Motor detection data acquisition method and system based on simulation environment

Abstract

The invention relates to the field of data processing, in particular to a motor detection data acquisition method and system based on a simulation environment, motor detection data is acquired through a preset acquisition frequency, and the local reachable density of the motor detection data is acquired by preprocessing the acquired motor detection data. Furthermore, according to the local reachable density of the motor detection data, the adjustment factor of the data acquisition frequency is calculated, and according to the adjustment factor, the acquisition frequency of the motor detection data is adjusted to carry out the next round of acquisition of the motor detection data, so that the acquisition frequency of the motor detection data is adaptively adjusted according to the abnormal degree of the motor detection data.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular to a method and system for collecting motor detection data based on a simulation environment. Background Art

[0002] In modern industry, motors, as key driving devices, are used in various fields such as manufacturing, transportation, and energy. With the rapid development of industrial automation, the requirements for motor performance reliability, efficiency, and fault diagnosis are increasing.

[0003] In the process of collecting motor detection data, there is a higher degree of concern for data that may contain abnormalities, and more detailed data collection is required. In the process of collecting motor detection data, a fixed data collection frequency often cannot meet the needs of data collection. Collecting detection data at a higher collection frequency will consume a large amount of additional calculation when collecting relatively stable normal data, while a lower collection frequency will result in the inability to fully obtain the changes in the abnormal data that is the focus of attention, resulting in deviations in the analysis of motor detection data.

[0004] For example, a Chinese patent application document with publication number CN114966396A discloses a motor data acquisition device, which is connected to the motor through data acquisition modules such as a motor three-phase current acquisition module, a motor vibration data acquisition module, a motor temperature acquisition module and a motor analog quantity acquisition module to collect motor data such as current data, vibration data and temperature data. However, due to the lack of an efficient and reasonable data acquisition method, the collected data cannot effectively reflect the operating conditions of the motor, and it is difficult to detect abnormalities in the motor operation in a timely manner. Summary of the invention

[0005] In view of the problem that the above-mentioned collected data cannot effectively reflect the operation status of the motor and it is difficult to timely discover the abnormality of the motor operation, in the first aspect, the present invention proposes a motor detection data collection method based on a simulation environment, comprising the following steps: collecting a number of motor detection data, and calculating the local reachable density of the collected motor detection data by using a LOF algorithm; calculating the adjustment factor of the data collection frequency according to the local reachable density, and the calculation method of the adjustment factor is specifically as follows: ; Among them, A i represents the adjustment factor of the ith data point, P i represents the density anomaly degree of the ith data point, which is limited by the local reachable density and distance of the ith data point and its surrounding data points. irepresents the local fluctuation degree of the ith data point, which is positively correlated with the change degree of the value of the ith data point and the values ​​of multiple data points around it. P Represents the preset density anomaly threshold, T F represents a preset local fluctuation degree threshold; and adjusting the acquisition frequency of the motor detection data according to the adjustment factor to collect the next round of motor detection data.

[0006] The local reachable density of the collected motor detection data is calculated through the LOF algorithm, and the abnormality degree of each data point can be quantified according to the local reachable density to adaptively adjust the collection frequency. This method has higher sensitivity to abnormal data and can adaptively increase the collection frequency of abnormal data and reduce the collection frequency of normal data, thereby improving the collection efficiency.

[0007] Furthermore, a number of motor detection data are collected, including: collecting two or more of the following signals: rotation speed, temperature, current, and voltage.

[0008] By collecting two or more signals, it is possible to prevent detection errors caused by collecting only one signal.

[0009] Furthermore, the calculation method of the density anomaly degree is specifically as follows: ; Among them, P i Indicates the density anomaly of the i-th data point, C i represents the number of all other data points within the local reachable distance of the ith data point, exp() represents the natural exponential function, I k Represents the gap in data collection timing, ρ i,k represents the local reachable density of the kth data point among multiple data points around the i-th data point, d(i,k) represents the distance between the kth data point and the i-th data point, represents the average value of the local reachability density of other data points within the local reachability distance of the i-th data point, It represents the average distance between the i-th data point and other data points within the local reachable distance.

[0010] By calculating the local reachable density of the collected motor detection data and the distance between data points, the density abnormality can be quantified to calculate the adjustment factor. The exp (-I k ) item takes into account the gap in the timing of data collection, which means that this method can take into account the changing trend of data points over time, thereby identifying anomalies more accurately.

[0011] Furthermore, the distance measurement method between the data points is configured as one or more of the following: Euclidean distance, Manhattan distance, and Mahalanobis distance.

[0012] When the number of data items included in the data points is too large, conventional distance measurement methods are difficult to directly measure the distance between the data points of multiple data items. For example, when the data points include four data items: speed, temperature, current, and voltage, it is no longer possible to calculate the relative distance of the data points through conventional two-dimensional or three-dimensional distance measurement methods. By selecting a suitable distance measurement method (such as Euclidean distance), the relative distance of multi-dimensional data points can be calculated.

[0013] Furthermore, the calculation method of the local fluctuation degree is specifically as follows: ; Among them, F i represents the local fluctuation degree of the ith data point, N represents the number of data items contained in each data point, and K represents the number of other data points in the neighborhood of each data point. represents the difference in acquisition time between the i-th data point and the k-th data point among the multiple data points around it, D j (i,k) represents the jth data item of the kth data point among multiple data points around the i-th data point. It represents the average value of the jth data point among multiple data points around the i-th data point. Represents the standard deviation of the j-th data point among all data points around the i-th data point.

[0014] Further, adjusting the collection frequency of the motor detection data according to the adjustment factor includes: taking an average value of several adjustment factors calculated from several motor detection data collected in this round, and multiplying the average value by the collection frequency of this round as the collection frequency of the next round of motor detection data.

[0015] By averaging the adjustment factors of multiple data points, we can comprehensively consider the changes in the overall data and avoid errors in the calculated adjustment factors due to errors in a single data point. By taking the average value, we can more reasonably derive a sampling frequency.

[0016] Furthermore, the adjustment factors calculated from the motor detection data collected in this round are averaged, including: averaging the adjustment factors calculated from the motor detection data within a preset time period, or averaging the adjustment factors calculated from the motor detection data of a preset number of times.

[0017] Furthermore, the calculation of the local reachable density includes setting a K value, where the K value represents the number of neighboring data points considered when calculating the local reachable density, and the K value is set to 10.

[0018] By setting a reasonable K value, we can take into account both the amount of data calculation and the calculation efficiency, avoid the impact of excessive calculation on the calculation efficiency due to setting a K value that is too large, and avoid the calculation results being unrepresentative due to a K value that is too small. Setting a reasonable K value can efficiently and reasonably calculate the local reachable density of the motor detection data.

[0019] Furthermore, the motor detection data acquisition method based on a simulation environment also includes filtering, amplifying and analog-to-digital conversion operations on the collected motor detection data to convert the collected analog data signal into a digital signal suitable for analysis and processing.

[0020] In a second aspect, a motor detection data acquisition system based on a simulation environment is provided, comprising an acquisition device and a controller, wherein the controller comprises a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the motor detection data acquisition method based on a simulation environment of the present invention is implemented.

[0021] The technical effects of the present invention are: The motor detection data is collected at a preset collection frequency, and the local reachable density of the motor detection data is obtained by preprocessing the collected motor detection data. Further, an adjustment factor of the data collection frequency is calculated according to the local reachable density of the motor detection data, and the collection frequency of the motor detection data is adjusted according to the adjustment factor to collect the next round of motor detection data, thereby realizing adaptive adjustment of the collection frequency of the motor detection data according to the abnormality of the motor detection data. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein: Figure 1 is a flow chart schematically showing a method for collecting motor detection data based on a simulation environment according to an embodiment of the present invention; Figure 2 It is a block diagram schematically showing the structure of a motor detection data acquisition system based on a simulation environment according to an embodiment of the present invention. DETAILED DESCRIPTION

[0023] 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 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.

[0024] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0025] Motor detection data collection method embodiment: like Figure 1 As shown, the motor detection data acquisition method based on the simulation environment of the present invention includes: S1. Collect motor detection data, and pre-process the motor detection data to obtain local reachable density of the motor detection data.

[0026] As an important power equipment in industrial production, the operating status of the motor is the key detection object in the production process. Usually, the motor's performance monitoring, fault diagnosis or predictive maintenance are carried out by detecting the motor's operating parameters. Among them, common parameters include speed, temperature, current, voltage, etc. Select the corresponding sensor according to the parameters to be detected, such as temperature sensor, vibration sensor, current sensor, etc. During the operation of the motor, the initial acquisition frequency of the motor detection data can be preset, for example, the initial acquisition frequency is 1 time / hour, and the data acquisition system is used to collect sensor signals. In some cases, it is also necessary to filter, amplify and perform analog-to-digital conversion on the collected motor detection data (for example, the sensor outputs an analog signal) to make it suitable for analysis and processing.

[0027] Furthermore, in order to obtain the local reachable density of the motor detection data, the collected motor detection data is preprocessed by using the LOF (Local Outlier Factor) algorithm.

[0028] In the present invention, the data collected each time is called a data point, and the set of data points collected in each round is called a data set. For example, in this embodiment, 24 data points are collected in each round, and the 24 data points constitute the data set collected in this round.

[0029] A K value of local reachable density is preset for each data point. The K value refers to the number of neighbor points of the data point considered when calculating the local reachable density. The neighbor points refer to a group of points that are closest to a specific data point under a given distance metric. For example, the K value is preset to 5, which means that under a given distance metric, the distance between a specific data point and other data points is calculated, and the 5 data points with the smallest distance value are selected. The K value can be preset to 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 or other values. Preferably, the K value is preset to 10.

[0030] Further, the local reachable density of each data point in the data set is obtained, and the local reachable density refers to the reciprocal of the average reachable distance from the data point to its neighboring points. In one embodiment, the K value is preset to 10, and the frequency of the first round of collection is 1 time / hour, and one data point is collected each time, and each round of collection is 24 times.

[0031] In this embodiment, after the first round of 24 motor detection data points are collected, one of the data points is selected as a specific data point (for example, the first data point is selected), and the local reachable density of the data point is calculated, that is, the reciprocal of the average of the reachable distances between the data point and its 10 neighboring data points (the K value is preset to 10). In this embodiment, the distance measurement method selects Euclidean distance. In other embodiments, the distance measurement method can also select other distance measurement methods such as Manhattan distance and Mahalanobis distance, and the present invention does not limit this. Similarly, the local reachable density of the remaining 23 data points in the data set can be calculated.

[0032] In an embodiment of the present invention, the collected data points include collecting more than two signals of motor speed, temperature, current, and voltage, that is, one data point includes more than two data items.

[0033] S2. Calculate an adjustment factor of data acquisition frequency according to the local reachable density of the motor detection data.

[0034] In one embodiment, the calculation formula of the adjustment factor is: ; Among them, A i represents the adjustment factor of the ith data point, P i represents the density anomaly degree of the ith data point, which is limited by the local reachable density and distance of the ith data point and its surrounding data points. i represents the local fluctuation degree of the ith data point, which is positively correlated with the change degree of the value of the ith data point and the values ​​of multiple data points around it. P Represents the preset density anomaly threshold, T F Indicates the preset local fluctuation degree threshold.

[0035] For any data point, the density trend of the data point is obtained according to its local reachable density and the local reachable density of other surrounding data, and the degree of outlier of the data point is judged according to the density trend of the data point.

[0036] The specific calculation process of the density anomaly degree is as follows: obtain all other data points within the local reachable distance of the ith data point, where the local reachable distance represents the distance between the ith data point and the data point that is K farthest from it (for example, if the K value is 10, the distance between the ith data point in the data set and the data point that is 10th farthest from the ith data point is taken as the local reachable distance), and the number of all other data points within the local reachable distance is recorded as C i (For example, if K is 10, There are at least 10 data points within the local reachable distance of a data point. If there is another data point outside these 10 data points whose distance is exactly equal to the local reachable distance, then C i =11), sort all other data points according to the Euclidean distance between them and the i-th data point, obtain the Euclidean distance between the k-th data point and the i-th data point, record it as d(i,k), and the local reachability density of the k-th data point (for example, K value is 10, k represents the sequence number of these 10 data points, that is, k ranges from 1 to 10) is recorded as ρ i,k , the difference in data collection timing between the kth data point and the ith data point is recorded as I k (For example, the collection time of the kth data point is 09:00, the collection time of the ith data point is 12:00, and the time difference is 3, then I k The above data is used to reflect the distribution of the local reachable density of data points within the local reachable distance range of the i-th data point, and then calculate the density anomaly degree P of the i-th data point i In this embodiment, the density anomaly degree P of the i-th data point i The calculation formula is: ; Among them, C i represents the number of all other data points within the local reachable distance of the ith data point, and exp() represents the natural exponential function. Indicates The average value of the local reachability density of other data points within the local reachability distance of a data point, Indicates The average value of the Euclidean distance between a data point and other data points within the local reachable distance.

[0037] It should be noted that in the formula of The part represents the ratio of the local reachability density of the data point within the local reachability distance range of the data point to its Euclidean distance, and the ratio of this ratio to the average of the two data. The larger the difference is, the more likely the data point is an abnormal data point. Then, by using the exp(-I k ) Partially weakened and The degree of influence of the ratio difference on the data points whose collection time is far apart.

[0038] The above formula can be used to obtain the density anomaly degree of each data point in the data set.

[0039] In this embodiment, in addition to the density abnormality, the degree of data abnormality is also measured based on the data point collection timing to obtain the degree of fluctuation of data that is close to the data point collection time. Data points with greater fluctuations may also be data points that need to be focused on.

[0040] According to the K value of the local reachable density, the K data points closest to each data point are obtained, and the j-th data value of the k-th (k ranges from 1 to K) local range data point among the K data points is recorded as D j (i,k).

[0041] Specifically, the local fluctuation degree of the data point is obtained according to various data within the local range of the data point: ; Among them, F i represents the local fluctuation degree of the ith data point, N represents the number of data items contained in each data point, and K represents the number of other data points in the neighborhood of each data point. represents the difference in acquisition time between the i-th data point and the k-th data point among the multiple data points around it, D j (i,k) represents the jth data item of the kth data point among multiple data points around the i-th data point. It represents the average value of the jth data point among multiple data points around the i-th data point. Represents the standard deviation of the jth data point among all the data points around the ith data point. The above formula can be used to obtain the local fluctuation degree of each data point in the data set.

[0042] Furthermore, by adjusting the factor calculation formula The adjustment factor of each data point in this round of data set can be obtained.

[0043] The sampling frequency of the data is adjusted according to the adjustment factor. The larger the adjustment factor, the greater the sampling frequency required at the data point to ensure that the detailed changes in the data can be collected.

[0044] The above method can be used to calculate the 24 adjustment factors (A1…A 24 ).

[0045] S3. Adjust the acquisition frequency of the motor detection data according to the adjustment factor to collect the motor detection data for the next time.

[0046] Specifically, the 24 adjustment factors calculated from the motor detection data collected in this round are averaged, and the average value is multiplied by the collection frequency of this round to serve as the collection frequency of the motor detection data of the next round.

[0047] Among them, each collection round can be determined by time period (for example, if 24 hours is one collection round, then the collection is performed at the same collection frequency in 24 hours), or it can be determined by the number of collection times (for example, if the collection of 24 data points is one collection round, then the collection frequency used for the 24 data points in this round is the same).

[0048] The local reachable density of the collected motor detection data is calculated through the LOF algorithm, and the abnormality degree of each data point can be quantified according to the local reachable density to adaptively adjust the collection frequency. This method has higher sensitivity to abnormal data and can adaptively increase the collection frequency of abnormal data and reduce the collection frequency of normal data, thereby improving the collection efficiency.

[0049] Embodiment of the motor detection data acquisition system based on the simulation environment: On the other hand, the present invention also provides a motor detection data acquisition system based on a simulation environment. Figure 2 As shown, the motor detection data acquisition system based on the simulation environment includes an acquisition device and a controller, the acquisition device includes a motor speed acquisition device (for example, a speed acquisition device of model DT2235B can be used), a temperature acquisition device (for example, a temperature acquisition device of model PX05006 can be used), a current acquisition device (for example, a current acquisition device of model INA282AQDRQ1 can be used), and a voltage acquisition device (for example, a voltage acquisition device of model YK-DCD-DS-16-DV-500 can be used), the controller includes a processor and a memory, the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a motor detection data acquisition method based on the simulation environment according to the first aspect of the present invention is implemented.

[0050] The motor detection data acquisition system based on the simulation environment also includes other components well known to those skilled in the art, such as a communication interface, and the configuration and functions thereof are known in the art, so they will not be described in detail here.

[0051] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, apparatus or device. For example, a computer-readable storage medium may be any appropriate magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM (Resistive Random Access Memory), a dynamic random access memory DRAM (Dynamic Random Access Memory), a static random access memory SRAM (Static Random-Access Memory), an enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), a high-bandwidth memory HBM (High-Bandwidth Memory), a hybrid memory cube HMC (Hybrid Memory Cube), etc., or any other medium that can be used to store the required information and can be accessed by an application, a module, or both. Any such computer storage medium may be part of a device or accessible or connectable to a device. Any application or module described in the present invention may be implemented by computer-readable / executable instructions stored or otherwise maintained by such a computer-readable medium.

[0052] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0053] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. A motor detection data acquisition method based on a simulation environment, characterized in that: The method comprises: Collect a number of motor detection data, and calculate the local reachable density of the collected motor detection data by using the LOF algorithm; The adjustment factor of the data acquisition frequency is calculated according to the local reachable density, and the calculation method of the adjustment factor is specifically as follows: ； Among them, A i represents the adjustment factor of the ith data point, P i represents the density anomaly degree of the ith data point, which is limited by the local reachable density and distance of the ith data point and its surrounding data points. i represents the local fluctuation degree of the ith data point, which is positively correlated with the change degree of the value of the ith data point and the values ​​of multiple data points around it. P Represents the preset density anomaly threshold, T F Indicates the preset local fluctuation degree threshold; The frequency of collecting the motor detection data is adjusted according to the adjustment factor to collect the next round of motor detection data.

2. The motor detection data acquisition method based on a simulation environment according to claim 1 is characterized in that: Collect a number of motor detection data, including: collect two or more of the following signals: speed, temperature, current, voltage.

3. The motor detection data acquisition method based on a simulation environment according to claim 1 is characterized in that: The calculation method of the density anomaly degree is specifically as follows: ； Among them, P i Indicates the density anomaly of the i-th data point, C i represents the number of all other data points within the local reachable distance of the ith data point, exp() represents the natural exponential function, I k Represents the gap in data collection timing, ρ i,k represents the local reachable density of the kth data point among multiple data points around the i-th data point, d(i,k) represents the distance between the kth data point and the i-th data point, represents the average value of the local reachability density of other data points within the local reachability distance of the i-th data point, It represents the average distance between the i-th data point and other data points within the local reachable distance.

4. The motor detection data acquisition method based on a simulation environment according to claim 3 is characterized in that: The distance measurement method between the data points is configured as one or more of the following: Euclidean distance, Manhattan distance, and Mahalanobis distance.

5. The motor detection data acquisition method based on a simulation environment according to claim 1 is characterized in that: The calculation method of the local fluctuation degree is specifically as follows: ； Among them, F i represents the local fluctuation degree of the ith data point, N represents the number of data items contained in each data point, and K represents the number of other data points in the neighborhood of each data point. represents the difference in acquisition time between the i-th data point and the k-th data point among the multiple data points around it, D j (i,k) represents the jth data item of the kth data point among multiple data points around the i-th data point. It represents the average value of the jth data point among multiple data points around the i-th data point. Represents the standard deviation of the j-th data point among all data points around the i-th data point.

6. The motor detection data acquisition method based on a simulation environment according to claim 1 is characterized in that: The acquisition frequency of the motor detection data is adjusted according to the adjustment factor, including: taking an average value of several adjustment factors calculated from several motor detection data collected in this round, and multiplying the average value by the acquisition frequency of this round to obtain the acquisition frequency of the next round of motor detection data.

7. The motor detection data acquisition method based on a simulation environment according to claim 6 is characterized in that: The adjustment factors calculated from the motor detection data collected in this round are averaged, including: averaging the adjustment factors calculated from the motor detection data within a preset time period, or averaging the adjustment factors calculated from the motor detection data of a preset number of times.

8. The motor detection data acquisition method based on a simulation environment according to claim 1 is characterized in that: The calculation of the local reachable density includes setting a K value, where the K value represents the number of neighboring data points considered when calculating the local reachable density, and the K value is set to 10.

9. The motor detection data acquisition method based on a simulation environment according to claim 1, characterized in that: It also includes filtering, amplifying and analog-to-digital conversion operations on the collected motor detection data to convert the collected analog data signals into digital signals.

10. A motor detection data acquisition system based on a simulation environment, characterized in that: It includes an acquisition device and a controller, the controller includes a processor and a memory, the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the motor detection data acquisition method based on a simulation environment as described in any one of claims 1 to 9 is implemented.

Images