Equipment fundamental frequency data generation system and method based on equipment fundamental frequency prediction

Abstract

The invention relates to an equipment fundamental frequency data generation system and method based on equipment fundamental frequency prediction. The system comprises a data acquisition module, an anomaly screening module, a weight configuration module and a fundamental frequency prediction module. Wherein the data acquisition module is used for acquiring vibration signals and fundamental frequency data under various operation conditions; the abnormity screening module is used for identifying and removing abnormal values in the fundamental frequency data; the weight configuration module is used for configuring a training weight according to the working condition characteristics and the abnormal data analysis result; the fundamental frequency prediction module is used for training a fundamental frequency predictor and generating fundamental frequency data of target equipment. The technical problem that in the prior art, abnormal data interference exists in the fundamental frequency data acquisition process of rotating equipment, and the fundamental frequency prediction precision is low due to the fact that the data reliability difference under different working conditions is large is solved, and abnormal data screening and differential weight configuration driven by working condition characteristics are achieved; and the fundamental frequency prediction precision and robustness of the equipment are improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular to a system and method for generating device fundamental frequency data based on device fundamental frequency prediction. Background Art

[0002] Rotating equipment is widely used in industrial production, and the monitoring of its operating status is of great significance to ensure the safe operation of equipment and production efficiency. The fundamental frequency of rotating equipment is an important parameter of the equipment's operating status, and its accurate acquisition is of great value for equipment status monitoring, fault diagnosis and predictive maintenance.

[0003] At present, the fundamental frequency data of rotating equipment is mainly acquired directly through sensors or analyzed and extracted through vibration signals. However, in actual industrial environments, equipment will produce different degrees of wear and noise interference under different operating conditions, resulting in large errors in the collected fundamental frequency data. In particular, when the equipment is running for a long time or working under high load conditions, the collected fundamental frequency data may contain a large number of outliers, which seriously affects the accuracy and reliability of the fundamental frequency data. Traditional fundamental frequency prediction methods usually adopt a unified data processing standard, ignoring the differences in data reliability under different working conditions, and cannot effectively screen out abnormal data. At the same time, the existing technology has also failed to perform differentiated processing according to the operating conditions and the degree of data abnormality, resulting in abnormal data in the model training process. The interference to the prediction results has a great influence, and ultimately affects the accuracy and robustness of the fundamental frequency prediction. Summary of the invention

[0004] The present invention aims to solve the technical problems in the prior art of low fundamental frequency prediction accuracy caused by abnormal data interference in the fundamental frequency data collection process of rotating equipment and large differences in data reliability under different working conditions, and provides a system and method for generating equipment fundamental frequency data based on equipment fundamental frequency prediction to solve the problems.

[0005] The technical solution of the present invention to solve the above technical problems is as follows: In a first aspect, the present invention provides a device fundamental frequency data generation system based on device fundamental frequency prediction, including: a data acquisition module, used to obtain multiple operating conditions, collect vibration signals of the device under the multiple operating conditions in historical time and perform fundamental frequency processing and marking, to obtain multiple sample operating condition vibration signal sets and multiple sample operating condition fundamental frequency data sets; an abnormal screening module, used to obtain operating condition characteristic information of multiple operating conditions, configure multiple abnormal fundamental frequency screening ratios, isolate abnormal fundamental frequencies for the multiple sample operating condition fundamental frequency data sets respectively, obtain multiple abnormal fundamental frequency data sets and delete them, obtain multiple normal operating condition vibration signal sets and multiple normal operating condition fundamental frequency data sets; a weight configuration module, used to calculate the abnormality of the multiple abnormal fundamental frequency data sets, and configure multiple initial training weights in combination with multiple operating condition characteristic information; a fundamental frequency prediction module, used to train a device fundamental frequency predictor according to the multiple initial training weights, using the multiple normal operating condition vibration signal sets and the multiple normal operating condition fundamental frequency data sets, collect vibration signals of the current target device, input the equipment fundamental frequency predictor, and predict and output to obtain the equipment fundamental frequency data.

[0006] Optionally, the data acquisition module includes: an operating condition acquisition unit, used to acquire multiple operating conditions of the target equipment; a vibration signal acquisition unit, used to collect vibration signals of similar equipment under the multiple operating conditions based on historical operation monitoring data of similar equipment, and obtain multiple sample operating condition vibration signal sets; a fundamental frequency processing and labeling unit, used to process each sample operating condition vibration signal using Fourier transform and harmonic product spectrum to obtain fundamental frequency data, and label it to obtain multiple sample operating condition fundamental frequency data sets.

[0007] Optionally, the abnormal screening module includes: an operating condition characteristic acquisition unit, used to obtain the operating condition operating time and average operating load of similar equipment under multiple operating conditions as multiple operating condition characteristic information; a coefficient calculation unit, used to calculate the ratio of each operating condition operating time to the average of multiple operating condition operating times within multiple operating condition characteristic information, to obtain multiple time coefficients, and calculate the ratio of each average operating load to the average of multiple average operating loads, to obtain multiple load coefficients; an adjustment coefficient calculation unit, used to calculate and obtain multiple screening ratio adjustment coefficients based on multiple time coefficients and multiple load coefficients; a preset ratio acquisition unit, used to obtain a preset abnormal base frequency screening ratio; an abnormal ratio configuration unit, used to respectively use the multiple screening ratio adjustment coefficients to configure and calculate the preset abnormal base frequency screening ratio to obtain multiple abnormal base frequency screening ratios.

[0008] Optionally, the abnormal screening module also includes: a fundamental frequency interval acquisition unit, which is used to acquire the fundamental frequency data interval of the equipment; an isolated data screening unit, which is used to randomly generate a first fundamental frequency data within the fundamental frequency data interval, perform binary classification on the multiple sample operating condition fundamental frequency data sets, and determine whether there is sample operating condition fundamental frequency data that is classified as isolated fundamental frequency data; an abnormal data classification unit, which is used to classify the isolated fundamental frequency data as abnormal fundamental frequency data if yes, and if not, continue to randomly generate the second fundamental frequency data to perform isolated abnormal fundamental frequency screening; a screening ratio control unit, which is used to stop the isolated abnormal fundamental frequency screening until the proportion of abnormal fundamental frequency data divided in each sample operating condition fundamental frequency data set reaches the multiple abnormal fundamental frequency screening ratios, and obtain multiple abnormal fundamental frequency data sets that have been screened; a normal data acquisition unit, which is used to obtain multiple abnormal operating condition vibration signal sets corresponding to the multiple abnormal fundamental frequency data sets, delete them, and obtain multiple normal operating condition vibration signal sets and multiple normal operating condition fundamental frequency data sets.

[0009] Optionally, the weight configuration module includes: a mean calculation unit, used to respectively calculate the means of the multiple normal operating fundamental frequency data sets to obtain multiple average normal fundamental frequency data; an abnormality degree calculation unit, used to respectively calculate the deviation amplitudes of the abnormal fundamental frequency data in the multiple abnormal fundamental frequency data sets and the multiple average normal fundamental frequency data to obtain multiple abnormal deviation amplitude sets, and calculate the mean to obtain multiple abnormality degrees.

[0010] Optionally, the weight configuration module also includes: a time weight calculation unit, which is used to calculate the ratio of each operating condition operating time to the sum of multiple operating condition operating times within the multiple operating condition feature information, as multiple operating time weights; a load weight calculation unit, which is used to calculate the ratio of each average operating load to the sum of multiple average operating loads, as multiple load weights; an abnormal weight allocation unit, which is used to allocate multiple abnormal training weights for multiple operating conditions according to the multiple abnormality degrees, wherein the size of the abnormality degree is negatively correlated with the size of the abnormal training weight; a weight fusion calculation unit, which is used to perform weight fusion calculation based on multiple operating time weights, multiple load weights and multiple abnormal training weights to obtain multiple initial training weights.

[0011] Optionally, the fundamental frequency prediction module includes: a training sample combination unit, which is used to combine the multiple normal operating vibration signal sets and the multiple normal operating fundamental frequency data sets to obtain multiple groups of normal fundamental frequency training sample sets and multiple normal sample data amounts; a sample weight allocation unit, which is used to respectively use the multiple initial training weights divided by the multiple normal sample data amounts to perform data-level weight allocation to obtain multiple sample data weight sets; a predictor construction unit, which is used to construct a device fundamental frequency predictor based on machine learning; a predictor training unit, which is used to use the multiple groups of normal fundamental frequency training sample sets, allocate training resources according to the multiple sample data weight sets, and supervise the training and testing of the device fundamental frequency predictor until convergence; a fundamental frequency data generation unit, which is used to collect the vibration signal of the current target device, input it into the device fundamental frequency predictor, and predict and output to obtain the device fundamental frequency data.

[0012] In a second aspect, the present invention provides a method for generating equipment fundamental frequency data based on equipment fundamental frequency prediction, including: obtaining multiple operating conditions, collecting vibration signals of the equipment under the multiple operating conditions in historical time and performing fundamental frequency processing and marking, obtaining multiple sample operating condition vibration signal sets and multiple sample operating condition fundamental frequency data sets; obtaining operating condition characteristic information of multiple operating conditions, configuring multiple abnormal fundamental frequency screening ratios, performing isolated abnormal fundamental frequency screening on the multiple sample operating condition fundamental frequency data sets respectively, obtaining multiple abnormal fundamental frequency data sets and deleting them, obtaining multiple normal operating condition vibration signal sets and multiple normal operating condition fundamental frequency data sets; calculating the abnormality of the multiple abnormal fundamental frequency data sets, configuring multiple initial training weights in combination with multiple operating condition characteristic information; according to the multiple initial training weights, using the multiple normal operating condition vibration signal sets and multiple normal operating condition fundamental frequency data sets, training an equipment fundamental frequency predictor, collecting vibration signals of the current target equipment, inputting the equipment fundamental frequency predictor, and predicting and outputting to obtain equipment fundamental frequency data.

[0013] The beneficial effects of the present invention are: Through the data acquisition module, various operating conditions are obtained, and the vibration signals of the equipment under various operating conditions in the historical time are collected and marked for fundamental frequency processing, so as to obtain multiple sample operating condition vibration signal sets and multiple sample operating condition fundamental frequency data sets. By collecting sample data under different operating conditions, basic data support is provided for subsequent fundamental frequency prediction to ensure the diversity and representativeness of data sources; through the abnormal screening module, the operating condition characteristic information of various operating conditions is obtained, and multiple abnormal fundamental frequency screening ratios are configured. The fundamental frequency data sets of multiple sample operating conditions are isolated and screened for abnormal fundamental frequencies respectively, and multiple abnormal fundamental frequency data sets are obtained and deleted, so as to obtain multiple normal operating condition vibration signal sets and multiple normal operating condition fundamental frequency data sets, so as to target the possible wear of the equipment under different operating conditions. The sample fundamental frequency data error problem caused by damage, noise, etc. is solved by isolating abnormal data and screening to improve the quality of sample data; the abnormality of multiple abnormal fundamental frequency data sets is calculated through the weight configuration module, and multiple initial training weights are configured in combination with multiple working condition feature information, so that the working condition with a larger abnormality has a smaller corresponding weight; this differentiated weight configuration ensures the reasonable influence of different working condition data in model training; the fundamental frequency prediction module is used according to multiple initial training weights, using multiple normal working condition vibration signal sets and multiple normal working condition fundamental frequency data sets to train the equipment fundamental frequency predictor, collect the vibration signal of the current target equipment, input the equipment fundamental frequency predictor, and predict the output to obtain the equipment fundamental frequency data, thereby accurately predicting the equipment fundamental frequency data.

[0014] Through the above technical solution, the present invention realizes abnormal data screening and differentiated weight configuration based on operating condition characteristics, effectively improving the accuracy and robustness of equipment fundamental frequency prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A schematic diagram of the structure of a device fundamental frequency data generation system based on device fundamental frequency prediction provided by the present invention; Figure 2 A schematic flow chart of a method for generating device fundamental frequency data based on device fundamental frequency prediction provided by the present invention.

[0016] In the accompanying drawings, the components represented by the reference numerals are as follows: Data collection module 11, abnormal screening module 12, weight configuration module 13, fundamental frequency prediction module 14. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0018] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0019] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or explanation". Any embodiment described as "for example" in the present invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present invention.

[0020] Embodiment 1:

[0021] like Figure 1 As shown, an embodiment of the present invention provides a device fundamental frequency data generation system based on device fundamental frequency prediction, including a data acquisition module 11, an abnormality screening module 12, a weight configuration module 13 and a fundamental frequency prediction module 14.

[0022] Among them, the data acquisition module 11 is used to obtain multiple operating conditions, collect vibration signals of the equipment under the multiple operating conditions in historical time and perform fundamental frequency processing and labeling to obtain multiple sample operating condition vibration signal sets and multiple sample operating condition fundamental frequency data sets.

[0023] The abnormal screening module 12 is used to obtain the operating condition characteristic information of various operating conditions, configure multiple abnormal fundamental frequency screening ratios, and perform isolated abnormal fundamental frequency screening on the multiple sample operating condition fundamental frequency data sets respectively, obtain multiple abnormal fundamental frequency data sets and delete them, and obtain multiple normal operating condition vibration signal sets and multiple normal operating condition fundamental frequency data sets.

[0024] The weight configuration module 13 is used to calculate the abnormality of the multiple abnormal fundamental frequency data sets, and configure multiple initial training weights in combination with multiple operating condition feature information.

[0025] The fundamental frequency prediction module 14 is used to train the equipment fundamental frequency predictor according to the multiple initial training weights, using the multiple normal operating vibration signal sets and the multiple normal operating fundamental frequency data sets, collect the vibration signal of the current target equipment, input it into the equipment fundamental frequency predictor, and predict and output to obtain the equipment fundamental frequency data.

[0026] Specifically, the data acquisition module 11 is responsible for acquiring the historical vibration data of the equipment under different working conditions and performing fundamental frequency extraction and labeling processing. Specifically, first, the data acquisition module 11 obtains various operating conditions of the target equipment during the historical operation through the equipment monitoring system. These operating conditions may include different load conditions, different operating durations, and different environmental parameters. For example, for a certain rotating equipment, there are many different operating states including full-load operating conditions, 75% load operating conditions, and 50% load operating conditions. Secondly, for each identified operating condition, the data acquisition module 11 collects the vibration signal generated by the equipment under the corresponding condition to form multiple sample condition vibration signal sets. For example, under full-load operating conditions, a vibration signal sample set within a period of time can be collected; under 75% load operating conditions, another set of vibration signal sample sets is collected, and so on. These vibration signals are usually obtained through vibration sensors on the equipment and stored after appropriate digital processing. Afterwards, the data acquisition module 11 performs fundamental frequency processing and labeling on the collected vibration signals. For example, the fundamental frequency information is extracted from the vibration signal using methods such as Fourier transform and harmonic product spectrum, and the extracted fundamental frequency data is associated with the corresponding vibration signal and labeled to form multiple sample operating condition fundamental frequency data sets. For example, after analyzing the vibration signal under full-load operating conditions, a data labeling result with a fundamental frequency of 49.8 Hz may be obtained. Multiple sample operating condition vibration signal sets and multiple sample operating condition fundamental frequency data sets are obtained through the data acquisition module 11, which provides the necessary original data basis for subsequent abnormal data screening and prediction model training, ensuring that the system can perform subsequent predictions based on real historical operating data.

[0027] The abnormal screening module 12 is responsible for screening and removing abnormal data in the sample fundamental frequency data set to ensure the accuracy and reliability of subsequent training data. Specifically, the abnormal screening module 12 first obtains the operating condition characteristic information of the equipment under various operating conditions, such as the operating time of the equipment, the average operating load and other key parameters. These characteristic information reflects the operating status and wear degree of the equipment under different operating conditions, and is an important basis for determining the abnormal screening ratio. For example, for a certain rotating equipment, its cumulative operating time, average temperature and other characteristic information under different loads are obtained. Secondly, the abnormal screening module 12 configures multiple abnormal fundamental frequency screening ratios according to the operating condition characteristic information. The abnormal data ratios under different operating conditions are often different. For example, the operating conditions with longer operating time may accumulate more wear and noise interference, resulting in a higher abnormal data ratio; the equipment is subjected to greater force under high-load operating conditions, and more abnormal vibration signal data may also be generated. Therefore, according to different operating condition characteristics, the system will configure corresponding abnormal fundamental frequency screening ratios. Thirdly, the abnormal screening module 12 performs isolated abnormal fundamental frequency screening on each sample operating condition fundamental frequency data set to identify the abnormal fundamental frequency data set. Subsequently, the abnormal screening module 12 deletes the identified abnormal fundamental frequency data set and its corresponding vibration signal from the original data set, thereby obtaining multiple normal operating condition vibration signal sets and multiple normal operating condition fundamental frequency data sets. The screened data set can more accurately reflect the normal fundamental frequency characteristics of the equipment under various operating conditions, providing a more reliable data basis for subsequent weight configuration and prediction model training. Through the processing of the abnormal screening module 12, the system effectively removes abnormal data caused by factors such as equipment wear and environmental noise, significantly improving the accuracy and stability of fundamental frequency prediction.

[0028] The weight configuration module 13 is responsible for analyzing the characteristics of abnormal data and intelligently configuring the training weights in combination with the working condition information. Specifically, the weight configuration module 13 first calculates the abnormality of multiple abnormal fundamental frequency data sets screened by the abnormal screening module. The abnormality is a quantitative indicator to measure the degree of deviation of abnormal data from normal data, and is obtained by calculating the deviation amplitude between the abnormal fundamental frequency data and the mean of normal fundamental frequency data. For example, for an abnormal fundamental frequency data set under a certain working condition, if the normal fundamental frequency mean is 50Hz and the abnormal data mean is 45Hz, the abnormality under this working condition is large; if the abnormal data mean is 49Hz under another working condition, its abnormality is relatively small. Secondly, the weight configuration module 13 combines a variety of operating condition characteristic information obtained from the data collection link, such as operating time, average load, etc., to perform weight configuration calculation. The operating condition characteristic information reflects the operating status and data reliability of the equipment under different working conditions, and is an important reference for weight configuration. For example, data collected under working conditions with longer operating time and more stable load may be more representative and should be given a higher weight. Afterwards, the weight configuration module 13 configures a plurality of initial training weights by comprehensively considering the abnormality and the operating condition characteristic information. Among them, the operating condition with a large abnormality usually indicates that the data quality under the operating condition is relatively low, and the corresponding training weight will be appropriately reduced; on the contrary, the data quality is high under the operating condition with a small abnormality, and the corresponding training weight will be increased accordingly. For example, a training weight of 0.8 is configured for the operating condition with an abnormality of 10%, while a training weight of only 0.4 is configured for the operating condition with an abnormality of 30%. Through the weight configuration module 13, the system can reasonably allocate training resources according to data quality and operating condition characteristics, focusing on high-quality and high-reliability data samples, and effectively improving the training efficiency and prediction accuracy of the subsequent fundamental frequency prediction model.

[0029] The fundamental frequency prediction module 14 is responsible for constructing and training the equipment fundamental frequency predictor based on the data processed by the above modules and the configured weights, so as to achieve accurate prediction of the fundamental frequency of the target equipment. Specifically, the fundamental frequency prediction module 14 first receives multiple initial training weights from the weight configuration module 13, as well as multiple normal working condition vibration signal sets and multiple normal working condition fundamental frequency data sets screened by the abnormal screening module 12, as input data for training the equipment fundamental frequency predictor. For example, for a certain rotating equipment, it includes multiple sets of training data such as a normal vibration signal set under full load conditions and its corresponding fundamental frequency data set, a normal vibration signal set under 75% load conditions and its corresponding fundamental frequency data set. Secondly, the module weights the training data according to the initial training weights, and refines the training weights to the data sample level, so that the training process pays more attention to high-quality and high-reliability data samples. For example, more training resources may be allocated to working condition data with higher weights, such as more iterations or higher learning rates, while the training resource allocation is reduced accordingly for working condition data with lower weights. Again, the fundamental frequency prediction module 14 constructs a device fundamental frequency predictor based on a machine learning algorithm, and the predictor can establish a prediction model from a vibration signal to fundamental frequency data by learning the mapping relationship between the vibration signal and the fundamental frequency data. Among them, the construction of the predictor can adopt algorithms such as deep neural networks, support vector machines or random forests, and supervised training is performed according to weight distribution until the prediction performance reaches the preset convergence standard. After the predictor training is completed, the fundamental frequency prediction module 14 collects the real-time vibration signal of the current target device, inputs it into the trained device fundamental frequency predictor, and quickly obtains the corresponding device fundamental frequency data output. For example, for a certain rotating device in operation, its vibration sensor data is collected, and after preprocessing, it is input into the predictor, and a prediction result such as "the current device fundamental frequency is 49.8Hz" can be directly output. Through the fundamental frequency prediction module 14, the system realizes the whole process from historical data training to real-time fundamental frequency prediction, providing accurate and reliable fundamental frequency data support for equipment status monitoring, fault warning and maintenance decision-making, and significantly improving the intelligence level and prediction accuracy of equipment monitoring.

[0030] Furthermore, the data acquisition module 11 includes a working condition acquisition unit, a vibration signal acquisition unit and a fundamental frequency processing and labeling unit. The working condition acquisition unit is used to acquire multiple operating conditions of the target equipment; the vibration signal acquisition unit is used to collect vibration signals of similar equipment under the multiple operating conditions based on the historical operation monitoring data of similar equipment, and obtain multiple sample working condition vibration signal sets; the fundamental frequency processing and labeling unit is used to process each sample working condition vibration signal using Fourier transform and harmonic product spectrum to obtain fundamental frequency data, and label to obtain multiple sample working condition fundamental frequency data sets.

[0031] In an optional implementation, the data acquisition module 11 includes three functional units: a working condition acquisition unit, a vibration signal acquisition unit, and a fundamental frequency processing and labeling unit. The units work together to form a complete data acquisition process.

[0032] The working condition acquisition unit is responsible for identifying and acquiring various operating condition information of the target equipment. The unit obtains the working condition parameters of the equipment under different operating conditions by exchanging data with the equipment control system, including but not limited to key information such as load rate, speed range, ambient temperature, and operating duration. For example, for the rotating equipment of a power plant, the unit identifies a variety of typical operating conditions such as full load (100%), high load (80%-90%), medium load (50%-70%), and low load (30%-40%). Subsequently, the vibration signal acquisition unit, based on the acquired working condition information, specifically collects vibration signals of similar equipment under the above-mentioned various operating conditions. The vibration signal acquisition unit accesses the historical operation monitoring database to extract vibration signal records of similar equipment under corresponding working conditions, and performs preliminary denoising and standardization processing on these signals. For example, for full load conditions, the vibration signal acquisition unit collects 200 groups of vibration signal samples; for medium load conditions, 200 groups of vibration signal samples may be collected, thereby forming multiple sample vibration signal sets corresponding to the working conditions. Afterwards, the fundamental frequency processing and labeling unit receives multiple sample working condition vibration signal sets output by the vibration signal acquisition unit, and performs in-depth processing and analysis on each sample vibration signal. Specifically, the fundamental frequency processing and labeling unit first applies Fourier transform to the vibration signal, converts the time domain signal into a frequency domain signal, and displays the various frequency components contained in the signal; then, through the harmonic product spectrum analysis technology, the fundamental frequency components in the signal are identified and extracted, and interference and noise effects are filtered out; then, the extracted fundamental frequency data is associated and labeled with the corresponding vibration signal and working condition information to form a structured sample working condition fundamental frequency data set. For example, after processing the vibration signal under a full-load working condition, it is labeled as "full load working condition, vibration signal ID_123, fundamental frequency 49.8Hz".

[0033] Through the collaborative work of the working condition acquisition unit, the vibration signal acquisition unit and the fundamental frequency processing and labeling unit, the data acquisition module 11 can efficiently acquire vibration signals and fundamental frequency data under various working conditions, and process and label them, laying a data foundation for subsequent abnormal screening and prediction model training.

[0034] Furthermore, the abnormal screening module 12 includes a working condition characteristic acquisition unit, a coefficient calculation unit, an adjustment coefficient calculation unit, a preset ratio acquisition unit and an abnormal ratio configuration unit. Among them, the working condition characteristic acquisition unit is used to obtain the working condition operation time and average operating load of the same type of equipment under multiple operating conditions as multiple working condition characteristic information; the coefficient calculation unit is used to calculate the ratio of each working condition operation time to the average of multiple working condition operation times in multiple working condition characteristic information to obtain multiple time coefficients, and calculate the ratio of each average operating load to the average of multiple average operating loads to obtain multiple load coefficients; the adjustment coefficient calculation unit is used to calculate and obtain multiple screening ratio adjustment coefficients based on multiple time coefficients and multiple load coefficients; the preset ratio acquisition unit is used to obtain a preset abnormal base frequency screening ratio; the abnormal ratio configuration unit is used to respectively use the multiple screening ratio adjustment coefficients to configure and calculate the preset abnormal base frequency screening ratio to obtain multiple abnormal base frequency screening ratios.

[0035] In a preferred embodiment, the abnormal screening module 12 includes an operating condition feature acquisition unit, a coefficient calculation unit, an adjustment coefficient calculation unit, a preset ratio acquisition unit and an abnormal ratio configuration unit to achieve adaptive abnormal data screening based on operating condition features.

[0036] Specifically, the working condition characteristic acquisition unit is responsible for obtaining the key characteristic information of the same type of equipment under various operating conditions, mainly including the working condition operating time and the average operating load. For example, for a certain rotating equipment, the working condition characteristic acquisition unit obtains the working condition characteristic information such as the cumulative operating time under full load conditions is 5000 hours and the average operating load is 95%; the cumulative operating time under medium load conditions is 8000 hours and the average operating load is 65%. These characteristic information directly reflects the use intensity and wear degree of the equipment under various working conditions, and is an important basis for judging the proportion of abnormal data. After receiving the working condition characteristic information obtained by the working condition characteristic acquisition unit, the coefficient calculation unit performs standardization processing on the working condition characteristics. The coefficient calculation unit calculates the ratio of each working condition operating time to the average of multiple working condition operating times to obtain multiple time coefficients; at the same time, it calculates the ratio of each working condition average operating load to the average of multiple working condition average operating loads to obtain multiple load coefficients. For example, if the operating time of the three working conditions is 5000 hours, 8000 hours and 2000 hours respectively, the average is 5000 hours, and the corresponding time coefficients are 1.0, 1.6 and 0.4 respectively; if the average operating load of the three working conditions is 95%, 65% and 40% respectively, the average is 66.7%, and the corresponding load factors are 1.42, 0.97 and 0.60 respectively.

[0037] Subsequently, the adjustment coefficient calculation unit calculates the final screening ratio adjustment coefficient according to the time coefficient and load coefficient output by the coefficient calculation unit. This calculation adopts the method of multiplying the time coefficient and the load coefficient, which reflects the compound influence of the operating time and the load on the abnormal data. For example, for the above three working conditions, the screening ratio adjustment coefficients are 1.0×1.42=1.42, 1.6×0.97=1.55 and 0.4×0.60=0.24 respectively. This means that the first and second working conditions may require a higher abnormal data screening ratio, while the abnormal data ratio of the third working condition may be lower. At the same time, the preset ratio acquisition unit obtains the abnormal base frequency screening ratio reference value preset by the system. This ratio is determined based on historical statistical data and represents the average proportion of abnormal base frequency data of the equipment under standard working conditions. For example, according to the historical monitoring records of the equipment, the preset abnormal base frequency screening ratio is set to 5%, indicating that under standard working conditions, about 5% of the base frequency data are abnormal due to various interference factors. Afterwards, the abnormal proportion configuration unit applies the screening proportion adjustment coefficient output by the adjustment coefficient calculation unit to the preset abnormal base frequency screening proportion, and obtains the actual abnormal base frequency screening proportion under each working condition through multiplication. For example, if the preset abnormal base frequency screening proportion is 5%, the abnormal base frequency screening proportions of the above three working conditions are 5%×1.42=7.1%, 5%×1.55=7.75% and 5%×0.24=1.2% respectively. This means that when screening abnormal base frequency data, 7.1% of the data will be screened out as abnormal data for the first working condition, 7.75% of the data will be screened out as abnormal data for the second working condition, and only 1.2% of the data will be screened out as abnormal data for the third working condition.

[0038] Through the coordinated work of the above five functional units, the abnormal screening module 12 can adaptively configure the abnormal base frequency screening ratio under each working condition based on the working condition characteristic information, realize the precision and personalization of abnormal data screening, and provide a proportional basis for subsequent abnormal base frequency screening operations.

[0039] Furthermore, the abnormal screening module 12 also includes a fundamental frequency interval acquisition unit, an isolated data screening unit, an abnormal data classification unit, a screening ratio control unit and a normal data acquisition unit. Among them, the fundamental frequency interval acquisition unit is used to obtain the fundamental frequency data interval of the equipment; the isolated data screening unit is used to randomly generate the first fundamental frequency data within the fundamental frequency data interval, and perform binary classification on the multiple sample working condition fundamental frequency data sets to determine whether there is sample working condition fundamental frequency data that is classified as isolated fundamental frequency data; the abnormal data classification unit is used to classify the isolated fundamental frequency data as abnormal fundamental frequency data if yes, and if not, continue to randomly generate the second fundamental frequency data to perform isolated abnormal fundamental frequency screening; the screening ratio control unit is used to stop the isolated abnormal fundamental frequency screening until the proportion of abnormal fundamental frequency data divided in each sample working condition fundamental frequency data set reaches the multiple abnormal fundamental frequency screening ratios, and obtain multiple abnormal fundamental frequency data sets that have been screened; the normal data acquisition unit is used to obtain multiple abnormal working condition vibration signal sets corresponding to the multiple abnormal fundamental frequency data sets, delete them, and obtain multiple normal working condition vibration signal sets and multiple normal working condition fundamental frequency data sets.

[0040] Specifically, the abnormal screening module 12 also includes a base frequency interval acquisition unit, an isolated data screening unit, an abnormal data classification unit, a screening ratio control unit and a normal data acquisition unit, which together constitute a complete abnormal base frequency screening execution system to achieve accurate abnormal data screening function.

[0041] The fundamental frequency interval acquisition unit is responsible for determining the effective interval range of the fundamental frequency data of the equipment, and obtaining the theoretical range and actual distribution range of the fundamental frequency of the equipment by analyzing the equipment specification parameters and historical operation data. For example, for a certain model of rotating equipment, the theoretical value of the fundamental frequency may be 50Hz. Considering the fluctuations in actual operation, the fundamental frequency interval may be determined to be 45Hz-55Hz. The determination of this interval provides the data space range for the subsequent abnormal data screening. After that, the isolated data screening unit uses a combination of random sampling and binary classification to identify isolated abnormal points in the fundamental frequency data set. The isolated data screening unit first randomly generates a fundamental frequency data point (called the first fundamental frequency data) within the determined fundamental frequency data interval, and then uses this point as the boundary to perform binary classification on the fundamental frequency data sets of multiple sample working conditions, and determines whether there is isolated fundamental frequency data formed by this division. For example, if the randomly generated fundamental frequency data is 47Hz, and most of the fundamental frequency data under a certain working condition are distributed in the 49Hz-51Hz interval, and only a small amount of data are distributed in the 46Hz-48Hz interval, then these small amounts of data may be classified as isolated data. Afterwards, the abnormal data classification unit processes the judgment result of the isolated data screening unit. When the judgment result is "yes", that is, there are fundamental frequency data that are classified as isolated, and the abnormal data classification unit classifies these isolated fundamental frequency data as abnormal fundamental frequency data; when the judgment result is "no", that is, no obvious isolated data is found, the unit will instruct the isolated data screening unit to continue to randomly generate new fundamental frequency data points (called second fundamental frequency data) and re-execute the isolated abnormal fundamental frequency screening process. For example, if the first randomly generated 47Hz fails to effectively divide the isolated data, the system will generate a second random point 52Hz and re-perform the division judgment.

[0042] The screening ratio control unit is responsible for monitoring the progress and ratio of abnormal data screening, calculating the ratio of abnormal fundamental frequency data divided in each sample working condition fundamental frequency data set in real time, and comparing it with the previously configured abnormal fundamental frequency screening ratio. When the ratio of abnormal data under a certain working condition reaches its corresponding screening ratio, the isolated abnormal fundamental frequency screening process of the working condition stops; when the screening of abnormal data of all working conditions reaches the predetermined ratio, the entire screening process ends, and the system obtains multiple abnormal fundamental frequency data sets that have been screened. For example, if the abnormal fundamental frequency screening ratio of a certain working condition is 7.1%, then when the fundamental frequency data marked as abnormal under the working condition reaches 7.1% of the total data volume, the screening process of the working condition stops. After completing the abnormal data screening, the normal data acquisition unit performs a data separation operation. The normal data acquisition unit first obtains the vibration signals corresponding to the multiple abnormal fundamental frequency data sets to form multiple abnormal working condition vibration signal sets; then deletes these abnormal data from the original sample set, and finally obtains multiple normal working condition vibration signal sets and multiple normal working condition fundamental frequency data sets without abnormal data. For example, if there are 100 sets of fundamental frequency data under a certain working condition, of which 7 sets are marked as abnormal, after deleting these 7 sets of data and their corresponding vibration signals, 93 sets of normal data will be retained under this working condition for subsequent processing.

[0043] Through the above processing, the abnormal screening module 12 can screen out abnormal fundamental frequency data under various working conditions according to a predetermined ratio, and remove it from the training data set, effectively improving the data quality and prediction accuracy of subsequent model training.

[0044] Furthermore, the weight configuration module 13 includes a mean value calculation unit and an abnormality calculation unit. The mean value calculation unit is used to respectively calculate the mean values ​​of the multiple normal operating condition fundamental frequency data sets to obtain multiple average normal fundamental frequency data; the abnormality calculation unit is used to respectively calculate the deviation amplitudes of the abnormal fundamental frequency data in the multiple abnormal fundamental frequency data sets and the multiple average normal fundamental frequency data to obtain multiple abnormal deviation amplitude sets, and calculate the mean value to obtain multiple abnormality degrees.

[0045] In a preferred embodiment, the weight configuration module 13 includes a mean value calculation unit and an abnormality calculation unit to achieve accurate analysis of the characteristics of abnormal fundamental frequency data to quantify the degree of deviation of abnormal data under various working conditions.

[0046] The mean calculation unit is responsible for statistical analysis of the normal fundamental frequency data processed by the abnormal screening module. The mean calculation unit receives multiple normal working condition fundamental frequency data sets as input, and calculates the arithmetic mean for the normal fundamental frequency data sets under each working condition to obtain the average normal fundamental frequency data under each working condition. For example, for the normal fundamental frequency data set under full load conditions, if it contains 93 sets of data, whose values ​​are 49.8Hz, 50.1Hz, 49.9Hz, etc., the average normal fundamental frequency data calculated is 50.0Hz; for medium load conditions, the average value is 49.5Hz, and so on. These average values ​​represent the normal fundamental frequency level of the equipment under each working condition and are the reference values ​​for subsequent abnormality calculations. Afterwards, the abnormality calculation unit quantitatively evaluates the degree of deviation of the abnormal fundamental frequency data based on the output result of the mean calculation unit. The abnormality calculation unit first calculates the deviation amplitude between each data point in the abnormal fundamental frequency data set under each working condition and the average normal fundamental frequency data of the corresponding working condition (absolute difference or relative difference can be used) to form multiple abnormal deviation amplitude sets. For example, if the abnormal base frequency data under full load condition is 45.0Hz, and the average normal base frequency data of this condition is 50.0Hz, then the deviation amplitude of this abnormal data is 5.0Hz or 10% (depending on whether the absolute difference or relative difference calculation is adopted). Subsequently, the abnormality calculation unit calculates the mean of each abnormal deviation amplitude set to obtain the comprehensive abnormality index under each condition. For example, if the deviation amplitude of 7 groups of abnormal data under full load condition is 4.5Hz, then the abnormality of this condition is 4.5Hz; if the abnormality under medium load condition is 2.8Hz, it indicates that the abnormal data under full load condition deviates more.

[0047] Through the mean calculation unit and the abnormality calculation unit, the weight configuration module 13 can quantify the severity of abnormal data under various working conditions, providing a scientific basis for subsequent weight configuration. The greater the degree of abnormality, the lower the data quality may be, and a lower weight should be given in training; the smaller the degree of abnormality, the higher the data quality, and a higher training weight should be given. This adaptive weight configuration mechanism based on data quality can improve the pertinence and effectiveness of model training.

[0048] Furthermore, the weight configuration module 13 also includes a time weight calculation unit, a time weight calculation unit, an abnormal weight allocation unit and a weight fusion calculation unit. Among them, the time weight calculation unit is used to calculate the ratio of each operating condition operation time to the sum of multiple operating condition operation times within the multiple operating condition feature information as multiple operating time weights; the load weight calculation unit is used to calculate the ratio of each average operating load to the sum of multiple average operating loads as multiple load weights; the abnormal weight allocation unit is used to allocate multiple abnormal training weights for multiple operating conditions according to the multiple abnormal degrees, wherein the magnitude of the abnormal degree is negatively correlated with the magnitude of the abnormal training weight; the weight fusion calculation unit is used to perform weight fusion calculation according to multiple operating time weights, multiple load weights and multiple abnormal training weights to obtain multiple initial training weights.

[0049] In a feasible implementation manner, the weight configuration module 13 further includes a time weight calculation unit, a load weight calculation unit, an abnormal weight allocation unit and a weight fusion calculation unit to implement weight configuration.

[0050] The time weight calculation unit is responsible for assigning corresponding weight coefficients based on the equipment operation time. The unit extracts the operation time data of each working condition from the working condition characteristic information, calculates the ratio of the operation time of each working condition to the sum of the operation time of all working conditions, and uses the ratio as the operation time weight of the corresponding working condition. For example, if the operation time of the three working conditions is 5000 hours, 8000 hours and 2000 hours respectively, the total operation time is 15000 hours, and the corresponding operation time weights are 5000 / 15000=0.33, 8000 / 15000=0.53 and 2000 / 15000=0.13 respectively. This weight allocation method reflects that the working conditions with longer operation time may contain more valuable operation information and should obtain relatively higher weights in training. The load weight calculation unit adopts a similar proportional allocation method and configures the weight coefficient based on the average operation load. The unit calculates the ratio of the average operation load of each working condition to the sum of the average operation loads of all working conditions as the load weight of the corresponding working condition. For example, if the average operating loads of the three working conditions are 95%, 65% and 40% respectively, the total load is 200%, and the corresponding load weights are 95 / 200=0.475, 65 / 200=0.325 and 40 / 200=0.2 respectively. This configuration method reflects that the data under higher load conditions more fully characterizes the operating characteristics of the equipment and should obtain a relatively higher training weight.

[0051] The abnormal weight allocation unit allocates abnormal training weights by reverse mapping based on the abnormality index calculated previously. The unit allocates weights according to the abnormality of each working condition and the principle of negative correlation between abnormality and weight, that is, the larger the abnormality of the working condition, the smaller the abnormality of the working condition, and the smaller the abnormality of the working condition, the larger the abnormality of the working condition. For example, if the abnormality of the three working conditions is 4.5Hz, 2.8Hz and 1.5Hz, they can be mapped to abnormal training weights of 0.15, 0.25 and 0.45 by a certain inverse function relationship (the specific mapping function can be designed according to actual needs). This configuration ensures that the working conditions with less abnormal data and higher data quality have greater influence in training. Afterwards, the weight fusion calculation unit is responsible for integrating the weights of the above three different dimensions to form the final initial training weight. The unit first calculates the sum of the running time weight, load weight and abnormal training weight of each working condition, and then calculates the ratio of the sum to the sum of the three weights of all working conditions as the initial training weight of the working condition. For example, if the sum of the three weights of the first working condition is 0.33+0.475+0.15=0.955, the second working condition is 0.53+0.325+0.25=1.105, and the third working condition is 0.13+0.2+0.45=0.78, then the sum of the weights of the three working conditions is 2.84, and the corresponding initial training weights are 0.955 / 2.84=0.336, 1.105 / 2.84=0.389, and 0.78 / 2.84=0.275. This fusion method takes into account the three key factors of the working condition's operating time, load level, and data quality to form a weight distribution scheme.

[0052] Through the weight calculation unit, load weight calculation unit, abnormal weight allocation unit and weight fusion calculation unit, the weight configuration module 13 can comprehensively evaluate the importance and reliability of each operating condition data from multiple dimensions, provide accurate weight configuration for subsequent fundamental frequency prediction model training, and effectively improve training efficiency and prediction accuracy.

[0053] Furthermore, the fundamental frequency prediction module 14 includes a training sample combination unit, a sample weight allocation unit, a predictor construction unit, a predictor training unit and a fundamental frequency data generation unit. Among them, the training sample combination unit is used to combine the multiple normal working condition vibration signal sets and the multiple normal working condition fundamental frequency data sets to obtain multiple groups of normal fundamental frequency training sample sets and multiple normal sample data volumes; the sample weight allocation unit is used to respectively use the multiple initial training weights divided by the multiple normal sample data volumes to perform data-level weight allocation to obtain multiple sample data weight sets; the predictor construction unit is used to construct a device fundamental frequency predictor based on machine learning; the predictor training unit is used to use the multiple groups of normal fundamental frequency training sample sets, allocate training resources according to the multiple sample data weight sets, and supervise the training and testing of the device fundamental frequency predictor until convergence; the fundamental frequency data generation unit is used to collect the vibration signal of the current target device, input it into the device fundamental frequency predictor, and predict and output to obtain the device fundamental frequency data.

[0054] In a preferred embodiment, the fundamental frequency prediction module 14 includes five functional units, namely a training sample combination unit, a sample weight allocation unit, a predictor construction unit, a predictor training unit and a fundamental frequency data generation unit, forming a complete predictor training and application process to achieve accurate and efficient device fundamental frequency prediction function.

[0055] The training sample combination unit is responsible for integrating the normal data processed by the previous module and constructing a standardized training sample set. The training sample combination unit receives multiple normal working condition vibration signal sets and multiple normal working condition fundamental frequency data sets as input, and pairs and combines the vibration signals under each working condition with the corresponding fundamental frequency data to form multiple groups of normal fundamental frequency training sample sets with clear "input-output" correspondence. At the same time, the training sample combination unit also calculates the number of normal samples under each working condition to obtain multiple normal sample data volumes. For example, the full load working condition may contain 93 groups of normal sample data, the medium load working condition may contain 88 groups of normal sample data, and the low load working condition may contain 95 groups of normal sample data. These data will be organized into a structured training sample set to provide basic data support for subsequent model training. The sample weight allocation unit refines the working condition-level initial training weights output by the weight configuration module to the data sample level. The sample weight allocation unit uses multiple initial training weights divided by the corresponding normal sample data volume to achieve uniform dispersion of weights and obtain multiple sample data weight sets. For example, if the initial training weight of the full-load condition is 0.336 and the normal sample data volume is 93, then the weight of each sample under this condition is 0.336 / 93=0.0036; if the initial training weight of the medium-load condition is 0.389 and the normal sample data volume is 88, then the weight of each sample under this condition is 0.389 / 88=0.0044. This refined weight distribution ensures that the influence of each condition in the training process is consistent with its preset weight, regardless of the number of samples.

[0056] The predictor construction unit is responsible for building the basic framework of the device fundamental frequency predictor based on machine learning technology. The predictor construction unit can select appropriate machine learning algorithms, such as support vector regression, random forest, deep neural network, etc., according to the characteristics of the prediction task and the complexity of the data, to build a prediction model structure that can map from vibration signal features to fundamental frequency data. For example, for complex nonlinear mapping relationships, a deep neural network with multiple hidden layers is constructed, the input layer receives the vibration signal features, and the output layer predicts the fundamental frequency data; for simpler mapping relationships, a support vector regression model is used to achieve nonlinear mapping through kernel functions. The predictor training unit performs the core process of model training. The predictor training unit first divides multiple groups of normal fundamental frequency training sample sets into training sets and test sets, and then allocates training resources (such as the number of iterations, learning rate, etc.) according to the sample data weight set, and performs supervised gradient descent training on the predictor. For example, a sample with a weight of 0.0044 may obtain more iterations of training or a higher learning rate, while a sample with a weight of 0.0036 will have a corresponding reduction in training resource allocation. During the training process, the predictor training unit continuously monitors the performance of the model on the test set, evaluates the prediction error and convergence status, until the model performance reaches the preset convergence standard, and completes the parameter optimization of the predictor. The fundamental frequency data generation unit is responsible for applying the trained predictor to the actual prediction scenario. The fundamental frequency data generation unit collects the vibration signal of the current target device in real time through the vibration sensor, preprocesses the signal in accordance with the training data, and then inputs the processed vibration signal characteristics into the trained device fundamental frequency predictor to quickly obtain the corresponding device fundamental frequency data output. For example, for a certain rotating device in operation, after collecting its vibration signal and inputting it into the predictor, the prediction result of "the current device fundamental frequency is 49.8Hz" can be directly obtained, providing real-time data support for equipment status monitoring and fault diagnosis.

[0057] Through the collaborative work of the above-mentioned functional units, the fundamental frequency prediction module 14 realizes the complete process from training sample organization to model training to actual prediction, enabling the system to intelligently learn the intrinsic mapping relationship between vibration signals and fundamental frequency data based on historical data, and apply this relationship to the fundamental frequency prediction of new equipment, thereby improving the accuracy and robustness of equipment fundamental frequency prediction.

[0058] Embodiment 2:

[0059] like Figure 2 As shown, based on the same inventive concept as the device fundamental frequency data generation system based on device fundamental frequency prediction provided in Embodiment 1, the embodiment of the present invention further provides a device fundamental frequency data generation method based on device fundamental frequency prediction, including: Acquire multiple operating conditions, collect vibration signals of the equipment under the multiple operating conditions in historical time and perform fundamental frequency processing and marking, and obtain multiple sample operating condition vibration signal sets and multiple sample operating condition fundamental frequency data sets; Acquire the working condition characteristic information of multiple operating conditions, configure multiple abnormal fundamental frequency screening ratios, perform isolated abnormal fundamental frequency screening on the multiple sample working condition fundamental frequency data sets respectively, obtain multiple abnormal fundamental frequency data sets and delete them, and obtain multiple normal working condition vibration signal sets and multiple normal working condition fundamental frequency data sets; Calculating the abnormality of the multiple abnormal fundamental frequency data sets, combining multiple operating condition feature information, and configuring multiple initial training weights; According to the multiple initial training weights, the multiple normal operating vibration signal sets and the multiple normal operating fundamental frequency data sets are used to train the equipment fundamental frequency predictor, collect the vibration signal of the current target equipment, input the equipment fundamental frequency predictor, and predict and output to obtain the equipment fundamental frequency data.

[0060] Furthermore, a plurality of operating conditions are obtained, vibration signals of the equipment under the plurality of operating conditions in historical time are collected and fundamental frequency processing and marking are performed, and a plurality of sample operating condition vibration signal sets and a plurality of sample operating condition fundamental frequency data sets are obtained, including: Obtain various operating conditions of the target equipment; According to the historical operation monitoring data of similar equipment, vibration signals of similar equipment under the multiple operating conditions are collected to obtain multiple sample condition vibration signal sets; The vibration signal of each sample working condition is processed by Fourier transform and harmonic product spectrum to obtain the fundamental frequency data, which are then labeled to obtain multiple sample working condition fundamental frequency data sets.

[0061] Furthermore, the operating characteristic information of various operating conditions is obtained, and multiple abnormal base frequency screening ratios are configured, including: Obtain the operating time and average operating load of similar equipment under various operating conditions as multiple operating condition feature information; In the plurality of working condition characteristic information, a ratio of each working condition operation time to an average of the plurality of working condition operation times is calculated to obtain a plurality of time coefficients, and a ratio of each average operation load to an average of the plurality of average operation loads is calculated to obtain a plurality of load coefficients; According to multiple time coefficients and multiple load coefficients, multiple screening ratio adjustment coefficients are calculated; Obtain the preset abnormal fundamental frequency screening ratio; The plurality of screening ratio adjustment coefficients are respectively used to calculate the preset abnormal base frequency screening ratio configuration to obtain a plurality of abnormal base frequency screening ratios.

[0062] Further, the multiple sample working condition fundamental frequency data sets are respectively screened for isolated abnormal fundamental frequencies, multiple abnormal fundamental frequency data sets are obtained and deleted, and multiple normal working condition vibration signal sets and multiple normal working condition fundamental frequency data sets are obtained, including: Get the baseband data interval of the device; Randomly generate first fundamental frequency data in the fundamental frequency data interval, perform binary classification on the plurality of sample operating condition fundamental frequency data sets, and determine whether there is any sample operating condition fundamental frequency data that is classified as isolated fundamental frequency data; If yes, the isolated fundamental frequency data is classified as abnormal fundamental frequency data; if no, the second fundamental frequency data is randomly generated to perform isolated abnormal fundamental frequency screening; Until the proportion of abnormal fundamental frequency data divided in each sample operating condition fundamental frequency data set reaches the multiple abnormal fundamental frequency screening ratios, the isolated abnormal fundamental frequency screening is stopped to obtain multiple abnormal fundamental frequency data sets after screening; A plurality of abnormal operating condition vibration signal sets corresponding to the plurality of abnormal fundamental frequency data sets are obtained, and the sets are deleted to obtain a plurality of normal operating condition vibration signal sets and a plurality of normal operating condition fundamental frequency data sets.

[0063] Furthermore, it is characterized in that calculating the abnormality of the plurality of abnormal fundamental frequency data sets comprises: Calculating the average values ​​of the multiple normal operating condition fundamental frequency data sets respectively to obtain multiple average normal fundamental frequency data; The deviation amplitudes of the abnormal fundamental frequency data in the multiple abnormal fundamental frequency data sets and the multiple average normal fundamental frequency data are calculated respectively to obtain multiple abnormal deviation amplitude sets, and the average is calculated to obtain multiple abnormality degrees.

[0064] Furthermore, it is characterized in that, in combination with multiple operating condition characteristic information, multiple initial training weights are configured, including: In the plurality of operating condition characteristic information, a ratio of each operating condition operation time to the sum of the operating times of the plurality of operating conditions is calculated as a plurality of operating time weights; Calculate the ratio of each average operating load to the sum of multiple average operating loads as multiple load weights; According to the multiple abnormality degrees, multiple abnormality training weights for multiple operating conditions are allocated, wherein the magnitude of the abnormality degree is negatively correlated with the magnitude of the abnormality training weight; According to multiple running time weights, multiple load weights and multiple abnormal training weights, weight fusion calculation is performed to obtain multiple initial training weights.

[0065] Further, it is characterized in that, according to the multiple initial training weights, the multiple normal working condition vibration signal sets and the multiple normal working condition fundamental frequency data sets are used to train the equipment fundamental frequency predictor, collect the vibration signal of the current target equipment, input the equipment fundamental frequency predictor, and predict and output to obtain the equipment fundamental frequency data, including: Combining the plurality of normal operating condition vibration signal sets and the plurality of normal operating condition fundamental frequency data sets to obtain a plurality of normal fundamental frequency training sample sets and a plurality of normal sample data volumes; The multiple initial training weights are divided by the multiple normal sample data amounts to perform data-level weight allocation to obtain multiple sample data weight sets; Build a device fundamental frequency predictor based on machine learning; Using the multiple groups of normal fundamental frequency training sample sets, allocating training resources according to the multiple sample data weight sets, and performing supervised training and testing on the device fundamental frequency predictor until convergence; The vibration signal of the current target device is collected, input into the device fundamental frequency predictor, and the device fundamental frequency data is obtained by prediction output.

[0066] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0067] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0068] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0069] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0070] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0071] Although preferred embodiments of the present invention have been described, additional changes and modifications may occur to these embodiments once those skilled in the art understand the basic inventive concepts.

[0072] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention belong to the scope of the present invention and its equivalent technologies, the present invention is also intended to include these changes and variations.

Claims

1. A device fundamental frequency data generation system based on device fundamental frequency prediction, characterized in that: The system comprises: A data acquisition module is used to acquire multiple operating conditions, collect vibration signals of the equipment under the multiple operating conditions in historical time and perform fundamental frequency processing and marking to obtain multiple sample operating condition vibration signal sets and multiple sample operating condition fundamental frequency data sets; An abnormal screening module is used to obtain the working condition characteristic information of various operating conditions, configure multiple abnormal fundamental frequency screening ratios, and perform isolated abnormal fundamental frequency screening on the multiple sample working condition fundamental frequency data sets respectively, obtain multiple abnormal fundamental frequency data sets and delete them, and obtain multiple normal working condition vibration signal sets and multiple normal working condition fundamental frequency data sets; A weight configuration module, used to calculate the abnormality of the multiple abnormal fundamental frequency data sets, and configure multiple initial training weights in combination with multiple operating condition feature information; The fundamental frequency prediction module is used to train the equipment fundamental frequency predictor according to the multiple initial training weights, using the multiple normal operating vibration signal sets and the multiple normal operating fundamental frequency data sets, collect the vibration signal of the current target equipment, input it into the equipment fundamental frequency predictor, and predict and output to obtain the equipment fundamental frequency data.

2. The device fundamental frequency data generation system based on device fundamental frequency prediction according to claim 1, characterized in that: The data acquisition module comprises: A working condition acquisition unit, used to acquire various operating conditions of the target device; A vibration signal collection unit, used to collect vibration signals of similar equipment under the multiple operating conditions based on historical operation monitoring data of similar equipment, and obtain multiple sample condition vibration signal sets; The fundamental frequency processing and labeling unit is used to process each sample working condition vibration signal by Fourier transform and harmonic product spectrum to obtain fundamental frequency data, and label it to obtain multiple sample working condition fundamental frequency data sets.

3. The device fundamental frequency data generation system based on device fundamental frequency prediction according to claim 1, characterized in that: The abnormal screening module comprises: A working condition characteristic acquisition unit is used to acquire working condition operation time and average operating load of the same type of equipment under various working conditions as multiple working condition characteristic information; A coefficient calculation unit, used to calculate the ratio of each operating condition operation time to the average of multiple operating condition operation times within multiple operating condition characteristic information to obtain multiple time coefficients, and calculate the ratio of each average operating load to the average of multiple average operating loads to obtain multiple load coefficients; An adjustment coefficient calculation unit, used for calculating and obtaining a plurality of screening ratio adjustment coefficients according to a plurality of time coefficients and a plurality of load coefficients; A preset ratio acquisition unit, used to acquire a preset abnormal fundamental frequency screening ratio; The abnormal proportion configuration unit is used to respectively use the multiple screening proportion adjustment coefficients to configure and calculate the preset abnormal base frequency screening proportion to obtain multiple abnormal base frequency screening proportions.

4. The device fundamental frequency data generation system based on device fundamental frequency prediction according to claim 1, characterized in that: The abnormal screening module also includes: A baseband interval acquisition unit, used to acquire a baseband data interval of a device; An isolated data screening unit, used for randomly generating first fundamental frequency data in the fundamental frequency data interval, performing binary classification on the plurality of sample operating condition fundamental frequency data sets, and determining whether there is any sample operating condition fundamental frequency data classified as isolated fundamental frequency data; An abnormal data classification unit is used to classify the isolated fundamental frequency data as abnormal fundamental frequency data if yes, and if no, continue to randomly generate the second fundamental frequency data to perform isolated abnormal fundamental frequency screening; A screening ratio control unit, used to stop the isolated abnormal fundamental frequency screening until the proportion of abnormal fundamental frequency data divided in each sample working condition fundamental frequency data set reaches the multiple abnormal fundamental frequency screening ratios, and obtain multiple abnormal fundamental frequency data sets that have been screened; The normal data acquisition unit is used to acquire multiple abnormal working condition vibration signal sets corresponding to the multiple abnormal fundamental frequency data sets, delete them, and obtain multiple normal working condition vibration signal sets and multiple normal working condition fundamental frequency data sets.

5. The device fundamental frequency data generation system based on device fundamental frequency prediction according to claim 1, characterized in that: The weight configuration module includes: A mean value calculation unit, used to respectively calculate the means of the plurality of normal operating condition fundamental frequency data sets to obtain a plurality of average normal fundamental frequency data; The abnormality degree calculation unit is used to respectively calculate the deviation amplitudes of the abnormal fundamental frequency data in the multiple abnormal fundamental frequency data sets and the multiple average normal fundamental frequency data to obtain multiple abnormal deviation amplitude sets, and calculate the average to obtain multiple abnormality degrees.

6. The device fundamental frequency data generation system based on device fundamental frequency prediction according to claim 1, characterized in that: The weight configuration module also includes: A time weight calculation unit, used to calculate the ratio of each operating condition running time to the sum of the operating times of the multiple operating conditions in the multiple operating condition feature information as multiple operating time weights; A load weight calculation unit, used for calculating the ratio of each average operating load to the sum of multiple average operating loads as multiple load weights; An abnormal weight allocation unit, used for allocating a plurality of abnormal training weights for a plurality of operating conditions according to the plurality of abnormal degrees, wherein the magnitude of the abnormal degree is negatively correlated with the magnitude of the abnormal training weight; The weight fusion calculation unit is used to perform weight fusion calculation according to multiple running time weights, multiple load weights and multiple abnormal training weights to obtain multiple initial training weights.

7. The device fundamental frequency data generation system based on device fundamental frequency prediction according to claim 1, characterized in that: The fundamental frequency prediction module comprises: A training sample combining unit, used for combining the plurality of normal working condition vibration signal sets and the plurality of normal working condition fundamental frequency data sets to obtain a plurality of normal fundamental frequency training sample sets and a plurality of normal sample data volumes; A sample weight allocation unit, configured to respectively divide the multiple initial training weights by the multiple normal sample data amounts to perform data-level weight allocation to obtain multiple sample data weight sets; A predictor building unit, used to build a device fundamental frequency predictor based on machine learning; A predictor training unit, configured to use the plurality of normal fundamental frequency training sample sets, allocate training resources according to the plurality of sample data weight sets, and perform supervised training and testing on the device fundamental frequency predictor until convergence; The fundamental frequency data generating unit is used to collect the vibration signal of the current target device, input the vibration signal into the device fundamental frequency predictor, and predict and output the fundamental frequency data of the device.

8. A method for generating device fundamental frequency data based on device fundamental frequency prediction, characterized in that: The method is performed by a device fundamental frequency data generation system based on device fundamental frequency prediction according to any one of claims 1 to 7, and the method comprises: Acquire multiple operating conditions, collect vibration signals of the equipment under the multiple operating conditions in historical time and perform fundamental frequency processing and marking, and obtain multiple sample operating condition vibration signal sets and multiple sample operating condition fundamental frequency data sets; Acquire the working condition characteristic information of multiple operating conditions, configure multiple abnormal fundamental frequency screening ratios, perform isolated abnormal fundamental frequency screening on the multiple sample working condition fundamental frequency data sets respectively, obtain multiple abnormal fundamental frequency data sets and delete them, and obtain multiple normal working condition vibration signal sets and multiple normal working condition fundamental frequency data sets; Calculating the abnormality of the multiple abnormal fundamental frequency data sets, combining multiple operating condition feature information, and configuring multiple initial training weights; According to the multiple initial training weights, the multiple normal operating vibration signal sets and the multiple normal operating fundamental frequency data sets are used to train the equipment fundamental frequency predictor, collect the vibration signal of the current target equipment, input the equipment fundamental frequency predictor, and predict and output to obtain the equipment fundamental frequency data.

Images