Intelligent control method and system for electric actuator

Abstract

The present invention discloses an intelligent control method and system for an electric actuator, which specifically relates to the field of intelligent control technology. By performing time series analysis on the valve position feedback signal, the stability of the valve response is evaluated, and the small signal chaotic characteristic quantity of the electric actuator is analyzed, and whether the chaotic critical state is reached is judged based on the chaos prediction algorithm. Combined with the judgment results of the valve response stability and the chaotic critical state, it is determined whether to trigger the advance intervention control strategy. The degree of precision degradation of the mechanical transmission components is evaluated by analyzing the mechanical displacement curve, and the influence of abnormal energy consumption on the operating efficiency is analyzed by the energy consumption data of the actuator. By comprehensively analyzing the valve response stability, the degree of precision degradation and the influence of abnormal energy consumption, the operating risk of the electric actuator is evaluated, and it is judged whether the control parameters need to be adjusted, thereby realizing accurate monitoring and intelligent optimization of the operating state of the electric actuator and improving the reliability of the operation of the electric actuator.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control technology, and more specifically, to an intelligent control method and system for an electric actuator. Background Art

[0002] Electric actuators are widely used in industrial automation control systems to accurately control the opening of valves. During long-term operation, electric actuators often suffer from reduced control accuracy and operating efficiency due to factors such as mechanical wear, abnormal vibration, and energy consumption fluctuations. Existing control methods mainly rely on simple valve position feedback signals for closed-loop control, which makes it difficult to timely identify potential anomalies in equipment operation, such as small signal chaotic vibration and mechanical precision degradation, and lack the ability to comprehensively evaluate the health status of the equipment and intelligently adjust it. In addition, traditional control strategies cannot implement advance intervention for abnormal conditions, which can easily lead to the accumulation of faults, increase equipment maintenance costs and downtime risks, and fail to achieve multi-dimensional real-time risk assessment and active control optimization.

[0003] In order to solve the above problems, an intelligent control method and system of an electric actuator are provided. Summary of the invention

[0004] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an intelligent control method of an electric actuator to solve the problems raised in the above-mentioned background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] An intelligent control method for an electric actuator comprises the following steps:

[0007] Perform time series analysis on the valve position feedback signal of the electric actuator to evaluate the stability of the valve response;

[0008] The small signal chaotic characteristic quantity of the electric actuator under the state of slight vibration is analyzed, and whether the vibration state of the electric actuator reaches the chaotic critical state is judged based on the chaos prediction algorithm;

[0009] Based on the stability of the valve response and the judgment result of whether the chaotic critical state is reached, determine whether to trigger the advance intervention control strategy;

[0010] When the advance intervention control strategy is triggered: by analyzing the mechanical displacement curve of the electric actuator, the accuracy degradation of the mechanical transmission components is evaluated using the curve fitting algorithm; by analyzing the energy consumption data of the electric actuator's actuator components, the impact of the electric actuator's abnormal energy consumption on the operating efficiency is evaluated;

[0011] Comprehensively analyze the stability of valve response, the degree of precision degradation of mechanical transmission components, and the impact of abnormal energy consumption of electric actuators on operating efficiency, evaluate the operating risks of electric actuators, and determine whether control parameters need to be adjusted.

[0012] In a preferred embodiment, a time series analysis is performed on the valve position feedback signal of the electric actuator to evaluate the stability of the valve response, specifically:

[0013] The valve position feedback signal of the electric actuator during operation is collected in real time through a high sampling rate sensor to generate high-precision time series data;

[0014] Adaptive filtering algorithm is applied to denoise the collected time series data;

[0015] Use machine learning prediction models to perform pattern recognition on filtered time series data to identify potential abnormal vibration patterns;

[0016] Based on the identified vibration modes, the dynamic stability of the valve response is evaluated.

[0017] In a preferred embodiment, the small signal chaotic characteristic quantity of the electric actuator in a slight vibration state is analyzed, and whether the vibration state of the electric actuator reaches a chaotic critical state is determined based on a chaos prediction algorithm, specifically:

[0018] Use high-precision acceleration sensors to collect high-frequency vibration signals of electric actuators in a state of slight vibration;

[0019] The collected vibration signals are de-noised and effective chaotic feature data are separated using wavelet transform method.

[0020] The Lyapunov exponent and phase trajectory distribution of the vibration signal are extracted using phase space reconstruction technology;

[0021] Apply the chaos prediction algorithm based on machine learning to perform trend analysis on the extracted feature quantities and predict the evolution of future vibration states;

[0022] The prediction result is compared with the preset small signal chaos critical threshold to determine whether the electric actuator has reached the chaos critical state: the predicted vibration signal value is compared with the preset small signal chaos critical threshold. When the predicted vibration signal value is greater than the preset small signal chaos critical threshold, it is determined that the electric actuator has reached the chaos critical state; when the predicted vibration signal value is less than or equal to the preset small signal chaos critical threshold, it is determined that the electric actuator has not reached the chaos critical state.

[0023] In a preferred embodiment, based on the stability of the valve response and the judgment result of whether the chaotic critical state is reached, it is determined whether to trigger the advance intervention control strategy, specifically:

[0024] Preset the volatility index threshold and compare the volatility index with the volatility index threshold:

[0025] When the volatility index is greater than the volatility index threshold, it means that the valve response has large volatility and the advance intervention control strategy needs to be triggered;

[0026] When the volatility index is less than or equal to the volatility index threshold, it means that the valve response is in a stable state and there is no need to trigger the advance intervention control strategy;

[0027] When the predicted vibration signal value is greater than the preset small signal chaos critical threshold and the volatility index is greater than the volatility index threshold, it is determined that the advance intervention control strategy is triggered; otherwise, the advance intervention control strategy is not triggered.

[0028] In a preferred embodiment, the mechanical displacement curve of the electric actuator is analyzed and the degree of accuracy degradation of the mechanical transmission component is evaluated using a curve fitting algorithm, specifically:

[0029] The mechanical displacement curve of the electric actuator is collected through a high-precision displacement sensor;

[0030] The key feature points and motion trend information of the mechanical displacement curve are extracted using the piecewise interpolation method;

[0031] The polynomial fitting algorithm is used to fit the mechanical displacement curve and the fitting residual is calculated;

[0032] According to the fitting residual, the accuracy degradation degree of the mechanical transmission components is evaluated: the accuracy degradation index is defined, and the expression is: ;in, Indicates the accuracy degradation index; and The start time and end time of the time interval respectively; is the fitting residual.

[0033] In a preferred embodiment, the influence of abnormal energy consumption of the electric actuator on the operating efficiency is evaluated by analyzing the energy consumption data of the actuator components of the electric actuator, specifically:

[0034] Collect energy consumption data of each actuator of the electric actuator, including static energy consumption and dynamic energy consumption;

[0035] Decompose the energy consumption data and extract the benchmark energy consumption characteristic value of each component under normal working conditions;

[0036] Calculate the current energy consumption deviation value of each execution component and compare the difference between the baseline energy consumption and the real-time energy consumption;

[0037] Evaluate the impact of abnormal energy consumption on the overall operating efficiency of the actuator: define the operating efficiency deviation index, and the calculation formula is: ;in, is the operating efficiency deviation index; For the The benchmark energy consumption characteristic value of each execution component; For the The energy consumption deviation value of each execution component; Indicates the number of execution components.

[0038] In a preferred embodiment, the stability of valve response, the degree of precision degradation of mechanical transmission components, and the influence of abnormal energy consumption of electric actuators on operating efficiency are comprehensively analyzed to evaluate the operating risk of the electric actuators and determine whether the control parameters need to be adjusted, specifically:

[0039] The volatility index corresponding to the stability of valve response, the precision degradation index corresponding to the degree of precision degradation of mechanical transmission components, and the operating efficiency deviation index corresponding to the impact of abnormal energy consumption of electric actuators on operating efficiency are normalized, and the normalized volatility index, precision degradation index, and operating efficiency deviation index are calculated to obtain the operating risk index. The calculation formula is: ;in, is the operational risk index; is the volatility index; is the accuracy degradation index; is the operating efficiency deviation index; is a non-zero positive number;

[0040] Preset the operation risk index threshold and compare the operation risk index with the operation risk index threshold:

[0041] When the operation risk index is less than or equal to the operation risk index threshold, it means that the operation status of the electric actuator is within the normal range, and there is no need to adjust the control parameters;

[0042] When the operation risk index is greater than the operation risk index threshold, it indicates that there is an abnormal risk in the operation state of the electric actuator and the control parameters need to be adjusted immediately.

[0043] On the other hand, the present invention provides an intelligent control system for an electric actuator, including a feedback signal analysis module, a chaotic characteristic quantity analysis module, a control strategy determination module, a displacement curve analysis module, an energy consumption data analysis module, and an operation risk assessment module;

[0044] Feedback signal analysis module: performs time series analysis on the valve position feedback signal of the electric actuator to evaluate the stability of the valve response;

[0045] Chaos characteristic quantity analysis module: analyzes the small signal chaos characteristic quantity of the electric actuator under the state of slight vibration, and judges whether the vibration state of the electric actuator reaches the chaotic critical state based on the chaos prediction algorithm;

[0046] Control strategy determination module: determines whether to trigger the advance intervention control strategy based on the stability of the valve response and the judgment result of whether the chaotic critical state is reached;

[0047] When the advance intervention control strategy is triggered: the displacement curve analysis module analyzes the mechanical displacement curve of the electric actuator and uses the curve fitting algorithm to evaluate the degree of accuracy degradation of the mechanical transmission components; the energy consumption data analysis module analyzes the energy consumption data of the electric actuator's actuator components to evaluate the impact of the electric actuator's abnormal energy consumption on its operating efficiency;

[0048] Operation risk assessment module: Comprehensively analyze the stability of valve response, the degree of precision degradation of mechanical transmission components, and the impact of abnormal energy consumption of electric actuators on operating efficiency, evaluate the operation risk of electric actuators, and determine whether control parameters need to be adjusted.

[0049] The technical effects and advantages of the intelligent control method and system of an electric actuator of the present invention are as follows:

[0050] Through the time series analysis of the valve position feedback signal and the detection of the small signal chaotic characteristic quantity, the accurate evaluation of the valve response stability and vibration anomalies is achieved, and early signs of dynamic instability can be identified. By introducing the advance intervention control strategy, the control strategy can be actively adjusted before the abnormal trend is detected to prevent further deterioration of equipment performance. In addition, through curve fitting analysis of the mechanical displacement curve and energy consumption data decomposition calculation, the degree of precision degradation of mechanical transmission components and the impact of abnormal energy consumption on operating efficiency can be quantitatively evaluated, thereby realizing multi-dimensional operating status monitoring. By comprehensively analyzing the impact of valve response, precision degradation and abnormal energy consumption, the operating risk of the electric actuator is effectively evaluated, and the control parameters are dynamically adjusted, which improves the early fault identification capability, operating stability and overall control accuracy of the electric actuator, and has high engineering practical value and reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 A schematic diagram of an intelligent control method for an electric actuator of the present invention;

[0052] Figure 2 The figure is a schematic diagram of the structure of an intelligent control system of an electric actuator of the present invention. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0054] Example 1

[0055] Figure 1 The present invention provides an intelligent control method for an electric actuator, which comprises the following steps:

[0056] Perform time series analysis on the valve position feedback signal of the electric actuator to evaluate the stability of the valve response;

[0057] The small signal chaotic characteristic quantity of the electric actuator under the state of slight vibration is analyzed, and whether the vibration state of the electric actuator reaches the chaotic critical state is judged based on the chaos prediction algorithm;

[0058] Based on the stability of the valve response and the judgment result of whether the chaotic critical state is reached, determine whether to trigger the advance intervention control strategy;

[0059] When the advance intervention control strategy is triggered: by analyzing the mechanical displacement curve of the electric actuator, the accuracy degradation of the mechanical transmission components is evaluated using the curve fitting algorithm; by analyzing the energy consumption data of the electric actuator's actuator components, the impact of the electric actuator's abnormal energy consumption on the operating efficiency is evaluated;

[0060] Comprehensively analyze the stability of valve response, the degree of precision degradation of mechanical transmission components, and the impact of abnormal energy consumption of electric actuators on operating efficiency, evaluate the operating risks of electric actuators, and determine whether control parameters need to be adjusted.

[0061] Specifically, the valve position feedback signal of the electric actuator is analyzed in time series to evaluate the stability of the valve response, including:

[0062] The high sampling rate sensor is used to collect the valve position feedback signal of the electric actuator in real time during operation to generate high-precision time series data: the high sampling rate sensor is fixedly installed on the valve movement part to collect the valve position feedback signal in real time. The sensor sampling rate is set to , in Hz, needs to satisfy > ,in, is the highest frequency of the target vibration signal. The time series data is represented by X=\left [ {{x}_{1}, {x}_{2}, …, {x}_{N}} \right ] ,in, At sampling time The feedback signal value of each sampling point, that is, the amplitude of the valve position feedback signal actually collected at the sampling time; is the sampling point number; is the total number of sampling points.

[0063] Sampling time By formula Calculate, where Time Series Data The resolution of Determine to ensure that critical information is not lost in high-frequency vibration scenarios.

[0064] Adaptive filtering algorithm is applied to denoise the collected time series data: Adaptive filtering algorithm is used to dynamically adjust the weight of the filter according to the frequency component of the feedback signal. The feedback signal after filtering is expressed as Y=\left [ {{y}_{1},{y}_{2},……,{y}_{N}} \right ] ,in, is the corresponding The feedback signal value of each sampling point is the corresponding feedback signal amplitude after filtering.

[0065] The filter model is: ;in, is the filter order; For the The weights of the filters are adaptively adjusted by the minimum mean square error criterion; is the feedback signal value at the sampling moment.

[0066] The weight update formula is: ;in, is the step size factor, which is used to control the weight adjustment rate; is the filtering error, and the calculation formula is: .

[0067] Use machine learning prediction models to perform pattern recognition on filtered time series data to identify potential abnormal vibration patterns: Use recursive neural networks to filter the feedback signals The analysis is performed to capture short-term and long-term dependencies in the time series. The model output is the vibration mode label Z=\left [ {{z}_{1}, {z}_{2}, …, {z}_{N}} \right ] .

[0068] Among them, the update formula of the recurrent neural network unit is: ; ;in, Represents the recurrent neural network at time The hidden state of represents the feature vector extracted from the time series; is the weight matrix of the hidden state, which indicates the influence weight of the previous hidden state on the current hidden state; is the weight matrix from the input signal to the hidden state, which represents the influence weight of the current input signal on the hidden state; is the hidden layer bias vector, which represents the fixed offset of the hidden layer and is used to improve the model fitting effect; For at the moment The classification label of is in the range of 0 or 1, indicating whether an abnormality is detected at the current moment, 1 indicates abnormality, and 0 indicates normality; is the weight matrix from the hidden state to the output layer, which represents the influence weight of the hidden state on the classification output; is the output layer bias, which represents the fixed offset of the classification model.

[0069] Binarization of abnormality determination labels: ;in, is the abnormality determination threshold.

[0070] Based on the identified vibration modes, the dynamic stability of the valve response is evaluated: the volatility index is defined, which is calculated as: ;in, is the volatility index; is the total number of sampling points; For the The classification label of each sampling point is 0 or 1; is the corresponding The absolute value of the feedback signal value at each sampling point.

[0071] The larger the volatility index, the more unstable the dynamic response of the valve is, and the larger the amplitude fluctuations are. The volatility index is obtained by time series analysis of the valve position feedback signal of the electric actuator, which reflects the amplitude of change and the degree of frequent fluctuations in the valve response during operation. A larger volatility index indicates that the fluctuation amplitude of the feedback signal is large, and more violent oscillations or periodic fluctuations may occur, which may cause the positioning accuracy of the valve to decrease and the control system to become unstable. At this time, the stability of the valve response is poor, and it is difficult to maintain a constant output, which may cause the control instructions to be unable to be accurately executed, and even affect long-term reliability. By monitoring the volatility index, potential problems in the valve response can be identified in a timely manner, and stability can be improved by adjusting the control strategy or optimizing the valve design to ensure that it can still operate stably under complex working conditions.

[0072] Specifically, the small signal chaotic characteristic quantity of the electric actuator in the state of slight vibration is analyzed, and whether the vibration state of the electric actuator reaches the chaotic critical state is judged based on the chaos prediction algorithm, including:

[0073] Use high-precision acceleration sensors to collect high-frequency vibration signals of electric actuators in a state of slight vibration: Use high-precision acceleration sensors (sampling frequency ≥ 20kHz) installed on the surface of the electric actuator to collect high-frequency vibration signals in a state of slight vibration. The collected vibration signals reflect the dynamic response characteristics of the electric actuator during operation.

[0074] The collected vibration signal is expressed as ;in, Indicates time in seconds; Represents the total number of sampled data points, which is determined by the sampling frequency and monitoring duration.

[0075] To ensure high-fidelity acquisition of vibration signals, the accelerometer adopts an automatic dynamic range adjustment function, which can adjust the sensitivity in real time in complex vibration environments to avoid signal distortion.

[0076] The collected vibration signal is denoised and the effective chaotic feature data is separated by wavelet transform method: the vibration signal is denoised by wavelet transform method. Perform multi-scale decomposition, expressed as: ;in, Indicates Layer-decomposed vibration signal components; is the noise component in the vibration signal; Represents the total number of decomposition scales.

[0077] The decomposed vibration signal components Apply soft threshold filtering method to filter out high-frequency noise and retain chaotic characteristic signals , whose expression is: ;in, and They represent the starting layer and the ending layer of the retained vibration signal component respectively; is the chaotic characteristic signal after denoising.

[0078] After processing, the filtered vibration signal is obtained .

[0079] Extraction of Lyapunov exponents and phase trajectory distribution of vibration signals using phase space reconstruction technology: Based on the delayed embedding method, Perform phase space reconstruction to obtain the phase space trajectory , whose expression is: ;in, Indicates time delay; represents the embedding dimension; is the reconstructed phase space trajectory matrix.

[0080] Compute the maximum Lyapunov exponent of a phase space trajectory , defined as: ;in, >0, indicating chaotic characteristics; represents the time step; Indicates time.

[0081] The distribution density of phase trajectories in phase space is statistically analyzed to obtain the phase trajectory density function , whose expression is: ;in, is any point in phase space; Indicate point The number of trajectories in the area; is the total number of trajectories.

[0082] Apply the chaos prediction algorithm based on machine learning to perform trend analysis on the extracted feature quantities and predict the evolution of future vibration states: The chaos prediction algorithm is based on machine learning, and the input features include: Lyapunov exponent , phase trajectory density function And the chaotic characteristic signal after denoising .

[0083] The prediction result is the vibration state at the future moment , whose expression is: ;in, is the mapping function of the chaos prediction model; is the predicted vibration signal value.

[0084] Through multiple iterative calculations, the model can generate vibration trends at several future time points and evaluate possible chaotic behavior.

[0085] The prediction result is compared with the preset small signal chaos critical threshold to determine whether the electric actuator has reached the chaos critical state: the predicted vibration signal value is compared with the preset small signal chaos critical threshold. When the predicted vibration signal value is greater than the preset small signal chaos critical threshold, it is determined that the electric actuator has reached the chaos critical state; when the predicted vibration signal value is less than or equal to the preset small signal chaos critical threshold, it is determined that the electric actuator has not reached the chaos critical state.

[0086] Specifically, based on the stability of the valve response and the judgment result of whether the chaotic critical state is reached, it is determined whether to trigger the advance intervention control strategy, including:

[0087] Preset the volatility index threshold and compare the volatility index with the volatility index threshold:

[0088] When the volatility index is greater than the volatility index threshold, it means that the valve response has a large volatility, which usually means that the dynamic stability of the valve under the current control state is poor, and there may be large feedback signal fluctuations, resulting in reduced positioning accuracy and ineffective operation. At this time, the valve response fluctuation exceeds the design allowable range and may cause adverse effects. Therefore, it is necessary to trigger the advance intervention control strategy, that is, take active adjustment measures and make timely corrections;

[0089] When the volatility index is less than or equal to the volatility index threshold, it means that the valve response is in a stable state, which means that the dynamic behavior of the valve is predictable and can run smoothly without large oscillations. At this time, there is no need to trigger the advance intervention control strategy.

[0090] The setting of the volatility index threshold is usually determined based on experimental data and actual application requirements. First, through historical data and a large number of experiments, the typical fluctuation range of the valve feedback signal under normal operation can be determined, which helps to define a reasonable range of the threshold. Secondly, the setting of the threshold takes into account the actual application requirements. For example, in precision control situations, such as optical equipment, the volatility index threshold is set lower because these scenarios require extremely high accuracy and stability. In some industrial applications, the threshold is set relatively loosely because these scenarios have a higher tolerance for small-scale fluctuations. In addition, the setting of the threshold should also take into account the impact of many aspects such as the response speed of the equipment, load changes, and environmental factors.

[0091] When the predicted vibration signal value is greater than the preset small signal chaos critical threshold and the volatility index is greater than the volatility index threshold, it is determined that the advance intervention control strategy is triggered; otherwise, the advance intervention control strategy is not triggered.

[0092] Specifically, by analyzing the mechanical displacement curve of the electric actuator, the accuracy degradation degree of the mechanical transmission components is evaluated using a curve fitting algorithm, including:

[0093] The mechanical displacement curve of the electric actuator is collected through a high-precision displacement sensor: The displacement signal of the mechanical transmission component during operation is collected in real time through a high-precision displacement sensor installed inside the electric actuator. The collected signal is the corresponding relationship between displacement and time, forming a mechanical displacement curve. The sensor has a nanometer-level resolution and can accurately capture the tiny displacement changes of the mechanical transmission component during movement, avoiding data distortion caused by mechanical clearance, vibration or external interference.

[0094] The collected displacement data is expressed as ;in, Indicates time in seconds; Indicates at a point in time The corresponding displacement value; is the total number of collected data points.

[0095] In order to ensure the high accuracy and stability of the collected data, high-frequency noise is filtered out through hardware filters, and real-time dynamic calibration is performed to correct the drift error and nonlinear response of the sensor.

[0096] The key feature points and motion trend information of the mechanical displacement curve are extracted using the segmented interpolation method: The collected mechanical displacement curve The curve may contain noise and redundant information. In order to accurately evaluate the motion characteristics of the transmission components, the segmented interpolation method is used to extract the key feature points and motion trends of the curve.

[0097] By analyzing the gradient change, the key points in the curve are determined, including extreme points (such as maximum displacement and minimum displacement points), inflection points, and uniform sampling points within the interval. Suppose the set of key feature points is ;in, Indicates The time coordinates of the key feature points; Indicates The displacement value of key feature points; The number of key feature points to be extracted is determined by the complexity of the curve.

[0098] The characteristic points are fitted using the piecewise interpolation method to reconstruct the motion trend of the mechanical displacement curve. The expression of the interpolation function is: ;in, is the displacement curve after interpolation fitting; For the The interpolation function for the interpolation interval.

[0099] Segmented interpolation can not only accurately reflect the local characteristics of the displacement curve, but also avoid the overfitting problem that may be caused by global fitting.

[0100] Use polynomial fitting algorithm to fit the mechanical displacement curve and calculate the fitting residual: Based on the curve feature extraction, use polynomial fitting algorithm to fit the mechanical displacement curve A global fit was performed, and the accuracy of the fit was assessed by residual analysis.

[0101] Assume that the fitting curve is , the expression is: ;in, is the displacement curve after polynomial fitting; The coefficients of the fitted polynomial are calculated by the least squares method; is the order of the polynomial, which is selected based on the fitting accuracy and curve complexity.

[0102] By using the least squares method, the calculation formula for the fitting polynomial coefficients is: ;in, Represents the polynomial coefficient vector ; is the matrix form of time data; is the corresponding displacement data vector.

[0103] The expression of the residual is: ;in, It is the fitting residual, reflecting the deviation between the fitting curve and the actual curve.

[0104] According to the fitting residual, the accuracy degradation degree of the mechanical transmission components is evaluated: the accuracy degradation index is defined, and its expression is: ;in, Indicates the accuracy degradation index; and The start time and end time of the time interval respectively; is the fitting residual.

[0105] The larger the precision degradation index, the more serious the precision degradation of the mechanical transmission components of the electric actuator, and the greater the degree to which its motion trajectory deviates from the ideal state. This indicates that the mechanical transmission components may have wear, looseness or other structural problems, resulting in an increase in the residual between the actual mechanical displacement curve and the fitting curve. The increase in the precision degradation index also reflects that the mechanical transmission components have large motion errors and dynamic instability during operation. This situation may have an adverse effect on the overall performance of the electric actuator, such as reduced control accuracy, increased energy loss, or reduced operating efficiency. When the degradation index exceeds the set threshold, the mechanical transmission components may be close to failure and there is a potential safety risk.

[0106] Specifically, by analyzing the energy consumption data of the electric actuator's actuator components, the impact of abnormal energy consumption of the electric actuator on its operating efficiency is evaluated, including:

[0107] Collect energy consumption data of each actuator of the electric actuator, including static energy consumption and dynamic energy consumption: The collected energy consumption data includes static energy consumption and dynamic energy consumption:

[0108] Static energy consumption refers to the power required to maintain the operation of the actuator in a non-moving state, such as the power consumption when standing by or holding a position;

[0109] Dynamic energy consumption refers to the power consumption of the actuator due to factors such as movement and load changes during actual operation.

[0110] The collected energy consumption data is represented as a set ;in, Represents the static energy consumption of each execution component, in watts; Represents the dynamic energy consumption of each execution component, in watts; Indicates the number of execution components; and Respectively represent The static and dynamic energy consumption of each execution unit.

[0111] Decompose the energy consumption data and extract the benchmark energy consumption characteristic values ​​of each component under normal working conditions: separate the energy consumption data into static and dynamic states, and the decomposition formula is as follows: ;in, For the The total energy consumption of each execution component; and Respectively The static energy consumption value and dynamic energy consumption value of each execution component.

[0112] Under normal working conditions, the energy consumption data in multiple working cycles are statistically analyzed to calculate the baseline energy consumption characteristic value of each actuator. : ;in, For the The benchmark energy consumption characteristic value of each execution component; For the In the second working cycle The total energy consumption of each execution component; is the number of operating cycles during statistics.

[0113] Calculate the current energy consumption deviation value of each execution component and compare the difference between the baseline energy consumption and the real-time energy consumption: In order to evaluate the operating status of the execution component, compare the current real-time energy consumption data with the baseline energy consumption characteristic value and calculate the energy consumption deviation value , the calculation formula is: ;in, For the The benchmark energy consumption characteristic value of each execution component; For the The total energy consumption of each execution component; For the The energy consumption deviation value of each execution component.

[0114] Energy consumption deviation The positive or negative value indicates the changing trend of energy consumption:

[0115] When the energy consumption deviation value is greater than 0, it indicates that the real-time energy consumption of the execution component is higher than the benchmark value, and there is abnormal energy consumption;

[0116] When the energy consumption deviation value is less than 0, it indicates that the real-time energy consumption of the execution component is higher than the reference value due to load reduction or state change.

[0117] Evaluate the impact of abnormal energy consumption on the overall operating efficiency of the actuator: define the operating efficiency deviation index to represent the efficiency change caused by abnormal energy consumption, and its calculation formula is: ;in, is the operating efficiency deviation index; For the The benchmark energy consumption characteristic value of each execution component; For the The energy consumption deviation value of each execution component; Indicates the number of execution components.

[0118] The larger the operating efficiency deviation index is, the more serious the decline in the operating efficiency of the electric actuator is, indicating that the actual energy consumption of the actuator deviates more from the baseline energy consumption characteristic value, reflecting that more electrical energy consumption has not been effectively converted into mechanical work output of the actuator, resulting in energy waste. An increase in the operating efficiency deviation index may indicate that some actuators have performance degradation, such as increased mechanical friction, aging of motor coils, or bearing wear. These degradations cause the actuators to consume additional energy to maintain the same working state, thereby reducing the overall operating efficiency. If the operating efficiency deviation index increases significantly and is accompanied by fluctuations in the operating state, it may reflect that the actuator is subjected to abnormal loads (such as valve jamming or overload operation), resulting in increased component wear and increased maintenance costs, and even the risk of equipment failure or shutdown of the electric actuator.

[0119] Specifically, the stability of valve response, the degree of precision degradation of mechanical transmission components, and the impact of abnormal energy consumption of electric actuators on operating efficiency are comprehensively analyzed to evaluate the operating risk of the electric actuator and determine whether the control parameters need to be adjusted, including:

[0120] The volatility index corresponding to the stability of valve response, the precision degradation index corresponding to the precision degradation degree of mechanical transmission components, and the operating efficiency deviation index corresponding to the impact of abnormal energy consumption of electric actuators on operating efficiency are normalized, and the normalized volatility index, precision degradation index and operating efficiency deviation index are calculated to obtain the operating risk index and determine whether the control parameters need to be adjusted.

[0121] The calculation formula of the operation risk index is: ;in, is the operational risk index; is the volatility index; is the accuracy degradation index; is the operating efficiency deviation index; A non-zero positive number.

[0122] Preset the operation risk index threshold and compare the operation risk index with the operation risk index threshold:

[0123] When the operation risk index is less than or equal to the operation risk index threshold, it means that the operation status of the electric actuator is within the normal range, and there is no need to adjust the control parameters;

[0124] When the operation risk index is greater than the operation risk index threshold, it means that there is an abnormal risk in the operation state of the electric actuator, and the control parameters need to be adjusted immediately to reduce the risk level. At this time, the abnormality may be caused by problems such as the deterioration of the accuracy of the mechanical transmission components, abnormal fluctuations in the energy consumption of the actuators, or maladjustment of the control response. The adjustment of control parameters includes reducing the control gain, adjusting the upper limit of the drive current of the actuator, or optimizing the proportional, integral and differential parameters of the PID controller to balance the control response speed and stability. In addition, it is also necessary to conduct a comprehensive self-inspection of the electric actuator in combination with the operation data to determine whether there is physical loss or sensor failure.

[0125] Example 2

[0126] The difference between Example 2 of the present invention and Example 1 is that this example introduces an intelligent control system of an electric actuator.

[0127] Figure 2 A structural schematic diagram of an intelligent control system of an electric actuator of the present invention is given, and an intelligent control system of an electric actuator includes a feedback signal analysis module, a chaos characteristic quantity analysis module, a control strategy determination module, a displacement curve analysis module, an energy consumption data analysis module, and an operation risk assessment module;

[0128] Feedback signal analysis module: performs time series analysis on the valve position feedback signal of the electric actuator to evaluate the stability of the valve response;

[0129] Chaos characteristic quantity analysis module: analyzes the small signal chaos characteristic quantity of the electric actuator under the state of slight vibration, and judges whether the vibration state of the electric actuator reaches the chaotic critical state based on the chaos prediction algorithm;

[0130] Control strategy determination module: determines whether to trigger the advance intervention control strategy based on the stability of the valve response and the judgment result of whether the chaotic critical state is reached;

[0131] When the advance intervention control strategy is triggered: the displacement curve analysis module analyzes the mechanical displacement curve of the electric actuator and uses the curve fitting algorithm to evaluate the degree of accuracy degradation of the mechanical transmission components; the energy consumption data analysis module analyzes the energy consumption data of the electric actuator's actuator components to evaluate the impact of the electric actuator's abnormal energy consumption on its operating efficiency;

[0132] Operation risk assessment module: Comprehensively analyze the stability of valve response, the degree of precision degradation of mechanical transmission components, and the impact of abnormal energy consumption of electric actuators on operating efficiency, evaluate the operation risk of electric actuators, and determine whether control parameters need to be adjusted.

[0133] The above formulas are dimensionless and numerical calculations. The formula is a formula obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions. The parts not described in the present invention are applicable to the prior art.

Claims

1. An intelligent control method for an electric actuator, characterized in that: The steps include: Perform time series analysis on the valve position feedback signal of the electric actuator to evaluate the stability of the valve response; The small signal chaotic characteristic quantity of the electric actuator under the state of slight vibration is analyzed, and whether the vibration state of the electric actuator reaches the chaotic critical state is judged based on the chaos prediction algorithm; Based on the stability of the valve response and the judgment result of whether the chaotic critical state is reached, determine whether to trigger the advance intervention control strategy; Preset the volatility index threshold and compare the volatility index with the volatility index threshold: When the volatility index is greater than the volatility index threshold, it means that the valve response has large volatility and the advance intervention control strategy needs to be triggered; When the volatility index is less than or equal to the volatility index threshold, it means that the valve response is in a stable state and there is no need to trigger the advance intervention control strategy; When the predicted vibration signal value is greater than the preset small signal chaos critical threshold, and the volatility index is greater than the volatility index threshold, it is determined that the advance intervention control strategy is triggered; otherwise, the advance intervention control strategy is not triggered; When the advance intervention control strategy is triggered: by analyzing the mechanical displacement curve of the electric actuator, the accuracy degradation of the mechanical transmission components is evaluated using the curve fitting algorithm; by analyzing the energy consumption data of the electric actuator's actuator components, the impact of the electric actuator's abnormal energy consumption on the operating efficiency is evaluated; Comprehensively analyze the stability of valve response, the degree of precision degradation of mechanical transmission components, and the impact of abnormal energy consumption of electric actuators on operating efficiency, evaluate the operating risks of electric actuators, and determine whether control parameters need to be adjusted.

2. The intelligent control method of an electric actuator according to claim 1, characterized in that: The valve position feedback signal of the electric actuator is analyzed in time series to evaluate the stability of the valve response, specifically: The valve position feedback signal of the electric actuator during operation is collected in real time through a high sampling rate sensor to generate high-precision time series data; Adaptive filtering algorithm is applied to denoise the collected time series data; Use machine learning prediction models to perform pattern recognition on filtered time series data to identify potential abnormal vibration patterns; Based on the identified vibration modes, the dynamic stability of the valve response is evaluated: the volatility index is defined and calculated as: Where STI is the volatility index; N is the total number of sampling points; z i is the classification label of the i-th sampling point; |y i | is the absolute value of the feedback signal value of the i-th sampling point after filtering.

3. The intelligent control method of an electric actuator according to claim 2, characterized in that: The small signal chaotic characteristic quantity of the electric actuator under the state of slight vibration is analyzed, and whether the vibration state of the electric actuator reaches the chaotic critical state is judged based on the chaos prediction algorithm. Specifically: Use high-precision acceleration sensors to collect high-frequency vibration signals of electric actuators in a state of slight vibration; The collected vibration signals are de-noised and effective chaotic feature data are separated using wavelet transform method. The Lyapunov exponent and phase trajectory distribution of the vibration signal are extracted using phase space reconstruction technology; Apply the chaos prediction algorithm based on machine learning to perform trend analysis on the extracted feature quantities and predict the evolution of future vibration states; The prediction result is compared with the preset small signal chaos critical threshold to determine whether the electric actuator has reached the chaos critical state: the predicted vibration signal value is compared with the preset small signal chaos critical threshold. When the predicted vibration signal value is greater than the preset small signal chaos critical threshold, it is determined that the electric actuator has reached the chaos critical state; when the predicted vibration signal value is less than or equal to the preset small signal chaos critical threshold, it is determined that the electric actuator has not reached the chaos critical state.

4. The intelligent control method of an electric actuator according to claim 3, characterized in that: By analyzing the mechanical displacement curve of the electric actuator, the accuracy degradation degree of the mechanical transmission components is evaluated using the curve fitting algorithm, specifically: The mechanical displacement curve of the electric actuator is collected through a high-precision displacement sensor; The key feature points and motion trend information of the mechanical displacement curve are extracted using the piecewise interpolation method; The polynomial fitting algorithm is used to fit the mechanical displacement curve and the fitting residual is calculated; According to the fitting residual, the accuracy degradation degree of the mechanical transmission components is evaluated: the accuracy degradation index is defined, and the expression is: Wherein, ADI represents the accuracy degradation index; T0 and T R are the start time and end time of the time interval respectively; R(T) is the fitting residual.

5. The intelligent control method of an electric actuator according to claim 4, characterized in that: By analyzing the energy consumption data of the electric actuator's actuator components, the impact of abnormal energy consumption of the electric actuator on its operating efficiency is evaluated, specifically: Collect energy consumption data of each actuator of the electric actuator, including static energy consumption and dynamic energy consumption; Decompose the energy consumption data and extract the benchmark energy consumption characteristic value of each component under normal working conditions; Calculate the current energy consumption deviation value of each execution component and compare the difference between the baseline energy consumption and the real-time energy consumption; Evaluate the impact of abnormal energy consumption on the overall operating efficiency of the actuator: define the operating efficiency deviation index, and the calculation formula is: Among them, OEI is the operating efficiency deviation index; B o is the benchmark energy consumption characteristic value of the oth execution component; D o is the energy consumption deviation value of the oth execution component; J represents the number of execution components.

6. The intelligent control method of an electric actuator according to claim 5, characterized in that: Comprehensively analyze the stability of valve response, the degree of precision degradation of mechanical transmission components, and the impact of abnormal energy consumption of electric actuators on operating efficiency, evaluate the operating risks of electric actuators, and determine whether control parameters need to be adjusted, specifically: The volatility index corresponding to the stability of valve response, the precision degradation index corresponding to the degree of precision degradation of mechanical transmission components, and the operating efficiency deviation index corresponding to the impact of abnormal energy consumption of electric actuators on operating efficiency are normalized, and the normalized volatility index, precision degradation index, and operating efficiency deviation index are calculated to obtain the operating risk index. The calculation formula is: Among them, ORI is the operation risk index; STI is the volatility index; ADI is the accuracy degradation index; OEI is the operation efficiency deviation index; δ is a non-zero positive number; Preset the operation risk index threshold and compare the operation risk index with the operation risk index threshold: When the operation risk index is less than or equal to the operation risk index threshold, it means that the operation status of the electric actuator is within the normal range, and there is no need to adjust the control parameters; When the operation risk index is greater than the operation risk index threshold, it indicates that there is an abnormal risk in the operation state of the electric actuator and the control parameters need to be adjusted immediately.

7. An intelligent control system for an electric actuator, used to implement an intelligent control method for an electric actuator according to any one of claims 1 to 6, characterized in that: It includes feedback signal analysis module, chaos characteristic quantity analysis module, control strategy determination module, displacement curve analysis module, energy consumption data analysis module and operation risk assessment module; Feedback signal analysis module: performs time series analysis on the valve position feedback signal of the electric actuator to evaluate the stability of the valve response; Chaos characteristic quantity analysis module: analyzes the small signal chaos characteristic quantity of the electric actuator under the state of slight vibration, and judges whether the vibration state of the electric actuator reaches the chaotic critical state based on the chaos prediction algorithm; Control strategy determination module: determines whether to trigger the advance intervention control strategy based on the stability of the valve response and the judgment result of whether the chaotic critical state is reached; When the advance intervention control strategy is triggered: the displacement curve analysis module analyzes the mechanical displacement curve of the electric actuator and uses the curve fitting algorithm to evaluate the degree of accuracy degradation of the mechanical transmission components; the energy consumption data analysis module analyzes the energy consumption data of the electric actuator's actuator components to evaluate the impact of the electric actuator's abnormal energy consumption on its operating efficiency; Operation risk assessment module: Comprehensively analyze the stability of valve response, the degree of precision degradation of mechanical transmission components, and the impact of abnormal energy consumption of electric actuators on operating efficiency, evaluate the operation risk of electric actuators, and determine whether control parameters need to be adjusted.

Images