Intelligent control system for pulse signal switching

Abstract

The invention provides an intelligent control system for pulse signal switching, which relates to the technical field of data processing and is used for acquiring main pulse signal data of a main reading head and auxiliary pulse signal data of an auxiliary reading head, calculating a differential sequence of the main pulse signal data and generating an abnormal signal indication. The method comprises the following steps: detecting duration and stability of an abnormal fluctuation signal, generating an abnormal confirmation instruction, calculating a numerical value difference between main pulse signal data and auxiliary pulse signal data, generating a switching trigger instruction, switching the main pulse signal data and the auxiliary pulse signal data, extracting an error sequence between pulse signals of a main reading head and an auxiliary reading head, and outputting the error sequence. The method comprises the following steps: acquiring an error correction formula, correcting switching signal data according to the error correction formula to obtain corrected signal data, calculating time offset, and performing time sequence alignment on the corrected signal data according to the time offset to obtain aligned signal data. According to the invention, switching of pulse signals can be intelligently controlled.

Description

Technical Field

[0001] The invention relates to the technical field of data processing, and in particular to an intelligent control system for pulse signal switching. Background Art

[0002] In the angle measurement of precision rotating equipment, the combination of high-precision reading heads and magnetic strips is a common measurement method. The existing technology usually adopts the automatic switching of dual reading heads to ensure seamless connection during rotation and improve the stability of angle measurement. This method relies on the signal on the sensing magnetic strip to achieve the acquisition and display of angle data. However, the signal switching of this system depends on the physical installation position of the sensing iron sheet and the relative position of the sensing switch and the reading head. Once there is a slight deviation in the installation process, the sensing distance may change, thereby affecting the accuracy of the signal switching and further affecting the accuracy of the angle measurement.

[0003] For example, in the angle measurement of a CNC rotary table, when the reading head passes through the magnetic strip joint area, the system may experience signal jumps or abnormal readings. This is because there may be tiny gaps in the magnetic strip during the splicing process, which interferes with the reading accuracy of the reading head at the joint. When the installation error of the induction switch is large, the induction switch may not be able to accurately detect the switching point, causing the system to switch signals at the wrong position, causing the angle measurement value to mutate or drift, affecting the accuracy of the angle measurement. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent control system for pulse signal switching, aiming to solve the problems mentioned in the background technology.

[0005] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0006] An intelligent control system for pulse signal switching, the system comprising:

[0007] The signal acquisition module is used to obtain the main pulse signal data of the main reading head and the auxiliary pulse signal data of the auxiliary reading head, and pre-process the data to remove redundant data to obtain a pulse signal data set;

[0008] A data continuity monitoring module is used to calculate the differential sequence of the main pulse signal data according to the pulse signal data set, and judge the continuity of the main pulse signal data according to the differential sequence, and generate an abnormal signal indication when an abnormal fluctuation signal is detected;

[0009] The abnormality detection module is used to detect the duration and stability of the abnormal fluctuation signal according to the abnormal signal indication, and judge whether the main pulse signal data has a failure risk based on it. If the result is yes, an abnormality confirmation indication is generated;

[0010] A signal difference calculation module, used to calculate the numerical difference between the main pulse signal data and the auxiliary pulse signal data according to the abnormal confirmation indication, and obtain signal difference data;

[0011] A signal switching determination module, used to calculate a signal switching threshold group according to the signal difference data, and generate a switching trigger indication when the signal difference data satisfies the signal switching threshold group;

[0012] A smooth signal switching module is used to switch the main pulse signal data and the auxiliary pulse signal data according to the switching trigger indication, and eliminate the signal mutation caused by the signal switching by calculating the smooth signal value to obtain the switching signal data;

[0013] An error correction module is used to extract the error sequence between the pulse signals of the main reading head and the auxiliary reading head according to the pulse signal data set, obtain the error correction formula, and correct the switching signal data according to the error correction formula to obtain the corrected signal data;

[0014] The signal alignment module is used to calculate the time offset of the pulse signal of the main reading head and the auxiliary reading head in the joint area, and align the correction signal data in time series according to the time offset to obtain the alignment signal data.

[0015] Furthermore, the data continuity monitoring module includes:

[0016] A differential calculation unit, used to calculate the first-order differential values ​​of adjacent pulse signals in the main pulse signal data according to the pulse signal data set, to obtain a pulse signal differential sequence;

[0017] A fluctuation detection unit is used to calculate the second-order difference values ​​of adjacent first-order difference values ​​according to the pulse signal difference sequence to obtain the pulse signal trend sequence;

[0018] The abnormal signal judgment unit is used to generate an abnormal signal indication based on the pulse signal trend sequence and the preset differential threshold group. When the first-order differential value is greater than the preset first-order differential threshold and the second-order differential value is greater than the preset second-order differential threshold, the continuity of the main pulse signal data is abnormal.

[0019] Furthermore, the anomaly detection module includes:

[0020] The abnormal duration calculation unit is used to extract the corresponding main pulse signal data according to the abnormal signal indication, obtain the first abnormal signal data set, and set the first time window according to it, and then count the abnormal fluctuation signal in the first time window, determine the duration of the abnormal fluctuation signal, and obtain the abnormal duration data;

[0021] The abnormal stability analysis unit is used to perform statistical analysis on the distribution of the abnormal fluctuation signal in the first time window according to the abnormal duration data, calculate the abnormal signal change rate, and compare it with a preset stability threshold. When the abnormal signal change rate is greater than the preset stability threshold, the abnormal fluctuation signal is unstable, and an abnormal stability determination result is obtained;

[0022] The abnormality confirmation unit is used to calculate the abnormality comprehensive judgment value according to the abnormality duration data and the abnormality stability judgment result, and compare it with the preset failure judgment threshold. When the abnormality comprehensive judgment value is greater than the failure judgment threshold, an abnormality confirmation indication is generated.

[0023] Furthermore, the calculation formula of the abnormal comprehensive judgment value is: ,

[0024] in, is the comprehensive abnormality judgment value, is the duration of the abnormal fluctuation signal in the first time window, is the index of the time point in the first time window, is the number of abnormal fluctuation signal points in the first time window, is the index of the abnormal fluctuation signal point, The first time window The pulse signal value of the abnormal fluctuation signal point, The first time window The pulse signal value of the abnormal fluctuation signal point, is a natural constant, is the coefficient.

[0025] Furthermore, the signal difference calculation module includes:

[0026] An abnormal signal extraction unit is used to extract the corresponding main pulse signal data and auxiliary pulse signal data according to the abnormal confirmation indication, obtain a second abnormal signal data set, and set a second time window according to the second abnormal signal data set;

[0027] A signal instantaneous difference calculation unit is used to calculate the instantaneous difference of the signals of the main reading head and the auxiliary reading head at the same time point in the second time window according to the second abnormal signal data set to obtain an instantaneous difference value;

[0028] A sliding average difference calculation unit, used to calculate the average signal instantaneous difference of all time points in the second time window according to the instantaneous difference value, to obtain a sliding difference value;

[0029] The signal difference change rate calculation unit is used to calculate the change rate of the instantaneous difference of the signals of the main reading head and the auxiliary reading head in the second time window according to the instantaneous difference value to obtain the difference change rate.

[0030] Furthermore, the signal switching determination module includes:

[0031] An instantaneous difference threshold calculation unit sets an instantaneous difference threshold according to the operation accuracy of the device, and adjusts the instantaneous difference threshold according to the operation speed of the device to obtain a first signal switching threshold;

[0032] A sliding difference threshold calculation unit, used to calculate the mean and variance of the instantaneous difference value according to the second time window, and set the sliding difference threshold according to the mean and variance, and then adjust the sliding difference threshold according to the external environment temperature and humidity to obtain the second signal switching threshold;

[0033] A difference change rate threshold calculation unit is used to analyze the short-term change trend of the signal difference according to the instantaneous difference value, and set the difference change rate threshold according to it, and then adjust the difference change rate threshold according to the device operation mode to obtain a third signal switching threshold;

[0034] The switching indication generating unit is used to compare the signal difference data with the first signal switching threshold, the second signal switching threshold and the third signal switching threshold respectively, and generate a switching trigger indication when at least one signal difference data exceeds its corresponding signal switching threshold.

[0035] Furthermore, the smooth signal switching module includes:

[0036] A switching control unit, used to determine the switching time point between the main pulse signal data and the auxiliary pulse signal data according to the switching trigger indication, and generate a switching execution indication;

[0037] A weighted calculation unit, used to set a smoothing time window before and after the switching time point according to the switching execution instruction, and perform weighted calculation on the main pulse signal data and the auxiliary pulse signal data according to the smoothing time window to obtain transition signal data;

[0038] A first smoothing calculation unit, used to calculate the mean value of the transition signal data within the smoothing time window to obtain first smoothed signal data;

[0039] A second smoothing calculation unit, used for calculating the exponential weighted value of the first smoothed signal data to obtain second smoothed signal data;

[0040] The signal fusion unit is used to linearly superimpose the main pulse signal data and the auxiliary pulse signal data according to the second smooth signal data to obtain the switching signal data.

[0041] Furthermore, the error correction module includes:

[0042] An error extraction unit, used to calculate the error sequence between the main pulse signal data and the auxiliary pulse signal data according to the pulse signal data set to obtain error signal data;

[0043] The error correction formula unit is used to select a fitting function according to the error signal data set, optimize its parameters, and generate an error correction formula;

[0044] The correction calculation unit is used to input the switching signal data into the error correction formula, calculate the correction amount, and perform error compensation on the switching signal data according to the correction amount to obtain the corrected signal data.

[0045] Furthermore, the error correction formula unit includes:

[0046] A data screening unit is used to extract data points whose error stability meets a preset error threshold according to the error signal data set, remove noise data, and obtain an error training data set;

[0047] A function selection unit is used to perform trend analysis on the error training data set, determine the characteristics of the error changing over time, obtain the error characteristics, and select a fitting function based on the error characteristics, the fitting function includes a linear function, a polynomial function and an exponential function;

[0048] The parameter optimization unit is used to perform least squares fitting on the fitting function according to the error training data set, calculate the optimal parameter value of the fitting function, and ensure that the fitting error of the error correction formula is minimized;

[0049] The error correction formula generating unit is used to generate an error correction formula according to the optimal parameter value.

[0050] Furthermore, the signal alignment module includes:

[0051] A signal sampling unit, used for sampling the pulse signal data set to obtain a main pulse signal sample and a sub-pulse signal sample when the pulse signals of the main reading head and the sub-reading head enter the joint area;

[0052] A feature point extraction unit is used to extract key feature points of the pulse signal according to the main pulse signal sample and the auxiliary pulse signal sample to obtain key feature point data, wherein the key feature point data includes a signal peak point, a zero crossing point and a specific phase point;

[0053] A time offset calculation unit, used to calculate the time difference between the key feature points of the main pulse signal sample and the auxiliary pulse signal sample according to the key feature point data, to obtain the time offset;

[0054] The time interpolation unit is used to perform interpolation processing on the correction signal data according to the time offset to ensure that the time axes of the main pulse signal data and the auxiliary pulse signal data are aligned to obtain the aligned signal data.

[0055] The above solution of the present invention includes at least the following beneficial effects:

[0056] The present invention can effectively solve the signal jump problem caused by the gap in the splicing of magnetic stripes or the installation deviation of the induction switch. In the prior art, when the reading head passes through the joint area of ​​the magnetic stripes, instantaneous signal loss may occur, thereby affecting the accuracy of angle measurement. By introducing a first-order differential sequence, the system can identify the discontinuity of signal data. Even if the signal loss duration is short in some cases, abnormal fluctuations can be effectively detected. The system can also calculate the second-order differential value and then analyze the signal trend to ensure that even gradually changing signal anomalies can be detected. By generating abnormal signal indications, accurate input data is provided for subsequent abnormality detection, avoiding false triggering of signal switching due to short-term fluctuations, ensuring that the system will not erroneously switch signals due to accidental interference, thereby improving the system's adaptability to complex signal environments.

[0057] The present invention can distinguish between accidental noise and real signal failure, which is difficult to achieve with traditional methods. In the prior art, magnetic stripe splicing errors or inductive switch position errors may cause signal jumps, but some of these jumps are short-lived, while others indicate that the signal is no longer available. The system calculates the duration of the abnormal fluctuation signal to ensure that only when the signal abnormality lasts for a period of time will it be judged as a real signal failure. Through abnormal signal change rate analysis, the system can distinguish between stable signal abnormalities and short-term fluctuations to ensure that only when the signal failure reaches a certain stability will the abnormal confirmation indication be triggered. This multi-level judgment method avoids the problem that traditional systems are easily affected by short-term interference, improves the accuracy and robustness of signal abnormality judgment, and ensures that signal switching occurs only when it is really needed.

[0058] The present invention can ensure that signal switching occurs at the optimal time point, avoiding measurement errors caused by signal drift or premature or late switching. Signal switching often relies on a preset fixed threshold and cannot adapt to changes in the operating state of the equipment. The system dynamically calculates the instantaneous difference threshold, the sliding difference threshold and the difference change rate threshold to ensure intelligent adjustment of the signal switching timing. The system can set stricter thresholds to avoid measurement instability caused by premature switching. In the case of high-speed operation, the threshold can be lowered to ensure that the signal can be switched in time and avoid signal lag. The system can also dynamically adjust the threshold according to external environmental parameters to improve the reliability of signal switching and ensure that the system can maintain optimal performance under different operating environments.

[0059] The present invention can effectively solve the problem of angle measurement value jump caused by signal mutation. In the traditional method, when the system switches the reading head, the measurement value may change suddenly due to a certain deviation between the main and auxiliary signals, thereby affecting the stability of the measurement. The system ensures a smooth data transition during the signal switching process through weighted calculation and smoothing calculation. The system can dynamically weight the main and auxiliary signals to make the switching transition more natural and eliminate the impact caused by the signal mutation. The system can also adjust the size of the smoothing window according to different measurement modes, use a longer smoothing window in high-precision measurement, and use a shorter window in fast measurement to meet the needs of different scenarios and improve the stability of the measurement results.

[0060] The present invention can solve the problem of measurement inconsistency caused by errors between the main and auxiliary reading heads. In traditional methods, even if the switching signal timing is correct, there may still be systematic errors, causing the final measurement results to deviate from the true value. The system can quantify the systematic deviations between the main and auxiliary reading heads by calculating the error sequence, and use the fitting function to establish an error correction formula. The system can ensure that the fitting error of the error correction formula is minimized, thereby improving the measurement accuracy. It can also automatically eliminate noise data to avoid abnormal points affecting the reliability of error correction, ensure the consistency of data after signal switching, and improve the measurement stability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 It is a flow chart of an intelligent control system for pulse signal switching provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0062] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0063] like Figure 1 As shown, an embodiment of the present invention provides an intelligent control system for pulse signal switching, the system comprising:

[0064] The signal acquisition module is used to obtain the main pulse signal data of the main reading head and the auxiliary pulse signal data of the auxiliary reading head, and pre-process the data to remove redundant data to obtain a pulse signal data set;

[0065] A data continuity monitoring module is used to calculate the differential sequence of the main pulse signal data according to the pulse signal data set, and judge the continuity of the main pulse signal data according to the differential sequence, and generate an abnormal signal indication when an abnormal fluctuation signal is detected;

[0066] The abnormality detection module is used to detect the duration and stability of the abnormal fluctuation signal according to the abnormal signal indication, and judge whether the main pulse signal data has a failure risk based on it. If the result is yes, an abnormality confirmation indication is generated;

[0067] A signal difference calculation module, used to calculate the numerical difference between the main pulse signal data and the auxiliary pulse signal data according to the abnormal confirmation indication, and obtain signal difference data;

[0068] A signal switching determination module, used to calculate a signal switching threshold group according to the signal difference data, and generate a switching trigger indication when the signal difference data satisfies the signal switching threshold group;

[0069] A smooth signal switching module is used to switch the main pulse signal data and the auxiliary pulse signal data according to the switching trigger indication, and eliminate the signal mutation caused by the signal switching by calculating the smooth signal value to obtain the switching signal data;

[0070] An error correction module is used to extract the error sequence between the pulse signals of the main reading head and the auxiliary reading head according to the pulse signal data set, obtain the error correction formula, and correct the switching signal data according to the error correction formula to obtain the corrected signal data;

[0071] The signal alignment module is used to calculate the time offset of the pulse signal of the main reading head and the auxiliary reading head in the joint area, and align the correction signal data in time series according to the time offset to obtain the alignment signal data.

[0072] In an embodiment of the present invention, a signal acquisition module is used to obtain the main pulse signal data of the main reading head and the sub-pulse signal data of the sub-reading head, and pre-process them to remove redundant data to obtain a pulse signal data set, thereby ensuring the integrity of the input signal and providing high-quality data for subsequent signal analysis; a data continuity monitoring module is used to calculate the differential sequence of the main pulse signal data according to the pulse signal data set, and judge the continuity of the main pulse signal data based on it. When an abnormal fluctuation signal is detected, an abnormal signal indication is generated, which can monitor the continuity of the main pulse signal data in real time to ensure that the data will not be misjudged due to signal jumps; an abnormal detection module is used to detect the duration and stability of the abnormal fluctuation signal according to the abnormal signal indication, and judge whether the main pulse signal data has a risk of failure based on it. When the result is yes, an abnormal confirmation indication is generated, which can accurately distinguish between instantaneous noise and real signal abnormalities to avoid misjudgment.

[0073] The signal difference calculation module is used to calculate the numerical difference between the main pulse signal data and the auxiliary pulse signal data according to the abnormal confirmation indication, obtain the signal difference data, provide an objective numerical basis, and ensure that the signal switching can be judged based on the real-time signal difference; the signal switching judgment module is used to calculate the signal switching threshold group according to the signal difference data. When the signal difference data meets the signal switching threshold group, a switching trigger indication is generated. Through multi-level threshold judgment, the robustness of the signal switching decision is ensured and false triggering is reduced; the smooth signal switching module is used to switch the main pulse signal data and the auxiliary pulse signal data according to the switching trigger indication, and eliminate the signal mutation caused by the signal switching by calculating the smooth signal value to obtain the switching signal data, solve the signal mutation problem in the traditional signal switching, and make the switching process smooth transition.

[0074] The error correction module is used to extract the error sequence between the pulse signals of the main reading head and the auxiliary reading head according to the pulse signal data set, obtain the error correction formula, and correct the switching signal data according to it to obtain the corrected signal data, so as to solve the system error between the main and auxiliary signals, improve the data consistency after signal switching, and eliminate the deviation caused by the difference in characteristics of different reading heads; the signal alignment module is used to calculate the time offset of the pulse signals of the main reading head and the auxiliary reading head in the joint area, and align the time series of the corrected signal data according to it to obtain the aligned signal data, solve the time synchronization problem in the signal switching process, ensure that the measurement results do not have phase drift, and reduce the measurement error caused by time offset.

[0075] The signal acquisition module is used to obtain the main pulse signal data of the main reading head and the auxiliary pulse signal data of the auxiliary reading head, and pre-process them to remove redundant data to obtain a pulse signal data set, which specifically includes:

[0076] First, the system obtains pulse signal data through the main reading head and the auxiliary reading head respectively. Due to changes in the measurement environment and the influence of different reading head hardware characteristics, the original signal data may contain electromagnetic interference noise, instantaneous signal jitter or time offset caused by sampling error. To ensure data quality, the signal acquisition module adopts a fixed-frequency sampling mechanism. While ensuring a sufficiently high sampling rate, it dynamically adjusts the sampling frequency according to the operating speed of the rotating equipment to improve the validity of the data. When running at low speed, the sampling frequency is appropriately reduced to reduce redundant data; when running at high speed, the sampling frequency is increased to ensure the continuity of the signal. The collected data will be accompanied by an accurate timestamp and stored in the main signal buffer and the auxiliary signal buffer for subsequent processing.

[0077] The collected raw signal data may be subject to external interference and require noise filtering. The signal acquisition module first uses a bandpass filter to pre-process the signal to remove high-frequency noise and low-frequency interference in the environment. High-frequency noise is usually caused by electromagnetic interference and may appear as random bursts of signal fluctuations, while low-frequency interference may come from mechanical vibrations or temperature changes. To further improve signal stability, the system uses mean filtering or Kalman filtering to smooth the signal to make the signal curve more continuous and avoid short-term mutations that affect subsequent analysis. In addition, the signal quality assessment mechanism is used to calculate the mean and variance of the signal to filter out outliers and mark noise points to ensure that only stable signal data enters the subsequent processing flow.

[0078] The signal acquisition module further performs redundant data removal to reduce unnecessary data storage and computing burdens. Since the signals at adjacent time points may change slightly during the signal sampling process, some data points may belong to repeated sampling data. In order to remove redundant data, the system uses a sliding window detection algorithm to calculate the change rate of adjacent data points and set a threshold. When the change rate is less than the set value, the system automatically deletes duplicate data points. In addition, the system will filter out invalid data segments that remain unchanged for a long time, and adaptively adjust the error signal to ensure that the de-redundancy strategy can adapt to different working environments. When it is found that too much data removal leads to information loss, the system will automatically lower the de-redundancy standard. When there is too much redundant data, the system will increase the de-redundancy standard to ensure the integrity and validity of the data set.

[0079] After completing the above processing, the system finally generates a structured pulse signal data set. The formatted storage of the data set contains multiple fields, including timestamp, main pulse signal data, auxiliary pulse signal data and signal quality indicators (such as noise level, synchronization error, etc.). In addition, the system will calculate data integrity indicators, such as data missing rate, error mean square error, etc., to ensure that the quality of the final stored data meets the standards. If the data integrity is found to be insufficient, the system can automatically adjust the parameters of the previous steps to optimize the data acquisition process. Finally, the generated pulse signal data set will serve as the input of the subsequent signal processing module, providing a high-quality data foundation for signal continuity monitoring, anomaly detection, signal switching, etc.

[0080] In a preferred embodiment of the present invention, the data continuity monitoring module includes:

[0081] A differential calculation unit, used to calculate the first-order differential values ​​of adjacent pulse signals in the main pulse signal data according to the pulse signal data set, to obtain a pulse signal differential sequence;

[0082] A fluctuation detection unit is used to calculate the second-order difference values ​​of adjacent first-order difference values ​​according to the pulse signal difference sequence to obtain the pulse signal trend sequence;

[0083] The abnormal signal judgment unit is used to generate an abnormal signal indication based on the pulse signal trend sequence and the preset differential threshold group. When the first-order differential value is greater than the preset first-order differential threshold and the second-order differential value is greater than the preset second-order differential threshold, the continuity of the main pulse signal data is abnormal.

[0084] In an embodiment of the present invention, a differential calculation unit is used to calculate the first-order differential values ​​of adjacent pulse signals in the main pulse signal data according to the pulse signal data set to obtain a pulse signal differential sequence. By calculating the first-order differential sequence, the change amplitude between adjacent signal points can be quantified, thereby intuitively identifying whether the signal has undergone a mutation; a fluctuation detection unit is used to calculate the second-order differential values ​​of adjacent first-order differential values ​​according to the pulse signal differential sequence to obtain a pulse signal trend sequence. By calculating the second-order differential sequence, the change trend of the pulse signal can be analyzed, so that the system can not only detect mutation signals but also identify abnormal change rates; an abnormal signal determination unit is used to determine, according to the pulse signal trend sequence and a preset differential threshold group, that when the first-order differential value is greater than the preset first-order differential threshold and the second-order differential value is greater than the preset second-order differential threshold, the continuity of the main pulse signal data is abnormal, an abnormal signal indication is generated, and the signal change amplitude and trend are combined to ensure that the abnormal signal indication is triggered only when the signal actually undergoes a mutation.

[0085] The calculation formula of the first-order difference value is: ,

[0086] in, The first The first-order difference value at a time point, The first The pulse signal value at a time point, The first The pulse signal value at a time point.

[0087] The calculation formula of the second-order difference value is: ,

[0088] in, The first The second-order difference value at each time point, The first The first-order difference value at a time point, The first The first-order difference value at a time point.

[0089] In a preferred embodiment of the present invention, the anomaly detection module includes:

[0090] The abnormal duration calculation unit is used to extract the corresponding main pulse signal data according to the abnormal signal indication, obtain the first abnormal signal data set, and set the first time window according to it, and then count the abnormal fluctuation signal in the first time window, determine the duration of the abnormal fluctuation signal, and obtain the abnormal duration data;

[0091] The abnormal stability analysis unit is used to perform statistical analysis on the distribution of the abnormal fluctuation signal in the first time window according to the abnormal duration data, calculate the abnormal signal change rate, and compare it with a preset stability threshold. When the abnormal signal change rate is greater than the preset stability threshold, the abnormal fluctuation signal is unstable, and an abnormal stability determination result is obtained;

[0092] The abnormality confirmation unit is used to calculate the abnormality comprehensive judgment value according to the abnormality duration data and the abnormality stability judgment result, and compare it with the preset failure judgment threshold. When the abnormality comprehensive judgment value is greater than the failure judgment threshold, an abnormality confirmation indication is generated.

[0093] In an embodiment of the present invention, an abnormal duration calculation unit is used to extract the corresponding main pulse signal data according to the abnormal signal indication to obtain a first abnormal signal data set, and set a first time window according to the abnormal signal indication, and then perform statistics on the abnormal fluctuation signal in the first time window to determine the duration of the abnormal fluctuation signal to obtain abnormal duration data. By limiting the time window, the change characteristics of the abnormal signal can be accurately captured to avoid misjudging the abnormal situation due to the influence of short-term noise; the abnormal stability analysis unit is used to perform statistical analysis on the distribution of the abnormal fluctuation signal in the first time window according to the abnormal duration data, calculate the abnormal signal change rate, and compare it with a preset stability threshold. When the abnormal signal change rate is greater than the preset stability threshold, the abnormal fluctuation signal is unstable, and an abnormal stability determination result is obtained. By calculating the change rate of the abnormal signal, the method can further refine the classification of the abnormal signal to avoid misjudging short-term fluctuations as faults; the abnormal confirmation unit is used to calculate the abnormal comprehensive determination value according to the abnormal duration data and the abnormal stability determination result, and compare it with a preset failure threshold. When the abnormal comprehensive determination value is greater than the failure determination threshold, an abnormal confirmation indication is generated. By calculating the abnormal comprehensive determination value, the module can quantify the degree of abnormality to avoid misjudgment caused by subjective setting standards.

[0094] The abnormal stability analysis unit is used to perform statistical analysis on the distribution of the abnormal fluctuation signal in the first time window according to the abnormal duration data, calculate the abnormal signal change rate, and compare it with the preset stability threshold. When the abnormal signal change rate is greater than the preset stability threshold, the abnormal fluctuation signal is unstable, and the abnormal stability determination result is obtained, which specifically includes:

[0095] The unit receives abnormal signal data provided by the abnormal duration calculation unit, the data including all abnormal signal points detected in the first time window and their time stamp information;

[0096] The data is stored in the abnormal signal data buffer area and sorted in chronological order to ensure that all abnormal signal points are arranged along the time axis to form a time series data set;

[0097] Calculate the signal change rate between adjacent abnormal signal points;

[0098] Calculate the average rate of change, that is, the mean rate of change of abnormal signal points;

[0099] Calculate the variance of the rate of change to measure the degree of dispersion of abnormal signal changes;

[0100] The stability coefficient of the abnormal signal change rate is calculated using the stability determination formula:

[0101] Stability coefficient , is the minimum constant;

[0102] Comparison is done by pre-set stability thresholds:

[0103] If the stability coefficient is greater than the preset stability threshold, the abnormal fluctuation signal is unstable, and the abnormal stability determination result is generated as unstable;

[0104] If the stability coefficient is not greater than the preset stability threshold, the abnormal fluctuation signal is stable, and the abnormal stability determination result generated is stable.

[0105] In a preferred embodiment of the present invention, the calculation formula of the abnormal comprehensive judgment value is: ,

[0106] in, is the comprehensive abnormality judgment value, is the duration of the abnormal fluctuation signal in the first time window, is the index of the time point in the first time window, is the number of abnormal fluctuation signal points in the first time window, is the index of the abnormal fluctuation signal point, The first time window The pulse signal value of the abnormal fluctuation signal point, The first time window The pulse signal value of the abnormal fluctuation signal point, is a natural constant, is the coefficient.

[0107] In an embodiment of the present invention, the calculation formula of the abnormal comprehensive judgment value is to obtain a quantitative index, namely, the abnormal comprehensive judgment value, by comprehensively analyzing the duration of the abnormal fluctuation signal and the signal value change. The index is used to judge whether the current abnormal fluctuation is a short-term interference or a long-term abnormality that affects the signal switching, and provide an accurate judgment basis for the abnormal detection module; the formula can more accurately identify long-term anomalies and improve the stability of signal switching by dynamically calculating the duration of the abnormal signal; by summing up the abnormal signal values, it can identify those abnormal situations that are short-term but have drastic changes in signal amplitude; the introduction of the exponential weight factor makes the abnormal signals at closer time points have a greater impact on the comprehensive judgment value, while the abnormal impact at farther time points gradually weakens; through the calculation of this formula, the system can more intelligently judge the signal abnormality, avoid false triggering of switching or delayed switching, and ensure that the system can maintain the best measurement accuracy under different operating conditions.

[0108] In a preferred embodiment of the present invention, the signal difference calculation module includes:

[0109] An abnormal signal extraction unit is used to extract the corresponding main pulse signal data and auxiliary pulse signal data according to the abnormal confirmation indication, obtain a second abnormal signal data set, and set a second time window according to the second abnormal signal data set;

[0110] A signal instantaneous difference calculation unit is used to calculate the instantaneous difference of the signals of the main reading head and the auxiliary reading head at the same time point in the second time window according to the second abnormal signal data set to obtain an instantaneous difference value;

[0111] A sliding average difference calculation unit, used to calculate the average signal instantaneous difference of all time points in the second time window according to the instantaneous difference value, to obtain a sliding difference value;

[0112] The signal difference change rate calculation unit is used to calculate the change rate of the instantaneous difference of the signals of the main reading head and the auxiliary reading head in the second time window according to the instantaneous difference value to obtain the difference change rate.

[0113] In an embodiment of the present invention, an abnormal signal extraction unit is used to extract the corresponding main pulse signal data and auxiliary pulse signal data according to the abnormal confirmation indication, obtain the second abnormal signal data set, and set the second time window according to it, so as to ensure that the signal analysis is only for the abnormal area without affecting the normal signal data, thereby improving the calculation efficiency; a signal instantaneous difference calculation unit is used to calculate the instantaneous difference of the signal of the main reading head and the auxiliary reading head at the same time point in the second time window according to the second abnormal signal data set, and obtain the instantaneous difference value. By calculating the signal difference point by point, the measurement deviation of the main and auxiliary reading heads at each time point can be intuitively reflected, and basic data can be provided for signal switching judgment; a sliding average difference calculation unit is used to calculate the average signal instantaneous difference of all time points in the second time window according to the instantaneous difference value, and obtain the sliding difference value. By sliding average calculation, short-term fluctuations in the signal can be smoothed out and a stable signal deviation trend can be extracted; a signal difference change rate calculation unit is used to calculate the change rate of the instantaneous difference of the signal of the main reading head and the auxiliary reading head in the second time window according to the instantaneous difference value, and obtain the difference change rate. By calculating the difference change rate, the dynamic characteristics of the signal deviation can be quantified, so as to ensure that measurement abnormalities will not be caused by sudden changes when the signal is switched.

[0114] In a preferred embodiment of the present invention, the signal switching determination module includes:

[0115] An instantaneous difference threshold calculation unit sets an instantaneous difference threshold according to the operation accuracy of the device, and adjusts the instantaneous difference threshold according to the operation speed of the device to obtain a first signal switching threshold;

[0116] A sliding difference threshold calculation unit, used to calculate the mean and variance of the instantaneous difference value according to the second time window, and set the sliding difference threshold according to the mean and variance, and then adjust the sliding difference threshold according to the external environment temperature and humidity to obtain the second signal switching threshold;

[0117] A difference change rate threshold calculation unit is used to analyze the short-term change trend of the signal difference according to the instantaneous difference value, and set the difference change rate threshold according to it, and then adjust the difference change rate threshold according to the device operation mode to obtain a third signal switching threshold;

[0118] The switching indication generating unit is used to compare the signal difference data with the first signal switching threshold, the second signal switching threshold and the third signal switching threshold respectively, and generate a switching trigger indication when at least one signal difference data exceeds its corresponding signal switching threshold.

[0119] In the embodiment of the present invention, the instantaneous difference threshold calculation unit sets the instantaneous difference threshold according to the operation accuracy of the device, and adjusts the instantaneous difference threshold according to the operation speed of the device to obtain the first signal switching threshold, which can dynamically adjust the switching condition according to the current state of the device, avoiding the problem that the fixed threshold cannot adapt to environmental changes; the sliding difference threshold calculation unit is used to calculate the mean and variance of the instantaneous difference value according to the second time window, and set the sliding difference threshold according to the mean and variance, and then adjust the sliding difference threshold according to the external environment temperature and humidity to obtain the second signal switching threshold. By combining the environmental data, the threshold can be automatically adjusted when the external conditions change, ensuring the reliability and stability of the signal switching; the difference change rate threshold A value calculation unit is used to analyze the short-term change trend of the signal difference according to the instantaneous difference value, and set the difference change rate threshold according to it, and then adjust the difference change rate threshold according to the equipment operation mode to obtain the third signal switching threshold, and dynamically adjust the threshold in combination with the equipment operation mode to ensure that the system can accurately judge the timing of signal switching under different working conditions; a switching indication generation unit is used to compare the signal difference data with the first signal switching threshold, the second signal switching threshold and the third signal switching threshold respectively, and generate a switching trigger indication when at least one signal difference data exceeds its corresponding signal switching threshold, thereby ensuring the robustness of the signal switching, avoiding switching based on only a single threshold, and reducing the possibility of false triggering.

[0120] The instantaneous difference threshold calculation unit sets the instantaneous difference threshold according to the operation accuracy of the device, and adjusts the instantaneous difference threshold according to the operation speed of the device to obtain the first signal switching threshold, which specifically includes:

[0121] During the system initialization phase, an initial instantaneous difference threshold is set based on the design accuracy of the equipment, environmental factors, and historical operating data.

[0122] Collect the pulse signal data of the main reading head and the auxiliary reading head, and calculate their instantaneous difference at the same time point;

[0123] Calculate the maximum and minimum instantaneous differences over a period of time, set a safety factor, and dynamically adjust the threshold accordingly;

[0124] If an abnormal signal occurs during the operation of the equipment, the safety factor will be adjusted to adapt to different operating conditions.

[0125] The sliding difference threshold calculation unit is used to calculate the mean and variance of the instantaneous difference value according to the second time window, and set the sliding difference threshold according to the mean and variance, and then adjust the sliding difference threshold according to the external environment temperature and humidity to obtain the second signal switching threshold, which specifically includes:

[0126] Setting a second time window, and extracting all instantaneous difference values ​​within the time window from the data of the primary and secondary reading heads;

[0127] Calculate the mean and standard deviation of the instantaneous difference values ​​within the time window, set the threshold factor, and calculate the sliding difference threshold;

[0128] Collect external environmental data, such as temperature, humidity, and vibration intensity, and adjust thresholds based on environmental parameters.

[0129] The difference change rate threshold calculation unit is used to analyze the short-term change trend of the signal difference according to the instantaneous difference value, and set the difference change rate threshold according to it, and then adjust the difference change rate threshold according to the device operation mode to obtain the third signal switching threshold, which specifically includes:

[0130] Calculate the first derivative of the instantaneous difference of the signal, i.e. the rate of change;

[0131] Set the time window and calculate the mean and standard deviation of the rate of change;

[0132] Set the threshold factor and calculate the difference change rate threshold;

[0133] The threshold is adjusted dynamically based on the device operation mode, such as high-speed rotation mode and low-speed precision mode.

[0134] In a preferred embodiment of the present invention, the smooth signal switching module includes:

[0135] A switching control unit, used to determine the switching time point between the main pulse signal data and the auxiliary pulse signal data according to the switching trigger indication, and generate a switching execution indication;

[0136] A weighted calculation unit, used to set a smoothing time window before and after the switching time point according to the switching execution instruction, and perform weighted calculation on the main pulse signal data and the auxiliary pulse signal data according to the smoothing time window to obtain transition signal data;

[0137] A first smoothing calculation unit, used to calculate the mean value of the transition signal data within the smoothing time window to obtain first smoothed signal data;

[0138] A second smoothing calculation unit, used for calculating the exponential weighted value of the first smoothed signal data to obtain second smoothed signal data;

[0139] The signal fusion unit is used to linearly superimpose the main pulse signal data and the auxiliary pulse signal data according to the second smooth signal data to obtain the switching signal data.

[0140] In an embodiment of the present invention, a switching control unit is used to determine the switching time point between the main pulse signal data and the auxiliary pulse signal data according to the switching trigger indication, generate a switching execution indication, and avoid switching when there is a large deviation between the main signal and the auxiliary signal by optimizing the switching timing, thereby reducing the impact of signal mutation; a weighted calculation unit is used to set a smoothing time window before and after the switching time point according to the switching execution indication, and perform weighted calculation on the main pulse signal data and the auxiliary pulse signal data according to the smoothing time window to obtain transition signal data, and ensure that the signals are smoothly connected at the switching point by dynamically adjusting the weights of the main and auxiliary signals, effectively eliminating errors caused by mutations; a first smoothing calculation unit is used to calculate the transition signal data in the smoothing time window. The first smoothed signal data is obtained by calculating the mean value within the time window, and the high-frequency noise in the signal is reduced through sliding average calculation, the signal stability is improved, and it is ensured that there will be no drastic fluctuations in the signal transition process; the second smoothing calculation unit is used to calculate the exponential weighted value of the first smoothed signal data to obtain the second smoothed signal data, which can reduce the instantaneous fluctuation of the signal while retaining the overall trend information of the signal to avoid the signal delay problem caused by over-smoothing; the signal fusion unit is used to linearly superimpose the main pulse signal data and the auxiliary pulse signal data according to the second smoothed signal data to obtain the switching signal data, so as to ensure that the final signal data takes into account both smoothness and real-time performance, avoid signal lag due to over-smoothing, and avoid mutation problems caused by directly using the original signal.

[0141] Among them, the weighted calculation unit is used to set a smooth time window before and after the switching time point according to the switching execution instruction, and perform weighted calculation on the main pulse signal data and the auxiliary pulse signal data according to the smooth time window to obtain the transition signal data, specifically including:

[0142] After receiving the switching execution instruction, a smoothing time window is set, that is, a certain time range before and after the switching, such as 10 to 50 pulse signal cycles, to ensure that there are enough data points for smoothing calculation.

[0143] The weighted average calculation method is used to calculate the main pulse signal data and the auxiliary pulse signal data point by point to ensure the smoothness of the signal transition:

[0144] In the first half of the smoothing window, the main signal has a higher weight and the secondary signal has a lower weight;

[0145] Near the switching point, the two weights are equal;

[0146] In the second half of the smoothing window, the weight of the secondary signal gradually increases, while the weight of the primary signal decreases;

[0147] The transition signal data is calculated and input into the subsequent processing module as the smoothed signal data.

[0148] The second smoothing calculation unit is used to calculate the exponential weighted value of the first smoothed signal data to obtain the second smoothed signal data, specifically including:

[0149] First, set the smoothing factor, whose value range is between 0 and 1, which determines the weighted ratio of the current signal data to the historical smoothing data. When the smoothing factor is small, the historical data has a greater impact on the current smoothing result, making the signal change smoother; when the smoothing factor is large, the current data has a stronger impact on the smoothing result, making the signal respond faster. In order to ensure a smooth transition of the signal, a suitable value is usually selected, such as 0.2 to 0.3, so that the signal can respond quickly to changes without being affected by sudden noise.

[0150] Through the recursive formula calculation, each new smoothed signal value is not only affected by the current signal, but also constrained by the historical smoothed signal, thereby ensuring the smoothness of the signal transition. In the actual calculation process, each data point is recursively calculated in turn until all data points are processed.

[0151] The signal fusion unit is used to linearly superimpose the main pulse signal data and the auxiliary pulse signal data according to the second smooth signal data to obtain the switching signal data, which specifically includes:

[0152] First, the second smoothed signal data is received, and the current main pulse signal data and auxiliary pulse signal data are obtained at the same time, and the signal fusion window is determined. The size of the signal fusion window depends on the measurement accuracy requirements of the system. If the signal smooth transition time is required to be longer, the signal fusion window can be appropriately increased to improve the smoothness; if a faster response is required, the window length can be shortened to reduce signal lag. After determining the window range, the unit uses a linear superposition calculation method to weightedly fuse the second smoothed signal data with the original pulse signal data.

[0153] In order to adapt to different signal switching scenarios, the unit adopts a dynamic weight adjustment strategy to make the weight factor change gradually over time. For example, in the initial stage of signal switching, the weight factor takes a smaller value, so that the original signal data occupies a larger proportion to maintain the real-time nature of the signal; in the middle stage of signal switching, the weight factor gradually increases, so that the contribution ratio of the second smooth signal increases, thereby reducing the impact of mutations on the measured signal; after the switching is completed, the weight factor is close to 1, so that the signal is completely dominated by the smooth signal, thereby ensuring the stability of the signal after the switching is completed.

[0154] In a preferred embodiment of the present invention, the error correction module includes:

[0155] An error extraction unit, used to calculate the error sequence between the main pulse signal data and the auxiliary pulse signal data according to the pulse signal data set to obtain error signal data;

[0156] The error correction formula unit is used to select a fitting function according to the error signal data set, optimize its parameters, and generate an error correction formula;

[0157] The correction calculation unit is used to input the switching signal data into the error correction formula, calculate the correction amount, and perform error compensation on the switching signal data according to the correction amount to obtain the corrected signal data.

[0158] In the embodiment of the present invention, the error extraction unit is used to calculate the error sequence between the main pulse signal data and the auxiliary pulse signal data according to the pulse signal data set to obtain the error signal data, thereby ensuring that the errors of the main and auxiliary reading head signals can be accurately quantified and providing basic data for subsequent error correction;

[0159] The error correction formula unit is used to select a fitting function according to the error signal data set, optimize its parameters, and generate an error correction formula. By selecting the fitting function, the error correction model can accurately predict and compensate for the deviation between the main signal and the auxiliary signal;

[0160] The correction calculation unit is used to input the switching signal data into the error correction formula, calculate the correction amount, and perform error compensation on the switching signal data according to the correction amount to obtain the corrected signal data, thereby ensuring that after the main and auxiliary signals are switched, the signal curve will not jump or deviate, thereby improving the continuity and consistency of the signal data.

[0161] In a preferred embodiment of the present invention, the error correction formula unit includes:

[0162] A data screening unit is used to extract data points whose error stability meets a preset error threshold according to the error signal data set, remove noise data, and obtain an error training data set;

[0163] A function selection unit is used to perform trend analysis on the error training data set, determine the characteristics of the error changing over time, obtain the error characteristics, and select a fitting function based on the error characteristics, the fitting function includes a linear function, a polynomial function and an exponential function;

[0164] The parameter optimization unit is used to perform least squares fitting on the fitting function according to the error training data set, calculate the optimal parameter value of the fitting function, and ensure that the fitting error of the error correction formula is minimized;

[0165] The error correction formula generating unit is used to generate an error correction formula according to the optimal parameter value.

[0166] In an embodiment of the present invention, a data screening unit is used to extract data points whose error stability meets a preset error threshold according to an error signal data set, remove noise data, and obtain an error training data set, which can effectively remove accidental errors and sudden noise, and ensure that the error data has good continuity and stability; a function selection unit is used to perform trend analysis on the error training data set, determine the characteristics of the error changing over time, obtain the error characteristics, and select a fitting function based on the trend analysis, and the fitting function includes a linear function, a polynomial function, and an exponential function. Through mathematical analysis of the error signal, the most suitable fitting function can be accurately selected, so that the error correction model is more in line with the actual situation; a parameter optimization unit is used to perform least squares fitting on the fitting function according to the error training data set, calculate the optimal parameter value of the fitting function, ensure that the fitting error of the error correction formula is minimized, and ensure that the error correction formula can accurately reflect the change law of the error signal, thereby effectively compensating for the error, reducing the measurement deviation caused by equipment drift or system noise, and improving the stability and consistency of the measurement; an error correction formula generation unit is used to generate an error correction formula according to the optimal parameter value, and the generation of the error correction formula enables the measurement system to adapt to the error change and effectively compensate for the system error between the primary and secondary signals.

[0167] The function selection unit is used to perform trend analysis on the error training data set, determine the characteristics of the error changing over time, obtain the error characteristics, and select the fitting function based on the error characteristics, specifically including:

[0168] First, the system performs a time series analysis on the error training data set to extract the changing trend of the error signal. The system extracts the error signal values ​​at multiple time points from the error training data set and constructs the error change curve in chronological order. In order to evaluate the growth pattern of the error, the system calculates the first-order derivative of the error signal to analyze the instantaneous rate of change of the error. If the first-order derivative of the error is a constant, it indicates that the error change shows a linear growth or linear decay trend, which is suitable for fitting with a linear function. If the first-order derivative changes with time, the second-order derivative of the error is further calculated to determine the acceleration change trend of the error. If the second-order derivative is close to a constant, it means that the error changes with time in accordance with the quadratic polynomial trend, which is suitable for quadratic or higher-order polynomial fitting; if the second-order derivative decays or grows exponentially, it indicates that the error trend may conform to the change pattern of the exponential function, which is suitable for fitting with an exponential function.

[0169] Secondly, in order to improve the accuracy of error feature analysis, the system smoothes the error data. Common methods include moving average or exponential smoothing, which are used to reduce the impact of random noise and make the error change trend more obvious. On the smoothed error signal data, the system calculates the mean and variance of the error distribution to evaluate the fluctuation range of the error data. If the fluctuation range of the error value is small and the error mean is stable in a certain interval, it means that the error conforms to the simple linear model; if the variance of the error value is large and the error mean changes greatly over time, a more complex function needs to be used for fitting.

[0170] After completing the error characteristic analysis, the system selects a suitable fitting function based on the time variation law of the error. The linear function is suitable for the case where the error value grows or decays uniformly over time. The fitting function form is ; Polynomial functions are suitable for situations where the error value changes in a curve, such as the quadratic polynomial function ; The exponential function is suitable for situations where the error grows or decays exponentially over time. The fitting function form is .

[0171] The parameter optimization unit is used to perform least squares fitting on the fitting function according to the error training data set, calculate the optimal parameter value of the fitting function, and ensure that the fitting error of the error correction formula is minimized, specifically including:

[0172] First, the system constructs the error fitting function and initializes the fitting parameters. For example, the error fitting function selected is , are the parameters to be optimized. The system needs to calculate the optimal values ​​of these parameters to minimize the fitting error.

[0173] In order to solve the optimal parameters, the system first uses the least squares method to calculate the initial values ​​of the parameters. The least squares method constructs a set of linear equations to minimize the sum of squared errors.

[0174] Constructing the error equation ,

[0175] in, , , , obtained from the error training set, is the time point matrix, is the actual value of the error sequence;

[0176] Solving the error equation yields, , , ;

[0177] The error formula of the preliminary fitting is ;

[0178] After obtaining the parameters estimated by the least squares method, the system further uses the gradient descent algorithm to optimize the parameters. The core idea of ​​gradient descent is to calculate the partial derivatives of the objective function with respect to each parameter and update the parameters according to the gradient direction.

[0179] Update the parameters by gradient descent: ,

[0180] in, For the parameters, is the index of the parameter, is the learning rate, , is the loss function, and the loss function is: ,

[0181] By calculating the loss function relative to The gradient of , after the preset number of optimizations, the final optimization parameter is , , 8;

[0182] The final error formula is .

[0183] In a preferred embodiment of the present invention, the signal alignment module includes:

[0184] A signal sampling unit, used for sampling the pulse signal data set to obtain a main pulse signal sample and a sub-pulse signal sample when the pulse signals of the main reading head and the sub-reading head enter the joint area;

[0185] A feature point extraction unit is used to extract key feature points of the pulse signal according to the main pulse signal sample and the auxiliary pulse signal sample to obtain key feature point data, wherein the key feature point data includes a signal peak point, a zero crossing point and a specific phase point;

[0186] A time offset calculation unit, used to calculate the time difference between the key feature points of the main pulse signal sample and the auxiliary pulse signal sample according to the key feature point data, to obtain the time offset;

[0187] The time interpolation unit is used to perform interpolation processing on the correction signal data according to the time offset to ensure that the time axes of the main pulse signal data and the auxiliary pulse signal data are aligned to obtain the aligned signal data.

[0188] In an embodiment of the present invention, a signal sampling unit is used to sample a pulse signal data set when the pulse signals of the main reading head and the auxiliary reading head enter the joint area to obtain a main pulse signal sample and an auxiliary pulse signal sample. By setting a fixed sampling window, the complete change of the main and auxiliary signals in the joint area can be captured, providing reliable data for subsequent feature point extraction and time alignment; a feature point extraction unit is used to extract key feature points of the pulse signal according to the main pulse signal sample and the auxiliary pulse signal sample to obtain key feature point data. The key feature point data includes a signal peak point, a zero crossing point and a specific phase point. By extracting the key feature points, the periodic characteristics of the signal can be accurately characterized, and the feature point extraction and time alignment can be provided. High signal alignment accuracy; a time offset calculation unit, which is used to calculate the time difference between the key feature points of the main pulse signal sample and the auxiliary pulse signal sample according to the key feature point data, and obtain the time offset. By calculating the time offset of the key feature point, the time difference between the main and auxiliary signals can be accurately measured to provide basic data for subsequent signal alignment; a time interpolation unit, which is used to interpolate the corrected signal data according to the time offset to ensure that the time axis of the main pulse signal data and the auxiliary pulse signal data are aligned to obtain the aligned signal data. Through time interpolation, the time reference of the signal before and after switching is guaranteed to be consistent, avoiding measurement errors caused by signal phase asynchrony.

[0189] Among them, the key features include:

[0190] Signal peak point: the position where the pulse signal reaches the maximum or minimum value;

[0191] Zero crossing point: The crossing point where the signal changes from positive to negative or from negative to positive;

[0192] Specific phase point: Select the signal value of a specific phase based on the periodic changes of the main and auxiliary signals.

[0193] The calculation formula of the time offset is: ,

[0194] in, is the time offset, is the total number of key feature points, is the index of the key feature point, The first sample of the main pulse signal The timestamp of the key feature points, is the main pulse signal sample, is the sub-pulse signal sample The timestamp of the key feature points, is the sub-pulse signal sample, is the coefficient.

[0195] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. An intelligent control system for pulse signal switching, characterized in that: The system comprises: The signal acquisition module is used to obtain the main pulse signal data of the main reading head and the auxiliary pulse signal data of the auxiliary reading head, and pre-process the data to remove redundant data to obtain a pulse signal data set; A data continuity monitoring module is used to calculate the differential sequence of the main pulse signal data according to the pulse signal data set, and judge the continuity of the main pulse signal data according to the differential sequence, and generate an abnormal signal indication when an abnormal fluctuation signal is detected; The abnormality detection module is used to detect the duration and stability of the abnormal fluctuation signal according to the abnormal signal indication, and judge whether the main pulse signal data has a failure risk based on it. If the result is yes, an abnormality confirmation indication is generated; A signal difference calculation module, used to calculate the numerical difference between the main pulse signal data and the auxiliary pulse signal data according to the abnormal confirmation indication, and obtain signal difference data; A signal switching determination module, used to calculate a signal switching threshold group according to the signal difference data, and generate a switching trigger indication when the signal difference data satisfies the signal switching threshold group; A smooth signal switching module is used to switch the main pulse signal data and the auxiliary pulse signal data according to the switching trigger indication, and eliminate the signal mutation caused by the signal switching by calculating the smooth signal value to obtain the switching signal data; The error correction module is used to extract the error sequence between the pulse signals of the main reading head and the auxiliary reading head according to the pulse signal data set, obtain the error correction formula, and correct the switching signal data according to the error correction formula to obtain the corrected signal data; The signal alignment module is used to calculate the time offset of the pulse signal of the main reading head and the auxiliary reading head in the joint area, and align the correction signal data in time series according to the time offset to obtain the alignment signal data.

2. An intelligent control system for pulse signal switching according to claim 1, characterized in that: The data continuity monitoring module includes: A differential calculation unit, used to calculate the first-order differential values ​​of adjacent pulse signals in the main pulse signal data according to the pulse signal data set, to obtain a pulse signal differential sequence; A fluctuation detection unit is used to calculate the second-order difference values ​​of adjacent first-order difference values ​​according to the pulse signal difference sequence to obtain the pulse signal trend sequence; The abnormal signal judgment unit is used to generate an abnormal signal indication based on the pulse signal trend sequence and the preset differential threshold group. When the first-order differential value is greater than the preset first-order differential threshold and the second-order differential value is greater than the preset second-order differential threshold, the continuity of the main pulse signal data is abnormal.

3. An intelligent control system for pulse signal switching according to claim 2, characterized in that: The anomaly detection module comprises: The abnormal duration calculation unit is used to extract the corresponding main pulse signal data according to the abnormal signal indication, obtain the first abnormal signal data set, and set the first time window according to it, and then count the abnormal fluctuation signal in the first time window, determine the duration of the abnormal fluctuation signal, and obtain the abnormal duration data; The abnormal stability analysis unit is used to perform statistical analysis on the distribution of the abnormal fluctuation signal in the first time window according to the abnormal duration data, calculate the abnormal signal change rate, and compare it with a preset stability threshold. When the abnormal signal change rate is greater than the preset stability threshold, the abnormal fluctuation signal is unstable, and an abnormal stability determination result is obtained; The abnormality confirmation unit is used to calculate the abnormality comprehensive judgment value according to the abnormality duration data and the abnormality stability judgment result, and compare it with the preset failure judgment threshold. When the abnormality comprehensive judgment value is greater than the failure judgment threshold, an abnormality confirmation indication is generated.

4. An intelligent control system for pulse signal switching according to claim 3, characterized in that: The calculation formula of the abnormal comprehensive judgment value is: ， in, is the comprehensive abnormality judgment value, is the duration of the abnormal fluctuation signal in the first time window, is the index of the time point in the first time window, is the number of abnormal fluctuation signal points in the first time window, is the index of the abnormal fluctuation signal point, The first time window The pulse signal value of the abnormal fluctuation signal point, The first time window The pulse signal value of the abnormal fluctuation signal point, is a natural constant, is the coefficient.

5. An intelligent control system for pulse signal switching according to claim 4, characterized in that: The signal difference calculation module includes: An abnormal signal extraction unit is used to extract the corresponding main pulse signal data and auxiliary pulse signal data according to the abnormal confirmation indication, obtain a second abnormal signal data set, and set a second time window according to the second abnormal signal data set; A signal instantaneous difference calculation unit is used to calculate the instantaneous difference of the signals of the main reading head and the auxiliary reading head at the same time point in the second time window according to the second abnormal signal data set to obtain an instantaneous difference value; A sliding average difference calculation unit, used to calculate the average signal instantaneous difference of all time points in the second time window according to the instantaneous difference value, to obtain a sliding difference value; The signal difference change rate calculation unit is used to calculate the change rate of the instantaneous difference of the signals of the main reading head and the auxiliary reading head in the second time window according to the instantaneous difference value to obtain the difference change rate.

6. An intelligent control system for pulse signal switching according to claim 5, characterized in that: The signal switching determination module comprises: An instantaneous difference threshold calculation unit sets an instantaneous difference threshold according to the operation accuracy of the device, and adjusts the instantaneous difference threshold according to the operation speed of the device to obtain a first signal switching threshold; A sliding difference threshold calculation unit, used to calculate the mean and variance of the instantaneous difference value according to the second time window, and set the sliding difference threshold according to the mean and variance, and then adjust the sliding difference threshold according to the external environment temperature and humidity to obtain the second signal switching threshold; A difference change rate threshold calculation unit is used to analyze the short-term change trend of the signal difference according to the instantaneous difference value, and set the difference change rate threshold according to it, and then adjust the difference change rate threshold according to the device operation mode to obtain a third signal switching threshold; The switching indication generating unit is used to compare the signal difference data with the first signal switching threshold, the second signal switching threshold and the third signal switching threshold respectively, and generate a switching trigger indication when at least one signal difference data exceeds its corresponding signal switching threshold.

7. An intelligent control system for pulse signal switching according to claim 6, characterized in that: The smooth signal switching module comprises: A switching control unit, used to determine the switching time point between the main pulse signal data and the auxiliary pulse signal data according to the switching trigger indication, and generate a switching execution indication; A weighted calculation unit, used to set a smoothing time window before and after the switching time point according to the switching execution instruction, and perform weighted calculation on the main pulse signal data and the auxiliary pulse signal data according to the smoothing time window to obtain transition signal data; A first smoothing calculation unit, used to calculate the mean value of the transition signal data within the smoothing time window to obtain first smoothed signal data; A second smoothing calculation unit, used for calculating the exponential weighted value of the first smoothed signal data to obtain second smoothed signal data; The signal fusion unit is used to linearly superimpose the main pulse signal data and the auxiliary pulse signal data according to the second smooth signal data to obtain the switching signal data.

8. An intelligent control system for pulse signal switching according to claim 7, characterized in that: The error correction module comprises: An error extraction unit, used to calculate the error sequence between the main pulse signal data and the auxiliary pulse signal data according to the pulse signal data set to obtain error signal data; The error correction formula unit is used to select a fitting function according to the error signal data set, optimize its parameters, and generate an error correction formula; The correction calculation unit is used to input the switching signal data into the error correction formula, calculate the correction amount, and perform error compensation on the switching signal data according to the correction amount to obtain the corrected signal data.

9. An intelligent control system for pulse signal switching according to claim 8, characterized in that: The error correction formula unit includes: A data screening unit is used to extract data points whose error stability meets a preset error threshold according to the error signal data set, remove noise data, and obtain an error training data set; A function selection unit is used to perform trend analysis on the error training data set, determine the characteristics of the error changing over time, obtain the error characteristics, and select a fitting function based on the error characteristics. The fitting function includes a linear function, a polynomial function, and an exponential function. The parameter optimization unit is used to perform least squares fitting on the fitting function according to the error training data set, calculate the optimal parameter value of the fitting function, and ensure that the fitting error of the error correction formula is minimized; The error correction formula generating unit is used to generate an error correction formula according to the optimal parameter value.

10. An intelligent control system for pulse signal switching according to claim 9, characterized in that: The signal alignment module comprises: A signal sampling unit, used for sampling the pulse signal data set to obtain a main pulse signal sample and a sub-pulse signal sample when the pulse signals of the main reading head and the sub-reading head enter the joint area; A feature point extraction unit is used to extract key feature points of the pulse signal according to the main pulse signal sample and the auxiliary pulse signal sample to obtain key feature point data, wherein the key feature point data includes a signal peak point, a zero crossing point and a specific phase point; A time offset calculation unit, used to calculate the time difference between the key feature points of the main pulse signal sample and the auxiliary pulse signal sample according to the key feature point data, to obtain the time offset; The time interpolation unit is used to perform interpolation processing on the correction signal data according to the time offset to ensure that the time axes of the main pulse signal data and the auxiliary pulse signal data are aligned to obtain the aligned signal data.

Images