An abnormal detection method for an electrotome motor

Abstract

The present invention discloses a method for detecting abnormal conditions of an electrosurgical knife motor, which relates to the technical field of detecting abnormal conditions of an electrosurgical knife motor. The method includes: extracting the sample frequency domain features of the motor vibration signal samples of the electrosurgical knife, and obtaining a plurality of first feature sets based on the sample frequency domain features; obtaining any two first feature sets, and respectively obtaining the features of the sets to obtain a first feature and a second feature; obtaining a distance threshold based on the distance between the first feature and the second feature; extracting the to-be-detected frequency domain features of the to-be-detected vibration signal of the to-be-detected electrosurgical knife, and obtaining a plurality of third sets based on the to-be-detected frequency domain features; respectively obtaining the features of the third set and the first feature set to obtain a third feature and a fourth feature; obtaining a to-be-detected distance based on the distance between the third feature and the fourth feature; and obtaining a first detection result of the motor based on the distance threshold and the to-be-detected distance, which can solve the problem that the feature fusion of the traditional vibration detection method is only simple splicing, resulting in low accuracy of abnormal condition detection.

Description

Technical Field

[0001] The present invention relates to the field of abnormal detection of electrosurgical unit motors, and specifically, to a method for detecting abnormal conditions of electrosurgical unit motors. Background Art

[0002] As an important medical surgical instrument, the normal operation of the electrosurgical unit motor is crucial for the safety and effectiveness of the surgery. During the operation of the electrosurgical unit motor, due to the influence of various factors, abnormal vibrations may occur, such as bearing wear, rotor imbalance, motor winding faults, etc. These abnormal vibrations not only affect the cutting accuracy and performance of the electrosurgical unit, but also may increase the surgical risk.

[0003] The traditional vibration detection method is to extract the characteristics of the vibration data of the electrosurgical unit motor, normalize all the characteristics and then superimpose them to obtain the superimposed characteristics, and perform abnormal detection by identifying the superimposed characteristics. However, since this method only simply concatenates the characteristics, that is, the weights of each characteristic are the same or the weights are designed manually through experience, the characteristics cannot be fully fused, resulting in low accuracy of abnormal detection. Moreover, due to the influence of different surgical types and individual differences, etc., this method has poor adaptability to complex working conditions and cannot effectively meet the actual needs, further leading to low accuracy of abnormal detection. Summary of the Invention

[0004] In order to solve the problem that the feature fusion of the traditional vibration detection method only simply concatenates the features and the features cannot be fully fused, resulting in low accuracy of abnormal detection, the present invention provides a method for detecting abnormal conditions of an electrosurgical unit motor, and the method includes:

[0005] Obtain a plurality of motor vibration signal samples of the electrosurgical unit, extract the sample frequency domain features of each of the motor vibration signal samples, obtain a sample frequency domain feature matrix based on the sample frequency domain features, and obtain a plurality of first feature sets based on the sample frequency domain feature matrix, and each of the first feature sets includes a plurality of the sample frequency domain features after normalization processing;

[0006] Obtain any two of the first feature sets, obtain a first set and a second set, respectively obtain the features of the first set and the second set, and obtain a plurality of first features and a plurality of second features;

[0007] Based on a first weight coefficient, obtain the Jaccard distance between the first feature and the second feature, obtain a plurality of first Jaccard distances, and obtain a first distance threshold based on the maximum of the first Jaccard distances;

[0008] Obtain the vibration signal to be detected of the electrosurgical unit, extract the frequency domain features to be detected of the vibration signal to be detected, obtain the frequency domain feature matrix to be detected based on the frequency domain features to be detected, obtain a number of third sets based on the frequency domain feature matrix to be detected, and each of the third sets includes a number of the frequency domain features to be detected after normalization processing;

[0009] Respectively obtain the features of the third set and the first feature set, and obtain a number of third features and a number of fourth features;

[0010] Based on the second weight coefficient, obtain the Jaccard distance between the third feature and the fourth feature, obtain a number of second Jaccard distances, and obtain the first distance to be detected based on the maximum of the second Jaccard distances;

[0011] Based on the first distance threshold and the first distance to be detected, obtain the first detection result of the motor of the electrosurgical unit to be detected.

[0012] Determining the weights of each feature can more accurately reflect the importance of different features in anomaly detection and improve the accuracy and reliability of detection; different vibration features have different degrees of importance in reflecting the operating state of the motor. The present invention introduces a weight coefficient. By assigning different weight coefficients to each vibration feature, the key features that have a greater impact on the operating state of the motor can be highlighted, and the interference of secondary features can be suppressed; the weight coefficient can adjust the contribution degree of each feature in the fusion process according to the actual situation, so that the fused features can more accurately reflect the relationship between the operating state of the motor and each feature, and thus the fused features can more accurately reflect the operating state of the motor. The Jaccard distance is used to measure the degree of difference between two sets. In vibration feature fusion, it can be used to measure the similarity or difference between different vibration features. By calculating the Jaccard distance, the present invention can understand the correlation degree between each vibration feature and provide a basis for feature fusion. The present invention combines the weight coefficient and the Jaccard distance, which can highlight the key features that have a greater impact on the operating state of the motor while being able to capture the non-linear relationship between different vibration features. Even if there are complex interactions between features, the Jaccard distance can reflect their internal connection to a certain extent, so as to more comprehensively and fully fuse the feature information, make the fused features more accurately reflect the operating state of the motor, and thus improve the accuracy and precision of anomaly detection.

[0013] Among the multiple characteristics of motor vibration, different characteristics have different sensitivities to reflecting minute changes. By reasonably setting weight coefficients, those characteristics that are more sensitive to minute changes will be assigned higher weights. In this way, when fusing characteristics, the minute changes of these key characteristics will be highlighted and amplified, making them easier to detect. The Jaccard distance can consider the correlations between vibration characteristics as a whole. When there are minute changes in motor vibration, it may cause coordinated changes among multiple characteristics. The Jaccard distance can capture the differences in the set of characteristics brought about by this coordinated change, rather than just the changes of individual characteristics, thus reflecting the minute changes in motor vibration more comprehensively and sensitively.

[0014] After combining the weight coefficient and the Jaccard distance, it can be adaptively adjusted according to the actual situation of the motor vibration characteristics. When capturing minute changes, the weight coefficient will guide the attention to the changes of key characteristics, while the Jaccard distance makes a comprehensive judgment from the perspective of the relationships between characteristics. When there is a minute change in a certain key characteristic, the weight coefficient makes it occupy an important position in the fused characteristics. At the same time, the Jaccard distance will examine the relationship between this change and other characteristics, and comprehensively evaluate whether it is indeed an effective minute change in motor vibration, rather than a pseudo-change caused by noise or other interferences, so as to more accurately and sensitively capture the minute changes in motor vibration, which is more conducive to the detection of early weak faults.

[0015] Furthermore, the method further includes:

[0016] Extract the sample time-domain characteristics of each of the motor vibration signal samples, obtain a sample time-domain characteristic matrix based on the sample time-domain characteristics, and obtain a number of second characteristic sets based on the sample time-domain characteristic matrix. Each of the second characteristic sets includes a number of the sample time-domain characteristics after normalization processing;

[0017] Obtain any two of the second characteristic sets, obtain a fourth set and a fifth set, respectively obtain the characteristics of the fourth set and the fifth set, and obtain a number of fifth characteristics and a number of sixth characteristics;

[0018] Based on a third weight coefficient, obtain the Jaccard distances of the fifth characteristics and the sixth characteristics, obtain a number of third Jaccard distances, and obtain a second distance threshold based on the maximum of the third Jaccard distances;

[0019] Extract the time-domain characteristics to be detected of the vibration signal to be detected, obtain a time-domain characteristic matrix to be detected based on the time-domain characteristics to be detected, and obtain a number of sixth sets based on the time-domain characteristic matrix to be detected. Each of the sixth sets includes a number of the time-domain characteristics to be detected after normalization processing;

[0020] Respectively obtain the features of the sixth set and the second feature set to obtain a number of seventh features and a number of eighth features; based on the fourth weight coefficient, obtain the Jaccard distance between the seventh feature and the eighth feature to obtain a number of fourth Jaccard distances, and based on the maximum of the fourth Jaccard distances, obtain the second distance to be detected;

[0021] Based on the second distance threshold and the second distance to be detected, obtain the second detection result of the motor of the electrosurgical knife to be detected;

[0022] Based on the first detection result and the second detection result, obtain the third detection result of the motor, and update the first detection result of the motor to the third detection result.

[0023] Obtain the time-domain features of the motor vibration. By combining time-domain and frequency-domain analysis, the characteristic information of the vibration signal can be obtained more comprehensively, and the abnormal information detection is more accurate.

[0024] Further, the acquisition methods of the first weight coefficient, the second weight coefficient, the third weight coefficient, and the fourth weight coefficient are all: obtain the first mean of the features, and obtain the first variance based on the first mean; obtain the first covariance of any two features, and obtain the first correlation coefficient based on the first covariance and the first variance; based on the first correlation coefficient and the first variance, obtain the weight coefficient of each feature.

[0025] The variance reflects the degree of dispersion of each feature in different samples. The greater the degree of dispersion, the richer the information contained in the feature; the correlation coefficient measures the linear correlation degree between two features. The closer it is to 1, the stronger the correlation between the two features, and the closer it is to 0, the weaker the correlation; determining the weight according to the variance and correlation comprehensively considers and normalizes the information content of each feature, takes into account various characteristics and historical data of the motor, ensures the scientificity and rationality of the weight, and makes the entire detection method more adaptable and effective.

[0026] Further, the specific steps for obtaining the sample frequency-domain feature matrix include:

[0027] Normalize the sample frequency-domain features, and based on the normalized sample frequency-domain features, obtain the sample frequency-domain feature matrix;

[0028] The specific steps for obtaining the frequency-domain feature matrix to be detected include: normalizing the frequency-domain features to be detected, and based on the normalized frequency-domain features to be detected, obtain the frequency-domain feature matrix to be detected;

[0029] The specific steps for obtaining the time-domain feature matrix of the sample include: normalizing the time-domain features of the sample, and obtaining the time-domain feature matrix of the sample based on the normalized time-domain features of the sample;

[0030] The specific steps for obtaining the time-domain feature matrix to be detected include: normalizing the time-domain features to be detected, and obtaining the time-domain feature matrix to be detected based on the normalized time-domain features to be detected.

[0031] Normalization is to unify the value ranges of various feature parameters into the interval [0, 1].

[0032] Further, the calculation formula for normalizing the frequency-domain features of the sample is:

[0033] where represents the th normalized frequency-domain feature of the sample, represents the th frequency-domain feature of the sample before normalization, and respectively represent the minimum and maximum values.

[0034] Further, the calculation formula for the first Jaccard distance is:

[0035] where represents the first Jaccard distance between the first feature set A and the first feature set B in represents the first weight coefficient of the sample frequency-domain feature represents the th feature of the first feature set A, represents the th feature of the first feature set B, represents the number of features in the first feature set, represents the feature and the feature the number of intersection elements, represents the feature and the feature the number of union elements.

[0036] Further, the calculation formula for the first correlation coefficient is:

[0037] where represents the first correlation coefficient of the sample frequency-domain feature represents the sample frequency-domain feature and the sample frequency domain features covariance, and respectively represent the sample frequency domain features and the sample frequency domain features mean square error, represents the th normalized sample frequency domain feature, represents an integer greater than or equal to 1.

[0038] Furthermore, the calculation formula of the first weight coefficient is:

[0039] Wherein, represents the first weight coefficient of the sample frequency domain feature , and respectively represent the mean square errors of the sample frequency domain features and the sample frequency domain features , represents the number of features in the first feature set.

[0040] Furthermore, the method further includes:

[0041] Classify the motor vibration signal samples to obtain a number of classified sample signals, and obtain the category of the vibration signal to be detected;

[0042] Based on the category, obtain the audio signals corresponding to the classified sample signals and the vibration signal to be detected, respectively obtain the first audio signal and the second audio signal, and obtain the abnormal audio signal based on the first audio signal and the second audio signal;

[0043] Based on the second audio signal and the classified sample signals, respectively obtain the first abnormal vibration signal corresponding to the abnormal audio signal and the second abnormal vibration signal corresponding to the classified sample signals;

[0044] Respectively obtain the first waveform diagram and the second waveform diagram of the first abnormal vibration signal and the second abnormal vibration signal, obtain the abnormal similarity of the first waveform diagram and the second waveform diagram, obtain the abnormal value based on the abnormal similarity, and obtain the usage duration based on the abnormal value and the category.

[0045] When the motor is running, it will emit specific sounds. If there are faults (such as bearing damage, rotor imbalance, etc.) and vibrations, it will cause changes in the frequency, pitch or volume of the sound. According to the audio characteristics, abnormal vibration signals are obtained, and it is comprehensively judged whether it is abnormal. According to the degree of abnormality and the type of surgery, the remaining usable time of the electrotome is judged, so that during the operation, the user can arrange the replacement time more reasonably.

[0046] Further, the specific steps for obtaining the abnormal similarity include:

[0047] Obtain the extreme values of the second waveform diagram, and adjust the first waveform diagram based on the extreme values to obtain a third waveform diagram;

[0048] Based on a preset period, respectively obtain the first data and the second data of the second waveform diagram and the third waveform diagram. Based on the fluctuation directions of the first data and the second data, obtain the waveform similarity, and based on the numerical values of the first data and the second data, obtain a proportional value;

[0049] Based on the waveform similarity and the proportional value, obtain the abnormal similarity.

[0050] Adjust the two waveforms according to the extreme values to expand the differences between the two waveform diagrams, so that the similarity and the proportional value can be obtained more accurately, thereby making the abnormal detection more accurate and the service time more accurate.

[0051] One or more technical solutions provided by the present invention have at least the following technical effects or advantages:

[0052] 1. Assign different weight coefficients to each vibration feature, which can highlight the key features that have a greater impact on the motor operating state and suppress the interference of secondary features; adjust the contribution degree of each feature during the fusion process according to the actual situation, so that the fused features can more accurately reflect the relationship between the motor operating state and each feature, and can more accurately reflect the operating state of the motor.

[0053] 2. Calculate the Jaccard distance between features, which can understand the correlation degree between each vibration feature. In the vibration feature fusion, it can measure the similarity or difference between different vibration features and provide a basis for feature fusion.

[0054] 3. Combine the weight coefficient and the Jaccard distance. While highlighting the key features that have a greater impact on the motor operating state, it can capture the non-linear relationship between different vibration features, can more comprehensively and fully fuse the information of multiple vibration features, and make the fused features more accurately reflect the operating state of the motor, thereby improving the accuracy and precision of abnormal detection.

[0055] 4. Combine the weight coefficient and Jaccard distance. When capturing minute changes, the weight coefficient guides attention to the changes in key features, while the Jaccard distance makes a comprehensive judgment from the perspective of the relationship between features to confirm whether it is an effective minute change in motor vibration, so as to more accurately and sensitively capture the minute changes in motor vibration, which is more conducive to the detection of early weak faults.

[0056] 5. Determine the weights based on variance and correlation, comprehensively consider and normalize the information content of each feature, take into account various characteristics and historical data of the motor, ensure the scientificity and rationality of the weights, and make the entire detection method more adaptable and effective.

[0057] 6. Obtain the usage duration based on outliers and categories, obtain abnormal vibration signals according to audio features, make a comprehensive judgment on whether it is abnormal, and judge the remaining normal usage time of the electrosurgical unit according to the degree of abnormality and the type of surgery, so that during the operation, the user can more reasonably arrange the replacement time.

[0058] 7. Based on the similarity ratio value of waveforms, obtain the abnormal similarity, adjust the two waveforms according to the extreme values, and expand the difference between the two waveform diagrams, so as to more accurately obtain the similarity and ratio value, making the abnormal detection more accurate and the usage time more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, form a part of the present invention, and do not limit the embodiments of the present invention;

[0060] Figure 1 It is a schematic diagram of the detection process of an abnormal detection method for an electrosurgical unit motor in the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0061] In order to more clearly understand the above objects, features and advantages of the present invention, the present invention will be further described in detail below with reference to the drawings and specific embodiments. It should be noted that, without conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other.

[0062] In the following description, many specific details are set forth in order to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited by the specific embodiments disclosed below.

[0063] Embodiment 1

[0064] Refer to Figure 1 , this embodiment provides an abnormal detection method for an electrosurgical unit motor, and the method includes:

[0065] Obtain a number of motor vibration signal samples of the electrosurgical knife, and extract the sample frequency domain features of each of the motor vibration signal samples. For example, obtain the frequency components through methods such as fast Fourier transform. The frequency components can include the harmonic frequencies and their amplitudes of each order, etc., so as to obtain the sample frequency domain features;

[0066] Based on the sample frequency domain features, obtain a sample frequency domain feature matrix, and based on the sample frequency domain feature matrix, obtain a number of first feature sets. Each of the first feature sets contains a number of the sample frequency domain features after normalization processing;

[0067] For example, the sample frequency domain feature matrix is defined as , where is the number of features, is the vector of the th feature, m is the number of samples. Similarly, the sample frequency domain feature matrix after normalizing the features is defined as ;

[0068] Obtain any two of the first feature sets to obtain a first set and a second set, respectively obtain the features of the first set and the second set, and obtain a number of first features and a number of second features;

[0069] For example, the first set and respectively represent the set of vibration feature vectors of the motor in the normal state, and both belong to a sample in, where and are the th feature vectors corresponding to the first set A and the second set B respectively

[0070] Based on the first weight coefficient, obtain the Jaccard distance between the first feature and the second feature, and obtain a number of first Jaccard distances. Based on the maximum of the first Jaccard distances, obtain a first distance threshold. For example, calculate and of the Jaccard distance, obtain the maximum value of the Jaccard distance, and obtain the first distance threshold;

[0071] Obtain the vibration signal to be detected of the electrosurgical knife to be detected, extract the frequency domain features to be detected of the vibration signal to be detected, based on the frequency domain features to be detected, obtain a frequency domain feature matrix to be detected, and based on the frequency domain feature matrix to be detected, obtain a number of third sets. Each of the third sets contains a number of the frequency domain features to be detected after normalization processing; the processing method of the frequency domain feature matrix to be detected is the same as that of the sample frequency domain feature matrix;

[0072] Respectively obtain the features of the third set and the first feature set, and obtain a number of third features and a number of fourth features;

[0073] Based on the second weight coefficient, obtain the Jaccard distance between the third feature and the fourth feature to obtain a number of second Jaccard distances, and based on the maximum of the second Jaccard distances, obtain the first distance to be detected;

[0074] Based on the first distance threshold and the first distance to be detected, obtain the first detection result of the motor of the electrosurgical knife to be detected. That is, if the first distance to be detected is greater than the first distance threshold, it is determined that the motor of the electrosurgical knife to be detected is abnormal, and if not, it is normal.

[0075] Among them, the specific steps for obtaining the sample frequency domain feature matrix include: normalizing the sample frequency domain features, and based on the normalized sample frequency domain features, obtaining the sample frequency domain feature matrix;

[0076] The specific steps for obtaining the frequency domain feature matrix to be detected include: normalizing the frequency domain features to be detected, and based on the normalized frequency domain features to be detected, obtaining the frequency domain feature matrix to be detected.

[0077] For example, use min-max normalization to process the sample frequency domain feature matrix and the frequency domain feature matrix to be detected.

[0078] Embodiment 2

[0079] On the basis of Embodiment 1, in this embodiment, the method further includes:

[0080] Extract the sample time domain features of each motor vibration signal sample, obtain the sample time domain feature matrix based on the sample time domain features, and obtain a number of second feature sets based on the sample time domain feature matrix. Each of the second feature sets includes a number of normalized sample time domain features; in this embodiment, the sample time domain features may include mean, standard deviation, root mean square, peak value, peak-to-peak value, effective value, skewness, peak factor, and kurtosis index, etc.;

[0081] Obtain any two of the second feature sets to obtain a fourth set and a fifth set, respectively obtain the features of the fourth set and the fifth set, and obtain a number of fifth features and a number of sixth features;

[0082] Based on the third weight coefficient, obtain the Jaccard distance between the fifth feature and the sixth feature to obtain a number of third Jaccard distances, and based on the maximum of the third Jaccard distances, obtain the second distance threshold;

[0083] Extract the time domain features to be detected of the vibration signal to be detected, obtain the time domain feature matrix to be detected based on the time domain features to be detected, and obtain a number of sixth sets based on the time domain feature matrix to be detected. Each of the sixth sets includes a number of normalized time domain features to be detected;

[0084] Obtain the features of the sixth set and the second feature set respectively to obtain a number of seventh features and a number of eighth features; based on the fourth weight coefficient, obtain the Jaccard distance between the seventh feature and the eighth feature to obtain a number of fourth Jaccard distances, and based on the maximum of the fourth Jaccard distances, obtain the second distance to be detected.

[0085] Based on the second distance threshold and the second distance to be detected, obtain the second detection result of the motor of the electrosurgical knife to be detected; that is, if the second distance to be detected is greater than the second distance threshold, it is determined that the motor of the electrosurgical knife to be detected is abnormal, otherwise it is normal.

[0086] Based on the first detection result and the second detection result, obtain the third detection result of the motor, and update the first detection result of the motor to the third detection result. That is, if both detection results are normal, it is determined that the motor of the electrosurgical knife to be detected is normal, otherwise it is abnormal.

[0087] Among them, the specific steps for obtaining the sample time-domain feature matrix include: normalizing the sample time-domain features, and obtaining the sample time-domain feature matrix based on the normalized sample time-domain features.

[0088] The specific steps for obtaining the time-domain feature matrix to be detected include: normalizing the time-domain features to be detected, and obtaining the time-domain feature matrix to be detected based on the normalized time-domain features to be detected.

[0089] In this embodiment, the manner of obtaining the second detection result of the motor is the same as that of the first detection result of the motor in Embodiment 1, except that the former is the time-domain feature and the latter is the frequency-domain feature.

[0090] Embodiment 3

[0091] On the basis of the above implementation, in this embodiment, the obtaining methods of the first weight coefficient, the second weight coefficient, the third weight coefficient, and the fourth weight coefficient are all:

[0092] Obtain the first mean of the features, and obtain the first variance based on the first mean;

[0093] Obtain the first covariance of any two features, and obtain the first correlation coefficient based on the first covariance and the first variance;

[0094] Based on the first correlation coefficient and the first variance, obtain the weight coefficient of each feature.

[0095] Embodiment 4

[0096] On the basis of the above implementation, in this embodiment, the calculation formula for normalizing the sample frequency-domain features is:

[0097] Among them, represents the th sample frequency domain feature after normalization, represents the th sample frequency domain feature before normalization, and respectively represent the minimum value and the maximum value.

[0098] The calculation formula of the first Jaccard distance is:

[0099] Among them, represents the first Jaccard distance between the first feature set A and the first feature set B in represents the first weight coefficient of the sample frequency domain feature , represents the th feature of the first feature set A, represents the th feature of the first feature set B, represents the number of features in the first feature set, represents the feature and the feature the number of intersection elements, represents the feature and the feature the number of union elements.

[0100] The calculation formula of the first correlation coefficient is:

[0101] Among them, represents the first correlation coefficient of the sample frequency domain feature , represents the covariance of the sample frequency domain feature and the sample frequency domain feature , and respectively represent the mean square deviations of the sample frequency domain feature and the sample frequency domain feature , represents the th sample frequency domain feature after normalization, represents an integer greater than or equal to 1.

[0102] The calculation formula of the first weight coefficient is:

[0103] Among them, Represents the frequency-domain features of the sample The first weight coefficient of and respectively represent the frequency-domain features of the sample and the frequency-domain features of the sample The mean square error of represents the number of features in the first feature set.

[0104] In this formula, the denominator part comprehensively considers and normalizes the information content of each feature by summing the products of the variances of all features and (1 - the square of the correlation coefficient), making the sum of weights equal to 1.

[0105] Embodiment 5

[0106] Based on the above implementation, in this embodiment, the method further includes:

[0107] Classify the motor vibration signal samples, for example, classify them according to the surgical category, obtain several classified sample signals, and obtain the category of the vibration signal to be detected;

[0108] Based on the category, obtain the audio signals corresponding to the classified sample signals and the vibration signal to be detected, respectively obtain the first audio signal and the second audio signal, and obtain the abnormal audio signal based on the first audio signal and the second audio signal. For example, compare according to the frequency, pitch, and volume of the sound to obtain the audio signal outside the preset range, thereby obtaining the abnormal audio signal;

[0109] Based on the second audio signal and the classified sample signals, respectively obtain the first abnormal vibration signal corresponding to the abnormal audio signal and the second abnormal vibration signal corresponding to the classified sample signals;

[0110] Respectively obtain the first waveform diagram and the second waveform diagram of the first abnormal vibration signal and the second abnormal vibration signal, obtain the abnormal similarity of the first waveform diagram and the second waveform diagram, obtain the abnormal value based on the abnormal similarity, and obtain the usage duration based on the abnormal value and the category. For example, preset the available time of the electrosurgical unit under different abnormal values and different surgical categories, and thus obtain the available time of the current electrosurgical unit according to the current abnormal value.

[0111] Among them, the specific steps for obtaining the abnormal similarity include:

[0112] Obtain the extreme values of the second waveform diagram, adjust the first waveform diagram based on the extreme values to obtain the third waveform diagram. For example, according to the extreme values of the second waveform diagram, adjust the extreme values of the first waveform diagram to the same value;

[0113] Based on a preset period, obtain the first data and the second data of the second waveform diagram and the third waveform diagram respectively, such as obtaining the first data and the second data within one waveform period, and obtain the waveform similarity based on the fluctuation directions of the first data and the second data. For example, judge whether the fluctuation direction is upward or downward according to the numerical value, and then obtain the tangent point according to the derivative of the numerical value, so as to calculate the slope of the waveform according to the numerical value of the tangent point. If the fluctuation directions are different, the waveform similarity can be directly defined as -1. If they are the same, the waveform similarity can be obtained according to the slope;

[0114] Obtain a ratio value based on the numerical values of the first data and the second data, such as ratio value = first data / second data;

[0115] Obtain the anomaly similarity based on the waveform similarity and the ratio value. For example, assign different weights to them and obtain the anomaly similarity after adding them up.

[0116] Although the preferred embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications to these embodiments once they learn the basic creative concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications falling within the scope of the present invention.

[0117] 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 fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these changes and modifications.

Claims

1. A method for detecting abnormality of an electric knife motor, characterized in that: The method comprises: Acquire a plurality of motor vibration signal samples of the electric knife, extract the sample frequency domain features of each of the motor vibration signal samples, obtain a sample frequency domain feature matrix based on the sample frequency domain features, obtain a plurality of first feature sets based on the sample frequency domain feature matrix, wherein the first feature sets all include a plurality of normalized sample frequency domain features, and the motor vibration signal samples include normal motor vibration signals and abnormal motor vibration signals; Obtain any two of the first feature sets to obtain a first set and a second set, obtain features of the first set and the second set respectively, and obtain a plurality of first features and a plurality of second features; Based on a first weight coefficient, obtaining a Jaccard distance between the first feature and the second feature, obtaining a plurality of first Jaccard distances, and obtaining a first distance threshold based on a maximum of the first Jaccard distances; Acquire a vibration signal to be detected of the electric knife to be detected, extract frequency domain features to be detected of the vibration signal to be detected, obtain a frequency domain feature matrix to be detected based on the frequency domain features to be detected, and obtain a plurality of third sets based on the frequency domain feature matrix to be detected, wherein the third sets all contain a plurality of normalized frequency domain features to be detected; Respectively acquiring features of the third set and the first feature set to obtain a plurality of third features and a plurality of fourth features; Based on the second weight coefficient, obtaining the Jaccard distance between the third feature and the fourth feature, obtaining a plurality of second Jaccard distances, and obtaining a first distance to be detected based on the maximum of the second Jaccard distances; Based on the first distance threshold and the first distance to be detected, obtaining a first detection result of the motor of the electric knife to be detected; The first weight coefficient and the second weight coefficient are both obtained by: obtaining a first mean of the feature, and obtaining a first variance based on the first mean; obtaining a first covariance of any two features, and obtaining a first correlation coefficient based on the first covariance and the first variance; and obtaining a weight coefficient for each feature based on the first correlation coefficient and the first variance.

2. The method for detecting abnormality of an electric knife motor according to claim 1, characterized in that: The method further comprises: Extracting sample time domain features of each motor vibration signal sample, obtaining a sample time domain feature matrix based on the sample time domain features, and obtaining a plurality of second feature sets based on the sample time domain feature matrix, wherein the second feature sets each include a plurality of normalized sample time domain features; Obtain any two of the second feature sets, obtain a fourth set and a fifth set, obtain features of the fourth set and the fifth set respectively, and obtain a plurality of fifth features and a plurality of sixth features; Based on the third weight coefficient, obtaining the Jaccard distance between the fifth feature and the sixth feature, obtaining a plurality of third Jaccard distances, and obtaining a second distance threshold based on the maximum third Jaccard distance; Extracting the time domain features to be detected of the vibration signal to be detected, obtaining a time domain feature matrix to be detected based on the time domain features to be detected, and obtaining a plurality of sixth sets based on the time domain feature matrix to be detected, wherein the sixth sets each include a plurality of normalized time domain features to be detected; Acquire the features of the sixth set and the second feature set respectively, and obtain a plurality of seventh features and a plurality of eighth features; acquire the Jaccard distance between the seventh feature and the eighth feature based on a fourth weight coefficient, and obtain a plurality of fourth Jaccard distances; and acquire a second distance to be detected based on the maximum fourth Jaccard distance; Based on the second distance threshold and the second distance to be detected, obtaining a second detection result of the motor of the electric knife to be detected; Based on the first detection result and the second detection result, obtaining a third detection result of the motor, and updating the first detection result of the motor to the third detection result of the motor; The third weight coefficient and the fourth weight coefficient are both obtained by: obtaining the first mean of the feature, and obtaining the first variance based on the first mean; obtaining the first covariance of any two features, and obtaining the first correlation coefficient based on the first covariance and the first variance; and obtaining the weight coefficient of each feature based on the first correlation coefficient and the first variance.

3. The method for detecting abnormality of an electric knife motor according to claim 2, characterized in that: The specific steps of obtaining the sample frequency domain feature matrix include: Normalizing the sample frequency domain features, and obtaining the sample frequency domain feature matrix based on the normalized sample frequency domain features; The specific steps of obtaining the frequency domain feature matrix to be detected include: normalizing the frequency domain features to be detected, and obtaining the frequency domain feature matrix to be detected based on the normalized frequency domain features to be detected; The specific steps of obtaining the sample time domain feature matrix include: normalizing the sample time domain features, and obtaining the sample time domain feature matrix based on the normalized sample time domain features; The specific steps of obtaining the time domain feature matrix to be detected include: normalizing the time domain features to be detected, and obtaining the time domain feature matrix to be detected based on the normalized time domain features to be detected.

4. The method for detecting abnormality of an electric knife motor according to claim 3, characterized in that: The calculation formula for normalizing the sample frequency domain characteristics is: in, Indicates The normalized sample frequency domain features, Indicates The frequency domain features of samples before normalization, and Respectively The minimum and maximum values ​​of .

5. The method for detecting abnormality of an electric knife motor according to claim 4, characterized in that: The calculation formula of the first Jaccard distance is: in, express The first Jaccard distance between the first feature set A and the first feature set B in , Represents the sample frequency domain characteristics The first weight coefficient of The first feature set A represents Features, represents the first feature set B Features, represents the number of features in the first feature set, Representation characteristics With features The number of intersection elements, Representation characteristics With features The number of elements in the union of .

6. The method for detecting abnormality of an electric knife motor according to claim 5, characterized in that: The calculation formula of the first correlation coefficient is: in, Represents the sample frequency domain characteristics The first correlation coefficient of Represents the sample frequency domain characteristics And the sample frequency domain characteristics The covariance of and Represent the sample frequency domain characteristics And the sample frequency domain characteristics The mean square error, Indicates The normalized sample frequency domain features, Represents an integer greater than or equal to 1.

7. The method for detecting abnormality of an electric knife motor according to claim 6, characterized in that: The calculation formula of the first weight coefficient is: in, Represents the sample frequency domain characteristics The first weight coefficient of and Represent the sample frequency domain characteristics And the sample frequency domain characteristics The mean square error, Indicates the number of features in the first feature set.

8. The method for detecting abnormality of an electric knife motor according to claim 3, characterized in that: The method further comprises: Classifying the motor vibration signal samples to obtain a number of classified sample signals, and obtaining the category of the vibration signal to be detected; Based on the category, obtaining audio signals corresponding to the classified sample signal and the vibration signal to be detected, respectively obtaining a first audio signal and a second audio signal, and obtaining an abnormal audio signal based on the first audio signal and the second audio signal; Based on the second audio signal and the classified sample signal, respectively acquiring a first abnormal vibration signal corresponding to the abnormal audio signal and a second abnormal vibration signal corresponding to the classified sample signal; The first waveform graph and the second waveform graph of the first abnormal vibration signal and the second abnormal vibration signal are respectively obtained, the abnormal similarity of the first waveform graph and the second waveform graph is obtained, the abnormal value is obtained based on the abnormal similarity, and the usage time is obtained based on the abnormal value and the category.

Images