A method for online data measurement and monitoring of anchor drilling rig
By calculating the temperature, torque, speed and acceleration data of the anchor drill rig, dynamically determine the number of nearest neighbors, combined with the abnormal detection algorithm, the problem of low efficiency and large error in the monitoring of anchor drill rigs is solved, achieving more efficient and accurate abnormal detection.
Patent Information
- Application Number
- CN202411088156.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-09
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-08-09
AI Technical Summary
The LOF abnormality detection algorithm needs to determine the nearest neighbor number k in the anchor drilling rig monitoring, resulting in low efficiency or large error, and the fixed k value cannot adapt to the variable working environment and state of the anchor drilling rig.
By collecting the temperature, torque, speed and acceleration data of the anchor drilling rig, the degree of change, correlation and abnormal coefficients are calculated, the number of nearest neighbors is dynamically determined, and the abnormality detection algorithm is used for monitoring.
It improves the accuracy and efficiency of abnormal detection of anchor drilling rigs, adapts to the changing working environment and state, and enhances the reliability of abnormal judgment.
Smart Images

Figure CN118643446B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of anchor drilling rig monitoring, and in particular to an online data measurement and monitoring method for an anchor drilling rig. Background Art
[0002] In the field of modern mining, tunnel construction and geotechnical engineering, anchor drilling rigs, as a key construction equipment, can not only improve the stability of underground projects, but also effectively prevent safety accidents such as landslides and landslides, and protect the lives of workers. Online data measurement and monitoring can monitor the operating status of equipment in real time, predict maintenance needs, and optimize operating efficiency. It has become a key technical means to improve project quality and safety.
[0003] The LOF anomaly detection algorithm is often used to detect anomalies in monitoring data when the bolter is working. However, in LOF anomaly detection, the number of nearest neighbors k needs to be determined in advance. When the k value is set too large, it can provide a global perspective and help enhance the ability to capture anomalies, but it will make the efficiency of anomaly detection too slow, which is not conducive to real-time monitoring of the bolter's working process; when the k value is small, although the anomaly detection speed is improved, it is more susceptible to noise, resulting in detection errors; and because the working environment and working state of the bolter are changeable, the use of a fixed k value is not conducive to real-time anomaly detection during the working process of the bolter. Summary of the invention
[0004] In order to solve the above technical problems, the present application provides an online data measurement and monitoring method for an anchor drilling rig to solve the existing problems.
[0005] The online data measurement and monitoring method of an anchor drilling rig of the present application adopts the following technical solution:
[0006] An embodiment of the present application provides an online data measurement and monitoring method for an anchor drilling rig, the method comprising the following steps:
[0007] The temperature, torque, speed and acceleration at each moment when the anchor drilling rig is working are collected, and all the normalized temperatures, torques, speeds and accelerations before each moment are respectively formed into a temperature sequence, a torque sequence, a speed sequence and an acceleration sequence;
[0008] Based on the change trend between the first and last elements in the trend sequence of the temperature sequence, determine the degree of change of the temperature sequence; analyze the correlation between the torque sequence and the speed sequence to determine the first correlation at each moment; analyze the correlation between the change trend between all adjacent elements in the trend sequence of the temperature sequence and the speed sequence to determine the second correlation at each moment;
[0009] Determine the temperature anomaly coefficient at each moment by combining the degree of change, the first correlation, and the second correlation; analyze the autocorrelation of the acceleration sequence, determine the central tendency amount and the degree of fit of the acceleration sequence, and obtain the vibration anomaly coefficient at each moment by combining the temperature anomaly coefficient;
[0010] Based on the difference between the vibration anomaly coefficient at each moment and the vibration anomaly coefficient at the historical moment, the number of nearest neighbors at each moment is determined, and the bolter rig is monitored during operation using the anomaly detection algorithm.
[0011] In one embodiment, the method for determining the degree of change is:
[0012] Each element in the trend sequence of the temperature sequence is used as the ordinate and the serial number of each element is used as the abscissa, and each element in the trend sequence is converted into each coordinate point. The degree of change is the slope of the straight line determined by the first coordinate point and the last coordinate point in the trend sequence.
[0013] In one embodiment, the first correlation is a Pearson correlation coefficient between a torque sequence and a rotation speed sequence.
[0014] In one embodiment, the second correlation is determined as follows:
[0015] The slopes of all the straight lines between two adjacent coordinate points in the trend sequence of the temperature sequence are determined to form a slope sequence, and the second correlation is the Pearson correlation coefficient between the slope sequence and the rotation speed sequence.
[0016] In one embodiment, the process of determining the temperature anomaly coefficient is as follows:
[0017] Calculating the product of the first correlation and the second correlation, and determining the temperature anomaly coefficient based on the product and the degree of change;
[0018] The temperature anomaly coefficient is negatively correlated with the product, and positively correlated with the degree of change.
[0019] In one embodiment, determining the amount of central tendency includes:
[0020] The autocorrelation function is used to calculate the autocorrelation coefficient of the acceleration sequence at different lags, and the autocorrelation coefficient greater than a preset threshold is recorded as a significant peak. The central tendency value is the mean of all significant peaks of the acceleration sequence.
[0021] In one embodiment, the fit measure determination includes:
[0022] The hysteresis values corresponding to all significant peaks are combined into a hysteresis value sequence, and each element in the hysteresis value sequence is used as the ordinate and the sequence number of each element is used as the abscissa to obtain a fitting curve corresponding to the hysteresis value sequence. The fitting degree measure is the determination coefficient of the fitting curve.
[0023] In one embodiment, the determination of the vibration abnormality coefficient includes:
[0024] The multiplication result of the central tendency quantity and the fitting degree metric is obtained, and the vibration abnormality coefficient is the ratio of the temperature abnormality coefficient to the multiplication result.
[0025] In one embodiment, the number of nearest neighbors is calculated as follows:
[0026] ; In the formula, for The number of nearest neighbors at time, , They are time, The vibration anomaly coefficient at the moment, for The set of all historical moments before the moment, is the preset benchmark nearest neighbor number, is the maximum value function, is the floor function.
[0027] In one embodiment, the use of an anomaly detection algorithm to monitor the bolter during operation includes:
[0028] The temperature, torque, speed, and acceleration at each moment are combined into the feature vector of each moment. The feature vectors of the current moment and all previous moments and the number of nearest neighbors at the current moment are used as the input of the anomaly detection algorithm. The output is the anomaly value at each moment. The segmentation threshold of the anomaly values at all moments is obtained. If the anomaly value at the current moment is greater than the segmentation threshold, it is determined that the anchor drill is abnormal. Otherwise, the anchor drill is normal.
[0029] This application has at least the following beneficial effects:
[0030] The present application collects the temperature, torque, speed, and acceleration at each moment when the anchor drilling rig is working, and respectively forms a temperature sequence, a torque sequence, a speed sequence, and an acceleration sequence with all the normalized temperatures, torques, speeds, and accelerations before each moment; determines the degree of change of the temperature sequence based on the change trend between the first and last elements in the trend sequence of the temperature sequence; reflects the overall change trend of the temperature when the anchor drilling rig is working; analyzes the correlation between the torque sequence and the speed sequence, determines the first correlation at each moment, and reflects the correlation between the torque and the speed when the anchor drilling rig is working; analyzes the correlation between the change trend between all adjacent elements in the trend sequence of the temperature sequence and the speed sequence, and determines the second correlation at each moment, and reflects the correlation between the temperature and the speed when the anchor drilling rig is working; combines the degree of change, the first correlation, and the second correlation The temperature anomaly coefficient at each moment is determined by analyzing the correlation, which reflects the abnormal degree of temperature at each moment when the anchor drill is working, and improves the reliability of anchor drill anomaly detection; the autocorrelation relationship of the acceleration sequence is analyzed to determine the central tendency amount and the fitting degree measurement of the acceleration sequence; combined with the temperature anomaly coefficient, the vibration anomaly coefficient at each moment is obtained, which reflects the abnormal degree of vibration at each moment when the anchor drill is working, and further enhances the accuracy of anchor drill anomaly judgment; finally, based on the difference between the vibration anomaly coefficient at each moment and the maximum value of the vibration anomaly coefficient at the historical moment, the number of nearest neighbors at each moment is determined, so that the determination of the number of nearest neighbors is combined with the abnormal degree of the anchor drill when working, and the accuracy of the determination of the number of nearest neighbors is improved; the anomaly detection algorithm is used to monitor the anchor drill when working; the efficiency and accuracy of anchor drill anomaly detection are improved. BRIEF DESCRIPTION OF THE DRAWINGS
[0031] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0032] Figure 1 A flowchart of the steps of an online data measurement and monitoring method for an anchor drilling rig provided in the present application;
[0033] Figure 2 Flowchart for determining the number of nearest neighbors. DETAILED DESCRIPTION
[0034] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following is a detailed description of the online data measurement and monitoring method of an anchor drilling rig proposed in accordance with the present application, its specific implementation method, structure, features and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.
[0035] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.
[0036] The specific scheme of the online data measurement and monitoring method of the anchor drilling rig provided by the present application is described in detail below with reference to the accompanying drawings.
[0037] An embodiment of the present application provides an online data measurement and monitoring method for an anchor drilling rig. Specifically, the following online data measurement and monitoring method for an anchor drilling rig is provided. Figure 1 , the method comprises the following steps:
[0038] Step S001, collecting the temperature, torque, speed, and acceleration at each moment when the anchor drilling rig is working, and forming a temperature sequence, a torque sequence, a speed sequence, and an acceleration sequence with all the normalized temperatures, torques, speeds, and accelerations before each moment.
[0039] A temperature sensor is installed on the gearbox housing of the anchor drilling rig to collect temperature data when the anchor drilling rig is working; a torque sensor is installed between the motor output end and the reducer input end on the main drive shaft of the anchor drilling rig to collect torque data when the anchor drilling rig is working; a rotary encoder is installed close to the torque sensor to collect speed data when the anchor drilling rig is working; an acceleration sensor is installed on the main drive shaft of the anchor drilling rig to collect acceleration data when the anchor drilling rig is working.
[0040] It should be noted that vibration is essentially a phenomenon in which an object moves back and forth near a balanced position. This movement can be described as a change in acceleration, so that the acceleration data and the vibration data have consistent changes, which can reflect the vibration changes of the anchor drill when working. Temperature, torque, speed, and acceleration are collected once a minute. In chronological order, the data collected before each moment are respectively formed into a data sequence corresponding to each moment. Each data sequence is normalized to the maximum and minimum values. The normalized sequences are recorded as temperature sequence X, torque sequence L, speed sequence V, and acceleration sequence H.
[0041] In this embodiment, taking the temperature sequence [35, 39, 43, 48, 52, 58, 67, 72, 75, 78] as an example, the specific calculation formula for the maximum-minimum value normalization process is:
[0042] ; In the formula, is the sth temperature value in the temperature sequence, is the sth temperature value in the data sequence corresponding to the temperature, mi is the minimum temperature value in the data sequence corresponding to the temperature, and ma is the maximum temperature value in the data sequence corresponding to the temperature.
[0043] The temperature sequence after maximum-minimum normalization is [0, 0.093, 0.186, 0.291, 0.395, 0.535, 0.791, 0.93, 0.988, 1].
[0044] The maximum-minimum value normalization is a known technology, and the specific process will not be described in detail. The implementer can select other feasible normalization processing algorithms according to the actual situation, and this application does not limit it. Among them, the implementer can set the time interval of data collection according to the actual situation, and this application does not limit it.
[0045] It should be noted that, in order to determine the accuracy of subsequent anchor drilling rig anomaly detection, this embodiment performs data analysis 10 minutes after the temperature, torque, rotation speed, and acceleration are collected. The implementer can determine the time of data collection by himself.
[0046] Step S002, based on the change trend between the first and last elements in the trend sequence of the temperature sequence, determine the degree of change of the temperature sequence; analyze the correlation between the torque sequence and the speed sequence to determine the first correlation at each moment; analyze the correlation between the change trend between all adjacent elements in the trend sequence of the temperature sequence and the speed sequence to determine the second correlation at each moment.
[0047] As a mechanical system, the anchor drilling rig has interactions between its internal parts, so there is also a correlation between the data. Changes in one type of data may affect other data. Take the temperature anomaly of the anchor drilling rig as an example. Under normal circumstances, the temperature of the anchor drilling rig increases with the increase of working time during operation. When the anchor drilling rig is overloaded or lacks lubrication, it will also cause an increase in temperature. If the anchor drilling rig needs to be shut down for rest and heat dissipation only based on the overheating of the temperature, the anomalies of overload or lack of lubrication will be ignored. If these anomalies are not handled in time, they will continue to accumulate and eventually cause more serious faults such as parts damage and mechanical damage. Therefore, it is necessary to determine in advance whether the temperature anomaly is caused by long-term work or due to overload or lack of lubrication before causing serious faults. The anomaly should be discovered and checked in time before the parts are damaged, so as to avoid unnecessary wear of the parts and even losses caused by repairs.
[0048] First of all, under normal circumstances, during the operation of the anchor drilling rig, the temperature will rise due to friction, power consumption, normal mechanical collision, and increased torque, which increases the friction between mechanical parts. This temperature rise accumulates with the working time and becomes larger and larger. When the anchor drilling rig is overloaded or lacks lubrication, in addition to the temperature rise caused by normal operation, the abnormal increase in torque will also lead to increased contact pressure between mechanical parts, deformation of the transmission system, and wear. These problems will increase the energy loss during the operation of the anchor drilling rig, and the lost energy will be converted into heat, which will further lead to a temperature rise. Therefore, when an abnormal situation occurs, the temperature rise trend is more intense than the temperature rise trend in normal situations.
[0049] It should be noted that although there is a positive correlation between the change in torque and temperature, since both normal and abnormal changes in torque will lead to an increase in temperature, it is impossible to determine whether the temperature is abnormally increased based on the correlation between temperature and torque.
[0050] Secondly, for the anchor drill when it is working, the smaller the speed, the greater the torque. The increase in torque causes the temperature to rise, which makes the temperature increase faster; conversely, the larger the speed, the smaller the torque. The smaller the torque, the smaller the friction between the machines, thereby reducing the heat generated by friction, which makes the temperature increase rate relatively slower.
[0051] When the anchor drill is overloaded or lacks lubrication, the torque will increase abnormally. In order to maintain power conservation, the speed should theoretically be reduced rapidly and significantly. However, due to the inertia factor in the operation of the mechanical device, the speed adjustment requires a certain reaction time, and it is difficult to reduce it significantly in a short time, which makes the speed change lag; and the increase in deformation and wear of the transmission system leads to energy loss, which will also affect the change in speed, making the lag of the speed change more obvious than the torque change, and the degree of decline is smaller, thereby weakening the negative relationship between speed and torque, and the temperature will increase faster at this time.
[0052] Based on the above analysis, the temperature sequence is used as input, and the STL (Seasonal and Trend decomposition using Loess) time series decomposition algorithm is used to output the trend sequence of the temperature sequence. As a well-known technology, the specific process of the STL time series decomposition algorithm will not be repeated. The trend sequence reflects the overall trend of temperature change. The changes between adjacent elements in the trend sequence can reflect the local growth rate of temperature. Taking the elements in the trend sequence as the ordinate and the sequence number of the element in the sequence as the abscissa, a series of coordinate points can be obtained, and the slope of the straight line determined between two adjacent coordinate points in the trend sequence is calculated. All the slopes obtained are formed into a slope sequence according to the abscissa order of the coordinate points. Each element in the slope sequence can reflect the local growth rate of temperature.
[0053] In this embodiment, the steps of obtaining the trend sequence of the temperature series by using the STL time series decomposition algorithm are as follows:
[0054] ; is the sth temperature value in the temperature sequence, is the trend value of the sth temperature value in the temperature sequence, is the seasonal term value of the sth temperature value in the temperature series, is the residual value of the sth temperature value in the temperature sequence.
[0055] The trend values of all temperature values in the temperature sequence are arranged in the order of the corresponding temperature values to form a trend sequence of the temperature sequence.
[0056] Calculate the slope of the straight line between the first coordinate point and the last coordinate point in the trend sequence, and record it as the degree of change of the temperature sequence corresponding to the trend sequence; calculate the Pearson correlation coefficient between the torque sequence and the speed sequence at each moment as the first correlation at each moment, and calculate the Pearson correlation coefficient between the slope sequence and the speed sequence at each moment as the second correlation at each moment.
[0057] In this embodiment, the first correlation is calculated as follows:
[0058] ; In the formula, is the first correlation at time t, L is the torque sequence at time t, V is the speed sequence at time t, cov() is the function for calculating covariance, is the standard deviation of the torque series at time t, is the standard deviation of the speed series at time t.
[0059] The calculation of the covariance, standard deviation and Pearson correlation coefficient are all existing well-known technologies, and will not be described in detail in this embodiment.
[0060] The same calculation method as the first correlation is used to obtain the second correlation at each moment for the slope sequence and the speed sequence.
[0061] Step S003, combining the degree of change of the temperature sequence, the first correlation, and the second correlation, determining the temperature anomaly coefficient at each moment; analyzing the autocorrelation of the acceleration sequence to determine the central tendency amount and the degree of fitting measure of the acceleration sequence; combining the temperature anomaly coefficient to obtain the vibration anomaly coefficient at each moment.
[0062] The temperature anomaly coefficient at each moment is determined by combining the degree of change of the temperature sequence, the first correlation, and the second correlation, wherein the temperature anomaly coefficient is negatively correlated with the product of the first correlation and the second correlation, and is positively correlated with the degree of change.
[0063] It should be understood that a positive correlation means that the dependent variable will increase as the independent variable increases, and the dependent variable will decrease as the independent variable decreases. A negative correlation means that the dependent variable will decrease as the independent variable increases, and the dependent variable will increase as the independent variable decreases.
[0064] In this embodiment, the calculation method of the temperature anomaly coefficient at each moment can be:
[0065] :In the formula, is the temperature anomaly coefficient at time t, g is the degree of change of the temperature series at time t, is the first correlation at time t, is the second correlation at time t, and e is a natural constant.
[0066] In another embodiment, the temperature anomaly coefficient at each moment may be calculated as follows:
[0067] :In the formula, is a constant greater than 0 to avoid the denominator being 0. The implementer can set it according to the actual situation, and this embodiment does not limit it.
[0068] As the heat accumulation of the anchor drill increases gradually during operation, the more drastic the temperature change, the greater the change in the temperature sequence, that is, the greater the g value. When the anchor drill is normal, the temperature change is weaker than that in abnormal conditions, so the g value is relatively small; and under normal conditions, there is a negative correlation between torque and speed, and there is also a negative correlation between torque and temperature growth rate, that is, there is a negative correlation between the torque sequence and the slope sequence, so , are close to -1, so The larger the value, the smaller the temperature anomaly coefficient. Conversely, the larger the temperature anomaly coefficient.
[0069] Although the relationship between temperature, torque, and speed under normal circumstances is different from the relationship between the three when the anchor drill rig is overloaded or underlubricated, it is still difficult to accurately determine whether there is an abnormality in the anchor drill rig. Under normal circumstances, when the anchor drill rig is working, the entire mechanical system is relatively stable, that is, the vibration signal of the anchor drill rig is periodic and has small up and down fluctuations. When the anchor drill rig is overloaded or underlubricated, the abnormal increase in torque will cause uneven force on mechanical parts and unbalanced drive system, which will intensify the vibration of the drill rig, thereby affecting the periodic changes in the vibration signal, and the abnormal torque will also cause abnormal temperature fluctuations.
[0070] Therefore, based on the above analysis, an autocorrelation function is used to obtain a series of autocorrelation coefficients of the acceleration sequence. Each autocorrelation coefficient corresponds to a different lag, and the difference between adjacent lags is 1. In this embodiment, the lag range is [1,720]. The implementer can set it according to the actual situation, and this embodiment does not limit it.
[0071] For all the autocorrelation coefficients obtained, a rectangular coordinate system is established with the autocorrelation coefficient as the ordinate and the lag amount corresponding to the autocorrelation coefficient as the abscissa. Each data point in the rectangular coordinate system is greater than the preset threshold. The autocorrelation coefficient is recorded as a significant peak. The implementer can set it according to the actual situation, and this embodiment does not limit it here.
[0072] Based on the calculation of the autocorrelation coefficient, it can be determined whether the acceleration sequence has periodic changes. The periodicity can be determined based on the size of the significant peak and the time when the significant peak appears. If the acceleration sequence has periodicity, a significant peak will appear at a fixed and repeatable time interval.
[0073] Then, the hysteresis corresponding to all significant peaks is used to form a hysteresis sequence in the order in which the significant peaks appear. Then, the elements in the hysteresis sequence are used as the ordinate and the sequence number of the elements is used as the abscissa, and the least square method is used to perform straight line fitting on all elements in the hysteresis sequence to obtain a fitting straight line. Among them, the least square method is an existing well-known technology, which is not described in detail in this embodiment, and the implementer can select other existing feasible straight line fitting algorithms according to actual conditions, which is not limited in this embodiment.
[0074] Calculate the vibration abnormality coefficient at each moment. The specific calculation method in this embodiment is:
[0075] : is the vibration anomaly coefficient at time t, is the temperature anomaly coefficient at time t, is the central tendency of the acceleration series at time t, It is the measure of the degree of fit of the acceleration sequence at time t. In this embodiment, the central tendency quantity is the mean of all significant peaks of the acceleration sequence at time t; and the measure of the degree of fit is the coefficient of determination of the fitting line corresponding to time t.
[0076] In this embodiment, the calculation process of the central tendency of the acceleration sequence is:
[0077] ; In the formula, is the central tendency of the acceleration series at time t, is the dth significant peak of the acceleration sequence at time t, and D is the number of significant peaks in the acceleration sequence at time t.
[0078] In this embodiment, the calculation process of the fitting degree measure of the acceleration sequence is:
[0079] ; In the formula, It is a measure of the degree of fit of the acceleration series at time t. SSR is the regression sum of squares of all data points of the fitting line corresponding to time t, and SSE is the residual sum of squares of all data points of the fitting line corresponding to time t.
[0080] The calculation of the determination coefficient of the fitting straight line is a known technique and will not be described in detail in this embodiment.
[0081] When the bolter is normal, the smaller the temperature anomaly coefficient is; the vibration signal has periodicity, the closer the significant peak value of the vibration signal is to 1, the better the linear fitting degree of the lag sequence is, that is, , Therefore, when the anchor drilling rig is normal, the vibration abnormality coefficient is smaller. On the contrary, the vibration abnormality coefficient is larger.
[0082] Step S004, based on the difference between the vibration abnormality coefficient at each moment and the maximum value of the vibration abnormality coefficient at the historical moment, the number of nearest neighbors at each moment is determined, and the bolter rig is monitored during operation using an abnormality detection algorithm.
[0083] Based on the difference between the vibration anomaly coefficient at each moment and the maximum value of the vibration anomaly coefficient at the historical moment, the number of nearest neighbors at each moment is determined. In this embodiment, the specific calculation method of the number of nearest neighbors is:
[0084] ; In the formula, for The number of nearest neighbors at time, , They are time, The vibration anomaly coefficient at the moment, for The set of all historical moments before the moment, is the preset benchmark nearest neighbor number. In this embodiment , the implementer can set it according to the actual situation, and this embodiment does not limit it; is the maximum value function, is a floor rounding function. The flowchart for determining the number of nearest neighbors is as follows Figure 2 shown.
[0085] The temperature, torque, speed, and acceleration data collected at each moment constitute the feature vector of each moment, and the feature vectors of the current moment and all previous moments, and the number of nearest neighbors at the current moment are used as the input of the LOF (Local Outliers Factor) anomaly detection algorithm, and the output is the anomaly value at each moment. Then all the anomaly values are used as input, and the Otsu threshold segmentation algorithm is used to output the segmentation threshold. If the anomaly value at the current moment is greater than the segmentation threshold, it is determined that the anchor drilling rig is abnormal, otherwise, the anchor drilling rig is determined to be normal. The LOF anomaly detection algorithm and the Otsu threshold segmentation are both existing well-known technologies, and the specific process will not be repeated, and the implementer can choose other existing feasible threshold segmentation algorithms according to actual conditions, and this embodiment is not limited here.
[0086] By adjusting the number of nearest neighbors in the LOF anomaly detection algorithm in real time, the possibility of anomalies at the current moment can be judged by the time series changes and correlations of the collected data. When the possibility is greater, a larger number of nearest neighbors is used to enhance the ability to capture global anomalies and improve the accuracy of anomaly judgment; when the possibility is smaller, a smaller number of nearest neighbors is used to improve the efficiency of anomaly detection. Since the number of nearest neighbors is adaptively determined based on the time series changes and correlations of the collected data, the limitations of setting a fixed number of nearest neighbors are avoided; adaptively setting the number of nearest neighbors also helps to adapt to the changing working environment and working state of the anchor drill rig and achieve flexible anomaly detection.
[0087] It should be noted that the above sequence of the embodiments of the present application is for description only and does not represent the advantages and disadvantages of the embodiments. The above is a description of a specific embodiment of this specification. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0088] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.
[0089] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Modifications to the technical solutions recorded in the aforementioned embodiments, or equivalent replacement of some of the technical features therein, do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.
Claims
1. An online data measurement and monitoring method for an anchor drilling rig, characterized in that: The method comprises the following steps: The temperature, torque, speed and acceleration at each moment when the anchor drilling rig is working are collected, and all the normalized temperatures, torques, speeds and accelerations before each moment are respectively formed into a temperature sequence, a torque sequence, a speed sequence and an acceleration sequence; Based on the change trend between the first and last elements in the trend sequence of the temperature sequence, determine the degree of change of the temperature sequence; analyze the correlation between the torque sequence and the speed sequence to determine the first correlation at each moment; analyze the correlation between the change trend between all adjacent elements in the trend sequence of the temperature sequence and the speed sequence to determine the second correlation at each moment; Determine the temperature anomaly coefficient at each moment by combining the degree of change, the first correlation, and the second correlation; analyze the autocorrelation of the acceleration sequence, determine the central tendency amount and the degree of fit of the acceleration sequence, and obtain the vibration anomaly coefficient at each moment by combining the temperature anomaly coefficient; Based on the difference between the vibration anomaly coefficient at each moment and the vibration anomaly coefficient at the historical moment, the number of nearest neighbors at each moment is determined, and the bolter rig is monitored during operation using an anomaly detection algorithm. The determination process of the temperature anomaly coefficient is as follows: Calculating the product of the first correlation and the second correlation, and determining the temperature anomaly coefficient based on the product and the degree of change; The temperature anomaly coefficient is negatively correlated with the product and positively correlated with the degree of change; The determination of the vibration abnormality coefficient includes: Obtaining a multiplication result of the central tendency quantity and the fitting degree measure, wherein the vibration anomaly coefficient is a ratio of the temperature anomaly coefficient to the multiplication result; The number of nearest neighbors is calculated as follows: ; In the formula, for The number of nearest neighbors at time, , They are time, The vibration anomaly coefficient at the moment, for The set of all historical moments before the moment, is the preset benchmark nearest neighbor number, is the maximum value function, is the floor function.
2. The online data measurement and monitoring method of an anchor drilling rig according to claim 1, characterized in that: The degree of change is determined by: Each element in the trend sequence of the temperature sequence is used as the ordinate and the sequence number of each element is used as the abscissa, and each element in the trend sequence is converted into each coordinate point. The degree of change is the slope of the straight line determined by the first coordinate point and the last coordinate point in the trend sequence.
3. The online data measurement and monitoring method of an anchor drilling rig according to claim 1, characterized in that: The first correlation is the Pearson correlation coefficient between the torque sequence and the speed sequence.
4. The online data measurement and monitoring method of an anchor drilling rig according to claim 2, characterized in that: The process of determining the second correlation is: The slopes of all the straight lines between two adjacent coordinate points in the trend sequence of the temperature sequence are determined to form a slope sequence, and the second correlation is the Pearson correlation coefficient between the slope sequence and the rotation speed sequence.
5. The online data measurement and monitoring method of an anchor drilling rig according to claim 1, characterized in that: The determination of the amount of central tendency includes: The autocorrelation function is used to calculate the autocorrelation coefficient of the acceleration sequence at different lags, and the autocorrelation coefficient greater than a preset threshold is recorded as a significant peak. The central tendency value is the mean of all significant peaks of the acceleration sequence.
6. The online data measurement and monitoring method of an anchor drilling rig according to claim 5, characterized in that: The fit measure determination includes: The hysteresis values corresponding to all significant peaks are grouped into a hysteresis value sequence, and each element in the hysteresis value sequence is used as the ordinate and the sequence number of each element is used as the abscissa to obtain a fitting curve corresponding to the hysteresis value sequence. The fitting degree measure is the determination coefficient of the fitting curve.
7. The online data measurement and monitoring method of an anchor drilling rig according to claim 1, characterized in that: The method of using an abnormality detection algorithm to monitor the bolter during operation includes: The temperature, torque, speed, and acceleration at each moment are combined into the feature vector of each moment. The feature vectors of the current moment and all previous moments and the number of nearest neighbors at the current moment are used as the input of the anomaly detection algorithm. The output is the anomaly value at each moment. The segmentation threshold of the anomaly values at all moments is obtained. If the anomaly value at the current moment is greater than the segmentation threshold, it is determined that the anchor drill is abnormal. Otherwise, the anchor drill is normal.
