Servo system data acquisition and analysis system

Abstract

The invention relates to the technical field of state monitoring, in particular to a servo system data acquisition and analysis system which comprises the following steps: a sampling identification module, a response extraction module, a behavior division module, a strain speculation module and an early warning trigger module. According to the method, the whole-process identification capability of the dynamic behavior of the servo system is enhanced by combining the control instruction and the multi-source response data, the identification path of the asymmetric behavior and the strain anomaly is optimized by utilizing the linkage analysis of the response offset, the phase drift and the amplitude difference trend, and a synchronous trend accumulation judgment mechanism is adopted, so that the dynamic behavior of the servo system is identified. Continuous labeling of structure disturbance evolution is achieved, segmentation accuracy and time sequence consistency of a key action process are improved, a periodic fluctuation direction judgment and continuous offset labeling strategy is used, the preposed range of perturbation state recognition is expanded, and the perturbation state recognition accuracy is improved. A servo execution state is expanded from a traditional single-point static judgment mode to a trend recognition mode based on a dynamic change process, and trend early warning of a servo system on abnormal behaviors is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of state monitoring, in particular to a servo system data acquisition and analysis system. BACKGROUND

[0002] The technical field of state monitoring includes systems and methods for real-time acquisition, analysis and evaluation of the operating state of industrial equipment, involving continuous tracking of physical quantities such as vibration, temperature, current, voltage, etc. of key components, and determining whether the equipment is in a normal, abnormal or fault state by setting thresholds or models, combining sensor deployment, data acquisition unit, signal preprocessing circuit, feature extraction unit, analysis unit, acquiring dynamic data during equipment operation to assist operators in realizing the cognition and early warning of equipment health status, and applying to multiple industries such as power, transportation, aviation and manufacturing, supporting equipment life cycle management and maintenance decision-making, and being an important foundation for realizing predictive maintenance and intelligent operation and maintenance. Among them, the servo system data acquisition and analysis system refers to the signals such as speed, displacement, acceleration and current related to the operating state of the servo system execution element, which are acquired by deploying multiple sensors including current sensors and angular displacement encoders, and then input to the analog-to-digital conversion unit in the data acquisition circuit for conversion, and then stored and classified by the micro-processing unit according to the fixed period. A multi-channel data input architecture is adopted to simultaneously acquire state information of different nodes of the servo driver, supplemented by a condition triggering mechanism based on numerical comparison and threshold discrimination to mark key data segments. After the acquired data is temporarily stored in the on-chip buffer, it is sent to the host computer through the serial bus for subsequent processing and evaluation by a fixed format structured analysis script to form a time series-based state change record. It specifically relates to embedded circuit design, parallel sampling mechanism, fixed period buffering and transmission scheduling, etc.

[0003] Traditional servo system data acquisition and analysis technology relies on single-point sampling and threshold judgment of speed and current signals, and limit value comparison based on instantaneous changes of component data in dynamic state monitoring. It lacks a joint modeling mechanism for hysteresis characteristics, response symmetry and multi-dimensional signal trend relationships in the signal change process. In the startup or state transition stage of the servo system, due to the non-synchronous time delay between the electric command and the mechanical response, there is a delay or misjudgment problem in state determination. Since the extreme value marking method does not introduce phase drift and amplitude difference dynamic characteristics, it lacks cumulative recognition ability for the evolution process of structural strain, and cannot effectively form a trend warning before the state reaches the abnormal threshold, making it difficult to mark the device perturbation state in advance or the trend judgment lags behind the actual load change. SUMMARY

[0004] To solve the technical problems existing in the prior art, the present application provides a servo system data acquisition and analysis system. The technical solution is as follows: In one aspect, a servo system data acquisition and analysis system is provided, the system comprising: The sampling recognition module calls the control signal data sequence, analyzes the control instruction trigger node before the driving execution, recognizes the starting point of the current rising section waveform and compares it with the motor speed response time point, identifies the hysteresis characteristics between the electric execution and the mechanical action, marks the state transition section, and obtains the transition judgment interval sequence; The response extraction module uses the transition judgment interval sequence to analyze the duration characteristics of the current rising and falling stages, calculates the waveform coverage area difference, constructs the correction proportion factor based on the offset direction, combines the current proportion of the peak section and the stable section, identifies the current response asymmetry behavior, and obtains the current offset correction configuration; The behavior division module uses the current offset correction configuration to analyze the curvature distribution formed by the angle change of the displacement response curve, identifies the change extreme value and calculates the distribution density, locates the speed direction reversal node and divides the displacement section, and obtains the behavior action division result; The strain estimation module calls the behavior action division result, identifies the speed phase change in the period and extracts the acceleration amplitude difference, analyzes the change trend of the phase shift and the amplitude difference, judges the synchronization enhancement relationship of the continuous period and calculates the cumulative change proportion, and obtains the strain gradient estimation output; The early warning trigger module calls the strain gradient estimation output, extracts the current period fluctuation range, analyzes the continuous period fluctuation difference and identifies the section with consistent direction, marks the continuous offset area combined with the strain trend, establishes the disturbance accumulation state and constructs the disturbance growth trend indication, and obtains the micro-disturbance state early warning trigger configuration.

[0005] As a further scheme of the application, the transition judgment interval sequence includes a state switching starting node, a current rising stage waveform starting point, and a speed response time interval, the current offset correction configuration includes a current rising stage area difference, a waveform offset direction factor, and a response correction proportion parameter, the behavior action division result includes a curvature change extreme point, a speed direction reversal node, and displacement section distribution information, the strain gradient estimation output includes a phase shift sequence, an amplitude difference change sequence, and a structural response abnormal interval, and the micro-disturbance state early warning trigger configuration includes a period fluctuation difference sequence, a continuous offset area identifier, and a disturbance growth trend indication.

[0006] As a further scheme of the application, the sampling recognition module comprises: The instruction trigger recognition sub-module obtains the control signal data sequence, locates the instruction trigger node before the driving execution action starts, filters the variable starting time in the instruction signal time sequence, establishes the instruction trigger time period label combined with the signal amplitude change trend in the time period before and after the target time; The signal contrast determination sub-module acquires the rising stage starting point of the current signal curve and acquires the positioning motor speed response time point according to the instruction trigger time period tag, analyzes the interval of the current signal starting point and the speed response point on the time axis, calls interval data for amplitude change comparison, and obtains an instruction response interval; The hysteresis characteristic calibration sub-module calls the instruction response interval, compares the amplitude change trend of the response stage of the electric execution and the mechanical action in the interval, judges the signal change strength and response time relationship in the target interval, combines the change rate change in the interval, marks the execution state conversion time period tag, and generates a transfer judgment interval sequence.

[0007] As a further scheme of the application, the response extraction module comprises: The area difference identification sub-module calls the transfer judgment interval sequence, obtains the current amplitude sequence and the duration sequence of the rising segment and the falling segment in the current signal, calculates the area of the two current waveform coverage areas according to the current change sequence in the coverage range of the rising segment and the falling segment, and generates an area symmetry offset degree index; The current offset ratio construction sub-module extracts the rising segment wave peak amplitude, the falling segment steady-state current amplitude, and the current fluctuation difference sequence in each segment in a continuous period based on the area symmetry offset degree index, extracts the difference value of the current peak value and the steady-state value, combines the area offset value, and calculates the offset dynamic quantitative ratio; The correction configuration generation sub-module calls the offset dynamic quantitative ratio, identifies the asymmetric behavior existing in the current response, constructs the correction configuration parameter of the adjustment response stage input item, and generates a current offset correction configuration.

[0008] As a further scheme of the application, the behavior division module comprises: The curvature distribution construction sub-module obtains the current offset correction configuration, extracts the displacement response sequence of the servo device, calculates the angle change between continuous data, establishes the continuous curvature distribution according to the angle change, extracts the extreme value point of the change rate, and generates a curvature extreme value distribution sequence; The distribution density judgment sub-module calls the curvature extreme value distribution sequence, calculates the extreme value point distribution density of multiple intervals, analyzes the extreme value point number and interval trend in the interval, and obtains a key change section set; The behavior segmentation generation sub-module calls the key change section set, positions the extreme value point position of the speed direction reversal in the section, takes the displacement process interval between adjacent extreme value points as a segmentation node, divides the displacement process into multiple sections, and obtains a behavior action division result.

[0009] As a further scheme of the application, the process of obtaining the key change section set is specifically: The time positions corresponding to each extreme point in the extreme value distribution sequence of curvature is acquired, the time interval between adjacent extreme points is calculated, and an extreme point interval sequence is constructed, a density determination threshold of extreme points is set as the sum of the median and the average number of extreme points in a unit time for the extreme point interval sequence, a section in which the density of extreme points in a plurality of continuous intervals is greater than the density determination threshold of extreme points is selected as a key change section, and all continuous sections satisfying the density of extreme points being greater than the density determination threshold of extreme points are defined as a key change section set.

[0010] As a further scheme of the present application, the strain estimation module comprises: The periodic signal extraction submodule calls the segmented node interval in the behavior action division result, collects the rotation speed data and acceleration response sequence in each interval corresponding to the period, constructs a phase change sequence for the rotation speed signal, extracts the peak value and valley value of the acceleration response, calculates the peak-to-peak fluctuation amplitude, and generates signal change amplitude difference data; The synchronous enhancement judgment submodule analyzes the change direction of the phase shift sequence and the amplitude difference sequence based on the signal change amplitude difference data, calculates the trend change of both in continuous periods, judges the signal change synchronous enhancement relationship, marks the signal synchronous growth section, and obtains a synchronous growth identification section; The strain index generation submodule calls the synchronous growth identification section, calculates the cumulative proportion of the phase shift and amplitude difference change in the corresponding section, establishes an internal structure response abnormal index, and obtains a strain gradient estimation output.

[0011] As a further scheme of the present application, the process of marking the signal synchronous growth section is specifically: The change direction comparison result of the phase shift sequence and the amplitude difference sequence in continuous periods is acquired, the judgment result of the change direction of both in each period is calculated based on the period index, a period consistency identification sequence is formed, the synchronous growth start judgment threshold is set as three consecutive periods, and the end judgment threshold is set as two periods of inconsistent direction, the period consistency identification sequence is traversed, a continuous period set satisfying the start judgment threshold condition and not triggering the end judgment threshold condition is marked, and is used as a signal synchronous growth section.

[0012] As a further scheme of the present application, the early warning triggering module comprises: The fluctuation amplitude extraction submodule calls the synchronous growth section marked by the strain gradient estimation output, extracts the maximum and minimum values of the current signal in a plurality of periods, calculates the maximum and minimum fluctuation amplitude in each period, establishes a continuous period fluctuation amplitude sequence, and generates a period fluctuation amplitude sequence; The offset consistent marking sub-module analyzes the fluctuation amplitude change trend in the continuous period based on the periodic fluctuation amplitude sequence, judges the consistency of the fluctuation direction in adjacent periods, combines the trend direction in the strain gradient speculation output, marks the continuous offset interval, and obtains the continuous offset interval label; The trend indication generation sub-module calls the continuous offset interval label, extracts the trend growth segment of the disturbance behavior according to the continuous continuity of the fluctuation amplitude in the interval, establishes a trend accumulation state label, and generates a micro-perturbation state early warning trigger configuration.

[0013] The technical scheme provided by the embodiment of the application has at least the following beneficial effects: By combining the control instruction with the multi-source response data, the full-process recognition ability of the servo system dynamic behavior is enhanced, the linkage analysis of response offset, phase drift and amplitude difference trend is used to optimize the identification path of asymmetric behavior and strain anomaly, the synchronous trend accumulation judgment mechanism is used to realize the continuous labeling of structure disturbance evolution, the segmentation accuracy and time sequence consistency of the key action process are improved, the periodic fluctuation direction determination and continuous offset marking strategy are used to expand the pre-position range of the micro-perturbation state recognition, the servo execution state is expanded from the traditional single-point static judgment to the trend recognition mode based on the dynamic change process, and the trend early warning of abnormal behavior of the servo system is realized. BRIEF DESCRIPTION OF DRAWINGS

[0014] In order to more clearly illustrate the technical scheme in the embodiment of the application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor.

[0015] Figure 1 The system flowchart of the application; Figure 2 The system framework schematic diagram of the application; Figure 3 The sampling recognition module flowchart of the application; Figure 4 The response extraction module flowchart of the application; Figure 5 The behavior division module flowchart of the application; Figure 6 The strain speculation module flowchart of the application; Figure 7 The early warning trigger module flowchart of the application. DETAILED DESCRIPTION

[0016] The technical scheme in the application will be described below with reference to the drawings.

[0017] In the embodiments of the present application, the words such as "example", "for example" are used to represent an example, illustration, or description. Any embodiment or design scheme described as "example" in the present application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. Rather, the word "example" is intended to present the concept in a specific manner. In addition, in the embodiments of the present application, the meaning expressed by "and / or" can be both, or can be one of the two.

[0018] In the embodiments of the present application, "image" and "picture" can be used interchangeably at times, and it should be pointed out that the meanings expressed are consistent when the distinction is not emphasized. "Of", "corresponding" and "corresponding" can be used interchangeably at times, and it should be pointed out that the meanings expressed are consistent when the distinction is not emphasized.

[0019] In the embodiments of the present application, sometimes the subscript such as W1 can be written in the form of non-subscript such as W1, and the meanings expressed are consistent when the distinction is not emphasized.

[0020] In order to make the technical problems, technical schemes and advantages to be solved by the present application more clear, specific embodiments will be described in detail below with reference to the drawings.

[0021] The embodiments of the present application provide a servo system data acquisition and analysis system, please refer to Figures 1 to 2 The present application provides a technical scheme, a servo system data acquisition and analysis system comprises: The sampling identification module calls the control signal data sequence, analyzes the driving execution front control instruction trigger node, identifies the current rising segment waveform starting point and compares with the motor speed response time point, identifies the hysteresis characteristics between the electric execution and the mechanical action, marks the state transition section, and obtains the transfer judgment interval sequence; The response extraction module uses the transfer judgment interval sequence to analyze the duration characteristics of the current rising and falling stages, calculates the waveform coverage area difference, constructs the correction proportion factor based on the offset direction, combines the current proportion of the wave peak segment and the stable segment, identifies the current response asymmetric behavior, and obtains the current offset correction configuration; The behavior division module uses the current offset correction configuration to analyze the angle change of the displacement response curve to form the curvature distribution, identifies the change extreme value and calculates the distribution density, locates the speed direction reversal node and divides the displacement section, and obtains the behavior action division result; The strain estimation module calls the behavior action division result, identifies the speed phase change in the period and extracts the acceleration amplitude difference, analyzes the change trend of the phase drift and the amplitude difference, judges the synchronization enhancement relationship of the continuous period and calculates the cumulative change proportion, and obtains the strain gradient estimation output. The early warning trigger module calls the strain gradient speculation output, extracts the current cycle fluctuation range, analyzes the difference between consecutive cycle fluctuations and identifies the section with consistent direction, marks the continuous deviation area in combination with the strain trend, establishes the disturbance accumulation state and constructs the disturbance growth trend indication, and obtains the micro-perturbation state early warning trigger configuration.

[0022] The transfer judgment interval sequence includes the state switching start node, the current rising stage waveform starting point, and the speed response time interval. The current deviation correction configuration includes the current rising stage area difference, the waveform deviation direction factor, and the response correction proportion parameter. The behavior action division result includes the curvature change extreme point, the speed direction reversal node, and the displacement section distribution information. The strain gradient speculation output includes the phase drift sequence, the amplitude difference change sequence, and the structure response abnormal interval. The micro-perturbation state early warning trigger configuration includes the cycle fluctuation difference sequence, the continuous deviation area identification, and the disturbance growth trend indication.

[0023] Please refer to Figure 2 and Figure 3 , the sampling identification module includes: The instruction trigger identification submodule obtains a control signal data sequence, locates the instruction trigger node before the driving execution action is started, screens the variation start time in the instruction signal time sequence, establishes the instruction trigger time period label in combination with the signal amplitude change trend in the time period before and after the target time. The obtained control signal data sequence is a group of voltage values collected at intervals of on the time axis, and the specific sequence is {t=0ms, V=0V; t=0.5ms, V=0V; t=1.0ms, V=0V; t=1.5ms, V=5V; t=2.0ms, V=5V; t=2.5ms, V=5V}. This submodule scans point by point from the head of the sequence, detects that the voltage amplitude jumps from to at , and records this time point as the instruction trigger node. Subsequently, the system extracts the signal amplitudes at (i.e. ) and (i.e. ) before and after the node, which are and respectively, confirms that the change trend of the signal amplitude is unidirectional stable growth, and establishes the instruction trigger time period label {trigger_timestamp:1.5ms, status:‘confirmed_rising_edge’}.

[0024] The signal comparison and judgment submodule collects the starting point of the rising phase of the current signal curve based on the instruction trigger time period label, and collects the positioning motor speed response time point. It analyzes the interval between the starting point of the current signal and the speed response point on the time axis, calls the interval interval data to compare the amplitude changes, and obtains the instruction response interval interval. The trigger time stamp {trigger_timestamp:1.5ms,status:'confirmed_rising_edge'} is triggered according to the command. Using the reference time, the module synchronously retrieves the U-phase current signal sequence of the motor driver and the speed feedback sequence of the motor encoder, which are strictly synchronized with the control signal. The module searches the current signal sequence and sets the condition for effective current start-up as the current value at three consecutive sampling points exceeding the system's static noise threshold. ,exist The current value collected at the location is , for , for The conditions are met, therefore... The starting point of the rising phase of the current signal curve is recorded. Simultaneously, the module searches the speed feedback sequence, setting the criterion for a valid speed response as the speed value first exceeding the minimum resolvable rotational speed. ,exist The speed value collected at that location is Therefore, this moment is recorded as the positioning motor speed response time point. Next, the starting point of the current signal is analyzed. With speed response point The absolute interval on the time axis is calculated to obtain and call this Within the time interval, that is All current and velocity sampling data are compared for amplitude changes, and the current amplitude within the interval is... linear growth to The speed value is from Growth to The instruction response interval range [{start:2.8ms,end:4.6ms}] is obtained.

[0025] The hysteresis feature calibration submodule calls the instruction response interval interval, compares the amplitude change trend of the response stages of electric execution and mechanical action within the interval interval, judges the relationship between signal change intensity and response time within the target interval, combines the change rate within the interval, marks the execution state transition period label, and generates a transfer judgment interval sequence. The calling instruction response interval interval [{start:2.8ms, end:4.6ms}] is called, and the response stage amplitude change trend of the electric execution (current) and the mechanical action (speed) is compared in the interval. Through calculation, the total change amount of the current amplitude of the electric execution is , the total change amount of the speed amplitude of the mechanical action is , and the relationship between the signal change intensity and the response time in the target interval is judged, that is, the average change rate of the current is , the average change rate of the speed is , and the dynamic change of the change rate in the interval is combined, wherein the current change rate is in the early stage of the interval, and decreases to in the last stage, and the speed change rate continuously increases from to , the interval is marked as an execution state conversion period label, and a transition judgment interval sequence [{interval_id:001, start_time:2.8ms, end_time:4.6ms, lag_type:‘electro_mechanical’}] is generated.

[0026] Please refer to Figure 2 and Figure 4 , the response extraction module includes: The area difference identification sub-module calls the transition judgment interval sequence, obtains the current amplitude sequence and the duration sequence of the rising segment and the falling segment in the current signal, calculates the area of the two current waveform coverage areas according to the current change sequence in the coverage range of the rising segment and the falling segment, and generates an area symmetry offset degree index; The transition judgment interval sequence [{interval_id:001, start_time:2.8ms, end_time:4.6ms, lag_type:‘electro_mechanical’}] is called and extended to the complete current waveform interval of the servo motor completing a complete “acceleration-deceleration” action. The starting time point of the interval is , and the ending time point is , the system identifies that the rising segment of the current is , and the duration is , the falling segment is , and the duration is , the system calculates the area of the two current waveform coverage areas by using the numerical integration method according to the current amplitude sequence collected with a sampling interval of in the two time periods. Specifically, the current value sequence of 50 sampling points of the rising segment is accumulated and multiplied by the sampling interval , and the rising segment area is Similarly, the 90-sample-point current value sequence of the falling section is calculated to obtain the falling section area Then, the difference degree is calculated, and the absolute difference value is The area symmetry offset degree index {value: 1.7, unit: 'A·ms'} is generated.

[0027] The current offset ratio construction submodule extracts the rising section peak amplitude, the falling section steady-state current amplitude, the current fluctuation difference sequence in each section within the continuous period, and the difference between the current peak value and the steady-state value based on the area symmetry offset degree index, and combines the area offset value to use the formula: ; The offset dynamic quantization ratio is calculated. Among them, is the rising section current normalized average value, which is obtained by normalizing the amplitude of each sample point in the rising section of the transfer judgment interval and then averaging, is the falling section current normalized average value, which is obtained by calculating the average value of the normalized sample value sequence of the falling section, is the normalized current peak value, which is obtained by extracting the maximum value of the rising section current and normalizing it, is the normalized steady-state current amplitude, which is obtained by extracting the average value of the stable section at the end of the falling section and normalizing it, is the current normalized difference value of the th sample point, which is obtained by normalizing the difference between the current value of the current sample point and the reference stable value, is the index number of the fluctuation section current sample point sequence, is the total number of current sample points in the current fluctuation section, is the offset dynamic quantization ratio, which represents the structural asymmetry degree between the rising section and the falling section of the current waveform; Based on the area symmetry offset degree index {value: 1.7, unit: 'A·ms'}, more detailed current characteristic parameters are extracted from the extended complete current waveform interval As shown in Table 1, it is a current characteristic sampling data table of a certain servo drive conversion period, Table 1 Current characteristic sampling data table of servo drive conversion period ; As shown in Table 1, the peak amplitude of the rising section is , the steady-state current amplitude at the end of the falling section is , and a continuous current fluctuation difference sequence is recorded in the falling section, and some of the values are , a total of 90 sampling points, the peak value of the current The difference between the steady-state value is extracted to obtain , combined with the area offset value, using the formula to calculate the offset dynamic quantization ratio, the formula is , the offset dynamic quantization ratio, dimensionless, represents the normalized current value, and the normalization reference value is the rated current of the servo motor, which is obtained by consulting the motor nameplate or technical manual, and in this example it is set to , the superscript indicates that the value is the normalized value, and the subscript and respectively refer to the rising section and the falling section, and the superscript indicates the peak value (peak), and the superscript indicates the steady-state value (steady-state), indicates the arithmetic mean value of all sampling points in the corresponding section, is the difference between the normalized current value of the th sampling point in the falling section (i.e. the fluctuation section) and the normalized steady-state current value of the section, is the index number of the sampling point sequence, starting from 1, is the total number of sampling points in the fluctuation section, which is 90 in this example, and the summation symbol indicates the accumulation of all sampling points, the numerator part of the formula adds the average current difference between the rising section and the falling section (representing area asymmetry) to the square root value of the peak steady-state difference (representing amplitude characteristics), and the denominator part adds the disturbance energy of all sampling points in the fluctuation section (characterized by the sum of squares of differences) to the peak steady-state difference. The final division operation constructs an index that measures relative asymmetry. The benefit of the formula is that by integrating the area symmetry, amplitude characteristics and microscopic fluctuation energy of the current waveform, a more comprehensive and more resistant to noise interference dynamic imbalance quantization method is created than a single-dimensional index. The parameters are obtained and calculated as follows. From the current sampling data, the current mean value of the rising section with 50 sampling points is , the current mean value of the falling section with 90 sampling points is , the peak current is , and the steady-state current is . Divide these values by the rated current to normalize and get , , , , and the reference stable value is the steady-state value Calculate point by point among the 90 sampling points in the descent segment. For example, if in Time sampling value ,but Its square is The sum obtained after calculating all 90 points The value is Substitute the values ​​into the formula: ; The offset dynamic quantization ratio is a dimensionless response offset index constructed to address the inconsistency of current morphology at different stages of the servo system's current response. This ratio integrates the amplitude difference between the current peak and the steady-state segment, the area symmetry of the regions before and after the curve, and the fluctuation scale of current disturbance energy within the period. It measures the non-equilibrium of the current response behavior during a specific action transition period. A larger value indicates a stronger asymmetric structure in the current morphology between the upper and lower stages, which may be accompanied by external disturbances or internal response deviations. As a bridging variable for current response feature identification and behavior adjustment configuration, the index can directly drive subsequent correction configuration selection logic, indicating whether the input signal deviates from the standard dynamic morphology in the time domain, and assisting in the identification of potential faults or atypical behavior segments. The result shows that the current current response waveform has an asymmetry quantization value of 1.782.

[0028] The correction configuration generation submodule calls the offset dynamic quantization ratio, identifies asymmetric behavior in the current response, constructs the correction configuration parameters for the input items in the adjustment response stage, and generates the current offset correction configuration. Call the offset dynamic quantization ratio It is then compared with a preset, tiered threshold range, which is set based on 5000 cycle tests of this servo system under standard load. Statistical analysis of the values ​​will be performed. The interval with values ​​less than the 90th percentile (1.20) is defined as the "normal zone," the interval between the 90th and 98th percentiles (1.20 to 1.80) is defined as the "zone of concern," and the interval greater than the 98th percentile (1.80) is defined as the "warning zone." This is based on the current calculation results. If the current falls within the "region of interest," the system identifies asymmetric behavior in the current response and determines its location within the region of interest based on this asymmetric behavior. This is mapped to a nonlinear correction parameter lookup table, constructs the correction configuration parameters for the input items in the adjustment response stage, and generates the current offset correction configuration {compensation_factor:1.18,timing_adjust:-12ms,alert_level:'watch'}.

[0029] Referring to Figure 2 and Figure 5 , the behavior division module comprises: The curvature distribution construction submodule obtains the current offset correction configuration, extracts the displacement response sequence of the servo device, calculates the angle change between consecutive data, establishes the continuous curvature distribution according to the angle change, extracts the extreme value point of the change rate, and generates the curvature extreme value distribution sequence; The current offset correction configuration {compensation_factor: 1.18, timing_adjust: -12ms, alert_level: 'watch'} is obtained, and the displacement response sequence of the servo device in the corresponding time period is extracted as a trigger, which is a group of angle data with a sampling interval of , specifically {…t=29ms, θ=35.1°; t=30ms, θ=36.0°; t=31ms, θ=36.8°; t=32ms, θ=37.5°;…}, the module calculates the angle change between consecutive data, that is, the approximate value of the angular velocity, for example, the angular velocity at is about , and the angular velocity at is about , according to the sequence of these angle change values, the change rate of the angle change, that is, the approximate value of the angular acceleration, is further calculated, which is the continuous curvature distribution, for example, the curvature at is about , the system traverses the curvature distribution of the entire displacement sequence, extracts the points where the local maximum and minimum values of the curvature change rate appear, and generates the curvature extreme value distribution sequence [{t:34, val:2.5}, {t:41, val:-1.1}, {t:46, val:1.9}, {t:50, val:1.5}, {t:58, val:-1.8}, {t:65, val:1.2}], with the unit of .

[0030] The distribution density judgment submodule calls the curvature extreme value distribution sequence to calculate the distribution density of the extreme value points in multiple intervals, analyzes the number and interval trend of the extreme value points in the interval, and obtains the key change section set; The curvature extreme value distribution sequence [{t:34, val:2.5}, {t:41, val:-1.1}, {t:46, val:1.9}, {t:50, val:1.5}, {t:58, val:-1.8}, {t:65, val:1.2}] is called to obtain the time position sequence {34, 41, 46, 50, 58, 65} corresponding to each extreme value point, with the unit of , and then the time interval between adjacent extreme value points is calculated to construct the extreme value point interval sequence {7, 5, 4, 8, 7}, with the unit of For this spacing sequence, an extreme point density threshold is set, which is defined as the number of extreme points per unit time. The calculation method is as follows: The median of the spacing sequence is The average is The density threshold is then determined as follows. Items / ms, the system uses a sliding window (window size is...) Traverse the time axis and filter for extreme point densities greater than a certain value in multiple consecutive intervals. A segment of 1 / ms, for example in this Within the interval, there are 3 extreme points (located at 41, 46, 50), and their density is... per ms, because Therefore, this segment is identified as a critical change segment. All continuous segments that meet the conditions are grouped together to obtain the critical change segment set {[41ms,50ms],[88ms,102ms]}.

[0031] The behavior segmentation generation submodule calls the set of key change segments, locates the extreme point where the velocity direction reverses within the segment, takes the displacement process interval between adjacent extreme points as segmentation nodes, divides the displacement process into multiple segments, and obtains the behavior action segmentation result. Call the set of key change segments {[41ms, 50ms], [88ms, 102ms]} and lock onto the first segment. Within this time period, the system synchronously retrieves the motor speed feedback sequence, which is {…t=46ms, v=3.1r / min; t=47ms, v=1.2r / min; t=48ms, v=-0.5r / min; t=49ms, v=-2.2r / min;…}. By detecting changes in the sign of the speed values, the system locates the extreme point where the speed direction reverses. and Between these points, linear interpolation was used to determine the time point at which the velocity reaches zero. The system will link the end point of the previous action segment with this reversal point. The displacement process interval between the reversal point and the starting point of the next action segment is used as a segmentation node to divide the complete displacement process into multiple segments with clear physical meaning, and the behavior action segmentation result is obtained [{segment_id:1,type:'deceleration',start:30.8ms,end:47.7ms},{segment_id:2,type:'reversal_acceleration',start:47.7ms,end:60.1ms},…].

[0032] Referring to Figure 2 and Figure 6 , the strain prediction module comprises: The periodic signal extraction submodule calls the segmented node interval in the behavior action division result, collects the rotation speed data and acceleration response sequence in each interval, constructs a phase change sequence for the rotation speed signal, extracts the peak and valley values of the acceleration response, calculates the peak-to-peak fluctuation amplitude, and generates signal change amplitude difference data; The segmented node interval in the behavior action division result is called, for example, the deceleration segment [30.8ms, 47.7ms], and the corresponding rotation speed data sequence and three-axis acceleration response sequence are collected in this interval. The system performs fast Fourier transform (FFT) on the rotation speed signal, and identifies that the main fluctuation frequency is , that is, the theoretical period is By detecting the zero-crossing point or peak point of the rotation speed signal, an actual phase change sequence is constructed. If the actual time lengths of three consecutive periods are {25.2ms, 25.5ms, 25.9ms}, the phase drifts relative to the theoretical period are {+0.2ms, +0.5ms, +0.9ms}. At the same time, the system extracts the peak and valley values of the Z-axis acceleration response signal in each period in this interval, for example, the peak value of the acceleration in the first period is , the valley value is , the peak-to-peak fluctuation amplitude is calculated as , and the amplitudes of the subsequent two periods are calculated as and , and a signal change amplitude difference data sequence {12.0, 12.4, 12.9, …} is generated, with the unit being .

[0033] The synchronous enhancement judgment submodule analyzes the change direction of the phase drift sequence and the amplitude difference sequence based on the signal change amplitude difference data, calculates the trend change of the two in consecutive periods, judges the signal change synchronous enhancement relationship, marks the signal synchronous growth segment, and obtains a synchronous growth identification segment; Based on the signal change amplitude difference data {12.0, 12.4, 12.9, …}, the phase drift sequence {+0.2, +0.5, +0.9, …} and the amplitude difference sequence are analyzed for the change direction, and the trend changes of both in consecutive periods are calculated. From the first period to the second period, the phase drift increases (+0.5 > +0.2), and the amplitude difference also increases (12.4 > 12.0), the change direction is consistent. From the second period to the third period, both increase and the direction is consistent. Based on this comparison result, the system forms a period consistency identification sequence. If the comparison result of 8 consecutive periods is {1, 1, 0, 1, 1, 1, 1, 0} (1 for consistent, 0 for inconsistent), the system sets the synchronization growth start judgment threshold to be consistent for three consecutive periods, and the end judgment threshold to be inconsistent for two consecutive periods. The system traverses the identification sequence, and from the fifth period, there are four consecutive “1”s, which meet the start threshold, and no two consecutive “0”s appear to trigger the end. Therefore, the period corresponding to the fifth to eighth periods is marked as a signal synchronization growth segment, and the synchronization growth identification section [{start_period: 5, end_period: 8, start_time: 130.8ms, end_time: 230.8ms}] is obtained.

[0034] The strain index generation submodule calls the synchronization growth identification section, calculates the cumulative proportion of the phase drift and amplitude difference change in the corresponding section, establishes the interval structure response abnormal index, and obtains the strain gradient prediction output; The synchronization growth identification section [{start_period: 5, end_period: 8, start_time: 130.8ms, end_time: 230.8ms}] is called. In this section (fifth to eighth periods), the phase drift value sequence is {+1.5ms, +1.9ms, +2.4ms, +3.0ms}, and the amplitude difference value sequence is {13.8, 14.5, 15.3, 16.0}, with a unit of The system accumulates the change amounts of these two sequences. The cumulative change amount of the phase drift is The cumulative change amount of the amplitude difference is At the same time, the system obtains the total change range of the phase drift in the entire behavior (for example, 0ms to 500ms) as The total change range of the amplitude difference is The cumulative proportion of the phase drift is The cumulative proportion of the amplitude difference is , the system establishes the abnormal index of the structure response in the interval, the index is obtained by weighting and summing two proportion values according to preset weights, the weight is calibrated according to the experiment, the acceleration signal is more sensitive to the structure strain, the weight is set to 0.6, and the weight of the rotating speed phase is 0.4, so that the abnormal index is , the strain gradient prediction output {region: [130.8ms, 230.8ms], strain_gradient_metric: 39.72%} is obtained.

[0035] Please refer to Figure 2 and Figure 7 , the early warning triggering module comprises: The fluctuation amplitude extraction submodule calls the synchronous growth segment marked by the strain gradient prediction output, extracts the maximum value and the minimum value of the current signal in multiple periods, calculates the maximum and minimum fluctuation amplitude in each period, establishes a continuous period fluctuation amplitude sequence, and generates the period fluctuation amplitude sequence; The synchronous growth segment [130.8ms, 230.8ms] marked by the strain gradient prediction output is called, and in this time period, the maximum value and the minimum value of the corresponding U-phase current signal in each period are re-extracted, in the fifth period (130.8ms-155.8ms), the maximum value of the current is , the minimum value is , and the fluctuation amplitude is calculated as , the calculation is repeated for the next three periods (the sixth, seventh and eighth periods) in the synchronous growth segment, and the values obtained are , so that the continuous period fluctuation amplitude sequence {5.4, 5.7, 5.9, 6.2} in the segment is established, and the period fluctuation amplitude sequence is generated.

[0036] The consistent offset marking submodule analyzes the fluctuation amplitude change trend in the continuous period based on the period fluctuation amplitude sequence, judges the consistency of the fluctuation direction in adjacent periods, and combines the trend direction in the strain gradient prediction output to mark the continuous offset interval, and obtains the continuous offset interval label; Based on the period fluctuation amplitude sequence {5.4, 5.7, 5.9, 6.2}, the fluctuation amplitude change trend in the continuous period is analyzed, and the change amount sequence is , the consistency of the fluctuation direction in the adjacent periods, since the change amount of the continuous three periods is all positive, that is, the fluctuation amplitude continues to increase, the direction is consistent, combined with the strain gradient to infer the abnormal growth trend indicated by the output {strain_gradient_metric: 39.72%}, mark the synchronous growth segment [130.8ms, 230.8ms] as a continuous offset interval, get the continuous offset interval label {interval: [130.8ms, 230.8ms], trend: 'consistent_increase'}.

[0037] The trend indication generation submodule calls the continuous offset interval label, extracts the trend growth segment of the disturbance behavior according to the continuous continuation of the fluctuation amplitude in the interval, establishes a trend cumulative state label, and generates a micro-disturbance state warning trigger configuration; Call the continuous offset interval label {interval: [130.8ms, 230.8ms], trend: 'consistent_increase'}; according to the continuous continuation of the fluctuation amplitude in the interval, that is, the continuous four periods all show an increasing trend, extract this growth behavior as the trend growth segment of the disturbance behavior, and the system further accumulates the amplitude growth amount in this growth segment to get the cumulative growth amount , compare this cumulative value with the preset micro-disturbance cumulative threshold , which is set by analyzing 100 groups of running data of the system before the critical failure, extracting the 80th percentile value of the cumulative growth amount of the current fluctuation within 500ms before the failure, and since , the system establishes the trend cumulative state label as "level_1_warning", and generates the micro-disturbance state warning trigger configuration {alert_code: 'MW-01', timestamp: 230.8ms, cumulative_value: 0.8A, threshold: 0.75A}.

[0038] The above-described embodiments can be implemented in whole or in part by software, hardware (such as a circuit), firmware, or any combination thereof. When implemented in software, the above-described embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are wholly or partially generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transferred from one website, computer, server, or data center to another website, computer, server, or data center through a wired (such as infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium accessible by a computer or a data storage device such as a server, data center, etc. containing one or more available medium collections. The available medium can be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state disk.

[0039] It should be understood that the term "and / or" herein merely describes an association relationship of associated objects, which means that there can be three relationships, for example, A and / or B can represent three cases of A alone, A and B together, and B alone, where A and B can be singular or plural. In addition, the character " / " herein generally represents an "or" relationship between the front and rear associated objects, but can also represent an "and / or" relationship, which can be understood in the context before and after.

[0040] In the present application, "at least one" means one or more, and "multiple" means two or more. "At least one of the following" or the like means any combination of the items, including any combination of single or multiple items. For example, at least one of a, b, or c can represent a, b, c, a-b, a-c, b-c, or a-b-c, where a, b, and c can be single or multiple.

[0041] It should be understood that in various embodiments of the present application, the size of the sequence number of the above-described processes does not mean the order of execution, and the execution order of the processes should be determined by their functions and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0042] Those skilled in the art can clearly understand that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0043] Those skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working processes of the devices, apparatuses and units described above can refer to the corresponding processes in the foregoing method embodiments, which will not be repeated here.

[0044] In several embodiments provided by the present application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely schematic, for example, the division of the units is only a logical function division, and actual implementation can have another division manner, for example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0045] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or can be distributed on multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0046] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically independently, or two or more units can be integrated into one unit.

[0047] If the functions are realized in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the parts of the present application that essentially contribute to the prior art or the parts of the technical solutions can be embodied in the form of software products. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the present application. The aforementioned storage medium includes a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0048] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A servo system data acquisition and analysis system, characterized by, The system comprises: The sampling identification module calls the control signal data sequence, analyzes the control instruction trigger node before driving execution, identifies the starting point of the current rising segment waveform and compares it with the motor speed response time point, identifies the lag feature between the electric execution and the mechanical action, marks the state transition section, and obtains the transition judgment interval sequence; The response extraction module uses the transition judgment interval sequence to analyze the duration characteristics of the current rising and falling stages, calculates the waveform coverage area difference, constructs the correction proportion factor based on the offset direction, combines the current proportion of the peak segment and the stable segment, identifies the current response asymmetry behavior, and obtains the current offset correction configuration; The behavior division module uses the current offset correction configuration to analyze the curvature distribution formed by the angle change of the displacement response curve, identifies the change extreme value and calculates the distribution density, locodes the speed direction reversal node and divides the displacement section, and obtains the behavior action division result; The strain estimation module calls the behavior action division result, identifies the speed phase change in the period and extracts the acceleration amplitude difference, analyzes the change trend of the phase drift and the amplitude difference, judges the synchronization enhancement relationship of the continuous period and calculates the cumulative change proportion, and obtains the strain gradient estimation output.

2. The servo system data acquisition and analysis system of claim 1, wherein, The transition judgment interval sequence includes the state switching starting node, the current rising stage waveform starting point, and the speed response time interval. The current offset correction configuration includes the current rising stage area difference, the waveform offset direction factor, and the response correction proportion parameter. The behavior action division result includes the curvature change extreme point, the speed direction reversal node, and the displacement section distribution information. The strain gradient estimation output includes the phase drift sequence, the amplitude difference sequence, and the structural response abnormal interval.

3. The servo system data acquisition and analysis system of claim 1, wherein, The sampling identification module comprises: The instruction trigger identification submodule obtains the control signal data sequence, locodes the instruction trigger node before the driving execution action starts, filters the variable starting time in the instruction signal time sequence, establishes the instruction trigger period label by combining the signal amplitude change trend in the time period before and after the target time, and locodes the instruction trigger node before the driving execution action starts. The signal comparison and judgment submodule acquires the starting point of the current signal curve rising stage according to the instruction trigger period label, locodes the motor speed response time point, analyzes the interval of the current signal starting point and the speed response point on the time axis, calls the interval data to compare the amplitude change, and obtains the instruction response interval. The lag feature labeling submodule calls the instruction response interval, compares the amplitude change trend of the response stage of the electric execution and the mechanical action in the interval, judges the signal change intensity and response time relationship in the target interval, combines the change rate change in the interval, marks the execution state transition period label, and generates the transition judgment interval sequence.

4. The servo system data acquisition and analysis system of claim 3, wherein, The response extraction module comprises: The area difference identification submodule calls the transition judgment interval sequence, obtains the current amplitude sequence and duration sequence of the rising segment and the falling segment in the current signal, calculates the area of the two current waveform coverage areas according to the current change sequence in the coverage range of the rising segment and the falling segment, and generates the area symmetry offset degree index. The current offset ratio constructing submodule extracts the rising section peak amplitude, the falling section steady current amplitude, the current fluctuation difference sequence in each section in a continuous period, and calculates the offset dynamic quantitative ratio value by means of the difference between the current peak value and the steady value, in combination with the area offset value, based on the area symmetry offset degree index; The correction configuration generating submodule generates the current offset correction configuration by identifying the asymmetric behavior existing in the current response, constructing the correction configuration parameter of the adjustment response stage input item, and calling the offset dynamic quantitative ratio value.

5. The servo system data acquisition and analysis system of claim 4, wherein, The behavior division module comprises: The curvature distribution constructing submodule extracts the displacement response sequence of the servo device, calculates the angle change between continuous data, establishes the continuous curvature distribution according to the angle change, extracts the extreme value point of the change rate, and generates the curvature extreme value distribution sequence, by acquiring the current offset correction configuration; The distribution density judging submodule calculates the extreme value point distribution density of multiple intervals, analyzes the interval extreme value point quantity and interval trend, and obtains the key change section set, by calling the curvature extreme value distribution sequence. The behavior segmentation generating submodule obtains the behavior action division result by locating the extreme value point position of the speed direction reversal in the section, taking the displacement process interval between adjacent extreme value points as a segmentation node, and dividing the displacement process into multiple sections, by calling the key change section set.

6. The servo system data acquisition and analysis system of claim 5, wherein, The process of obtaining the key change section set is specifically as follows: The time position corresponding to each extreme value point in the curvature extreme value distribution sequence is acquired, the time interval between adjacent extreme value points is calculated and an extreme value point interval sequence is constructed, the extreme value point density judgment threshold is set as the sum of the median and the average number of extreme value points in a unit time for the extreme value point interval sequence, the section in which the extreme value point density is greater than the extreme value point density judgment threshold in a plurality of continuous intervals is selected as the key change section, and all continuous section sets satisfying the extreme value point density greater than the extreme value point density judgment threshold are defined as the key change section set.

7. The servo system data acquisition and analysis system of claim 5, wherein, The strain estimation module comprises: The period signal extracting submodule acquires the speed data and acceleration response sequence in each corresponding period in the segmented node interval in the behavior action division result, constructs a phase change sequence for the speed signal, extracts the peak value and valley value of the acceleration response, calculates the peak-to-peak fluctuation amplitude, and generates signal change amplitude difference data; The synchronous enhancement judging submodule analyzes the change direction of the phase shift sequence and the amplitude difference sequence, calculates the trend change of both in a continuous period, judges the signal change synchronous enhancement relationship, marks the signal synchronous growth section, and obtains the synchronous growth identification section, based on the signal change amplitude difference data; The strain index generating submodule acquires the strain gradient estimation output by calling the synchronous growth identification section, and calculating the cumulative proportion of the phase shift and amplitude difference change in the corresponding section.

8. The servo system data acquisition and analysis system of claim 7, wherein, The process of marking the signal synchronous growth section is specifically as follows: The change direction comparison result of the phase drift sequence and the amplitude difference sequence in continuous periods is obtained, a judgment result of the change direction consistency of the two in each period is calculated based on a period index, a period consistency identification sequence is formed, a synchronization growth start judgment threshold is set as three continuous periods of consistency, and a termination judgment threshold is set as two periods of inconsistent direction, a continuous period set satisfying the start judgment threshold condition and not triggering the termination judgment threshold condition is marked by traversing the period consistency identification sequence, and the continuous period set is taken as a signal synchronization growth segment.

9. The servo system data acquisition and analysis system of claim 1, wherein, The system further comprises: The early warning trigger module calls the strain gradient speculation output, extracts a current period fluctuation range, analyzes the fluctuation difference of continuous periods and identifies a section with consistent direction, marks a continuous deviation region in combination with a strain trend, establishes a disturbance accumulation state and constructs a disturbance growth trend indication, and obtains a micro-perturbation state early warning trigger configuration; The micro-perturbation state early warning trigger configuration comprises a period fluctuation difference sequence, a continuous deviation region identification, and a disturbance growth trend indication.

10. The servo system data acquisition and analysis system of claim 9, wherein, The early warning trigger module comprises: A fluctuation amplitude extraction submodule calls the synchronization growth segment marked by the strain gradient speculation output, extracts the maximum value and the minimum value of the current signal in multiple periods, calculates the maximum-minimum fluctuation amplitude in each period, establishes a continuous period fluctuation amplitude sequence, and generates a period fluctuation amplitude sequence; A deviation consistency marking submodule analyzes the fluctuation amplitude trend in continuous periods based on the period fluctuation amplitude sequence, judges the consistency of the fluctuation direction in adjacent periods, marks a continuous deviation interval in combination with the trend direction in the strain gradient speculation output, and obtains a continuous deviation interval label; A trend indication generation submodule calls the continuous deviation interval label, extracts a trend growth segment of a disturbance behavior according to the continuous continuity of the fluctuation amplitude in the interval, establishes a trend accumulation state mark, and generates a micro-perturbation state early warning trigger configuration.

Images