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

Abstract

The invention discloses an industrial isolator health management method and system based on digital self-calibration, and relates to the technical field of industrial automation, and the method comprises the steps: continuously collecting the analog signal values of an input end and an output end of an industrial isolator, and calculating a real-time transmission error to form an error sequence; establishing a trend model according to the error sequence, and predicting a future transmission error value; when the current error is smaller than the precision threshold value and the prediction error is larger than the threshold value, a preventive compensation value is calculated; determining a sensing threshold according to the change rate of the output signal, decomposing the compensation value into a plurality of stepping quantities, and enabling the output change generated by each stepping quantity to be smaller than the sensing threshold; the stepping quantities are sequentially accumulated and written into a digital compensation register in multiple periods, the continuity of output signals is monitored after each time of writing, and undisturbed online self-calibration is completed. Through predictive compensation and progressive calibration, online precision maintenance of the industrial isolator is realized, and the problem of system interruption of traditional offline calibration is avoided.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial automation, and in particular to an industrial isolator health management method and system based on digital self-calibration. BACKGROUND

[0002] Currently, industrial isolators are widely used in industrial control systems to achieve electrical isolation and transmission of signals. During long-term operation, the transmission accuracy of industrial isolators will gradually decrease due to aging of components, temperature drift and environmental factors, resulting in zero drift and gain error. Traditional calibration methods mainly use offline periodic calibration methods, which require manual adjustment during shutdown. This not only causes production interruption and increases maintenance costs, but also makes it difficult to find and correct accuracy degradation problems in a timely manner.

[0003] Although existing online calibration techniques can avoid shutdown problems to some extent, they have the following shortcomings: on the one hand, most schemes use a post-compensation strategy, i.e., calibration is performed only after the error is detected to be out of limits, which may have already affected the control accuracy of the system; on the other hand, existing online compensation methods usually use one-time large-scale adjustment, which is easy to cause output signal mutation during calibration and cause disturbance to the downstream control system, affecting system stability. In addition, existing technologies lack long-term trend analysis and prediction capabilities for the health status of the isolator, and cannot predict the performance degradation trend in advance, making it difficult to implement preventive maintenance.

[0004] Therefore, the related art has problems such as disturbance to system operation during calibration, lack of predictive maintenance capability, and inability to achieve truly disturbance-free online self-calibration. SUMMARY

[0005] In view of the above problems, the present application is proposed.

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

[0007] To solve the above technical problems, the present application provides the following technical solutions: In a first aspect, the application provides an industrial isolator health management method based on digital self-calibration, comprising: continuously collecting input end analog signal values and output end analog signal values of an industrial isolator, calculating real-time transmission errors of the industrial isolator based on the input end analog signal values and the output end analog signal values, and forming a real-time transmission error sequence; establishing an error change trend model through a fitting algorithm according to the real-time transmission error sequence within a preset time window, and predicting a transmission error value at a future time based on the error change trend model; in the case that a 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, calculating a preventive compensation value according to a difference between the transmission error value at the future time and a target error value; calculating a change rate of the output end analog signal values, determining an output change perception threshold according to the change rate, decomposing the preventive compensation value into a plurality of compensation step sizes, each of the compensation step sizes generating an output change less than the output change perception threshold, and sequentially accumulating and writing the compensation step sizes into a digital compensation register of the industrial isolator in a plurality of adjustment periods, and monitoring output signal continuity of the industrial isolator after each writing to complete online self-calibration of the output signal without disturbance.

[0008] Preferably, the error change trend model is established through a fitting algorithm, and the transmission error value at the future time is predicted based on the error change trend model, comprising: separating a zero error sequence and a gain error sequence from the real-time transmission error sequence; performing linear fitting on the zero error sequence to obtain a zero error change rate and a fitting residual; in the case that the fitting residual exceeds a preset linear threshold, fitting the zero error sequence by using a high-order polynomial to obtain a zero error polynomial fitting result; in the case that the fitting residual does not exceed the preset linear threshold, taking the zero error change rate obtained by the linear fitting as a zero error fitting result; extrapolating and calculating a zero error prediction value after a preset prediction time based on the zero error polynomial fitting result or the zero error fitting result and a current value of the zero error sequence; similarly, processing the gain error sequence to obtain a gain error prediction value; and taking the zero error prediction value and the gain error prediction value as the transmission error value at the future time.

[0009] As preferred, the preventive compensation value is calculated according to the difference between the transmission error value at the future time and the target error value, including: setting a zero value as the target value of the zero error, taking the difference between the zero error prediction value and the zero value as the total amount of zero compensation; setting a nominal transmission ratio as the target value of the gain error, calculating the total amount of gain compensation required to reach the nominal transmission ratio; verifying whether the total amount of zero compensation and the total amount of gain compensation are within the adjustable range of the digital compensation register; in the case of exceeding the adjustable range, scaling down the total amount of zero compensation and the total amount of gain compensation to the upper limit value of the adjustable range; in the case of not exceeding the adjustable range, keeping the total amount of zero compensation and the total amount of gain compensation unchanged.

[0010] As preferred, the rate of change of the output analog signal value is calculated, and the output change perception threshold is determined according to the rate of change, including: obtaining a plurality of sampling values of the output analog signal value within a preset time interval, and calculating the rate of change of the sampling values; in the case that the rate of change is greater than a preset fast-changing signal threshold, setting the output change perception threshold as a first preset perception threshold; in the case that the rate of change is less than a preset slow-changing signal threshold, setting the output change perception threshold as a second preset perception threshold; wherein the second preset perception threshold is less than the first preset perception threshold; in the case that the rate of change is between the preset fast-changing signal threshold and the preset slow-changing signal threshold, interpolating and calculating the output change perception threshold between the first preset perception threshold and the second preset perception threshold according to the rate of change.

[0011] As preferred, the preventive compensation value is decomposed into a plurality of compensation step amounts, including: according to the output change perception threshold and the ratio of the input analog signal value to the output analog signal value, inversely calculating the upper limit of the register value change corresponding to a single compensation; dividing the total amount of zero compensation by the upper limit of the register value change, and taking the integer part to obtain the number of zero compensation steps; using a decreasing allocation method to allocate each zero compensation step amount, with the first step amount being the largest and subsequent step amounts decreasing; dividing the total amount of gain compensation by the upper limit of the register value change, and taking the integer part to obtain the number of gain compensation steps; using a decreasing allocation method to allocate each gain compensation step amount; ensuring that the larger value of the number of zero compensation steps and the number of gain compensation steps does not exceed a preset maximum step number.

[0012] As preferred, the step of sequentially accumulating the compensation step size in multiple adjustment cycles and monitoring the output signal continuity of the industrial isolator after each write includes: recording the current value of the output analog signal value as a reference value before writing; performing a register write operation to accumulate the compensation step size corresponding to the current adjustment cycle to 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 perception threshold, backtracking the current write operation, reducing the compensation step size corresponding to the current adjustment cycle and retrying; if the maximum deviation does not exceed the output change perception threshold, confirming the current write as valid; if an out-of-limit deviation still occurs after a predetermined number of retries, suspending the progressive compensation and generating an abnormal alarm; after confirming the output signal continuity of the industrial isolator, waiting for a signal stabilization time, which is determined according to the response time of the industrial isolator.

[0013] As preferred, the method further includes: establishing a compensation effect evaluation model based on the execution result 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 if the compensation effect deviation shows an increasing trend; calculating the compensation total amount growth rate per unit time, and shortening the preset time window and increasing the compensation frequency if the compensation total amount growth rate exceeds a preset aging threshold; generating a device replacement warning if the cumulative compensation total amount approaches the range limit of the digital compensation register.

[0014] In a second aspect, the application further provides an industrial isolator health management system based on digital self-calibration, comprising: an error monitoring module configured to continuously collect an input end analog signal value and an output end analog signal value of an industrial isolator, calculate a real-time transmission error of the industrial isolator based on the input end analog signal value and the output end analog signal value, and form a real-time transmission error sequence; a trend prediction module configured to establish an error change trend model by a fitting algorithm according to the real-time transmission error sequence within a preset time window, and predict a transmission error value at a future time based on the error change trend model; a compensation calculation module configured to calculate a preventive compensation value according to a difference between the transmission error value at the future time and a target error value in a case that a 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; a threshold determination module configured to calculate a change rate of the output end analog signal value, and determine an output change perception threshold according to the change rate; a step decomposition module configured to decompose the preventive compensation value into a plurality of compensation step amounts, and each of the compensation step amounts produces an output change less than the output change perception threshold; and a gradual calibration module configured to sequentially accumulate the compensation step amounts into a digital compensation register of the industrial isolator in a plurality of adjustment periods, monitor output signal continuity of the industrial isolator after each write, and complete online self-calibration of the output signal without disturbance.

[0015] In a third aspect, the application further provides a computer device, comprising a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program: continuously collecting an input end analog signal value and an output end analog signal value of an industrial isolator, calculating a real-time transmission error of the industrial isolator based on the input end analog signal value and the output end analog signal value, and forming a real-time transmission error sequence; establishing an error change trend model by a fitting algorithm according to the real-time transmission error sequence within a preset time window, and predicting a transmission error value at a future time based on the error change trend model; calculating a preventive compensation value according to a difference between the transmission error value at the future time and a target error value in a case that a 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; calculating a change rate of the output end analog signal value, determining an output change perception threshold according to the change rate, decomposing the preventive compensation value into a plurality of compensation step amounts, and each of the compensation step amounts produces an output change less than the output change perception threshold, and sequentially accumulating the compensation step amounts into a digital compensation register of the industrial isolator in a plurality of adjustment periods, monitoring output signal continuity of the industrial isolator after each write, and completing online self-calibration of the output signal without disturbance.

[0016] In a fourth aspect, the present application also provides a computer readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the following steps: continuously collecting an input end analog signal value and an output end analog signal value of an industrial isolator, calculating a real-time transmission error of the industrial isolator based on the input end analog signal value and the output end analog signal value, and forming a real-time transmission error sequence; establishing an error change trend model through a fitting algorithm according to the real-time transmission error sequence within a preset time window, predicting a transmission error value at a future time based on the error change trend model; in the case that a current value of the real-time transmission error sequence is less than a preset precision threshold value and the transmission error value at the future time is greater than the preset precision threshold value, calculating a preventive compensation value according to a difference between the transmission error value at the future time and a target error value; calculating a change rate of the output end analog signal value, determining an output change perception threshold value according to the change rate, decomposing the preventive compensation value into a plurality of compensation step sizes, each of the compensation step sizes generating an output change less than the output change perception threshold value, and sequentially accumulating and writing the compensation step sizes into a digital compensation register of the industrial isolator in a plurality of adjustment periods, and monitoring an output signal continuity of the industrial isolator after each writing, to complete online self-calibration of the output signal without disturbance.

[0017] The present application has the following beneficial effects: the present application provides an industrial isolator health management method and system based on digital self-calibration, realizes a technical change from passive response to active prevention by constructing a trend prediction model based on historical error data. The error trend prediction mechanism can accurately predict the future precision degradation trend before the actual error exceeds the limit, provides a time window for preventive compensation, and avoids the loss of control precision caused by calibration after the error exceeds the limit in the traditional scheme. The calculation of the preventive compensation value combines the independent analysis of the zero error and the gain error, improves the accuracy and effectiveness of the calibration by separating and compensating different error components. Further, the gradual compensation strategy of the present application dynamically adjusts the compensation step size according to the real-time change rate of the output signal, decomposes a single large-scale calibration into a plurality of small adjustment steps, and controls the output change of each step below the perception threshold value, so that the present application can ensure smooth transition of the output signal during calibration, and thus solve the disturbance problem of online calibration to the downstream system. At the same time, through real-time evaluation of the compensation effect and dynamic updating of the model parameters, the present application can continuously optimize the prediction accuracy and give an early warning for replacement when the compensation capacity approaches the limit, thereby completing the whole life cycle health management of the industrial isolator. Compared with the prior art, the present application not only realizes real online self-calibration without disturbance, but also prolongs the effective service life of the equipment through the predictive maintenance strategy and reduces the risk of unplanned downtime. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the drawings needed to be used in the description of the embodiments of the present application or the related art will be briefly introduced. Obviously, the drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other related drawings can also be obtained without creative labor.

[0019] Figure 1 is a schematic diagram of a digital self-calibration-based industrial isolator health management method related to the present application; Figure 2 is a schematic diagram of a digital self-calibration-based industrial isolator health management system related to the present application; Figure 3 is a computer device diagram of a digital self-calibration-based industrial isolator health management method related to the present application. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solutions and advantages of the present application more clear, the present application will be further described in detail below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0021] Industrial isolators are key components in industrial automation control systems, mainly used to realize electrical isolation transmission of analog signals, and to protect control systems from field interference and high voltage impact. Industrial isolators realize electrical isolation between input and output through internal isolation transformers or optoelectronic coupling devices, while maintaining the linear transmission relationship of signals. In an ideal state, the output signal should maintain a fixed transmission ratio relationship with the input signal, but in actual application, due to the influence of factors such as component characteristic drift, temperature change, mechanical stress, etc., the transmission accuracy will decrease over time.

[0022] The transmission error of industrial isolators mainly includes two types: zero error and gain error. Zero error is manifested as an offset at the output end when the input signal is zero, and gain error is manifested as the actual transmission ratio deviating from the rated value. Traditional isolator calibration methods include manual adjustment method and offline calibration method. The manual adjustment method requires technicians to use standard signal sources and measuring instruments to correct errors by adjusting the potentiometer or jumper inside the isolator, so the manual adjustment method not only consumes time and effort, but also requires shutdown operation. While the offline calibration method can improve calibration efficiency through automatic test equipment, it still requires the isolator to be disconnected from the system, affecting production continuity.

[0023] Online calibration techniques developed in recent years can correct errors without interrupting signal transmission by integrating digital compensation circuits inside the isolator. Such techniques usually include error detection circuits and digital compensation registers, which adjust the transmission characteristics by writing compensation values. However, existing online calibration schemes have the following limitations: first, most schemes use threshold triggering mechanism, which only starts calibration when the error exceeds the set limit, at which time the system control accuracy may have been affected; second, the writing of compensation values usually uses one-time update method, and the sudden change of compensation amount will produce a step signal at the output end, causing impact on the downstream control loop; third, there is a lack of error trend analysis capability, which cannot predict equipment performance degradation in advance and is difficult to implement preventive maintenance.

[0024] Based on the above, the present application provides an industrial isolator health management method based on digital self-calibration, which realizes forward-looking analysis of transmission error by constructing an error trend prediction model. The present application continuously monitors the input and output signals and calculates the real-time transmission error, establishes a change trend model based on historical error data, and predicts the future accuracy degradation when the error has not yet exceeded the limit. When it is predicted that the future error will exceed the threshold, the required compensation amount is calculated in advance, and the compensation process is divided into multiple small steps according to the dynamic characteristics of the output signal. By gradually writing compensation values and monitoring output continuity in real time, it is ensured that the calibration process does not cause perceptible disturbance to the system operation. At the same time, the compensation effect is continuously evaluated and the prediction model parameters are updated, and a maintenance warning is given in advance when the compensation capacity approaches the limit, realizing the whole life cycle management of the equipment. This predictive and gradual calibration strategy not only ensures the uninterrupted operation of the system, but also prolongs the service life of the equipment and improves the reliability and stability of the industrial control system.

[0025] According to an embodiment of the present application, an industrial isolator health management method based on digital self-calibration is provided. It should be noted that the steps shown in the flowchart of the drawings can be executed in a computer device with data processing capability, such as a computer, a server, etc., and although a logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in a different order.

[0026] In this embodiment, an industrial isolator health management method based on digital self-calibration is provided, which can be used in the above-mentioned computer device, Figure 1 The flowchart of the industrial isolator health management method based on digital self-calibration according to the embodiment of the present application is shown in Figure 1 The flowchart includes the following steps: Step S101: continuously collect the input end analog signal value and the output end analog signal value of the industrial isolator, calculate the real-time transmission error of the industrial isolator based on the input end analog signal value and the output end analog signal value, and form a real-time transmission error sequence.

[0027] The industrial isolator is used in the industrial control system to realize the electrical isolation between the input signal and the output signal while maintaining the linear transmission relationship of the signal. The input end receives the analog signal from the field sensor or the upstream control device, and the output end outputs the analog signal transmitted through isolation to the downstream control system or the actuator. In order to monitor the transmission accuracy of the industrial isolator, the input end analog signal value and the output end analog signal value need to be collected simultaneously.

[0028] The collection of the input end analog signal value and the output end analog signal value is realized through an analog-to-digital converter. The analog-to-digital converter converts continuous analog voltage or current signals into discrete digital quantities. The collection process is carried out according to a fixed sampling period, which is determined according to the application requirements of the industrial isolator. For slow-changing process variables such as temperature and pressure, the sampling period can be set to 1 second; for fast-changing process variables such as flow and liquid level, the sampling period can be set to 100 milliseconds; for rapidly changing signals such as vibration and acceleration, the sampling period can be set to 10 milliseconds or less.

[0029] The input end analog signal value is denoted as V in , and the output end analog signal value is denoted as V out . At each sampling time, the values of V in and V out are read simultaneously. To ensure the synchronization of the collected data, the sampling operations of the input end and the output end need to be carried out under the control of the same clock, and the time deviation should not exceed 1% of the sampling period.

[0030] The real-time transmission error includes two components: zero point error and gain error. The zero point error reflects the offset of the output end when the input signal is zero, and the gain error reflects the deviation of the actual transmission ratio from the rated transmission ratio.

[0031] The calculation method of the zero point error is: when the input end analog signal value V in is close to zero, the output end analog signal value V out is directly taken as the zero point error E z . The criterion for judging whether the input signal is close to zero is that the absolute value of V in is less than a certain proportion of the input range, for example, 1% or 0.5%. If the input range of the industrial isolator is 0-10V, when V in is less than 0.1V, it is considered that the input signal is close to zero.

[0032] The calculation method of the gain error is: when the input end analog signal value V in is close to the maximum value of the input range, the output end analog signal value V out is directly taken as the gain error E z .in When in normal working range, first calculate actual transmission ratio Then calculate gain error 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 determining whether the input signal is in the normal working range is that the input signal value V in is greater than a certain proportion of the input range and less than another proportion, for example, greater than 10% and less than 90%.

[0033] The real-time transmission error sequence is composed of multiple groups of error values collected continuously. Each group of error values contains a zero-point error value and a gain error value, forming an error value pair [E z , E g ]. The real-time transmission error sequence is stored in a first-in, first-out 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 earliest error value pair at the beginning of the sequence is removed, keeping the sequence length unchanged.

[0034] The setting of the sequence length needs to consider two factors: one is to include enough historical data for trend analysis, and the other is to limit the storage space occupation. For example, for applications that require linear fitting, the sequence length is not less than 30 data points; for applications that require high-order polynomial fitting, the sequence length is not less than 100 data points. In actual application, the sequence length can be selected between 100 and 1000 according to the storage resources and computing power.

[0035] In the process of forming the real-time transmission error sequence, the collected data needs to be checked for validity. If the input end analog signal value or the output end analog signal value exceeds the respective range, the data is considered invalid and is not included in the real-time transmission error sequence. If invalid data appears for multiple consecutive sampling periods, a data anomaly alarm is generated, prompting the operator to check the signal connection and the working state of the industrial isolator.

[0036] Step S102, according to the real-time transmission error sequence in the pre-set time window, an error change trend model is established through a fitting algorithm, and the transmission error value at a future time is predicted based on the error change trend model.

[0037] ​​​​It should be noted that the preset time window determines the range of historical data participating in the trend analysis. The error change speed of the industrial isolator in different environments is different, for example: the components age slowly in the indoor constant temperature environment, and the preset time window can be 48 hours; the temperature cycle accelerates the aging in the outdoor variable temperature environment, and the preset time window needs to be shortened to 12 hours. The number of data points in the preset time window directly affects the fitting accuracy, and practice shows that the fitting result fluctuates greatly when there are less than 30 data points.

[0038] Step S102 establishes an error change trend model through a fitting algorithm, and the future transmission error value is predicted based on the error change trend model, including steps A1 to A7.

[0039] It should be noted that the error change trend model is a mathematical model describing the law of change of the transmission error of the industrial isolator with time. The transmission error of the industrial isolator is affected by factors such as temperature drift, component aging, mechanical stress, and presents a certain time change law. The error change trend model establishes a functional relationship between error and time by mathematical fitting of historical error data, and is used to predict the future error development trend. The error change trend model includes a zero error trend model and a gain error trend model, which respectively describe the change law of zero error and gain error.

[0040] Step A1: Separate the zero error sequence and the gain error sequence from the real-time transmission error sequence.

[0041] The real-time transmission error sequence stores a plurality of error value pairs [E z ,E g ] in the preset time window. Suppose that there are n error value pairs in the preset time window, and the value of n is obtained by dividing the length of the preset time window 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 transmission error sequence is less than n, then the actual length is used as the value of n.

[0042] The separation operation extracts the zero error E z in the n error value pairs in time sequence to form a zero error sequence . At the same time, the gain error E g in the n error value pairs is extracted in time sequence to form a gain error sequence . Each data point is assigned a time label , where , is the sampling period.

[0043] Step A2: Linear fitting is performed on the zero error sequence to obtain the zero error change rate and the fitting residual.

[0044] Linear fitting assumes that the zero-point error changes linearly over time, and a linear trend model of the zero-point error is established: ; where, is the fitted value of the zero-point error at time t, is the rate of change of the zero-point error (unit: V / h or mA / h), representing the amount of change in the zero-point error per unit time, is the initial zero-point error (unit: V or mA), representing the error value of the time zero point, and t is the time variable (unit: h).

[0045] The parameters of the linear trend model are determined using the least squares method and : ; ; The root mean square error of the fitting residual is used to evaluate the accuracy of the linear fitting: ; where, is the linear fitted value of the i-th data point, is the measured zero-point error value of the i-th data point.

[0046] Step A3, if the fitting residual exceeds the preset linear threshold, a high-order polynomial is used to fit the zero-point error sequence, and the zero-point error polynomial fitting result is obtained.

[0047] The preset linear threshold reflects the upper limit of acceptable linear fitting error. When the rated accuracy of the industrial isolator is 0.1%, the preset linear threshold is usually set to one-fifth of the rated accuracy, i.e. 0.02%. If the output range of the industrial isolator is 4-20mA, the preset linear threshold is 0.0032mA.

[0048] When the fitting residual exceeds the preset linear threshold, it indicates that there is a nonlinear component in the change of the zero-point error, and a high-order polynomial trend model needs to be established. The quadratic polynomial trend model is: ; where, is the coefficient of the quadratic term (unit: V / h² or mA / h²), is the coefficient of the linear term (unit: V / h or mA / h), is the constant term (unit: V or mA).

[0049] The quadratic polynomial coefficients are solved by constructing a system of equations: ; If the quadratic polynomial fitting residual still exceeds the preset linear threshold, a cubic polynomial trend model is adopted: ; Polynomials higher than cubic are rarely needed in practical applications, because too high an order is prone to overfitting, leading to unstable prediction results.

[0050] Step A4, if the fitting residual does not exceed the preset linear threshold, the zero-point error rate of change obtained by linear fitting is taken as the zero-point error fitting result.

[0051] Fitting residual The fact that the fitting residual does not exceed the preset linear threshold indicates that the linear change assumption of the zero-point error is valid, and the zero-point error trend model adopts a linear model . The zero-point error fitting result contains two parameters: the zero-point error rate of change and the initial zero-point error .

[0052] Step A5, 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, the zero-point error prediction value after a preset prediction period is extrapolated and calculated.

[0053] The preset prediction period is determined according to the calibration period of the industrial isolator. For devices that are calibrated monthly, the preset prediction period is 720 hours; for devices that are calibrated quarterly, the preset prediction period is 2160 hours. The preset prediction period should be less than the calibration period, so that preventive compensation can be completed before the next calibration.

[0054] For the linear trend model, the zero-point error prediction value calculation formula is: ; wherein, is the zero-point error prediction value (unit: V or mA), is the current time (unit: h), is the preset prediction period (unit: h).

[0055] For the polynomial trend model, substitute into the corresponding polynomial formula to calculate the zero-point error prediction value.

[0056] The current value of the zero-point error sequence is used to verify the reasonableness of the prediction. The zero-point error prediction value should not exceed ±5% of the output range of the industrial isolator. If the prediction value exceeds the reasonable range, the prediction value is limited within the boundary value.

[0057] Step A6, the gain error sequence is processed in the same way to obtain the gain error prediction value.

[0058] The current value of the gain error sequence A linear fitting is performed to establish a linear trend model of the gain error: ; wherein, is the fitting value of the gain error at time t (unit: %), is the change rate of the gain error (unit: % / h), is the initial gain error (unit: %).

[0059] The fitting residual of the gain error is calculated and compared with a preset linear threshold. The preset linear threshold of the gain error is set according to the gain stability requirement of the industrial isolator. For an industrial isolator with a gain stability of 0.05%, the preset linear threshold is set to 0.01%.

[0060] According to the comparison result of the fitting residual and the preset linear threshold, the linear trend model or the polynomial trend model is selected. Based on the selected trend model, the gain error prediction value is extrapolated and calculated: ; The gain error prediction value should not exceed ±10%, and the prediction value exceeding the range needs to be limited within the boundary value.

[0061] Step A7, the zero error prediction value and the gain error prediction value are taken as the transmission error value at the future time.

[0062] The zero error prediction value and the gain error prediction value form an error prediction value pair , which represents the expected transmission error of the industrial isolator at the future time. The error prediction value pair is taken as the transmission error value at the future time, which is used for the preventive compensation value calculation in step S103.

[0063] Preferably, the present application can predict the development trend of the two error components respectively by establishing independent zero error trend model and gain error trend model. The zero error is mainly compensated by adjusting the bias value of the digital compensation register, and the gain error is mainly compensated by adjusting the gain value of the digital compensation register. Separating the processing of the two errors can improve the compensation accuracy and efficiency.

[0064] Step S103, in the case that 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, the preventive compensation value is calculated according to the difference between the transmission error value at the future time and the target error value.

[0065] The preset accuracy threshold is the maximum transmission error limit allowed by the industrial isolator. The preset accuracy threshold is determined according to the accuracy level of the industrial isolator. For an industrial isolator with 0.1 level accuracy, the preset accuracy threshold is set to 0.1% of the full scale; for an industrial isolator with 0.2 level accuracy, the preset accuracy threshold is set to 0.2% of the full scale. For example, for a 0.1 level industrial isolator with an output range of 4-20 mA, the preset accuracy threshold is 0.016 mA.

[0066] The current value of the real-time transmission error sequence includes the current zero point error and the current gain error . The judgment condition needs to meet two requirements at the same time: first, the current error is not out of limit, that is and ; second, the predicted error will be out of limit, that is or . Meeting the above conditions indicates that the current accuracy of the industrial isolator is qualified but the accuracy will degrade in the future, and preventive compensation is needed.

[0067] The target error value is the ideal transmission error expected to be achieved. For the zero point error, the target error value is zero; for the gain error, the target error value corresponds to the error under the rated transmission ratio, which is usually also zero. The preventive compensation value is the compensation amount needed to adjust the transmission error value at the future time to the target error value.

[0068] In step S103, the preventive compensation value is calculated according to the difference between the transmission error value at the future time and the target error value, including steps B1 to B5.

[0069] In step B1, zero is set as the target value of the zero point error, and the difference between the zero point error prediction value and zero is taken as the total zero point compensation amount.

[0070] The ideal state of the zero point error is that the output is also zero when the input is zero, so the target value of the zero point error is set to 0V or 0mA. The total zero point compensation amount calculation formula is: ; wherein is the total zero point compensation amount (unit: V or mA), is the zero point error prediction value obtained in step S102.

[0071] The sign of the total zero point compensation amount indicates the compensation direction. 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 +0.08 mA, the total zero point compensation amount is +0.08 mA, and the output needs to be adjusted downward by 0.08 mA.

[0072] In step B2, the rated transmission ratio is set as the target value of the gain error, and the total gain compensation amount needed to reach the rated transmission ratio is calculated.

[0073] The rated transfer ratio is the input-output ratio determined in the design of the industrial isolator. Common rated transfer ratios include 1:1, 2:1, 1:2, etc. When the gain error is zero, the actual transfer ratio is equal to the rated transfer ratio. The calculation of the total gain compensation needs to consider the current gain error prediction value: ; where, is the total gain compensation (dimensionless or expressed in percentage), is the gain error prediction value (unit: %), is the rated transfer ratio.

[0074] For example, the rated transfer ratio is 1:1, and the gain error prediction value is +2%, which means the actual transfer ratio is 1.02:1. The gain needs to be reduced by 2%, and the total gain compensation is -2%.

[0075] Step B3, verify whether the total zero-point compensation and the total gain compensation are within the adjustable range of the digital compensation register.

[0076] 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 register bits and the resolution. For example, a 12-bit digital compensation register can represent 4096 different compensation values, and if the full-scale corresponds to a ±10% adjustment range, the resolution is 20% / 4096≈0.0049%.

[0077] The adjustable range of 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 adjustable range of the gain compensation register is usually ±5% to ±20%. The verification process checks: ; ; where, is the maximum adjustable range of the zero-point compensation register, is the maximum adjustable range of the gain compensation register.

[0078] Step B4, if the total zero-point compensation and the total gain compensation exceed the adjustable range, scale down the total zero-point compensation and the total gain compensation to the upper limit value of the adjustable range.

[0079] When the total compensation exceeds the adjustable range of the digital compensation register, the predicted error cannot be completely compensated. In order to delay the degradation of accuracy to the greatest extent, the total compensation needs to be limited within the adjustable range. The scaling method is: If , then: ; If , then: ; where, and are the reduced total compensation for zero and gain respectively, is the sign function.

[0080] The reduction operation keeps the compensation direction unchanged, only limits the compensation amplitude. When reduction is needed, a pre-warning information should also be generated to prompt the industrial isolator that it is approaching the compensation capacity limit.

[0081] Step B5, keep the total zero compensation and the total gain compensation unchanged if they are not beyond the adjustable range.

[0082] When both the total zero compensation and the total gain compensation are within the adjustable range of the digital compensation register, the predicted error can be completely compensated. Keep the calculated and values unchanged and directly used for subsequent compensation step decomposition.

[0083] Step S104, calculate the rate of change of the output analog signal value, and determine the output change perception threshold according to the rate of change.

[0084] The rate of change of the output analog signal value reflects the dynamic characteristics of the output signal of the industrial isolator. The signal that changes rapidly is not sensitive to the disturbance caused by compensation and can accept a larger single-step compensation amount; the signal that changes slowly is sensitive to the disturbance and needs to use a smaller single-step compensation amount. The output change perception threshold defines the minimum output change that can be perceived by the downstream control system.

[0085] Step S104 includes steps S1041 to S1044.

[0086] Step S1041, obtain multiple sample values of the output analog signal value within a preset time interval, and calculate the rate of change of the sample values.

[0087] The preset time interval is the observation window for the rate of change calculation, usually 1-10 seconds. The output analog signal value is obtained within the preset time interval according to a fixed sampling frequency, and the sampling frequency is not less than 100 Hz to capture the rapid change of the signal. Suppose that m sample values are obtained within the preset time interval , the rate of change calculation formula is: ; where, is the rate of change of the output analog signal value (unit: V / s or mA / s), and take the maximum and minimum values respectively.

[0088] Another variation rate calculation method is to use the root mean square variation rate: ; wherein, is the time interval of adjacent sampling points.

[0089] Step S1042, in the case where the variation rate is greater than a preset fast variation signal threshold, setting the output variation perception threshold as a first preset perception threshold.

[0090] The preset fast variation signal threshold is used to identify a fast variation dynamic signal. For example, for a 4-20mA signal, the preset fast variation signal threshold is set as 1 mA / s. When , it is considered that the output signal is in a fast variation state.

[0091] The first preset perception threshold is the output variation perception threshold under the fast variation signal condition, and has a larger value. For a 4-20mA signal, the first preset perception threshold can be set as 0.16 mA, which is equivalent to 1% of the full scale. The fast variation signal itself has a large amplitude fluctuation, and the compensation step of 0.16 mA will not be recognized as an abnormality by the downstream system.

[0092] Step S1043, in the case where the variation rate is less than a preset slow variation signal threshold, setting the output variation perception threshold as a second preset perception threshold; wherein the second preset perception threshold is less than the first preset perception threshold.

[0093] The preset slow variation signal threshold is used to identify a slow variation or steady signal. For a 4-20mA signal, the preset slow variation signal threshold is usually set as 0.1 mA / s. When , it is considered that the output signal is in a slow variation or stable state.

[0094] The second preset perception threshold is the output variation perception threshold under the slow variation signal condition, and has a smaller value. For a 4-20mA signal, the second preset perception threshold can be set as 0.016 mA, which is equivalent to 0.1% of the full scale. The slow variation signal requires a more smooth compensation process to avoid producing perceptible step changes.

[0095] Step S1044, in the case where the variation rate is between the preset fast variation signal threshold and the preset slow variation signal threshold, determining the output variation perception threshold by interpolation calculation according to the variation rate between the first preset perception threshold and the second preset perception threshold.

[0096] When , the output signal is in a medium variation speed state. The output variation perception threshold is determined by linear interpolation: ; wherein, an output change perception threshold, a first preset perception threshold, a second preset perception threshold, a preset slow change signal threshold, a preset fast change signal threshold.

[0097] The linear interpolation ensures that the output change perception threshold varies continuously with the change rate, avoiding sudden changes in the threshold. For example, when the change rate is 0.5 mA / s, the output change perception threshold is calculated as: ; By adjusting the output change perception threshold according to the dynamic characteristics of the signal, the compensation process is adapted to different working conditions.

[0098] Step S105: decompose the preventive compensation value into multiple compensation steps, and each compensation step produces an output change less than the output change perception threshold.

[0099] The zero point compensation total and the gain compensation total calculated in step S103 will produce a sudden change at the output end of the industrial isolator if directly written into the digital compensation register. In order to avoid the downstream control system detecting abnormal signal changes, the compensation total needs to be decomposed into multiple smaller compensation steps, and the compensation is implemented gradually. Each compensation step must meet the output change perception threshold requirement determined in step S104.

[0100] Step S105 decomposing the preventive compensation value into multiple compensation steps includes steps C1 to C6.

[0101] Step C1: according to the output change perception threshold and the ratio of the input end analog signal value to the output end analog signal value, inversely calculate the upper limit of the register value change corresponding to a single compensation.

[0102] There is a corresponding relationship between the value stored in the digital compensation register and the actual output change. When the register value changes by one minimum unit, the output end analog signal value produces a corresponding change. According to the output change perception threshold obtained in step S104, the maximum allowed change of the register value can be calculated inversely. The calculation process needs to consider the proportional relationship between the current input end analog signal value V in and the output end analog signal value V out .

[0103] Step C2: divide the zero point compensation total by the upper limit of the register value change, and round up to get the number of zero point compensation steps.

[0104] The total amount of zero point compensation needs to be implemented in multiple steps to ensure that the output change of each step does not exceed the perception threshold. The number of zero point compensation steps represents the number of compensation operations that need to be performed. The ceiling function ensures that the total amount of compensation can be fully implemented, avoiding insufficient compensation due to rounding errors.

[0105] Step C3: The step size of each zero point compensation is allocated in a decreasing allocation manner, with the first step size being the largest and subsequent step sizes decreasing.

[0106] The decreasing allocation allows the compensation process to quickly approach the target value at the early stage and fine-tune at the later stage. A large portion of the total amount of the first compensation is implemented, for example, half of the remaining compensation amount, and the compensation amount decreases gradually with each subsequent step. This not only speeds up the compensation, but also ensures the accuracy of the compensation.

[0107] Step C4: Divide the total gain compensation amount by the upper limit of the register value change, and take the ceiling to get the number of gain compensation steps.

[0108] The decomposition method of the total gain compensation is similar to that of the zero point compensation, and the number of gain compensation steps is calculated. Gain compensation affects the transmission ratio of the industrial isolator, so the effect of register value change on the output is related to the size of the input signal.

[0109] Step C5: The step size of each gain compensation is allocated in a decreasing allocation manner.

[0110] The allocation principle of the gain compensation step size is the same as that of the zero point compensation, and the decreasing allocation is used to allocate step by step.

[0111] Step C6: Ensure that the larger value of the number of zero point compensation steps and the number of gain compensation steps does not exceed the preset maximum step number.

[0112] The preset maximum step number limits the length of the entire compensation process. Too many compensation steps will prolong the calibration time and increase the possibility of being disturbed externally. If the calculated step number exceeds the preset maximum step number, the compensation step size needs to be redistributed, or only part of the compensation needs to be implemented.

[0113] Step S106: The compensation step size is written into the digital compensation register of the industrial isolator one by one in multiple adjustment periods, and the output signal continuity of the industrial isolator is monitored after each write, completing the online self-calibration of the output signal without disturbance.

[0114] The compensation step size determined in step S105 needs to be written into the digital compensation register in time sequence. An adjustment period is interval between each write operation, and the length of the adjustment period should be greater than the response time of the industrial isolator to ensure that the effect of the previous adjustment is fully stable. Online self-calibration means that the industrial isolator performs calibration while transmitting signals normally, without interrupting work or switching modes.

[0115] Step S106 accumulates the compensation step size into the digital compensation register of the industrial isolator in multiple adjustment cycles, and monitors the output signal continuity of the industrial isolator after each write, including steps D1 to D7.

[0116] Step D1: Record the current value of the output analog signal value as a reference value before writing.

[0117] Before performing the register write operation, the current output analog signal value needs to be recorded as a comparison reference. To reduce the impact of noise, multiple output values can be continuously sampled and averaged. The reference value is used for subsequent judgment of whether the compensation has caused excessive output change.

[0118] Step D2: Perform the register write operation to accumulate the compensation step size corresponding to the current adjustment cycle to the existing value of the register.

[0119] The digital compensation register is read and written through a communication interface. First, read the current storage value of the register, add the compensation step size this time, and write the new value to the register after obtaining the new value. The write operation needs to be verified to ensure correct data transmission.

[0120] Step D3: Continuously sample multiple output analog signal values and calculate the maximum deviation from the reference value.

[0121] After the register value changes, the output analog signal value will change accordingly. Continuous sampling within the response time of the industrial isolator finds the sampling point with the maximum deviation from the reference value. The maximum deviation reflects the actual impact of the compensation operation on the output signal.

[0122] Step D4: If the maximum deviation exceeds the output change perception threshold, back up the current write operation and reduce the compensation step size corresponding to the current adjustment cycle and then retry.

[0123] If the maximum deviation exceeds the output change perception threshold determined in step S104, the current compensation step size is too large. The register value needs to be restored to the state before writing, and then the compensation step size is reduced to retry. The reduction can use the bisection method to gradually find the appropriate step size.

[0124] Step D5: If the maximum deviation does not exceed the output change perception threshold, confirm that the current write is valid.

[0125] The maximum deviation is within the allowed range, indicating that the compensation has not disturbed the downstream system, and the next step operation can continue.

[0126] Step D6: If the maximum deviation still exceeds the limit after a predetermined number of retries, suspend the gradual compensation and generate an abnormal alarm.

[0127] 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.

[0128] 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.

[0129] 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.

[0130] Step S107: After completing the online self-calibration in step S106, this includes steps S1071 to S1074.

[0131] 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.

[0132] 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.

[0133] Step S1072: When the deviation of the compensation effect shows an increasing trend, update the fitting parameters of the error change trend model.

[0134] 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.

[0135] 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.

[0136] 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 amount is The daily average growth rate is then calculated as follows: ; in, To compensate for the total growth rate (unit: % / day), Range is the output range of the industrial isolator.

[0137] The preset aging threshold depends on the total compensation capacity of the digital compensation register and the expected service life. If the adjustable range of the digital compensation register is ±10% of the full scale and the expected service life is 10 years, the preset aging threshold is set to: / day. In practical applications, a margin will be left, and the preset aging threshold is set to 0.003% / day.

[0138] When exceeds 0.003% / day, it indicates that the aging rate of the industrial isolator exceeds the normal range. At this time, the preset time window in step S102 is shortened from the original 48 hours to 24 hours, so that the error trend model is based on more recent data for prediction. At the same time, the compensation execution period is shortened from the original 7 days to 3 days, and the accuracy maintenance is more frequent during the accelerated aging period.

[0139] For example, a certain industrial isolator has a normal aging period, and the zero point compensation amount increases from 0.1 mA to 0.15 mA in 30 days, and the gain compensation amount increases from 0.5% to 0.6%, and the output range is 16 mA, then: ; At this time, the growth rate is lower than the preset aging threshold, and the original compensation period is maintained. If the growth rate reaches 0.004% / day, the above adjustment measures are executed.

[0140] Step S1074: Generate a device replacement warning when the cumulative compensation total approaches the range limit of the digital compensation register.

[0141] All compensation amounts accumulated since the device was put into operation reflect the overall aging degree of the industrial isolator. When the cumulative compensation total approaches the upper limit of the adjustable range mentioned in step B3, it indicates that the compensation capacity will soon be exhausted. At this time, a warning message is generated to remind the maintenance personnel to prepare for replacement of the device to avoid accuracy overrun due to insufficient compensation capacity.

[0142] In summary, the application realizes the technical transformation from passive response to active prevention by constructing a trend prediction model based on historical error data. The error trend prediction mechanism can accurately predict the future precision degradation trend before the actual error exceeds the limit, providing a time window for preventive compensation and avoiding the loss of control accuracy caused by calibration after the error exceeds the limit in traditional schemes. The calculation of the preventive compensation value combines the independent analysis of zero-point error and gain error, improving the accuracy and effectiveness of calibration by separating and compensating different error components. Further, the progressive compensation strategy of the application dynamically adjusts the compensation step size according to the real-time change rate of the output signal, decomposing a single large-scale calibration into multiple small adjustment steps, and the output change generated by each step is controlled below the perception threshold. Therefore, the application can ensure smooth transition of the output signal during calibration, thereby solving the disturbance problem of online calibration to the downstream system. At the same time, through real-time evaluation of the compensation effect and dynamic updating of the model parameters, the application can continuously optimize the prediction accuracy and issue a replacement warning in advance when the compensation capacity approaches the limit, thereby completing the whole life cycle health management of the industrial isolator. Compared with the prior art, the application not only realizes true non-disturbance online self-calibration, but also prolongs the effective service life of the equipment and reduces the risk of unplanned downtime through the predictive maintenance strategy.

[0143] It should be understood that, although each step in the flowchart involved in the above embodiments is shown in sequence according to the arrow, these steps are not necessarily executed in the order indicated by the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other orders. Moreover, at least part of the steps in the flowchart involved in the above embodiments can include multiple steps or stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily sequential, but can be executed in rotation or alternation with at least part of other steps or steps or stages in other steps.

[0144] Based on the same inventive concept, the application also provides an industrial isolator health management system based on digital self-calibration. The implementation scheme for solving the problem provided by the system is similar to the implementation scheme described in the above method, so the specific limitations in one or more embodiments of the industrial isolator health management system based on digital self-calibration provided below can refer to the limitations of the industrial isolator health management method based on digital self-calibration described above, which will not be repeated here.

[0145] In one exemplary embodiment, as shown in Figure 3 a digital self-calibration-based industrial isolator health management system is provided, comprising: an error monitoring module configured to continuously collect an input analog signal value and an output analog signal value of the industrial isolator, and calculate a 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; a trend prediction module configured to establish an error change trend model by a fitting algorithm according to the real-time transmission error sequence within a preset time window, and predict a transmission error value at a future time based on the error change trend model; a compensation calculation module configured to calculate a preventive compensation value according to a difference between the transmission error value at the future time and a target error value, in a case that a current value of the real-time transmission error sequence is less than a preset precision threshold and the transmission error value at the future time is greater than the preset precision threshold; a threshold determination module configured to calculate a rate of change of the output analog signal value, and determine an output change perception threshold according to the rate of change; a step decomposition module configured to decompose the preventive compensation value into a plurality of compensation step amounts, each of which produces an output change less than the output change perception threshold; a gradual calibration module configured to sequentially write the compensation step amounts into a digital compensation register of the industrial isolator in a plurality of adjustment periods, and monitor output signal continuity of the industrial isolator after each writing, to complete online self-calibration of the output signal without disturbance.

[0146] The above-mentioned various modules in the industrial isolator health management system based on digital self-calibration can be realized by software, hardware, or a combination thereof. The above-mentioned various modules can be embedded in or independent of a processor in a computer device in hardware form, or stored in a memory in a computer device in software form, so as to be called and executed by a processor to perform operations corresponding to the above-mentioned various modules.

[0147] In an exemplary embodiment, a computer device is provided, which can be a terminal, and an internal structure diagram of the computer device can be as shown in Figure 3As shown in the figure. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. Among them, the processor, the memory and the input / output interface are connected through the system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capability. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner. The wireless manner can be realized through WIFI, mobile cellular network, near field communication (NFC) or other technologies. The computer program is executed by the processor to realize an industrial isolator health management method based on digital self-calibration. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0148] Those skilled in the art can understand that, Figure 3 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0149] In one embodiment, a computer device is also provided, including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to realize the steps in each of the above method embodiments.

[0150] In one embodiment, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to realize the steps in each of the above method embodiments.

[0151] In one embodiment, a computer program product is provided, including a computer program, and the computer program is executed by a processor to realize the steps in each of the above method embodiments.

[0152] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant regulations.

[0153] It can be understood by those skilled in the art that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing related hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments of each method. In the embodiments provided in the present application, any reference to memory, database or other medium 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 storage, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive 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. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., without being limited thereto.

[0154] Any technical features in the above embodiments can be combined, and for the sake of brevity, not all possible combinations are described above, however, any combination of the technical features is deemed to be within the scope of the present application.

[0155] The above embodiments only express several implementation manners of the present application, and the description is relatively specific and detailed, but it should not be understood as a limitation on the patent scope of the present application. It should be pointed out that, for ordinary skilled persons in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A method for health management of an industrial isolator based on digitized self-calibration, characterized in that, The method comprises the following steps: continuously collecting input terminal analog signal values and output terminal analog signal values of an industrial isolator, calculating real-time transmission errors of the industrial isolator based on the input terminal analog signal values and the output terminal analog signal values, and forming a real-time transmission error sequence; establishing an error change trend model through a fitting algorithm based on the real-time transmission error sequence within a preset time window, and predicting a transmission error value at a future time based on the error change trend model; in a case where a current value of the real-time transmission error sequence is less than a preset precision threshold value and the transmission error value at the future time is greater than the preset precision threshold value, calculating a preventive compensation value based on a difference between the transmission error value at the future time and a target error value; calculating a change rate of the output terminal analog signal values, determining an output change perception threshold value based on the change rate, decomposing the preventive compensation value into a plurality of compensation step amounts, each of the compensation step amounts generating an output change less than the output change perception threshold value, and sequentially accumulating and writing the compensation step amounts into a digital compensation register of the industrial isolator in a plurality of adjustment periods, and monitoring output signal continuity of the industrial isolator after each writing to complete online self-calibration of the output signal without disturbance.

2. A method for health management of industrial isolators based on digitized self- calibration as claimed in claim 1, wherein: The method of establishing an error change trend model through a fitting algorithm and predicting a transmission error value at a future time based on the error change trend model comprises the following steps: separating a zero error sequence and a gain error sequence from the real-time transmission error sequence; linearly fitting the zero error sequence to obtain a zero error change rate and a fitting residual; in a case where the fitting residual exceeds a preset linear threshold value, fitting the zero error sequence by using a high-order polynomial to obtain a zero error polynomial fitting result; in a case where the fitting residual does not exceed the preset linear threshold value, taking the zero error change rate obtained by the linear fitting as a zero error fitting result; extrapolating and calculating a zero error prediction value after a preset prediction time based on the zero error polynomial fitting result or the zero error fitting result and a current value of the zero error sequence; Similarly, the gain error sequence is processed to obtain a gain error prediction value; and the zero error prediction value and the gain error prediction value are taken as the transmission error value at the future time.

3. A method for health management of industrial isolators based on digitized self- calibration as claimed in claim 2, wherein: The method of calculating a preventive compensation value based on a difference between a transmission error value at a future time and a target error value comprises the following steps: setting a zero value as a target value of a zero error, and taking a difference between the zero error prediction value and the zero value as a total zero compensation amount; setting a rated transmission ratio as a target value of a gain error, and calculating a total gain compensation amount required to reach the rated transmission ratio; verifying whether the total zero compensation amount and the total gain compensation amount are within an adjustable range of the digital compensation register; in a case where the total zero compensation amount and the total gain compensation amount exceed the adjustable range, proportionally reducing the total zero compensation amount and the total gain compensation amount to an upper limit value of the adjustable range; in a case where the total zero compensation amount and the total gain compensation amount do not exceed the adjustable range, keeping the total zero compensation amount and the total gain compensation amount unchanged.

4. A method for health management of industrial isolators based on digitized self- calibration as claimed in claim 3, wherein: The calculation of the rate of change of the output analog signal value and the determination of the output change sensing threshold based on the rate of change include: Acquire multiple sampled values ​​of the output analog signal value within a preset time interval, and calculate the rate of change of the sampled values; If the rate of change is greater than a preset fast-changing signal threshold, the output change sensing threshold is set to a first preset sensing threshold. When the rate of change is less than a preset slow-changing signal threshold, the output change sensing threshold is set to a second preset sensing threshold; wherein the second preset sensing threshold is less than the first preset sensing threshold. When the rate of change is between the preset fast-changing signal threshold and the 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.

5. A method for health management of industrial isolators based on digitized self- calibration as claimed in claim 4, wherein: The step of decomposing the preventive compensation value into multiple compensation steps includes: Based on the output change sensing threshold and the ratio of the input analog signal value to the output analog signal value, the upper limit of the register value change corresponding to a single compensation is calculated. Divide the total zero-point compensation amount by the upper limit of the register value change and round up to obtain the number of zero-point compensation steps; The zero-point compensation step size is allocated in a decreasing manner, with the first step size being the largest and decreasing successively thereafter. Divide the total gain compensation by the upper limit of the register value change and round up to obtain the number of gain compensation steps; The gain compensation step size is allocated using a decreasing allocation method. 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.

6. A method for health management of industrial isolators based on digitized self- calibration as claimed in claim 5, wherein: The step of sequentially incrementing and writing the compensation step amount within multiple adjustment cycles, and monitoring the continuity of the output signal of the industrial isolator after each write, includes: Before writing, record the current value of the analog signal at the output terminal as a reference value; Perform a register write operation to accumulate the compensation step amount corresponding to the current adjustment period to the existing value of the register; The maximum deviation from the reference value is calculated by continuously sampling multiple analog signal values ​​at the output terminal. If the maximum deviation exceeds the output change sensing threshold, roll back the current write operation, reduce the compensation step size corresponding to the current adjustment cycle, and then retry. If the maximum deviation does not exceed the output change sensing threshold, the write operation is confirmed to be valid. If an excessive deviation still occurs after a preset number of retries, the gradual compensation is paused and an abnormal alarm is generated. After confirming the continuity of the output signal of the industrial isolator, wait for the signal to stabilize for a period of time, which is determined based on the response time of the industrial isolator.

7. A method for health management of industrial isolators based on digitized self- calibration as claimed in claim 6, wherein: The method further includes: Based on the execution results of the incremental compensation, a compensation effect evaluation model is established, and the prediction error, actual error and compensation effect deviation of each incremental compensation are recorded. When the deviation of the compensation effect shows an increasing trend, the fitting parameters of the error change trend model are updated; statistical unit time, and if the total compensation growth rate exceeds a preset aging threshold, shorten the preset time window and increase the compensation frequency; if the cumulative total compensation amount approaches the range limit of the digital compensation register, generate a device replacement warning.

8. A digital self-calibration based industrial isolator health management system employing the digital self-calibration based industrial isolator health management method as claimed in any one of claims 1 to 7. The method comprises the following steps: an error monitoring module for continuously collecting input and output analog signal values of an industrial isolator, calculating real-time transmission errors of the industrial isolator based on the input and output analog signal values, and forming a real-time transmission error sequence; a trend prediction module for establishing an error change trend model through a fitting algorithm based on the real-time transmission error sequence within a preset time window, and predicting a transmission error value at a future time based on the error change trend model; a compensation calculation module for calculating a preventive compensation value based on the difference between the transmission error value at the future time and a target error value if the current value of the real-time transmission error sequence is less than a preset precision threshold and the transmission error value at the future time is greater than the preset precision threshold; a threshold determination module for calculating the rate of change of the output analog signal value and determining an output change perception threshold based on the rate of change; a step decomposition module for decomposing the preventive compensation value into a plurality of compensation step amounts, each of which produces an output change less than the output change perception threshold; a gradual calibration module for sequentially accumulating the compensation step amounts in the digital compensation register of the industrial isolator within a plurality of adjustment periods, and monitoring the continuity of the output signal of the industrial isolator after each write to complete 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, and the computer device is configured to perform the method according to any one of claims 1-8 when the computer program is executed by the processor. The processor executes the computer program to realize the steps of the industrial isolator health management method based on digital self-calibration according to any one of claims 1 to 7.

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

Images