A method for improving anti-interference capability of a PLC electric control system

Abstract

The present application relates to the technical field of servo control, and particularly relates to a method for improving the anti-interference capability of a PLC electrical control system. The method performs orthogonal decomposition on the DC bus voltage fluctuation rate sequence of all servo axes, extracts common mode fluctuation characteristics to quantify the source activity independence of each axis, combines the source activity independence, the tracking error fluctuation difference, and the non-linear form similarity between the DC bus voltage fluctuation rate and the clock compensation response to identify the dominant interference path in the current system; analyzes the main interference source activity concentration and predicts the risk to perform targeted compensation for the main victim axis. The present application realizes more accurate interference attribution by common mode decoupling of the underlying operation data and fusing multi-source evidence to identify the dominant interference path and predictive compensation, suppresses internal interference while ensuring system control stability, and improves the synchronization accuracy and anti-interference robustness of the control system in a complex electromagnetic environment.

Description

Technical Field

[0001] This invention relates to the field of servo control technology, and more specifically to a method for improving the anti-interference capability of a PLC electrical control system. Background Technology

[0002] In modern industrial automation systems, using programmable logic controllers (PLCs) to synchronously control multi-axis servo systems via fieldbus is a key technology for achieving high-precision motion tasks. The synchronization performance of such systems heavily relies on a distributed clock (DC) mechanism, which ensures synchronized operation of all motion axes by unifying the time base of all nodes in the network. Existing methods for improving system anti-interference capabilities primarily focus on physical layer protection, such as installing power filters or using highly shielded cables. These measures are effective against strong electromagnetic interference from outside the system.

[0003] However, in a tightly coupled multi-axis system, when a servo axis performs a highly dynamic, heavy-load task, its own power stage switching will generate internal electromagnetic interference. The intensity of this interference may not be enough to disrupt the integrity of bus communication, but it can affect the clock source circuit inside the axis slave controller through conduction or near-field coupling, causing a brief and non-linear frequency drift in its local clock.

[0004] The synchronization algorithm for distributed clocks is based on the fundamental assumption that the local clock frequencies of each slave station are stable and that the drift is linear. When this assumption is violated, the DC synchronization algorithm misinterprets this nonlinear drift and pollutes the common time base of the entire network through protocol logic. This ultimately manifests macroscopically as a decrease in the position tracking accuracy of other slave axes. In addition, factors such as instantaneous drops in the external power grid can cause synchronous fluctuations in the DC bus voltage of all servo axes in the system, forming common-mode interference artifacts. This makes it difficult for conventional signal correlation analysis to distinguish between the real internal interference sources and other axes affected by common-mode interference, which can easily lead to miscompensation or system instability. Summary of the Invention

[0005] To address the technical problem in existing multi-axis synchronous control systems that struggle to effectively distinguish and eliminate common-mode interference artifacts through conventional signal correlation analysis when facing common-mode events such as power grid fluctuations, thus hindering the accurate identification of the true internal electromagnetic interference source causing clock drift and leading to uncontrollable degradation of overall coordinated position tracking performance, this invention aims to provide a method for improving the anti-interference capability of PLC electrical control systems. The specific technical solution adopted is as follows: This invention provides a method for improving the anti-interference capability of a PLC electrical control system, the method comprising: The DC bus voltage and position tracking error data of each servo axis in the multi-axis servo system are collected. Based on the statistical distribution within the sliding window, the DC bus voltage fluctuation rate sequence is extracted, and the tracking error fluctuation difference sequence between any two servo axes is obtained. The clock drift compensation value of each servo axis is collected and the dynamic changes are analyzed to obtain the clock compensation response sequence. Orthogonal analysis is performed on the DC bus voltage fluctuation rate sequences of all servo axes to extract the global common-mode fluctuation characteristics of the system; based on the common-mode fluctuation characteristics, the source activity independence of the DC bus voltage fluctuation rate sequences of each servo axis after removing common-mode interference is analyzed. For any two servo axes, analyze the nonlinear morphological similarity between the DC bus voltage fluctuation rate sequence and the clock compensation response sequence. Combine the source activity independence and tracking error fluctuation difference sequence to identify the dominant interference path and determine the main interference source axis and the main victim axis. Based on the DC bus voltage fluctuation sequence of the main interference source axis and the concentration of the dominant interference path, the interference risk is assessed and predicted, and compensation control is performed for the main victim axis.

[0006] Furthermore, the method for obtaining the DC bus voltage fluctuation rate sequence includes: For each servo axis, the coefficient of variation of the DC bus voltage within a preset sliding window is calculated to obtain the DC bus voltage fluctuation rate; the DC bus voltage fluctuation rates are arranged in time order to obtain the DC bus voltage fluctuation rate sequence.

[0007] Furthermore, the method for obtaining the tracking error fluctuation difference sequence includes: For any two different servo axes, calculate the standard deviation of the position tracking error of the two servo axes within a preset sliding window; based on the difference between the standard deviations of the two servo axes within the preset sliding window, obtain the tracking error fluctuation difference degree; arrange the tracking error fluctuation difference degree in chronological order to obtain the tracking error fluctuation difference degree sequence.

[0008] Furthermore, the method for obtaining the clock compensation response sequence includes: The clock drift compensation value of each servo axis in the current analysis cycle is collected through the bus diagnostic interface; The clock drift compensation values ​​of adjacent analysis cycles are subjected to first-order differential operation, and the differential values ​​are arranged in time sequence to obtain the clock compensation response sequence of each servo axis.

[0009] Furthermore, the method for obtaining the common-mode fluctuation characteristics includes: The DC bus voltage fluctuation sequence of all servo axes is constructed into a feature matrix; principal component analysis is performed on the feature matrix, and the first principal component vector associated with the largest eigenvalue is extracted as the common mode fluctuation feature.

[0010] Furthermore, the method for obtaining the independence of the source activity includes: For any servo axis, calculate the orthogonal projection vector of the DC bus voltage fluctuation sequence of that servo axis onto the common-mode fluctuation characteristics; The source activity independence of the servo axis is obtained by the ratio between the square of the magnitude of the orthogonal projection vector and the square of the magnitude of the DC bus voltage fluctuation rate sequence of the servo axis. The ratio is negatively correlated with the source activity independence.

[0011] Furthermore, the method for obtaining the nonlinear morphological similarity includes: For any pair of axes consisting of two different servo axes, the DTW value between the DC bus voltage fluctuation rate sequence of one servo axis and the clock compensation response sequence of the other servo axis is analyzed using the dynamic time warping algorithm to obtain the nonlinear morphological similarity of the pair.

[0012] Furthermore, the method for determining the main interference source axis and the main victim axis includes: For any pair of axes consisting of two different servo axes, the mean of the tracking error fluctuation difference sequence of the pair within a preset analysis period is used as an asymmetry index; combined with the source activity independence, asymmetry index and nonlinear morphological similarity, the correlation cost of the pair is obtained. Construct an interference path cost matrix based on the correlation cost value; for any servo axis, transform the correlation cost value of that servo axis with respect to each other servo axis in the interference path cost matrix into a probability distribution; based on the information entropy of all probability distributions, obtain the concentration of the dominant interference path of that servo axis as the interference source, where information entropy and concentration are negatively correlated. The servo axis with the highest concentration is selected as the main interference source axis, and the servo axis with the lowest associated cost value with the main interference source axis in the interference path cost matrix is ​​selected as the main victim axis.

[0013] Furthermore, the method for determining the predicted interference risk includes: The predicted voltage fluctuation rate is obtained by using linear extrapolation on the DC bus voltage fluctuation rate sequence of the main interference source axis; the product of the predicted voltage fluctuation rate and the concentration of the main interference source axis is used as the predicted interference risk.

[0014] Furthermore, the execution of compensation control for the primary victim axis includes: When the predicted interference risk is greater than the preset high-risk threshold, the gain adjustment factor is set to the preset minimum gain factor; when the predicted interference risk is less than the preset adjustment risk threshold, the gain adjustment factor is set to the preset standard gain factor; the preset minimum gain factor is less than the preset standard gain factor; and the preset adjustment risk threshold is less than the preset high-risk threshold. Otherwise, the gain adjustment factor is obtained by using a linear decay function to predict the interference risk, and the predicted interference risk is negatively correlated with the gain adjustment factor. The main victim axis is controlled by using a gain adjustment factor.

[0015] The present invention has the following beneficial effects: This invention orthogonally decomposes the DC bus voltage fluctuation rate sequences of all servo axes and extracts the global common-mode fluctuation characteristics of the system. This process separates the true independent activity of each servo axis from common-mode interference artifacts, quantifying the source activity independence of each axis. This allows for the unique identification of the interference source as an internal interference axis whose activity is independent of the common-mode event during common-mode events such as grid voltage dips. Furthermore, by combining source activity independence, tracking error fluctuation differences, and the nonlinear morphological similarity between DC bus voltage fluctuation rate and clock compensation response, spurious correlation interference paths are effectively eliminated. This enables high-confidence identification of the dominant interference path and its concentration in the current system. Targeted compensation for the primary affected axis is then performed based on the concentration of the primary interference source activity and predicted risk, achieving a shift from passive defense to predictive avoidance. This avoids system impact caused by blind global compensation and dynamically matches compensation actions with risk levels. This invention achieves more accurate interference attribution by decoupling the common mode of the underlying operating data and integrating multi-source evidence to identify the dominant interference path and predictive compensation. It suppresses internal interference while ensuring the stability of system control and improves the synchronization accuracy and anti-interference robustness of the control system in complex electromagnetic environments. Attached Figure Description

[0016] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating a method for improving the anti-interference capability of a PLC electrical control system, as provided in one embodiment of the present invention. Detailed Implementation

[0018] To further illustrate the technical means and effects adopted by the present invention to achieve the intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a method for improving the anti-interference capability of a PLC electrical control system according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0020] The following describes in detail, with reference to the accompanying drawings, a specific scheme for improving the anti-interference capability of a PLC electrical control system provided by the present invention.

[0021] Taking an EtherCAT bus control system with N servo axes as a specific example, the PLC acts as the master station, and each servo drive acts as a slave station. The method for improving the anti-interference capability of a PLC electrical control system provided by this invention is executed in the PLC application. All calculations are performed cyclically with a fixed base analysis cycle Tbase (e.g., 10ms). To store the time-series data required for the calculations, a fixed-length circular buffer is maintained for each feature to be analyzed for each servo axis.

[0022] Please see Figure 1 The diagram illustrates a method for improving the anti-interference capability of a PLC electrical control system according to an embodiment of the present invention. The method includes the following steps: S1: Collect DC bus voltage and position tracking error data of each servo axis in the multi-axis servo system. Based on the statistical distribution within the sliding window, extract the DC bus voltage fluctuation rate sequence and obtain the tracking error fluctuation difference sequence between any two servo axes. Collect the clock drift compensation value of each servo axis and analyze the dynamic changes to obtain the clock compensation response sequence.

[0023] Data acquisition is performed via cyclic synchronous communication using the EtherCAT bus. In this embodiment of the invention, since the energy exchange activity of the servo axis power level is directly reflected in the fluctuation state of the DC bus voltage, the degree of voltage fluctuation can intuitively reflect the activity intensity of the interference source. Therefore, the activity state of potential interference sources on each axis is characterized by periodically acquiring the DC bus voltage. Since internal interference will eventually propagate through a distributed clock mechanism and be reflected in motion control deviations, the position tracking error can directly reflect the degradation of synchronization accuracy caused by interference. Therefore, the position tracking error is synchronously acquired to characterize the actual impact of interference on the motion control of each axis.

[0024] Because the raw collected data contains baseline differences and random noise, it cannot be directly used for time series feature comparison. Therefore, a sliding window statistical method is used to extract standardized dynamic features. Considering the balance between system response speed and analysis stability, a preset analysis period is set to ensure that the length of the time series data can support effective pattern recognition. The length can be set to, for example, 500ms, and can be flexibly adjusted according to the system cycle without limitation.

[0025] In this embodiment of the invention, for each servo axis, the coefficient of variation of the DC bus voltage within a preset sliding window is calculated to obtain the DC bus voltage fluctuation rate. This is used to eliminate the influence of the baseline value and unify the dimensions, quantifying only the severity of the fluctuation. Subsequently, the DC bus voltage fluctuation rates are arranged in chronological order to obtain a DC bus voltage fluctuation rate sequence. It should be noted that the coefficient of variation is a technique well-known to those skilled in the art; the coefficient of variation is the ratio of the standard deviation to the mean, and will not be elaborated upon here. The preset sliding window is used to extract local fluctuation features, and the length of the sliding window can be set to 100ms; the specific size can be adjusted according to the embodiment.

[0026] To further quantify the asymmetry of the impact of interference between axes, in this embodiment of the invention, for any two different servo axes, the standard deviation of their position tracking error within a preset sliding window is calculated to characterize their respective motion stability. Then, based on the difference between the standard deviations of the two servo axes within the preset sliding window, the tracking error fluctuation difference is obtained to reflect the asymmetry of motion deviation between axes caused by interference. Specifically, the absolute value of the difference between the standard deviations of the two servo axes within the preset sliding window is calculated, and the sum of the standard deviations of the two servo axes and a minimum positive compensation term are used as the denominator. The absolute value of the difference in standard deviations is then normalized to obtain the tracking error fluctuation difference. In one specific embodiment of the invention, the minimum positive compensation term is typically set to 10. -6 This is used to prevent calculation crashes caused by the denominator being zero when both axes are simultaneously in an absolutely static state.

[0027] Finally, the tracking error fluctuation difference is arranged in chronological order to obtain the tracking error fluctuation difference sequence.

[0028] Since the distributed clock mechanism compensates for the frequency drift of the local clock of each slave station by continuously sending time correction values ​​from the master station, when the clock source of a certain axis is interfered with by internal electromagnetic noise, its compensation value will inevitably change. Therefore, the clock drift compensation value of each servo axis in the current analysis cycle is collected through the bus diagnostic interface. For the initial startup state, the compensation value is initialized to the original reading of the corresponding register, and the differential value of the first cycle is initialized to 0.

[0029] In this embodiment of the invention, the clock drift compensation values ​​of adjacent analysis cycles are subjected to first-order differential operation, and the differential values ​​are arranged in time sequence to obtain the clock compensation response sequence of each servo axis. By quantifying the dynamic change rate of clock compensation, the degree of nonlinear drift of the local crystal oscillator of the servo axis under interference is reflected more intuitively.

[0030] S2: Perform orthogonal analysis on the DC bus voltage fluctuation rate sequences of all servo axes to extract the global common-mode fluctuation characteristics of the system; based on the common-mode fluctuation characteristics, analyze the source activity independence of the DC bus voltage fluctuation rate sequences of each servo axis after removing common-mode interference.

[0031] In multi-axis servo systems in industrial settings, external factors such as instantaneous drops in mains voltage or fluctuations in bus power can cause synchronous changes in the DC bus voltage of all servo axes, creating a global common-mode interference artifact. If this artifact is not isolated, conventional correlation analysis can easily misjudge innocent axes affected by the mains as sources of internal electromagnetic interference.

[0032] In this embodiment of the invention, the DC bus voltage fluctuation rate sequence of all servo axes is constructed into a feature matrix, which is a two-dimensional data matrix with each servo axis as the column vector and the sampling time as the row dimension.

[0033] To prevent algorithm degradation or misidentification of strong interference from a single axis as global common mode when the number of servo axes in the system is extremely small, in one specific embodiment of this invention, when the number of servo axes is greater than or equal to a preset threshold (e.g., a threshold of 3), principal component analysis is performed on the feature matrix. Before the analysis, the DC bus voltage fluctuation sequence of each servo axis is standardized, for example, by Z-score processing. After unifying the variance benchmark of each axis, the first principal component vector associated with the largest eigenvalue is extracted through eigenvalue decomposition as the common mode fluctuation feature. This common mode fluctuation feature represents the dominant synchronous fluctuation mode in the system during the current analysis period. When the number of servo axes is less than the preset threshold, the arithmetic mean vector of all vectors in the feature matrix is ​​directly extracted as the common mode fluctuation feature.

[0034] To determine whether the fluctuation of a particular axis is an independent behavior or follows global common-mode fluctuations, in this embodiment of the invention, for any servo axis, the orthogonal projection vector of the DC bus voltage fluctuation sequence of that servo axis onto the common-mode fluctuation characteristics is calculated to quantify the energy proportion of that axis following system-level common-mode fluctuations. It should be noted that the orthogonal projection vector represents the non-independent portion of the servo axis voltage fluctuations that can be explained by external common-mode events.

[0035] Furthermore, the source activity independence of the servo axis is obtained by the ratio between the square of the magnitude of the orthogonal projection vector and the square of the magnitude of the DC bus voltage fluctuation rate sequence of the servo axis. This ratio represents the proportion of energy explained by the common mode. The smaller the ratio, the stronger the autonomous activity of the servo axis independent of the common mode, and the more likely it is to become a real internal interference source. The ratio is negatively correlated with the source activity independence.

[0036] In one specific embodiment of the present invention, to prevent division-by-zero deadlock under ideal operating conditions where the servo axis is stationary and without fluctuations, the square of the magnitude of the orthogonal projection vector is divided by the sum of the square of the magnitude of the DC bus voltage fluctuation rate sequence of the servo axis and a preset bias term to obtain a ratio. The preset bias term can be 10. -6 As an example, the expression for source activity independence is: In the formula, Represented as the first Independence of source activity for each servo axis Represented as the first The square of the magnitude of the orthogonal projection vector of each servo axis onto the common-mode fluctuation characteristic. Represented as the first The square of the magnitude of the DC bus voltage fluctuation sequence of each servo axis. This is represented as a preset bias term.

[0037] S3: For any two servo axes, analyze the nonlinear morphological similarity between the DC bus voltage fluctuation rate sequence and the clock compensation response sequence. Combine the source activity independence and tracking error fluctuation difference sequence to identify the dominant interference path and determine the main interference source axis and the main victim axis.

[0038] Because internal electromagnetic interference involves unpredictable time delays and nonlinear morphological distortions throughout the entire link from its generation and propagation to affecting the victim axis clock and then to the main station issuing compensation, it is difficult to accurately capture the temporal correlation between the two using simple Euclidean distance.

[0039] In this embodiment of the invention, elastic alignment of sequences can be achieved by introducing a dynamic programming algorithm. For any pair of axes composed of two different servo axes, the DTW value between the DC bus voltage fluctuation rate sequence of one servo axis and the clock compensation response sequence of the other servo axis is analyzed using a dynamic time warping algorithm. To eliminate the dimensional interference of the accumulated DTW distance value on subsequent weight fusion, the DTW value is normalized, for example, by using the hyperbolic tangent function, to obtain the nonlinear morphological similarity of the axis pair. The smaller the nonlinear morphological similarity after normalization, the higher the consistency of the fluctuation patterns of the two after removing time delays, reflecting the high similarity of the causal propagation path.

[0040] To further eliminate common-mode spurious correlations, a multi-dimensional evidence fusion judgment is achieved by combining source activity independence and tracking error fluctuation difference sequence. In this embodiment of the invention, for any pair of axes composed of two different servo axes, the mean of the tracking error fluctuation difference sequence of the pair within a preset analysis period is used as an asymmetric index to measure the degree of difference in inter-axis motion stability caused by interference, reflecting the actual damage to tracking accuracy of the affected axis at the physical motion level caused by the potential interference path.

[0041] Furthermore, by combining source activity independence, asymmetry indices, and nonlinear morphological similarity, the association cost of the axis pair is obtained through a penalized fusion method. A smaller cost indicates a higher causal confidence level for the interference path. In one specific embodiment of the invention, a penalty denominator is constructed by multiplying the source activity independence and asymmetry indices. The nonlinear morphological similarity is then divided by the penalty denominator and its minimum positive compensation term to obtain the association cost. Only true paths with a more moderate and consistent morphology, highly independent sources, and significant motion degradation possess a minimal cost. As an example, the expression for the association cost value is: In the formula, Indicates the first The servo axis and the first The associated cost of an axis pair composed of individual servo axes Indicates the first The servo axis and the first Nonlinear morphological similarity of axis pairs composed of servo axes Indicates the first The servo axis and the first The asymmetry index of the axis pairs composed of servo axes. Represented as the first Independence of source activity for each servo axis and All are extremely small positive numbers to prevent division by zero, and can be set to 10. -6 The implementer can adjust it at their own discretion, and no restrictions are imposed here.

[0042] Then, based on the associated cost value, an interference path cost matrix is ​​constructed. The smaller the element value in the matrix, the higher the physical confidence of the corresponding row index servo axis applying internal electromagnetic interference to the column index servo axis.

[0043] For any given servo axis, the associated cost value of that servo axis with respect to each other servo axis in the interference path cost matrix is ​​transformed into a probability distribution. Specifically, the Softmax function is used for normalization to map smaller associated cost values ​​to a probability distribution range with a dominant advantage. At the same time, a preset sensitivity coefficient with a value between 5 and 20 is introduced as a function sharpness adjustment factor to weight the associated cost value, which is used to amplify the subtle differences between associated cost values. That is, the product of the preset sensitivity coefficient and the associated cost value is negative and then normalized to achieve the probability distribution transformation.

[0044] Based on the information entropy of all probability distributions, the concentration of the dominant interference path of the servo axis as the interference source is obtained, that is, to quantitatively determine whether the interference generated by the servo axis is widespread disordered noise or precisely targeted at a certain target. Information entropy and concentration are negatively correlated. The smaller the information entropy, the more concentrated the probability distribution, that is, the more concentrated the interference energy of the axis is directed towards a few victim axes. The larger the information entropy, the more dispersed the interference and the unclear the path. In a specific embodiment of the present invention, 1 minus the normalized information entropy is used as the concentration index. The closer its value is to 1, the more clearly the directionality of the interference is indicated. Its normalization can be performed using the hyperbolic tangent function, which will not be elaborated here.

[0045] Ultimately, the servo axis with the highest concentration was selected as the primary source of interference, indicating that the internal activity characteristics of this servo axis constitute the most significant source of electromagnetic damage in the current system. In the interference path cost matrix, the servo axis with the lowest associated cost with the primary source of interference was selected as the primary victim axis; this servo axis is the one most significantly affected by the interference source and experiences the most pronounced degradation in control accuracy.

[0046] S4: Based on the DC bus voltage fluctuation rate sequence of the main interference source axis and the concentration of the dominant interference path, assess and predict the interference risk, and perform compensation control for the main victim axis.

[0047] Because the power level activity of interference sources exhibits short-term continuity and trends, the intensity of future short-term interference can be reasonably extrapolated based on historical time-series data, enabling interference prediction rather than passive response. In this embodiment of the invention, the DC bus voltage fluctuation rate sequence of the main interference source axis is extrapolated using a linear extrapolation method to obtain the predicted voltage fluctuation rate, which is used to characterize the activity intensity of the interference source within a short time-domain window in the future. In a specific embodiment of the invention, multiple most recent consecutive sample points, such as five, in the DC bus voltage fluctuation rate sequence of the main interference source axis can be selected, and the trend slope can be fitted using the least squares method for linear extrapolation to obtain the predicted voltage fluctuation rate. This can improve the anti-disturbance capability against second-order characteristic high-frequency abrupt noise and reduce prediction distortion caused by single-point difference extrapolation.

[0048] The product of the predicted voltage fluctuation rate and the concentration of the main disturbance source axis is then used as the predicted disturbance risk. The risk value reflects both the disturbance intensity and the path confidence. When the intensity is high and the path is clear, the risk score will increase significantly. The higher the risk value, the more significant the impact of the impending disturbance on the control accuracy.

[0049] To achieve smooth compensation that matches the risk level and avoid system instability caused by abrupt adjustments, in this embodiment of the invention, when the predicted interference risk is greater than the preset high-risk threshold, it indicates that the internal electromagnetic coupling strength faced by the current system is approaching the system's tolerance limit, the interference intensity is extremely high, and the maximum compensation amplitude needs to be executed. The gain adjustment factor is set to the preset minimum gain factor to forcibly reduce the position loop servo following rigidity of the main victim axis, so that it acts as a low-pass filter in the control logic to desensitize the high-frequency jitter of the interfered clock.

[0050] When the predicted interference risk is less than the preset adjustment risk threshold, it indicates that the system is in a self-healing background noise environment or the interference source has subsided. The gain adjustment factor is set to the preset standard gain factor to ensure that each servo axis of the system maintains the required high-precision dynamic collaborative tracking capability.

[0051] It is understandable that excessively reducing the position gain will increase the static error of the servo axis following the command. Therefore, the preset minimum gain factor must be less than the preset standard gain factor. At the same time, the preset high-risk threshold must represent the extreme operating condition and should be greater than the upper limit of the fluctuation under normal operating conditions. The preset adjustment risk threshold must be less than the preset high-risk threshold. In this embodiment of the invention, the preset standard gain factor is set to 1, the preset minimum gain factor is set to 0.7, the preset adjustment risk threshold is set to the maximum value of the pre-evaluation benchmark of the system under no-load idling conditions, and the preset high-risk threshold is set to 3 to 5 times the preset adjustment risk threshold. The specific values ​​need to be pre-determined by the implementer based on the hardware mechanical resonance characteristics of the target motor, and are not limited here.

[0052] Otherwise, when the system is in the moderately dangerous transition range between the two thresholds, a gain adjustment factor is obtained using a linear decay function to predict the disturbance risk. The predicted disturbance risk is negatively correlated with the gain adjustment factor, thereby achieving a smooth transition of the compensation control intensity and avoiding the mechanical shock to the system caused by parameter abrupt changes. As an example, the expression for the gain adjustment factor is: In the formula, Represented as the gain adjustment factor, This is expressed as the difference between the preset standard gain factor and the preset minimum gain factor. This is expressed as the difference between the preset high-risk threshold and the preset adjusted risk threshold. This indicates a preset adjustment risk threshold. This represents the predicted risk of interference.

[0053] Finally, the gain adjustment factor is used to control the parameters of the primary affected axis. To prevent repeated modifications to the driver's underlying parameters from crippling the communication link, the absolute value of the difference between the current period's gain adjustment factor and the previous actual value is calculated. The position loop proportional gain within the primary affected axis servo driver is updated using the gain adjustment factor only if this absolute value is greater than a preset anti-jitter dead zone threshold (e.g., 0.05). After the risk is completely eliminated, a reset command is issued to restore the gain to the standard value, thus effectively isolating internal interference while ensuring that bus bandwidth is not abused.

[0054] In summary, this invention orthogonally decomposes the DC bus voltage fluctuation rate sequences of all servo axes and extracts the global common-mode fluctuation characteristics of the system. This process separates the true independent activity of each servo axis from common-mode interference artifacts, quantifying the source activity independence of each axis. This allows for the unique identification of the interference source as an internal interference axis whose activity is independent of the common-mode event during common-mode events such as grid voltage dips. Furthermore, by combining source activity independence, tracking error fluctuation differences, and the nonlinear morphological similarity between DC bus voltage fluctuation rate and clock compensation response, spurious interference paths are effectively eliminated. This enables high-confidence identification of the dominant interference path and its concentration in the current system. Targeted compensation for the primary affected axis is then performed based on the concentration of the primary interference source activity and predicted risk, achieving a shift from passive defense to predictive avoidance. This avoids system impact caused by blind global compensation and dynamically matches compensation actions with risk levels. This invention achieves more accurate interference attribution by decoupling the common mode of the underlying operating data and integrating multi-source evidence to identify the dominant interference path and predictive compensation. It suppresses internal interference while ensuring the stability of system control and improves the synchronization accuracy and anti-interference robustness of the control system in complex electromagnetic environments.

[0055] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0056] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for improving the anti-interference capability of a PLC electrical control system, characterized in that, The method includes: The DC bus voltage and position tracking error data of each servo axis in the multi-axis servo system are collected. Based on the statistical distribution within the sliding window, the DC bus voltage fluctuation rate sequence is extracted, and the tracking error fluctuation difference sequence between any two servo axes is obtained. The clock drift compensation value of each servo axis is collected and the dynamic changes are analyzed to obtain the clock compensation response sequence. Orthogonal analysis is performed on the DC bus voltage fluctuation rate sequences of all servo axes to extract the global common-mode fluctuation characteristics of the system. Based on the common-mode fluctuation characteristics, the source activity independence of the DC bus voltage fluctuation rate sequences of each servo axis after removing common-mode interference is analyzed, including: for any servo axis, the orthogonal projection vector of the DC bus voltage fluctuation rate sequence of the servo axis on the common-mode fluctuation characteristics is calculated; the source activity independence of the servo axis is obtained according to the ratio between the square of the magnitude of the orthogonal projection vector and the square of the magnitude of the DC bus voltage fluctuation rate sequence of the servo axis, and the ratio is negatively correlated with the source activity independence. For any two servo axes, analyze the nonlinear morphological similarity between the DC bus voltage fluctuation rate sequence and the clock compensation response sequence. Combine the source activity independence and tracking error fluctuation difference sequence to identify the dominant interference path and determine the main interference source axis and the main victim axis. Based on the DC bus voltage fluctuation sequence of the main interference source axis and the concentration of the dominant interference path, the interference risk is assessed and predicted, and compensation control is performed for the main victim axis.

2. The method for improving the anti-interference capability of a PLC electrical control system according to claim 1, characterized in that, The method for obtaining the DC bus voltage fluctuation rate sequence includes: For each servo axis, the coefficient of variation of the DC bus voltage within a preset sliding window is calculated to obtain the DC bus voltage fluctuation rate; the DC bus voltage fluctuation rates are arranged in time order to obtain the DC bus voltage fluctuation rate sequence.

3. The method for improving the anti-interference capability of a PLC electrical control system according to claim 1, characterized in that, The method for obtaining the tracking error fluctuation difference sequence includes: For any two different servo axes, calculate the standard deviation of the position tracking error of the two servo axes within a preset sliding window; based on the difference between the standard deviations of the two servo axes within the preset sliding window, obtain the tracking error fluctuation difference degree; arrange the tracking error fluctuation difference degree in chronological order to obtain the tracking error fluctuation difference degree sequence.

4. The method for improving the anti-interference capability of a PLC electrical control system according to claim 1, characterized in that, The method for obtaining the clock compensation response sequence includes: The clock drift compensation value of each servo axis in the current analysis cycle is collected through the bus diagnostic interface; The clock drift compensation values ​​of adjacent analysis cycles are subjected to first-order differential operation, and the differential values ​​are arranged in time sequence to obtain the clock compensation response sequence of each servo axis.

5. The method for improving the anti-interference capability of a PLC electrical control system according to claim 1, characterized in that, The method for obtaining the common-mode fluctuation characteristics includes: The DC bus voltage fluctuation sequence of all servo axes is constructed into a feature matrix; principal component analysis is performed on the feature matrix, and the first principal component vector associated with the largest eigenvalue is extracted as the common mode fluctuation feature.

6. The method for improving the anti-interference capability of a PLC electrical control system according to claim 1, characterized in that, The method for obtaining nonlinear morphological similarity includes: For any pair of axes consisting of two different servo axes, the DTW value between the DC bus voltage fluctuation rate sequence of one servo axis and the clock compensation response sequence of the other servo axis is analyzed using the dynamic time warping algorithm to obtain the nonlinear morphological similarity of the pair.

7. The method for improving the anti-interference capability of a PLC electrical control system according to claim 1, characterized in that, The method for determining the main interference source axis and the main victim axis includes: For any pair of axes consisting of two different servo axes, the mean of the tracking error fluctuation difference sequence of the pair within a preset analysis period is used as an asymmetry index; combined with the source activity independence, asymmetry index and nonlinear morphological similarity, the correlation cost of the pair is obtained. Construct an interference path cost matrix based on the correlation cost value; for any servo axis, transform the correlation cost value of that servo axis with respect to each other servo axis in the interference path cost matrix into a probability distribution; based on the information entropy of all probability distributions, obtain the concentration of the dominant interference path of that servo axis as the interference source, where information entropy and concentration are negatively correlated. The servo axis with the highest concentration is selected as the main interference source axis, and the servo axis with the lowest associated cost value with the main interference source axis in the interference path cost matrix is ​​selected as the main victim axis.

8. The method for improving the anti-interference capability of a PLC electrical control system according to claim 7, characterized in that, The method for determining the predicted interference risk includes: The predicted voltage fluctuation rate is obtained by using linear extrapolation on the DC bus voltage fluctuation rate sequence of the main interference source axis; the product of the predicted voltage fluctuation rate and the concentration of the main interference source axis is used as the predicted interference risk.

9. The method for improving the anti-interference capability of a PLC electrical control system according to claim 1, characterized in that, The execution of compensation control targeting the primary affected axis includes: When the predicted interference risk is greater than the preset high-risk threshold, the gain adjustment factor is set to the preset minimum gain factor; when the predicted interference risk is less than the preset adjustment risk threshold, the gain adjustment factor is set to the preset standard gain factor; the preset minimum gain factor is less than the preset standard gain factor; and the preset adjustment risk threshold is less than the preset high-risk threshold. Otherwise, the gain adjustment factor is obtained by using a linear decay function to predict the interference risk, and the predicted interference risk is negatively correlated with the gain adjustment factor. The main victim axis is controlled by using a gain adjustment factor.

Images