An industrial isolator health management method and system based on digital self-calibration

Abstract

The application discloses an industrial isolator health management method and system based on digital self-calibration, and relates to the technical field of industrial automation, comprising: continuously collecting input end and output end analog signal values of the industrial isolator, calculating real-time transmission error to form an error sequence; establishing a trend model according to the error sequence to predict future transmission error values; when the current error is less than the precision threshold value and the predicted error is greater than the threshold value, calculating a preventive compensation value; determining a perception threshold value according to the output signal change rate, decomposing the compensation value into multiple step amounts, so that the output change generated by each step amount is less than the perception threshold value; and in multiple cycles, sequentially accumulating the step amounts and writing them into a digital compensation register, monitoring the output signal continuity after each writing, and completing the undisturbed online self-calibration. Through the predictive compensation and gradual calibration, the application realizes the online precision maintenance of the industrial isolator, and avoids the system interruption problem in the traditional offline calibration.

Description

A Health Management Method and System for Industrial Isolators Based on Digital Self-calibration Technical Field

[0001] This application relates to the field of industrial automation technology, and in particular to a method and system for health management of industrial isolators based on digital self-calibration. Background Technology

[0002] Currently, industrial isolators are widely used in industrial control systems to achieve electrical isolation and transmission of signals. During long-term operation, due to component aging, temperature drift, and environmental factors, the transmission accuracy of industrial isolators gradually decreases, resulting in zero-point drift and gain errors. Traditional calibration methods mainly employ offline, periodic calibration, requiring manual adjustments with system downtime. This not only leads to production interruptions and increased maintenance costs but also makes it difficult to detect and correct accuracy degradation issues in a timely manner.

[0003] While existing online calibration technologies can mitigate downtime to some extent, they suffer from several shortcomings: First, most solutions employ a reactive compensation strategy, performing calibration only after errors exceed limits, which may have already impacted system control accuracy. Second, existing online compensation methods typically involve large, one-time adjustments, easily causing abrupt changes in output signals during calibration, disrupting downstream control systems and affecting system stability. Furthermore, current technologies lack the ability to analyze and predict long-term trends in isolator health, making it impossible to anticipate equipment performance degradation and hindering preventative maintenance.

[0004] Therefore, the relevant technologies have problems such as the calibration process disturbing the system operation, lack of predictive maintenance capabilities, and inability to achieve truly uninterrupted online self-calibration. Summary of the Invention

[0005] In view of the aforementioned problems, this application is hereby filed.

[0006] Therefore, this application provides a digital self-calibration-based industrial isolator health management method and system, which can solve the problems mentioned in the background art.

[0007] To solve the above-mentioned technical problems, this application provides the following technical solution:

[0008] In a first aspect, this application provides a health management method for industrial isolators based on digital self-calibration, comprising: continuously collecting the input analog signal value and the output analog signal value of the industrial isolator; calculating the real-time transmission error of the industrial isolator based on the input analog signal value and the output analog signal value to form a real-time transmission error sequence; establishing an error change trend model based on the real-time transmission error sequence within a preset time window using a fitting algorithm; predicting the transmission error value at a future time based on the error change trend model; calculating a preventive compensation value based on the difference between the transmission error value at the future time and the target error value when the current value of the real-time transmission error sequence is less than a preset accuracy threshold and the transmission error value at a future time is greater than the preset accuracy threshold; calculating the rate of change of the output analog signal value; determining an output change sensing threshold based on the rate of change; decomposing the preventive compensation value into multiple compensation steps, wherein the output change generated by each compensation step is less than the output change sensing threshold; and sequentially accumulating the compensation steps and writing them into the digital compensation register of the industrial isolator within multiple adjustment cycles; monitoring the continuity of the output signal of the industrial isolator after each write, thereby completing an online self-calibration of the output signal without disturbance.

[0009] Preferably, the step of establishing an error change trend model through a fitting algorithm and predicting the transmission error value at future times based on the error change trend model includes: separating the zero-point error sequence and the gain error sequence from the real-time transmission error sequence; performing linear fitting on the zero-point error sequence to obtain the zero-point error change rate and fitting residual; if the fitting residual exceeds a preset linear threshold, fitting the zero-point error sequence with a higher-order polynomial to obtain a zero-point error polynomial fitting result; if the fitting residual does not exceed the preset linear threshold, using the zero-point error change rate obtained by the linear fitting as the zero-point error fitting result; extrapolating the zero-point error prediction value after a preset prediction time based on the zero-point error polynomial fitting result or the zero-point error fitting result and the current value of the zero-point error sequence; similarly, processing the gain error sequence to obtain a gain error prediction value; and using the zero-point error prediction value and the gain error prediction value as the transmission error value at the future time.

[0010] Preferably, the step of calculating the preventive compensation value based on the difference between the transmission error value at the future time and the target error value includes: setting zero as the target value of the zero-point error, and using the difference between the predicted zero-point error value and the zero value as the total zero-point compensation; setting the rated transmission ratio as the target value of the gain error, and calculating the total gain compensation required to achieve the rated transmission ratio; verifying that the total zero-point compensation and the total gain compensation are within the adjustable range of the digital compensation register; if they exceed the adjustable range, proportionally reducing the total zero-point compensation and the total gain compensation to the upper limit of the adjustable range; and if they do not exceed the adjustable range, keeping the total zero-point compensation and the total gain compensation unchanged.

[0011] Preferably, calculating the rate of change of the output analog signal value and determining the output change sensing threshold based on the rate of change includes: acquiring multiple sampled values ​​of the output analog signal value within a preset time interval and calculating the rate of change of the sampled values; when the rate of change is greater than a preset fast-changing signal threshold, setting the output change sensing threshold as a first preset sensing threshold; when the rate of change is less than a preset slow-changing signal threshold, setting the output change sensing threshold as a second preset sensing threshold; wherein the second preset sensing threshold is less than the first preset sensing threshold; and when the rate of change is between the preset fast-changing signal threshold and the preset slow-changing signal threshold, determining the output change sensing threshold by interpolation calculation between the first preset sensing threshold and the second preset sensing threshold based on the rate of change.

[0012] Preferably, the step of decomposing the preventive compensation value into multiple compensation steps includes: calculating the upper limit of register value change corresponding to a single compensation based on the output change sensing threshold and the ratio of the input analog signal value to the output analog signal value; dividing the total zero-point compensation by the upper limit of register value change and rounding up to obtain the number of zero-point compensation steps; allocating each zero-point compensation step using a decreasing allocation method, with the first step being the largest and subsequent steps decreasing; dividing the total gain compensation by the upper limit of register value change and rounding up to obtain the number of gain compensation steps; allocating each gain compensation step using a decreasing allocation method; ensuring that the larger of the number of zero-point compensation steps and the number of gain compensation steps does not exceed a preset maximum number of steps.

[0013] Preferably, the step of sequentially accumulating and writing the compensation step amount within multiple adjustment cycles, and monitoring the continuity of the industrial isolator's output signal after each write, includes: recording the current value of the output analog signal as a reference value before writing; performing a register write operation to accumulate the compensation step amount corresponding to the current adjustment cycle into the existing value of the register; continuously sampling multiple output analog signal values ​​and calculating the maximum deviation from the reference value; if the maximum deviation exceeds the output change sensing threshold, reverting the current write operation, reducing the compensation step amount corresponding to the current adjustment cycle, and retrying; if the maximum deviation does not exceed the output change sensing threshold, confirming the validity of the current write; if an excessive deviation still occurs after a preset number of retries, pausing progressive compensation and generating an abnormal alarm; and after confirming the continuity of the industrial isolator's output signal, waiting for the signal stabilization time, the signal stabilization time being determined based on the industrial isolator's response time.

[0014] Preferably, the method further includes: establishing a compensation effect evaluation model based on the execution results of the progressive compensation, recording the prediction error, actual error, and compensation effect deviation for each progressive compensation; updating the fitting parameters of the error change trend model when the compensation effect deviation shows an increasing trend; statistically analyzing the growth rate of the total compensation amount per unit time, shortening the preset time window and increasing the compensation frequency when the growth rate of the total compensation amount exceeds a preset aging threshold; and generating a device replacement warning when the cumulative total compensation amount approaches the range limit of the digital compensation register.

[0015] Secondly, this application also provides a digital self-calibration-based industrial isolator health management system, comprising: an error monitoring module, used to continuously collect the input analog signal value and the output analog signal value of the industrial isolator, calculate the real-time transmission error of the industrial isolator based on the input analog signal value and the output analog signal value, and form a real-time transmission error sequence; a trend prediction module, used to establish an error change trend model based on the real-time transmission error sequence within a preset time window through a fitting algorithm, and predict the transmission error value at future times based on the error change trend model; and a compensation calculation module, used to calculate the transmission error value at future times when the current value of the real-time transmission error sequence is less than a preset accuracy threshold and the transmission error at future times is less than a preset accuracy threshold. If the transmission error value is greater than the preset accuracy threshold, a preventive compensation value is calculated based on the difference between the transmission error value at the future time and the target error value; a threshold determination module is used to calculate the rate of change of the output analog signal value and determine the output change perception threshold based on the rate of change; a step decomposition module is used to decompose the preventive compensation value into multiple compensation step amounts, where the output change generated by each compensation step amount is less than the output change perception threshold; a progressive calibration module is used to sequentially accumulate the compensation step amounts and write them into the digital compensation register of the industrial isolator in multiple adjustment cycles, and monitor the continuity of the output signal of the industrial isolator after each write to complete the online self-calibration of the output signal without disturbance.

[0016] Thirdly, this application also provides a computer device, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to perform the following steps: continuously acquiring the input analog signal value and the output analog signal value of an industrial isolator; calculating the real-time transmission error of the industrial isolator based on the input analog signal value and the output analog signal value to form a real-time transmission error sequence; establishing an error change trend model based on the real-time transmission error sequence within a preset time window using a fitting algorithm; predicting the transmission error value at future times based on the error change trend model; and determining when the current value of the real-time transmission error sequence is less than a preset value. If the transmission error value at the future time is greater than the preset accuracy threshold, a preventive compensation value is calculated based on the difference between the transmission error value at the future time and the target error value. The rate of change of the output analog signal value is calculated, and an output change sensing threshold is determined based on the rate of change. The preventive compensation value is decomposed into multiple compensation steps, and the output change generated by each compensation step is less than the output change sensing threshold. The compensation steps are sequentially accumulated and written into the digital compensation register of the industrial isolator within multiple adjustment cycles. After each write, the continuity of the output signal of the industrial isolator is monitored to complete the online self-calibration of the output signal without disturbance.

[0017] Fourthly, this application also provides a computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program performs the following steps: continuously acquiring the input analog signal value and the output analog signal value of an industrial isolator; calculating the real-time transmission error of the industrial isolator based on the input analog signal value and the output analog signal value to form a real-time transmission error sequence; establishing an error change trend model based on the real-time transmission error sequence within a preset time window using a fitting algorithm; predicting the transmission error value at future times based on the error change trend model; and determining when the current value of the real-time transmission error sequence is less than a preset accuracy threshold. If the transmission error value at the future time is greater than the preset accuracy threshold, a preventive compensation value is calculated based on the difference between the transmission error value at the future time and the target error value; the rate of change of the output analog signal value is calculated, and an output change sensing threshold is determined based on the rate of change; the preventive compensation value is decomposed into multiple compensation steps, and the output change generated by each compensation step is less than the output change sensing threshold; the compensation steps are sequentially accumulated and written into the digital compensation register of the industrial isolator within multiple adjustment cycles; the continuity of the output signal of the industrial isolator is monitored after each write, and the online self-calibration of the output signal without disturbance is completed.

[0018] Implementing this application offers the following advantages: This application provides a health management method and system for industrial isolators based on digital self-calibration. By constructing a trend prediction model based on historical error data, it achieves a technological shift from passive response to proactive prevention. The error trend prediction mechanism can accurately predict future accuracy degradation trends before actual errors exceed limits, providing a time window for preventative compensation and avoiding the control accuracy loss caused by calibration only after errors exceed limits in traditional solutions. The calculation of preventative compensation values ​​combines independent analysis of zero-point error and gain error. By separating and processing different error components and providing targeted compensation, the accuracy and effectiveness of calibration are improved. Furthermore, the progressive compensation strategy of this application dynamically adjusts the compensation step size according to the real-time rate of change of the output signal, decomposing a single large-amplitude calibration into multiple small adjustment steps. The output change generated by each step is controlled below the sensing threshold. Therefore, this application can ensure a smooth transition of the output signal during calibration, thereby solving the problem of disturbance to downstream systems caused by online calibration. Meanwhile, through real-time evaluation of the compensation effect and dynamic updating of model parameters, this application can continuously optimize prediction accuracy and issue early replacement warnings when the compensation capability approaches its limit, thereby completing the full lifecycle health management of industrial isolators. Compared with existing technologies, this application not only achieves true non-disruptive online self-calibration but also extends the effective service life of the equipment and reduces the risk of unplanned downtime through predictive maintenance strategies. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 is a schematic diagram of a health management method for industrial isolators based on digital self-calibration, which relates to this application.

[0021] Figure 2 is a schematic diagram of the overall structure of an industrial isolator health management system based on digital self-calibration according to this application;

[0022] Figure 3 is a computer device diagram of an industrial isolator health management method based on digital self-calibration involved in this application. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0024] Industrial isolators are key components in industrial automation control systems, primarily used to achieve electrical isolation transmission of analog signals, protecting the control system from field interference and high-voltage surges. Industrial isolators achieve electrical isolation between the input and output terminals through internal isolation transformers or optocouplers, while maintaining a linear signal transmission relationship. Ideally, the output signal should maintain a fixed transmission ratio with the input signal; however, in practical applications, due to factors such as component characteristic drift, temperature changes, and mechanical stress, transmission accuracy decreases over time.

[0025] The transmission errors of industrial isolators mainly fall into two categories: zero-point error and gain error. Zero-point error manifests as a deviation at the output when the input signal is zero, while gain error manifests as a deviation of the actual transmission ratio from the rated value. Traditional isolator calibration methods include manual adjustment and offline calibration. Manual adjustment requires technicians to use standard signal sources and measuring instruments to correct errors by adjusting potentiometers or jumpers inside the isolator. Therefore, manual adjustment is not only time-consuming and labor-intensive but also requires system shutdown. While offline calibration can improve calibration efficiency through automated testing equipment, it still requires disconnecting the isolator from the system, affecting production continuity.

[0026] In recent years, online calibration technology has been developed, which can correct errors without interrupting signal transmission by integrating digital compensation circuits inside the isolator. This type of technology typically includes an error detection circuit and a digital compensation register, adjusting transmission characteristics by writing compensation values. However, existing online calibration schemes have the following limitations: First, most schemes use a threshold trigger mechanism, initiating calibration only when the error exceeds a set limit, which may have already affected the system's control accuracy. Second, the compensation value is usually written in a one-time update manner; sudden changes in the compensation amount can generate a step signal at the output, impacting downstream control loops. Third, there is a lack of ability to analyze error trends, making it impossible to predict equipment performance degradation in advance and hindering preventative maintenance.

[0027] Based on the above, this application provides a health management method for industrial isolators based on digital self-calibration. By constructing an error trend prediction model, it achieves forward-looking analysis of transmission errors. This application continuously monitors input and output signals and calculates real-time transmission errors. Based on historical error data, it establishes a trend model to predict future accuracy degradation before errors exceed limits. When it predicts that future errors will exceed a threshold, it calculates the required compensation amount in advance and decomposes the compensation process into multiple small steps based on the dynamic characteristics of the output signal. By progressively writing compensation values ​​and monitoring output continuity in real time, it ensures that the calibration process does not cause perceptible disturbances to system operation. Simultaneously, it continuously evaluates the compensation effect and updates the prediction model parameters, issuing maintenance warnings in advance when the compensation capability approaches its limit, achieving full lifecycle management of the equipment. This predictive and progressive calibration strategy not only ensures uninterrupted system operation but also extends equipment lifespan through proactive maintenance, improving the reliability and stability of the industrial control system.

[0028] According to the embodiments of this application, an embodiment of health management for industrial isolators based on digital self-calibration is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer device with data processing capabilities, such as a computer, server, etc. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than that shown here.

[0029] This embodiment provides a health management method for industrial isolators based on digital self-calibration, which can be used in the aforementioned computer equipment. Figure 1 is a flowchart of the health management method for industrial isolators based on digital self-calibration according to an embodiment of this application. As shown in Figure 1, the process includes the following steps:

[0030] Step S101: Continuously collect the input and output analog signal values ​​of the industrial isolator, calculate the real-time transmission error of the industrial isolator based on the input and output analog signal values, and form a real-time transmission error sequence.

[0031] Industrial isolators are used in industrial control systems to achieve electrical isolation between input and output signals while maintaining linear signal transmission. The input receives analog signals from field sensors or upstream control equipment, while the output sends an isolated analog signal to the downstream control system or actuator. To monitor the transmission accuracy of the industrial isolator, both the input and output analog signal values ​​need to be acquired simultaneously.

[0032] The acquisition of analog input and output signal values ​​is achieved through an analog-to-digital converter (ADC). The ADC converts continuous analog voltage or current signals into discrete digital quantities. The acquisition process follows a fixed sampling period, determined based on the application requirements of the industrial isolator. For slowly changing process variables such as temperature and pressure, the sampling period can be set to 1 second; for rapidly changing process variables such as flow rate and liquid level, the sampling period can be set to 100 milliseconds; and for rapidly changing signals such as vibration and acceleration, the sampling period can be set to 10 milliseconds or less.

[0033] The input analog signal value is denoted as V. in The output analog signal value is denoted as V. out At each sampling time, V is read simultaneously. in and V out The numerical value. To ensure the synchronization of the acquired data, the sampling operations at the input and output ends need to be performed under the control of the same clock, with a time deviation not exceeding 1% of the sampling period.

[0034] Real-time transmission error includes two components: zero-point error and gain error. Zero-point error reflects the offset of the output when the input signal is zero, while gain error reflects the deviation between the actual transmission ratio and the rated transmission ratio.

[0035] The zero-point error is calculated as follows: when the input analog signal value V... in When the value is close to zero, directly output the analog signal value V. out As zero-point error E z The criterion for judging whether the input signal is close to zero is: V in The absolute value is less than a certain percentage of the input range, such as 1% or 0.5%. If the input range of the industrial isolator is 0-10V, then when V in When the value is less than 0.1V, the input signal is considered to be close to zero.

[0036] The gain error is calculated as follows: when the input analog signal value V...in When within the normal operating range, first calculate the actual transmission ratio. :

[0037] ;

[0038] Then calculate the gain error. :

[0039] ;

[0040] Where K0 is the rated transmission ratio of the industrial isolator. The rated transmission ratio K0 is determined by the design parameters of the industrial isolator, and common values ​​include 1:1, 2:1, 1:2, etc. The standard for judging whether the input signal is within the normal operating range is: V in It is greater than one percentage of the input range and less than another percentage, for example, greater than 10% and less than 90%.

[0041] The real-time transmission error sequence consists of multiple continuously acquired sets of error values. Each set of error values ​​includes a zero-point error value and a gain error value, forming an error value pair [E]. z E g The real-time transmitted error sequence is stored in a first-in, first-out (FIFO) manner, and the sequence length is pre-set according to the needs of subsequent trend analysis. When a new error value pair is generated, the new error value pair is added to the end of the sequence; if the sequence is full, the oldest error value pair at the beginning of the sequence is removed, keeping the sequence length unchanged.

[0042] Setting the sequence length requires consideration of two factors: first, it must include sufficient historical data for trend analysis; second, it must limit storage space usage. For example, for applications requiring linear fitting, the sequence length should be no less than 30 data points; for applications requiring higher-order polynomial fitting, the sequence length should be no less than 100 data points. In practical applications, the sequence length can be chosen between 100 and 1000 based on storage resources and computing power.

[0043] During the formation of the real-time transmission error sequence, the validity of the acquired data needs to be checked. If the input or output analog signal value exceeds its respective range, the data is considered invalid and is not included in the real-time transmission error sequence. If invalid data appears in multiple consecutive sampling periods, a data anomaly alarm is generated, prompting operators to check the signal connections and the operational status of the industrial isolator.

[0044] Step S102: Based on the real-time transmission error sequence within the preset time window, establish an error change trend model through a fitting algorithm, and predict the transmission error value at future times based on the error change trend model.

[0045] It should be noted that the preset time window determines the range of historical data used in trend analysis. The rate of error change of industrial isolators varies under different environments. For example, components age slowly in a constant indoor temperature environment, so a preset time window of 48 hours is suitable; however, in a variable outdoor temperature environment, temperature cycling accelerates aging, requiring the preset time window to be shortened to 12 hours. The number of data points within the preset time window directly affects the fitting accuracy; practice shows that fitting results fluctuate significantly when there are fewer than 30 data points.

[0046] Step S102 establishes an error change trend model through a fitting algorithm, and predicts the transmission error value at future times based on the error change trend model, including steps A1 to A7.

[0047] It should be noted that the error variation trend model is a mathematical model describing the change of transmission error of industrial isolators over time. The transmission error of industrial isolators is affected by factors such as temperature drift, component aging, and mechanical stress, exhibiting a certain time-varying pattern. The error variation trend model establishes a functional relationship between error and time by mathematically fitting historical error data, and is used to predict future error trends. The error variation trend model consists of two parts: a zero-point error trend model and a gain error trend model, which describe the variation patterns of zero-point error and gain error, respectively.

[0048] Step A1: Separate the zero-point error sequence and the gain error sequence from the real-time transmission error sequence.

[0049] The real-time transmission error sequence stores multiple pairs of error values ​​within a preset time window [E] z E g Let there be n error value pairs within a preset time window. The value of n is obtained by dividing the preset time window length by the sampling period. For example, if the preset time window is 48 hours and the sampling period is 1 hour, then n=48. If the actual length of the real-time transmitted error sequence is less than n, then the actual length is used as the value of n.

[0050] The separation operation separates the zero-point error E from n error value pairs. z Extracted in chronological order to form a zero-point error sequence. Simultaneously, the gain error E in the n error value pairs is... g Extracted in chronological order to form a gain error sequence. Assign time labels to each data point. ,in , The sampling period.

[0051] Step A2: Perform linear fitting on the zero-point error sequence to obtain the zero-point error change rate and fitting residual.

[0052] The linear fitting assumption is that the zero-point error changes linearly with time, and a linear trend model of the zero-point error is established:

[0053] ;

[0054] in, The zero-point error fitting value at time t, The rate of change of zero-point error (unit: V / h or mA / h) represents the amount of change in zero-point error per unit time. The initial zero-point error (unit: V or mA) represents the error value at the zero point of time, and t is the time variable (unit: h).

[0055] Determining the parameters of the linear trend model using the least squares method and :

[0056] ;

[0057] ;

[0058] The accuracy of the linear fit is evaluated using the root mean square error of the fitting residuals.

[0059] ;

[0060] in, Let i be the linear fit value for the i-th data point. Let be the measured zero-point error value of the i-th data point.

[0061] Step A3: If the fitting residual exceeds the preset linear threshold, a higher-order polynomial is used to fit the zero-point error sequence to obtain the zero-point error polynomial fitting result.

[0062] The preset linearity threshold reflects the upper limit of acceptable linear fitting error. When the rated accuracy of an industrial isolator is 0.1%, the preset linearity threshold is typically set to one-fifth of the rated accuracy, i.e., 0.02%. If the output range of the industrial isolator is 4-20mA, then the preset linearity threshold is 0.0032mA.

[0063] When the fitting residual When the error exceeds a preset linear threshold, it indicates that the zero-point error change contains a nonlinear component, requiring the establishment of a higher-order polynomial trend model. The quadratic polynomial trend model is as follows:

[0064] ;

[0065] in, The coefficient of the quadratic term (unit: V / h² or mA / h²). The coefficient for the linear term (unit: V / h or mA / h). This is a constant term (unit: V or mA).

[0066] The coefficients of the quadratic polynomial are solved by constructing a system of normal equations:

[0067] ;

[0068] If the residuals from the quadratic polynomial fitting still exceed the preset linear threshold, then a cubic polynomial trend model is used.

[0069] ;

[0070] In practical applications, polynomials of more than the third degree are rarely needed, because higher orders are prone to overfitting, leading to unstable prediction results.

[0071] Step A4: If the fitting residual does not exceed the preset linear threshold, the rate of change of the zero-point error obtained by linear fitting is taken as the zero-point error fitting result.

[0072] Fitting residuals If the zero-point error does not exceed the preset linear threshold, the assumption of linear change in the zero-point error holds, and the zero-point error trend model adopts a linear model. The zero-point error fitting result includes the rate of change of the zero-point error. and initial zero point error Two parameters.

[0073] Step A5: Based on the zero-point error polynomial fitting result or the current value of the zero-point error fitting result and the zero-point error sequence, extrapolate and calculate the predicted value of the zero-point error after the preset prediction time.

[0074] The preset prediction duration is determined based on the calibration cycle of the industrial isolator. For equipment calibrated monthly, the preset prediction duration is 720 hours; for equipment calibrated quarterly, the preset prediction duration is 2160 hours. The preset prediction duration should be less than the calibration cycle to allow for preventative compensation before the next calibration.

[0075] For the linear trend model, the formula for calculating the zero-point error prediction value is:

[0076] ;

[0077] in, This is the predicted value of the zero-point error (unit: V or mA). The current time (in hours). Preset prediction duration (unit: h).

[0078] For the multinomial trend model, Substitute the values ​​into the corresponding polynomial formula to calculate the zero-point error prediction value.

[0079] Current value of the zero-point error sequence Used to verify the reasonableness of the prediction. The zero-point error prediction value should not exceed ±5% of the industrial isolator's output range. If the prediction value exceeds the reasonable range, the prediction value should be limited within the boundary values.

[0080] Step A6: Process the gain error sequence in the same way to obtain the predicted gain error value.

[0081] For gain error sequence Perform linear fitting to establish a linear trend model for the gain error:

[0082] ;

[0083] in, The gain error fitting value at time t (unit: %). The gain error change rate (unit: % / h) Initial gain error (unit: %).

[0084] Calculate the fitting residual of the gain error It is then compared with a preset linearity threshold. The preset linearity threshold for gain error is set according to the gain stability requirements of the industrial isolator. For an industrial isolator with a gain stability of 0.05%, the preset linearity threshold is set to 0.01%.

[0085] Based on the comparison between the fitted residuals and the preset linear threshold, either a linear trend model or a polynomial trend model is selected. Based on the selected trend model, the predicted gain error is extrapolated and calculated.

[0086] ;

[0087] The gain error prediction value should not exceed ±10%, and prediction values ​​that exceed the range need to be limited to the boundary values.

[0088] Step A7: Use the zero-point error prediction value and the gain error prediction value as the transmission error value at future time.

[0089] Zero-point error prediction value and gain error prediction value Composition of error prediction values ​​[ , [ ] represents the expected transmission error of the industrial isolator at a future time. The error prediction value is used as the transmission error value at a future time for calculating the preventive compensation value in step S103.

[0090] Preferably, this application establishes independent zero-point error trend models and gain error trend models, enabling the prediction of the development trends of the two error components separately. Zero-point error is primarily compensated by adjusting the bias value of the digital compensation register, while gain error is primarily compensated by adjusting the gain value of the digital compensation register. Separate processing of the two errors improves compensation accuracy and efficiency.

[0091] Step S103: If the current value of the real-time transmission error sequence is less than the preset accuracy threshold and the transmission error value at a future time is greater than the preset accuracy threshold, calculate the preventive compensation value based on the difference between the transmission error value at the future time and the target error value.

[0092] The preset accuracy threshold is the maximum allowable transmission error limit for an industrial isolator. The preset accuracy threshold is determined based on the accuracy class of the industrial isolator. For an industrial isolator with an accuracy class of 0.1, the preset accuracy threshold is set to 0.1% of the full scale; for an industrial isolator with an accuracy class of 0.2, the preset accuracy threshold is set to 0.2% of the full scale. For example, for a 0.1 class industrial isolator with an output range of 4-20mA, the preset accuracy threshold is 0.016mA.

[0093] The current value of the real-time transmitted error sequence includes the current zero-point error. and current gain error The judgment condition requires that two conditions be met simultaneously: First, the current error has not exceeded the limit, that is... and Second, the prediction error will exceed the limit, that is... or Meeting the above conditions indicates that the industrial isolator's current accuracy is acceptable, but future accuracy degradation is likely, necessitating preventative compensation.

[0094] The target error value is the ideal transmission error to be achieved. For zero-point error, the target error value is zero; for gain error, the target error value corresponds to the error at the rated transmission ratio, and is usually also zero. The preventative compensation value is the amount of compensation required to adjust the transmission error value at future times to the target error value.

[0095] Step S103 involves calculating a preventative compensation value based on the difference between the transmission error value at a future time and the target error value, including steps B1 to B5.

[0096] Step B1: Set the zero value as the target value of the zero-point error, and use the difference between the predicted value of the zero-point error and the zero value as the total amount of zero-point compensation.

[0097] Ideally, zero-point error should occur when the input is zero, resulting in zero output. Therefore, the target value for zero-point error is set to 0V or 0mA. The formula for calculating the total zero-point compensation is:

[0098] ;

[0099] in, This represents the total amount of zero-point compensation (unit: V or mA). The zero-point error prediction value obtained in step S102.

[0100] The sign of the total zero-point compensation indicates the direction of compensation. A positive value indicates that the output bias needs to be reduced, and a negative value indicates that the output bias needs to be increased. For example, if the zero-point error prediction value is +0.08mA, then the total zero-point compensation is +0.08mA, and the output needs to be adjusted down by 0.08mA.

[0101] Step B2: Set the rated transmission ratio as the target value for the gain error, and calculate the total gain compensation required to achieve the rated transmission ratio.

[0102] The rated transmission ratio is the input-output ratio determined during the design of industrial isolators. Common rated transmission ratios include 1:1, 2:1, and 1:2. When the gain error is zero, the actual transmission ratio equals the rated transmission ratio. The calculation of the total gain compensation needs to consider the current predicted gain error value.

[0103] ;

[0104] in, The total gain compensation (dimensionless or expressed as a percentage). This represents the predicted gain error (in %). This is the rated transmission ratio.

[0105] For example, if the rated transmission ratio is 1:1 and the predicted gain error is +2%, it means that the actual transmission ratio is 1.02:1, and the gain needs to be reduced by 2%, with a total gain compensation of -2%.

[0106] Step B3: Verify whether the total zero-point compensation and total gain compensation are within the adjustable range of the digital compensation register.

[0107] The digital compensation register is a storage unit inside the industrial isolator used to store compensation values. The adjustable range of the digital compensation register is determined by the number of bits and the resolution. For example, a 12-bit digital compensation register can represent 4096 different compensation values. If the full scale corresponds to an adjustment range of ±10%, then the resolution is 20% / 4096≈0.0049%.

[0108] The zero-point compensation register can be set to ±5% to ±10% of the output range. For a 4-20mA output, the adjustable range is ±0.8mA to ±1.6mA. The gain compensation register is typically adjustable to ±5% to ±20%. The verification process checks the following:

[0109] ;

[0110] ;

[0111] in, This represents the maximum adjustable range of the zero-point compensation register. This represents the maximum adjustable range of the gain compensation register.

[0112] Step B4: If the value exceeds the adjustable range, proportionally reduce the total zero-point compensation and total gain compensation to the upper limit of the adjustable range.

[0113] When the total compensation exceeds the adjustable range of the digital compensation register, prediction errors cannot be fully compensated. To minimize accuracy degradation, the total compensation needs to be limited to the adjustable range. The reduction method is as follows:

[0114] if ,but: ;

[0115] if ,but: ;

[0116] in, and This is the reduced total compensation amount. It is a symbolic function.

[0117] The reduction operation maintains the compensation direction unchanged, only limiting the compensation magnitude. When reduction is required, a warning message should also be generated to indicate that the industrial isolator is approaching its compensation capacity limit.

[0118] Step B5: Keep the total zero-point compensation and total gain compensation unchanged, provided they do not exceed the adjustable range.

[0119] When both the total zero-point compensation and the total gain compensation are within the adjustable range of the digital compensation register, the prediction error can be completely compensated. The calculated values ​​are maintained. and The value remains unchanged and is directly used for subsequent compensation step decomposition.

[0120] Step S104: Calculate the rate of change of the analog signal value at the output end, and determine the output change sensing threshold based on the rate of change.

[0121] The rate of change of the analog signal value at the output terminal reflects the dynamic characteristics of the industrial isolator's output signal. Rapidly changing signals are insensitive to disturbances caused by compensation and can tolerate larger single-step compensation amounts; slowly changing signals are sensitive to disturbances and require smaller single-step compensation amounts. The output change sensing threshold defines the minimum output change that the downstream control system can sense.

[0122] Step S104 includes steps S1041 to S1044.

[0123] Step S1041: Obtain multiple sampled values ​​of the output analog signal within a preset time interval, and calculate the rate of change of the sampled values.

[0124] The preset time interval serves as the observation window for rate of change calculation, typically ranging from 1 to 10 seconds. Within this preset time interval, the analog signal value at the output terminal is acquired at a fixed sampling frequency of at least 100Hz to capture rapid changes in the signal. Let the preset time interval... m sample values ​​were obtained. The formula for calculating the rate of change is:

[0125] ;

[0126] in, The rate of change of the analog signal value at the output terminal (unit: V / s or mA / s). and Take the maximum and minimum values ​​respectively.

[0127] Another method for calculating the rate of change is to use the root mean square rate of change:

[0128] ;

[0129] in, This represents the time interval between adjacent sampling points.

[0130] Step S1042: When the rate of change is greater than the preset fast change signal threshold, the output change perception threshold is set to the first preset perception threshold.

[0131] A preset fast-changing signal threshold is used to identify rapidly changing dynamic signals. For example, for a 4-20mA signal, the preset fast-changing signal threshold is set to 1mA / s. When... At that time, it is assumed that the output signal is in a state of rapid change.

[0132] The first preset sensing threshold is the output change sensing threshold under rapidly changing signal conditions, and its value is relatively large. For a 4-20mA signal, the first preset sensing threshold can be set to 0.16mA, which is equivalent to 1% of the full scale. Rapidly changing signals themselves have large fluctuations, and a compensation step of 0.16mA will not be recognized as an anomaly by the downstream system.

[0133] Step S1043: When the rate of change is less than the preset slow-changing signal threshold, the output change perception threshold is set to the second preset perception threshold; wherein the second preset perception threshold is less than the first preset perception threshold.

[0134] A preset slow-changing signal threshold is used to identify slowly changing or steady-state signals. For 4-20mA signals, the preset slow-changing signal threshold is typically set to 0.1mA / s. At that time, the output signal is considered to be in a slowly changing or stable state.

[0135] The second preset sensing threshold is the output change sensing threshold under slowly varying signal conditions, and its value is relatively small. For a 4-20mA signal, the second preset sensing threshold can be set to 0.016mA, which is equivalent to 0.1% of the full scale. Slowly varying signals require a smoother compensation process to avoid perceptible step changes.

[0136] Step S1044: When the rate of change is between a preset fast-changing signal threshold and a preset slow-changing signal threshold, the output change sensing threshold is determined by interpolation calculation between the first preset sensing threshold and the second preset sensing threshold based on the rate of change.

[0137] when At this time, the output signal is in a state of moderate change rate. The output change sensing threshold is determined by linear interpolation:

[0138] ;

[0139] in, To determine the threshold for sensing changes in the output, The first preset perception threshold, The second preset perception threshold, To preset the threshold for slowly changing signals, This is a preset threshold for fast-changing signals.

[0140] Linear interpolation ensures that the output change sensing threshold changes continuously with the rate of change, avoiding abrupt threshold changes. For example, when the rate of change is 0.5 mA / s, the output change sensing threshold is calculated as follows:

[0141] ;

[0142] By adjusting the output change sensing threshold according to the dynamic characteristics of the signal, the compensation process can adapt to different operating conditions.

[0143] Step S105: Decompose the preventive compensation value into multiple compensation steps, where the output change generated by each compensation step is less than the output change perception threshold.

[0144] If the total zero-point compensation and total gain compensation calculated in step S103 are directly written into the digital compensation register, a sudden change will occur at the output of the industrial isolator. To prevent the downstream control system from detecting abnormal signal changes, the total compensation needs to be decomposed into multiple smaller compensation steps and compensation implemented step by step. Each compensation step must meet the output change sensing threshold requirement determined in step S104.

[0145] Step S105 decomposes the preventive compensation value into multiple compensation steps, including steps C1 to C6.

[0146] Step C1: Based on the output change sensing threshold and the ratio of the input analog signal value to the output analog signal value, calculate the upper limit of the register value change corresponding to a single compensation.

[0147] There is a correspondence between the values ​​stored in the digital compensation register and the actual output changes. For every minimum unit change in the register value, the analog signal value at the output terminal changes accordingly. Based on the output change sensing threshold obtained in step S104, the maximum allowable change in the register value can be calculated in reverse. The calculation process needs to consider the current analog signal value V at the input terminal. in and the analog signal value V at the output terminal out The proportional relationship between them.

[0148] Step C2: Divide the total zero-point compensation amount by the upper limit of register value change and round up to obtain the number of zero-point compensation steps.

[0149] The total zero-point compensation needs to be implemented in multiple steps to ensure that each output change does not exceed the sensing threshold. The zero-point compensation step count represents the number of compensation operations that need to be performed. Rounding up ensures that the total compensation is fully implemented, avoiding insufficient compensation due to rounding errors.

[0150] Step C3: Allocate the zero-point compensation step size in a decreasing manner, with the first step size being the largest and decreasing in subsequent steps.

[0151] Decreasing allocation allows the compensation process to quickly approach the target value in the early stages, followed by fine-tuning in the later stages. The first compensation covers a large portion of the total amount, such as half of the remaining compensation, with subsequent compensation amounts gradually decreasing. This approach speeds up the compensation process while ensuring accuracy.

[0152] Step C4: Divide the total gain compensation by the upper limit of register value change and round up to get the number of gain compensation steps.

[0153] The decomposition method for the total gain compensation is similar to that for zero-point compensation, calculating the number of gain compensation steps. Gain compensation affects the transmission ratio of the industrial isolator; therefore, the impact of register value changes on the output is related to the magnitude of the input signal.

[0154] Step C5: Allocate the gain compensation step size for each iteration using a decreasing allocation method.

[0155] The principle for allocating the gain compensation step size is the same as that for zero-point compensation, using a decreasing method for successive allocation.

[0156] Step C6: Ensure that the larger of the zero-point compensation step count and the gain compensation step count does not exceed the preset maximum step count.

[0157] The preset maximum number of steps limits the overall compensation process time. Too many compensation steps will prolong calibration time and increase the likelihood of external interference. If the calculated number of steps exceeds the preset maximum number of steps, the compensation step size needs to be reallocated, or only partial compensation should be performed.

[0158] Step S106: The compensation step amount is sequentially accumulated and written into the digital compensation register of the industrial isolator during multiple adjustment cycles. After each write, the continuity of the output signal of the industrial isolator is monitored to complete the online self-calibration of the output signal without disturbance.

[0159] The compensation step determined in step S105 needs to be written to the digital compensation register sequentially in chronological order. An adjustment cycle should be placed between each write operation, and the length of this adjustment cycle should be greater than the response time of the industrial isolator to ensure the effectiveness of the previous adjustment is completely stable. Online self-calibration means that the industrial isolator performs calibration while transmitting signals normally, without requiring interruption of operation or mode switching.

[0160] Step S106 involves sequentially accumulating the compensation step amount and writing it into the digital compensation register of the industrial isolator within multiple adjustment cycles, and monitoring the continuity of the output signal of the industrial isolator after each write, including steps D1 to D7.

[0161] Step D1: Record the current value of the analog signal at the output terminal as the reference value before writing.

[0162] Before performing a register write operation, the current output analog signal value needs to be recorded as a comparison benchmark. To reduce the influence of noise, multiple output values ​​can be sampled continuously and averaged. The benchmark value is used to subsequently determine whether compensation has caused excessive output changes.

[0163] Step D2: Perform a register write operation to add the compensation step amount corresponding to the current adjustment cycle to the existing value of the register.

[0164] The digital compensation register is read and written via a communication interface. First, the current value stored in the register is read, then the current compensation step is added to obtain the new value, which is then written to the register. The write operation requires verification to ensure correct data transmission.

[0165] Step D3: Continuously sample multiple output analog signal values ​​and calculate the maximum deviation from the reference value.

[0166] When the register value changes, the analog signal value at the output terminal will change accordingly. Continuous sampling is performed within the response time of the industrial isolator to identify the sampling point with the largest deviation from the reference value. The maximum deviation reflects the actual impact of the compensation operation on the output signal.

[0167] Step D4: If the maximum deviation exceeds the output change perception threshold, roll back the current write operation, reduce the compensation step size corresponding to the current adjustment cycle, and then retry.

[0168] If the maximum deviation exceeds the output change sensing threshold determined in step S104, it indicates that the compensation step size is too large. The register value needs to be restored to its state before being written, and then the compensation step size should be reduced to try again. The reduction can be achieved using a binary search method to gradually find a suitable step size.

[0169] Step D5: If the maximum deviation does not exceed the output change perception threshold, confirm that the write operation is valid.

[0170] If the maximum deviation is within the allowable range, it indicates that the compensation has not interfered with the downstream system, and the next step can be carried out.

[0171] Step D6: If an excessive deviation still occurs after a preset number of retries, pause the progressive compensation and generate an abnormal alarm.

[0172] If repeatedly reducing the compensation step size still fails to meet the requirements, there may be a hardware fault or external interference. In this case, the compensation process should be paused and an alarm signal should be sent to the operator. The preset number of retries should be determined based on the actual application; excessive retries will delay troubleshooting.

[0173] Step D7: After confirming the continuity of the output signal of the industrial isolator, wait for the signal to stabilize. The signal stabilization time is determined based on the response time of the industrial isolator.

[0174] After each register write, the output signal requires a certain amount of time to stabilize completely. This stabilization time is related to the circuit design and component parameters of the industrial isolator. Waiting for the signal to stabilize ensures that the next compensation is based on a stable state.

[0175] Step S107: After completing the online self-calibration in step S106, this includes steps S1071 to S1074.

[0176] Step S1071: Establish a compensation effect evaluation model based on the execution results of incremental compensation, and record the prediction error, actual error and compensation effect deviation of each incremental compensation.

[0177] After compensation is completed, the actual effect needs to be evaluated. The prediction error comes from the prediction result in step S102, and the actual error is obtained through measurement. The difference between the two is the compensation effect deviation. The compensation effect deviation reflects the accuracy of the prediction model. Recording these data helps to improve subsequent predictions and compensations.

[0178] Step S1072: When the deviation of the compensation effect shows an increasing trend, update the fitting parameters of the error change trend model.

[0179] If the deviation of the compensation effect gradually increases after multiple consecutive compensations, it indicates that the error change trend model established in step S102 can no longer accurately describe the actual state of the industrial isolator. It is necessary to refit the model using the latest error data and update the model parameters.

[0180] Step S1073: Calculate the growth rate of total compensation within a unit of time. If the growth rate of total compensation exceeds the preset aging threshold, shorten the preset time window and increase the compensation frequency.

[0181] The growth rate of total compensation is obtained by calculating the change in cumulative compensation over the most recent 30 days. Let the cumulative zero-point compensation on day i be... The cumulative gain compensation is The daily average growth rate is then calculated as follows:

[0182] ;

[0183] in, To compensate for the total growth rate (unit: % / day), Range is the output range of the industrial isolator.

[0184] The preset aging threshold depends on the total compensation capability of the digital compensation register and the expected service life. If the adjustable range of the digital compensation register is ±10% of full scale, and the expected service life is 10 years, then the preset aging threshold is set as follows: / day. In practical applications, a margin will be left, and the preset aging threshold will be set to 0.003% / day.

[0185] when When the rate exceeds 0.003% / day, it indicates that the aging rate of the industrial isolator is exceeding the normal range. In this case, the preset time window in step S102 is shortened from 48 hours to 24 hours, allowing the error trend model to make predictions based on more recent data. Simultaneously, the compensation execution cycle is shortened from 7 days to 3 days, enabling more frequent precision maintenance during accelerated aging.

[0186] For example, during the normal aging period of an industrial isolator, the zero-point compensation increases from 0.1mA to 0.15mA within 30 days, the gain compensation increases from 0.5% to 0.6%, and the output range is 16mA. Then:

[0187] ;

[0188] If the growth rate is below the preset aging threshold, the original compensation cycle will be maintained. If the growth rate reaches 0.004% / day, the above adjustment measures will be implemented.

[0189] Step S1074: When the total cumulative compensation is close to the range limit of the digital compensation register, generate a device replacement warning.

[0190] The total compensation accumulated since the equipment was put into operation reflects the overall aging of the industrial isolator. When the total accumulated compensation approaches the upper limit of the adjustable range mentioned in step B3, it indicates that the compensation capacity is about to be exhausted. At this time, a warning message is generated to remind maintenance personnel to prepare to replace the equipment and avoid exceeding the accuracy limit due to insufficient compensation capacity.

[0191] In summary, this application achieves a technological shift from passive response to proactive prevention by constructing a trend prediction model based on historical error data. The error trend prediction mechanism can accurately predict future accuracy degradation trends before actual errors exceed limits, providing a time window for preventative compensation and avoiding the control accuracy loss caused by calibration only after errors exceed limits in traditional solutions. The calculation of preventative compensation values ​​combines independent analysis of zero-point error and gain error. By separating and specifically compensating for different error components, the accuracy and effectiveness of calibration are improved. Furthermore, the progressive compensation strategy of this application dynamically adjusts the compensation step size according to the real-time rate of change of the output signal, decomposing a single large-amplitude calibration into multiple small adjustment steps. The output change generated by each step is controlled below the sensing threshold. Therefore, this application can ensure a smooth transition of the output signal during calibration, thereby solving the problem of disturbance to downstream systems caused by online calibration. Simultaneously, through real-time evaluation of the compensation effect and dynamic updating of model parameters, this application can continuously optimize prediction accuracy and issue early replacement warnings when the compensation capability approaches its limit, thus completing the full lifecycle health management of industrial isolators. Compared with existing technologies, this application not only achieves true non-disruptive online self-calibration, but also extends the effective service life of the equipment and reduces the risk of unplanned downtime through predictive maintenance strategies.

[0192] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0193] Based on the same inventive concept, this application also provides an industrial isolator health management system based on digital self-calibration. The solution provided by this system is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the industrial isolator health management system based on digital self-calibration provided below can be found in the limitations of the industrial isolator health management method based on digital self-calibration described above, and will not be repeated here.

[0194] In an exemplary embodiment, as shown in Figure 3, a digital self-calibration-based industrial isolator health management system is provided, comprising:

[0195] The error monitoring module is used to continuously collect the input and output analog signal values ​​of the industrial isolator, calculate the real-time transmission error of the industrial isolator based on the input and output analog signal values, and form a real-time transmission error sequence.

[0196] The trend prediction module is used to establish an error change trend model based on the real-time transmission error sequence within a preset time window through a fitting algorithm, and predict the transmission error value at future times based on the error change trend model.

[0197] The compensation calculation module is used to calculate a preventive compensation value based on the difference between the transmission error value at the future time and the target error value when the current value of the real-time transmission error sequence is less than the preset accuracy threshold and the transmission error value at the future time is greater than the preset accuracy threshold.

[0198] The threshold determination module is used to calculate the rate of change of the analog signal value at the output end and determine the output change sensing threshold based on the rate of change.

[0199] The step decomposition module is used to decompose the preventive compensation value into multiple compensation step amounts, and the output change generated by each compensation step amount is less than the output change perception threshold.

[0200] The progressive calibration module is used to sequentially accumulate the compensation step amount and write it into the digital compensation register of the industrial isolator in multiple adjustment cycles. After each write, it monitors the continuity of the output signal of the industrial isolator to complete the online self-calibration of the output signal without disturbance.

[0201] The modules in the aforementioned digital self-calibration-based industrial isolator health management system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the computer device's memory as software, allowing the processor to invoke and execute the corresponding operations of each module.

[0202] In an exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram is shown in Figure 3. The computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The input / output interface of the computer device is used for exchanging information between the processor and external devices. The communication interface of the computer device is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When the computer program is executed by the processor, it implements a digital self-calibration-based industrial isolator health management method. The display unit of the computer device is used to form a visually visible image and may be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0203] Those skilled in the art will understand that the structure shown in Figure 3 is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or may combine certain components, or may have different component arrangements.

[0204] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0205] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0206] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0207] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0208] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0209] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0210] The above embodiments are merely illustrative of several implementation methods of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A health management method for industrial isolators based on digital self-calibration, characterized in that, include: The input and output analog signal values ​​of the industrial isolator are continuously collected, and the real-time transmission error of the industrial isolator is calculated based on the input and output analog signal values ​​to form a real-time transmission error sequence. Based on the real-time transmission error sequence within a preset time window, an error change trend model is established using a fitting algorithm, and the transmission error value at future times is predicted based on the error change trend model. When the current value of the real-time transmission error sequence is less than a preset accuracy threshold, and the transmission error value at a future time is greater than the preset accuracy threshold, a preventive compensation value is calculated based on the difference between the transmission error value at the future time and the target error value. The rate of change of the output analog signal value is calculated, and an output change sensing threshold is determined based on the rate of change. The preventive compensation value is decomposed into multiple compensation steps, and the output change generated by each compensation step is less than the output change sensing threshold. The compensation steps are sequentially accumulated and written into the digital compensation register of the industrial isolator within multiple adjustment cycles. After each write, the continuity of the output signal of the industrial isolator is monitored to complete the online self-calibration of the output signal without disturbance.

2. The industrial isolator health management method based on digital self-calibration as described in claim 1, characterized in that: The step of establishing an error change trend model using a fitting algorithm and predicting the transmission error value at future times based on the error change trend model includes: separating the zero-point error sequence and the gain error sequence from the real-time transmission error sequence; performing linear fitting on the zero-point error sequence to obtain the zero-point error change rate and fitting residual; if the fitting residual exceeds a preset linear threshold, fitting the zero-point error sequence with a higher-order polynomial to obtain a zero-point error polynomial fitting result; if the fitting residual does not exceed the preset linear threshold, using the zero-point error change rate obtained by the linear fitting as the zero-point error fitting result; extrapolating the zero-point error prediction value after a preset prediction time based on the zero-point error polynomial fitting result or the zero-point error fitting result and the current value of the zero-point error sequence; similarly, processing the gain error sequence to obtain a gain error prediction value; and using the zero-point error prediction value and the gain error prediction value as the transmission error value at the future time.

3. The industrial isolator health management method based on digital self-calibration as described in claim 2, characterized in that: The step of calculating the preventive compensation value based on the difference between the transmission error value at the future time and the target error value includes: setting zero as the target value for zero-point error, and using the difference between the predicted zero-point error value and the zero value as the total zero-point compensation; setting the rated transmission ratio as the target value for gain error, and calculating the total gain compensation required to achieve the rated transmission ratio; verifying that the total zero-point compensation and the total gain compensation are within the adjustable range of the digital compensation register; if they exceed the adjustable range, proportionally reducing the total zero-point compensation and the total gain compensation to the upper limit of the adjustable range; and keeping the total zero-point compensation and the total gain compensation unchanged if they do not exceed the adjustable range.

4. The industrial isolator health management method based on digital self-calibration as described in claim 3, characterized in that: The step of calculating the rate of change of the output analog signal value and determining the output change sensing threshold based on the rate of change includes: acquiring multiple sampled values ​​of the output analog signal value within a preset time interval and calculating the rate of change of the sampled values; when the rate of change is greater than a preset fast-changing signal threshold, setting the output change sensing threshold as a first preset sensing threshold; when the rate of change is less than a preset slow-changing signal threshold, setting the output change sensing threshold as a second preset sensing threshold; wherein the second preset sensing threshold is less than the first preset sensing threshold; and when the rate of change is between the preset fast-changing signal threshold and the preset slow-changing signal threshold, determining the output change sensing threshold by interpolation calculation between the first preset sensing threshold and the second preset sensing threshold based on the rate of change.

5. The industrial isolator health management method based on digital self-calibration as described in claim 4, characterized in that: The step of decomposing the preventive compensation value into multiple compensation steps includes: calculating the upper limit of register value change corresponding to a single compensation based on the output change sensing threshold and the ratio of the input analog signal value to the output analog signal value; dividing the total zero-point compensation by the upper limit of register value change and rounding up to obtain the number of zero-point compensation steps; allocating each zero-point compensation step using a decreasing allocation method, with the first step being the largest and subsequent steps decreasing; dividing the total gain compensation by the upper limit of register value change and rounding up to obtain the number of gain compensation steps; allocating each gain compensation step using a decreasing allocation method; ensuring that the larger of the number of zero-point compensation steps and the number of gain compensation steps does not exceed the preset maximum number of steps.

6. The industrial isolator health management method based on digital self-calibration as described in claim 5, characterized in that: The step of sequentially accumulating and writing the compensation step amount within multiple adjustment cycles, and monitoring the continuity of the industrial isolator's output signal after each write, includes: recording the current value of the output analog signal as a reference value before writing; performing a register write operation to accumulate the compensation step amount corresponding to the current adjustment cycle into the existing value of the register; continuously sampling multiple output analog signal values ​​and calculating the maximum deviation from the reference value; if the maximum deviation exceeds the output change sensing threshold, reverting the current write operation, reducing the compensation step amount corresponding to the current adjustment cycle, and retrying; if the maximum deviation does not exceed the output change sensing threshold, confirming the validity of the current write; if an excessive deviation still occurs after a preset number of retries, pausing progressive compensation and generating an abnormal alarm; and after confirming the continuity of the industrial isolator's output signal, waiting for the signal stabilization time, which is determined based on the industrial isolator's response time.

7. The industrial isolator health management method based on digital self-calibration as described in claim 6, characterized in that: The method further includes: establishing a compensation effect evaluation model based on the execution results of the progressive compensation, recording the prediction error, actual error, and compensation effect deviation of each progressive compensation; updating the fitting parameters of the error change trend model when the compensation effect deviation shows an increasing trend; statistically analyzing the growth rate of the total compensation amount per unit time, shortening the preset time window and increasing the compensation frequency when the growth rate of the total compensation amount exceeds a preset aging threshold; and generating a device replacement warning when the cumulative total compensation amount approaches the range limit of the digital compensation register.

8. A health management system for industrial isolators based on digital self-calibration, employing the industrial isolator health management method based on digital self-calibration as described in any one of claims 1 to 7, characterized in that, include: The error monitoring module is used to continuously collect the input analog signal value and the output analog signal value of the industrial isolator, calculate the real-time transmission error of the industrial isolator based on the input analog signal value and the output analog signal value, and form a real-time transmission error sequence; The trend prediction module is used to establish an error change trend model based on the real-time transmission error sequence within a preset time window using a fitting algorithm, and to predict the transmission error value at future times based on the error change trend model. The compensation calculation module is used to calculate a preventive compensation value based on the difference between the transmission error value at the future time and the target error value when the current value of the real-time transmission error sequence is less than a preset accuracy threshold and the transmission error value at the future time is greater than the preset accuracy threshold. The threshold determination module is used to calculate the rate of change of the output analog signal value and determine the output change sensing threshold based on the rate of change. The step decomposition module is used to decompose the preventive compensation value into multiple compensation step amounts, wherein the output change generated by each compensation step amount is less than the output change perception threshold. The progressive calibration module is used to sequentially accumulate the compensation step amount and write it into the digital compensation register of the industrial isolator in multiple adjustment cycles. After each write, the continuity of the output signal of the industrial isolator is monitored to complete the online self-calibration of the output signal without disturbance.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the industrial isolator health management method based on digital self-calibration as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the industrial isolator health management method based on digital self-calibration as described in any one of claims 1 to 7.

Images