct a full-operating-condition efficiency monitoring database;

[0098] In one embodiment, constructing a multi-dimensional feature vector based on the tensile stress level data, the bending angle interval data and the temperature change mode data includes:

[0099] Aligning the tensile stress level data, the bending angle interval data, and the temperature change pattern data on a time axis, and performing data extraction through a second fixed time window to construct a multi-dimensional feature data matrix;

[0100] The multidimensional feature data matrix is ​​subjected to correlation calculation, and a first correlation coefficient between the tensile stress level data and the bending angle interval data, a second correlation coefficient between the tensile stress level data and the temperature change pattern data, and a third correlation coefficient between the bending angle interval data and the temperature change pattern data are weightedly summed to obtain the multidimensional feature vector; the multidimensional feature vector is used to reflect the working condition characteristics of the flexible device.

[0101] Specifically, the working characteristics of flexible equipment are reflected in three dimensions: tensile stress, bending angle and temperature change. The data of these three dimensions are synchronized in time. They are aligned on the time axis and segmented by a 1-second time window to capture the instantaneous working characteristics, so as to extract their corresponding fine-grained classification results. These data are uniformly converted to the range of 0 to 1 through minimum and maximum value normalization processing to construct a multidimensional feature data matrix; the Pearson correlation coefficient is used to calculate the correlation coefficient between the tensile stress level data and the bending angle interval data, the tensile stress level data and the temperature change mode data, and the bending angle interval data and the temperature change mode data in the multidimensional feature data matrix as the feature combination weight, and the weight value corresponding to the correlation coefficient greater than 0.8 is set to 0.4, the weight value corresponding to the correlation coefficient between 0.5 and 0.8 is set to 0.3, and the weight value corresponding to the correlation coefficient less than 0.5 is set to 0.2. The multidimensional feature vector is constructed by weighted summation to comprehensively reflect the working characteristics of the flexible equipment.

[0102] The present invention aligns different types of data (tensile stress level, bending angle range, temperature change pattern) on the time axis to ensure the consistency and comparability of the data, providing an accurate basis for subsequent analysis; uses a fixed time window for data extraction to effectively extract key information from a large amount of original data and construct a multi-dimensional feature data matrix to facilitate subsequent feature analysis and pattern recognition; calculates the correlation of the multi-dimensional feature data matrix to reveal the intrinsic connection between different operating parameters; and obtains a multi-dimensional feature vector by weighted summation of correlation coefficients, which can comprehensively reflect the operating characteristics of the flexible equipment and provide strong support for subsequent equipment monitoring.

[0103] In one embodiment, the constructing and classifying the multi-dimensional feature vector to obtain the feature classification result of the flexible device includes:

[0104] Roughly classifying the multi-dimensional feature vectors by a support vector machine to obtain a number of data clusters; the data clusters are used to characterize the multi-dimensional feature vectors corresponding to different working conditions;

[0105] The data cluster is refined and classified by a preset clustering algorithm to obtain multiple feature clusters, and the cluster center within the feature cluster is calculated to output as the feature classification result of the flexible device; the cluster center is the mean of all data points in the feature cluster, which is used to characterize the typical multi-dimensional feature vector of the corresponding working condition.

[0106] Specifically, the present invention uses a Gaussian kernel function through a support vector machine to realize nonlinear mapping of feature space to roughly classify multi-dimensional feature vectors; wherein, the kernel function parameter value of 0.1 ensures the smoothness of the mapping, and the penalty factor of 100 ensures the strictness of the classification boundary. In the three-dimensional feature space, the feature vectors corresponding to different working conditions form relatively independent data clusters, and the hyperplane established by maximizing the classification interval can effectively distinguish these data clusters. Then, the mean clustering algorithm based on Euclidean distance is used to refine the classification of data clusters, and the cluster distance threshold is set to 0.2, that is, the tensile stress difference is within 50 Pa, the bending angle difference is within 5 degrees, and the temperature difference is within 10 degrees Celsius. The data points within the threshold are classified into the same feature cluster, and the mean of all data points in the feature cluster is calculated, that is, the cluster center represents the typical characteristics of the corresponding working condition.

[0107] In practical applications, the working conditions of flexible equipment are often manifested as a combination of changes in multiple features. When an abnormality occurs in a flexible device, the tensile stress increases rapidly, the bending angle changes accordingly, and the temperature also rises accordingly. The characteristic values ​​of the three dimensions deviate from the normal range at the same time. The fused feature vector can comprehensively reflect this multi-dimensional change feature. Through the combined use of support vector machines and mean clustering, accurate classification of equipment working conditions is achieved. The present invention combines preliminary classification with detailed classification to improve the accuracy and meticulousness of classification; by calculating the cluster center within the feature cluster as a typical multi-dimensional feature vector of the corresponding working condition, it can intuitively reflect the working condition characteristics of the equipment and provide strong support for the subsequent monitoring of the operating efficiency of flexible equipment.

[0108] In one embodiment, obtaining the feature classification result of the flexible device to construct a full-operating efficiency monitoring database includes:

[0109] A working condition parameter matrix is ​​constructed based on the cluster centers, a linear quantization mapping is performed on the working condition parameter matrix to construct a working condition comprehensive feature space through three-dimensional vector space transformation, and the working condition comprehensive feature space is grid-divided to determine a working condition characteristic value according to the number of data points in each grid;

[0110] The operating condition characteristic values ​​are hierarchically divided through a three-layer decision tree to obtain an operating condition segmentation standard, and the operating condition parameter matrix is ​​calibrated according to the operating condition segmentation standard. A time sliding window is used to extract the steady-state component of the operating condition parameter change of the calibrated operating condition parameter matrix to construct a dynamic monitoring index for the operating condition parameters;

[0111] The dynamic monitoring indicators of the operating parameters are combined, and a four-layer feedforward neural network is used to extract the combined features of the operating parameters in the combination results. The combined features of the operating parameters are feature mapped through a linear rectification function, and the mapping results are graded to obtain the full operating efficiency monitoring database.

[0112] Specifically, the present invention constructs an operating condition parameter matrix based on the cluster center, and linearly quantizes and maps the operating condition parameter matrix from three dimensions of tensile stress, bending angle, and temperature change in the interval of 0 to 1, so as to construct an operating condition comprehensive feature space through three-dimensional vector space conversion. Taking tensile stress as an example, the original value is in the range of 0 to 1000 Pa, and is converted to the interval of 0 to 1 through linear mapping, which is convenient for unified processing with bending angle and temperature change data. The characteristic values ​​of the three dimensions form discrete point distribution in space, and the Euclidean distance between points reflects the similarity of the operating conditions. Since the point density analysis of the operating condition comprehensive feature space reflects the distribution characteristics of the operating condition parameters, the operating condition comprehensive feature space is gridded to obtain a number of cubic grids with a side length of 0.1, and the operating condition characteristic value, that is, the characteristic point in the grid, is determined according to the number of data points in each grid. Under normal operating conditions of the flexible equipment, the characteristic points are mainly concentrated in a specific grid, and the grid point density is high, while the characteristic points corresponding to abnormal conditions are relatively scattered, and the grid point density is low.

[0113] A three-layer decision tree is used to realize the hierarchical division of working conditions. The first layer divides the working conditions into high stress and low stress based on the tensile stress characteristics. The second layer introduces the bending angle characteristics to subdivide the working conditions into four categories. The third layer combines the temperature change characteristics to finally form the working condition subdivision standard. Each working condition category has a clear characteristic boundary value to achieve accurate identification of the working condition. The working condition parameter matrix is ​​calibrated according to the working condition subdivision standard. A 5-second time sliding window is used to calculate the rate of change of the working condition parameters. The change rate value is band-pass filtered from 0.1 Hz to 1 Hz to extract the steady-state component of the working condition parameter change. The mean square error of the working condition parameters is calculated based on the steady-state component sequence as a dynamic monitoring indicator of the working condition parameters.

[0114] The dynamic monitoring indicators of operating condition parameters are combined, and a four-layer feedforward neural network is used to extract the combined characteristics of operating condition parameters. The number of hidden layer neurons is set to 32, 16, 8, and 4 respectively. The 32 neurons in the first hidden layer realize the preliminary conversion of the combined characteristics of the operating condition parameters. After layer-by-layer mapping of 16, 8, and 4 neurons, a low-dimensional operating condition feature representation is finally obtained. The feature mapping is realized through a linear rectification function. The mapping results are divided into five quantization levels according to the interval of 0.2, and a full operating condition efficiency monitoring database is established to correspond to the different operating efficiency ranges of flexible equipment from low load to high load.

[0115] In practical applications, the working conditions of flexible equipment are often manifested as coupled changes in multiple parameters. When the equipment switches from low-load to high-load working conditions, the tensile stress rises rapidly, the bending angle increases accordingly, the temperature gradually rises, and the changes in working condition parameters have obvious time series characteristics. The present invention constructs a comprehensive working condition feature space and performs grid division, which can intuitively display the distribution of different working condition characteristics in three-dimensional space, providing a basis for subsequent decision tree division and neural network feature extraction; hierarchical division of working condition characteristic values ​​is performed using a three-layer decision tree to obtain detailed working condition subdivision standards, which helps to more accurately identify different operating states of the equipment; a time sliding window is used to extract the steady-state component in the calibrated working condition parameter matrix, and a working condition parameter dynamic monitoring index is constructed, which can reflect the operating state changes of the equipment in real time and provide important information for the evaluation of operating efficiency; the multi-dimensional characteristics, dynamic changes and characteristic differences of the equipment under different working conditions are comprehensively considered, and a comprehensive full-working condition efficiency monitoring database is obtained through multi-level and multi-step processing and analysis.

[0116] S6. Acquire real-time tensile stress data, real-time bending angle data, and real-time temperature change data of the flexible device to match them with the full-operating efficiency monitoring database, and optimize the operating efficiency monitoring process of the flexible device according to the comprehensive monitoring scoring matrix obtained by matching;

[0117] In one embodiment, step S6 includes:

[0118] Acquire real-time tensile stress data, real-time bending angle data, and real-time temperature change data of the flexible device, and calculate cosine similarity with the full-operating efficiency monitoring database, select corresponding real-time data based on the relationship between a preset similarity threshold and the calculation result, and determine a quantitative value of parameter matching degree based on the autocorrelation coefficient of the real-time data;

[0119] A comprehensive monitoring scoring matrix is ​​constructed according to the quantitative values ​​of the parameter matching degree, and the operation efficiency score of the comprehensive monitoring scoring matrix is ​​calculated by support vector regression of a Gaussian kernel function, so as to quantitatively calibrate the operation efficiency score through weighted combination;

[0120] A Kalman filter is used to smooth the quantified and calibrated operating efficiency score, the operating efficiency score of the current working condition is updated through recursive calculation, and the operating efficiency score of the current working condition is divided according to a preset efficiency division interval, and the division result is obtained as the output of the operating efficiency monitoring result of the flexible equipment under the current working condition.

[0121] Specifically, the present invention collects tensile stress values, bending angle values ​​and temperature values ​​of a flexible device every 50 milliseconds, and converts these real-time data to the range of 0 to 1 through maximum and minimum value normalization to eliminate the dimensional differences between different physical quantities, so that the tensile stress, bending angle and temperature data are compared in the same numerical range, and a 20-point sliding mean window corresponds to a time span of 1 second, which smoothes the noise while retaining the main trend of parameter changes; the vector inner product is divided by the vector modulus product to calculate the cosine similarity between the real-time parameters and the parameter template in the full-condition efficiency monitoring database, and a similarity threshold of 0.8 is set. The autocorrelation coefficients of 50 consecutive data points are calculated for parameter matching items above the threshold, and the stability of the parameter sequence is judged by the decay rate of the autocorrelation coefficient to obtain a quantitative value of the parameter matching degree.

[0122] A comprehensive monitoring scoring matrix was constructed based on the quantitative values ​​of the parameter matching degree, and the support vector regression of the Gaussian kernel function was used to calculate the operation efficiency score of the comprehensive monitoring scoring matrix. Specifically, the kernel function bandwidth parameter was set to 0.1, the penalty factor was set to 100, the tensile stress contribution value was weighted 0.4, the bending angle influence value was weighted 0.3, and the temperature change weight value was weighted 0.3. The operation efficiency score was quantitatively calibrated through weighted combination.

[0123] A Kalman filter with a two-dimensional state space is used to smooth the operating efficiency score after quantization and calibration to describe the dynamic characteristics of the operating efficiency. The state variables include the efficiency value and its rate of change, where the observation noise covariance is set to 0.1 and the process noise covariance is set to 0.01. The operating efficiency value of the current operating condition is updated by recursive calculation, and the operating efficiency of the current operating condition is fine-grainedly divided according to the interval division of 0 to 0.2 corresponding to low efficiency, 0.2 to 0.4 corresponding to relatively low efficiency, 0.4 to 0.6 corresponding to medium efficiency, 0.6 to 0.8 corresponding to relatively high efficiency, and 0.8 to 1 corresponding to high efficiency. The division result is obtained as the output of the operating efficiency monitoring result of the flexible equipment under the current working condition.

[0124] Generally, the operating efficiency of flexible equipment shows obvious dependence on working conditions. When the flexible equipment is operated at the optimal working point, the tensile stress is maintained within the rated range, the bending angle remains stable, the temperature changes gently, and the operating efficiency score is in the high efficiency range of 0.8 to 1. Under non-optimal working conditions, the parameter fluctuation increases and the efficiency score decreases. Through real-time monitoring and evaluation, the operating state of the flexible equipment can be accurately characterized and quantitatively evaluated. The present invention adopts a combination of cosine similarity and autocorrelation analysis to objectively reflect the degree of matching between data and current working conditions, and provides a basis for the construction of a comprehensive monitoring scoring matrix; the support vector regression (SVR) of the Gaussian kernel function is used to calculate the operating efficiency score of the comprehensive monitoring scoring matrix, which can make full use of the nonlinear relationship of the data and improve the accuracy and reliability of the score calculation; the operating efficiency score is quantitatively calibrated by weighted combination, and the Kalman filter is used for smoothing, which can reduce the noise and fluctuation in the score and improve the stability and credibility of the score; the operating efficiency score of the current working condition is divided according to the preset efficiency division interval, which can intuitively display the operating efficiency state of the flexible equipment under all working conditions, and provide strong support for the operation management and maintenance of the equipment.

[0125] In the embodiment of the present application, based on the problem of how to improve the monitoring of the operating efficiency of flexible equipment, a method for monitoring the operating efficiency of flexible equipment is designed, which realizes the acquisition of historical monitoring data of flexible equipment under various working conditions; the historical monitoring data includes historical tensile stress data, historical bending angle data and historical temperature change data; the historical tensile stress data is graded according to a preset tensile stress threshold to obtain tensile stress grade data of the flexible equipment under the corresponding working condition; based on the historical bending angle data, the angle change rate of the flexible equipment under the corresponding working condition is quantified to perform interval division to obtain the bending angle interval data of the flexible equipment under the corresponding working condition; the key features in the historical temperature change data are extracted, and the temperature change threshold and the key features are used to compare the historical tensile stress data. The historical temperature change data is divided into patterns to obtain the temperature change pattern data of the flexible device under the corresponding working conditions; a multi-dimensional feature vector is constructed and classified based on the tensile stress level data, the bending angle interval data and the temperature change pattern data to obtain the feature classification result of the flexible device to build a full-working efficiency monitoring database; the real-time tensile stress data, real-time bending angle data and real-time temperature change data of the flexible device are obtained to match them with the full-working efficiency monitoring database, and according to the comprehensive monitoring scoring matrix obtained by matching, a technical solution is provided to optimize the working condition operation efficiency monitoring process of the flexible device; the operating efficiency of the flexible device under different working conditions can be accurately evaluated, the monitoring and control accuracy of the flexible device can be improved, and the service life of the flexible device can be extended.

[0126] It should be noted that although the steps in the above flowchart are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders.

[0127] In another embodiment, if Figure 2 As shown, the second aspect of the present invention provides a working condition operation efficiency monitoring system for flexible equipment, comprising:

[0128] The data acquisition module 10 is used to acquire the historical monitoring data of the flexible device under various working conditions; the historical monitoring data includes historical tensile stress data, historical bending angle data and historical temperature change data;

[0129] A level classification module 20, configured to classify the historical tensile stress data according to a preset tensile stress threshold value, and obtain tensile stress level data of the flexible device under corresponding working conditions;

[0130] An interval division module 30, configured to quantify the angle change rate of the flexible device under the corresponding working condition based on the historical bending angle data to perform interval division, and obtain bending angle interval data of the flexible device under the corresponding working condition;

[0131] A mode division module 40 is used to extract key features from the historical temperature change data, and divide the historical temperature change data into modes according to a preset temperature change threshold and the key features, so as to obtain temperature change mode data of the flexible device under corresponding working conditions;

[0132] A database construction module 50 is used to construct and classify a multi-dimensional feature vector based on the tensile stress level data, the bending angle interval data and the temperature change mode data, to obtain a feature classification result of the flexible device, so as to construct a full-operating-condition efficiency monitoring database;

[0133] The efficiency monitoring module 60 is used to obtain the real-time tensile stress data, real-time bending angle data and real-time temperature change data of the flexible device to match them with the full-operating efficiency monitoring database, and optimize the operating efficiency monitoring process of the flexible device according to the comprehensive monitoring scoring matrix obtained by matching.

[0134] It should be noted that each module in the above-mentioned operating efficiency monitoring system for flexible equipment can be fully or partially implemented by software, hardware and their combination. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules. For the specific definition of a system for monitoring the operating efficiency of a flexible device, please refer to the definition of a method for monitoring the operating efficiency of a flexible device above. The two have the same functions and effects, which will not be repeated here.

[0135] In summary, the present invention relates to the field of data processing technology, and discloses a method and system for monitoring the operating efficiency of flexible equipment. The method includes obtaining the tensile stress, bending angle and temperature change of the flexible equipment under different working conditions, analyzing the change rate and frequency of the tensile stress to divide the tensile stress into several levels, and analyzing the change rate and duration of the bending angle to divide the bending angle into several intervals, and extracting the rate, amplitude and frequency of temperature change to divide the temperature change into multiple modes; comprehensively constructing a multi-dimensional feature vector based on the tensile stress level, bending angle interval and temperature change mode and classifying it, and establishing an efficiency monitoring database for flexible equipment under all working conditions; acquiring the tensile stress, bending angle and temperature data of the flexible equipment in real time, and matching them with the efficiency monitoring database, and optimizing the operating efficiency monitoring process of the flexible equipment according to the comprehensive monitoring scoring matrix obtained by matching.

[0136] Each embodiment in this specification is described in a progressive manner, and the same or similar parts of each embodiment can be directly referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. It should be noted that the technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above-mentioned embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0137] The above-mentioned embodiments only express several preferred implementation modes of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in the technical field, several improvements and substitutions can be made without departing from the technical principles of the present invention, and these improvements and substitutions should also be regarded as the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be based on the protection scope of the claims.

Claims

1. A method for monitoring the operating efficiency of flexible equipment, characterized in that: include: Obtain historical monitoring data of flexible equipment under various working conditions; The historical monitoring data includes historical tensile stress data, historical bending angle data and historical temperature change data; Classifying the historical tensile stress data according to a preset tensile stress threshold value to obtain tensile stress level data of the flexible device under corresponding working conditions; quantifying the angle change rate of the flexible device under the corresponding working condition based on the historical bending angle data to perform interval division, and obtaining the bending angle interval data of the flexible device under the corresponding working condition; Extracting key features from the historical temperature change data, and dividing the historical temperature change data into patterns according to a preset temperature change threshold and the key features, to obtain temperature change pattern data of the flexible device under corresponding working conditions; Based on the tensile stress level data, the bending angle interval data and the temperature change mode data, a multi-dimensional feature vector is constructed and classified to obtain a feature classification result of the flexible device to construct a full-operating-condition efficiency monitoring database; Real-time tensile stress data, real-time bending angle data and real-time temperature change data of the flexible device are obtained to match them with the full-operating efficiency monitoring database, and the operating efficiency monitoring process of the flexible device is optimized based on the comprehensive monitoring scoring matrix obtained by matching.

2. A method for monitoring the operating efficiency of flexible equipment according to claim 1, characterized in that: The step of classifying the historical tensile stress data according to a preset tensile stress threshold to obtain tensile stress level data of the flexible device under corresponding working conditions includes: constructing a stress reference data matrix through the historical tensile stress data, and quantifying stress change frequency data according to the stress reference data matrix; Constructing a two-dimensional stress frequency matrix according to the stress change frequency data, and performing a preliminary level division on the two-dimensional stress frequency matrix by a clustering algorithm based on Euclidean distance to obtain a preliminary stress level; The preliminary stress level is adjusted based on the preset tensile stress threshold to obtain tensile stress level data of the flexible device under corresponding working conditions.

3. A method for monitoring operating efficiency of flexible equipment according to claim 1, characterized in that: The step of quantifying the angle change rate of the flexible device under the corresponding working condition based on the historical bending angle data to divide the flexible device into intervals, and obtaining the bending angle interval data of the flexible device under the corresponding working condition, includes: The historical bending angle data is processed in sections using a first fixed time window, and the angle change trend curve of each section is calculated by cubic polynomial least squares fitting to quantify the angle change rate of the deformation area; The angle change rates of the deformation regions are grouped by a clustering algorithm, the slope of the angle change trend curve in each cluster group is quantified to divide the boundary interval, and an angle hierarchical quantization matrix is ​​constructed based on the boundary interval; The angle graded quantization matrix is ​​subjected to Fourier transform to obtain a change frequency characteristic, and the distance between the change frequency characteristic and a preset change frequency threshold is quantized to adjust the constructed angle graded quantization matrix to obtain the bending angle interval data of the flexible device under the corresponding working condition.

4. A method for monitoring operating efficiency of flexible equipment according to claim 1, characterized in that: The extracting key features from the historical temperature change data includes: The historical temperature change data is constructed into a historical temperature monitoring data matrix by a time series algorithm, and a first difference between adjacent data in the historical temperature monitoring data matrix is ​​quantified to obtain a historical temperature change rate based on the first difference and a time window; Performing Fourier transformation on the historical temperature monitoring data matrix to obtain the historical temperature frequency component, and quantifying the second difference between the minimum temperature value and the maximum temperature value of the historical temperature monitoring data matrix within the time window to obtain the historical temperature change amplitude; The historical temperature change rate, the historical temperature frequency component and the historical temperature change amplitude are combined to obtain the key feature.

5. The method for monitoring the operating efficiency of flexible equipment according to claim 1, characterized in that: The method of dividing the historical temperature change data into patterns by using the preset temperature change threshold and the key feature to obtain the temperature change pattern data of the flexible device under the corresponding working condition includes: quantifying the similarity between the key features by a preset clustering algorithm, and classifying the key features based on the similarity to obtain a number of cluster centers; Establishing an initial boundary of the historical temperature change data based on the cluster center, and calculating a temperature change quantization value of the historical temperature change data based on the initial boundary to obtain a temperature change pattern; Performing five-layer wavelet decomposition on the historical temperature change data to extract local temperature features, reconstructing the local temperature features through temperature feature scale coefficients to obtain temperature change reconstructed features; The variation deviation between the temperature variation reconstruction feature and the preset temperature variation threshold is quantified, and the temperature variation pattern is adjusted according to the variation deviation to obtain the temperature variation pattern data of the flexible device under the corresponding working condition.

6. A method for monitoring operating efficiency of flexible equipment according to claim 1, characterized in that: The constructing of a multi-dimensional feature vector based on the tensile stress level data, the bending angle interval data and the temperature change mode data comprises: Aligning the tensile stress level data, the bending angle interval data, and the temperature change pattern data on a time axis, and extracting data through a second fixed time window to construct a multi-dimensional feature data matrix; The multidimensional feature data matrix is ​​subjected to correlation calculation, and a first correlation coefficient between the tensile stress level data and the bending angle interval data, a second correlation coefficient between the tensile stress level data and the temperature change pattern data, and a third correlation coefficient between the bending angle interval data and the temperature change pattern data are weightedly summed to obtain the multidimensional feature vector; the multidimensional feature vector is used to reflect the working condition characteristics of the flexible device.

7. A method for monitoring operating efficiency of flexible equipment according to claim 1, characterized in that: The constructing and classifying the multi-dimensional feature vector to obtain the feature classification result of the flexible device includes: Roughly classifying the multi-dimensional feature vectors by a support vector machine to obtain a number of data clusters; the data clusters are used to characterize the multi-dimensional feature vectors corresponding to different working conditions; The data cluster is refined and classified by a preset clustering algorithm to obtain multiple feature clusters, and the cluster center within the feature cluster is calculated to output as the feature classification result of the flexible device; the cluster center is the mean of all data points in the feature cluster, which is used to characterize the typical multi-dimensional feature vector of the corresponding working condition.

8. A method for monitoring operating efficiency of flexible equipment according to claim 7, characterized in that: The feature classification result of the flexible device is obtained to construct a full-operating efficiency monitoring database, including: A working condition parameter matrix is ​​constructed based on the cluster centers, a linear quantization mapping is performed on the working condition parameter matrix to construct a working condition comprehensive feature space through three-dimensional vector space transformation, and the working condition comprehensive feature space is grid-divided to determine a working condition characteristic value according to the number of data points in each grid; The operating condition characteristic values ​​are hierarchically divided through a three-layer decision tree to obtain an operating condition segmentation standard, and the operating condition parameter matrix is ​​calibrated according to the operating condition segmentation standard. A time sliding window is used to extract the steady-state component of the operating condition parameter change of the calibrated operating condition parameter matrix to construct a dynamic monitoring index for the operating condition parameters; The dynamic monitoring indicators of the operating parameters are combined, and a four-layer feedforward neural network is used to extract the combined features of the operating parameters in the combination results. The combined features of the operating parameters are feature mapped through a linear rectification function, and the mapping results are graded to obtain the full operating efficiency monitoring database.

9. A method for monitoring operating efficiency of flexible equipment according to claim 1, characterized in that: The real-time tensile stress data, real-time bending angle data and real-time temperature change data of the flexible device are obtained to match with the full-operating efficiency monitoring database, and the operating efficiency monitoring process of the flexible device is optimized according to the comprehensive monitoring scoring matrix obtained by matching, including: Acquire real-time tensile stress data, real-time bending angle data, and real-time temperature change data of the flexible device, and calculate cosine similarity with the full-operating efficiency monitoring database, select corresponding real-time data based on the relationship between a preset similarity threshold and the calculation result, and determine a quantitative value of parameter matching degree based on the autocorrelation coefficient of the real-time data; A comprehensive monitoring scoring matrix is ​​constructed according to the quantitative values ​​of the parameter matching degree, and the operation efficiency score of the comprehensive monitoring scoring matrix is ​​calculated by support vector regression of a Gaussian kernel function, so as to quantitatively calibrate the operation efficiency score through weighted combination; A Kalman filter is used to smooth the quantified and calibrated operating efficiency score, the operating efficiency score of the current working condition is updated through recursive calculation, and the operating efficiency score of the current working condition is divided according to a preset efficiency division interval, and the division result is obtained as the output of the operating efficiency monitoring result of the flexible equipment under the current working condition.

10. A system for monitoring the operating efficiency of flexible equipment, characterized in that: include: A data acquisition module is used to obtain historical monitoring data of flexible equipment under various working conditions; The historical monitoring data includes historical tensile stress data, historical bending angle data and historical temperature change data; A level classification module, used for classifying the historical tensile stress data according to a preset tensile stress threshold value, and obtaining tensile stress level data of the flexible device under corresponding working conditions; An interval division module, used for quantifying the angle change rate of the flexible device under the corresponding working condition based on the historical bending angle data to perform interval division, and obtaining the bending angle interval data of the flexible device under the corresponding working condition; A mode division module is used to extract key features from the historical temperature change data, and divide the historical temperature change data into modes according to a preset temperature change threshold and the key features, so as to obtain temperature change mode data of the flexible device under corresponding working conditions; A database construction module, used to construct and classify a multi-dimensional feature vector based on the tensile stress level data, the bending angle interval data and the temperature change mode data, to obtain a feature classification result of the flexible device, so as to construct a full-operating-condition efficiency monitoring database; The efficiency monitoring module is used to obtain the real-time tensile stress data, real-time bending angle data and real-time temperature change data of the flexible device to match them with the full-operating efficiency monitoring database, and optimize the operating efficiency monitoring process of the flexible device according to the comprehensive monitoring scoring matrix obtained by matching.

Images