Intelligent monitoring method for servo valve failure

Abstract

The application discloses a kind of servo valve fault intelligent monitoring methods, by the following steps to establish real-time monitoring whether servo valve is safe operation: S1: based on servo amplifier input voltage and the historical data distribution characteristics of servo valve main valve spool position, set envelope curve function;S2: based on the envelope curve function of setting judgment model;S3: according to the data of real-time acquisition of set acquisition cycle is sent into judgment model, determines whether servo valve is in safe working state according to the judgment result of judgment model.The application discloses a kind of servo valve fault intelligent monitoring methods, can real-time monitoring whether the data point of servo valve is within envelope curve range, and zero drift condition, to realize servo valve real-time fault monitoring.It is the premise of realizing hydraulic servo system fault monitoring, and also the basis of realizing intelligent operation and maintenance of mechanical equipment containing hydraulic servo system.

Description

Technical Field

[0001] This invention belongs to the field of hydraulic fault monitoring technology, specifically relating to an intelligent monitoring method for servo valve faults. Background Technology

[0002] With the "Made in China 2025" initiative, intelligent manufacturing has become the main theme of industrial development. Many key pieces of equipment on factory production lines utilize hydraulic servo control systems, such as rolling mills, leveling machines, and coil trolleys. Servo valves are a crucial component of these systems; malfunctions can cause significant economic losses. Therefore, effective fault monitoring of servo valves is essential. Currently, troubleshooting servo valve malfunctions in industrial settings requires manual inspection, which is time-consuming, labor-intensive, and reduces production efficiency, severely impacting the company's economic benefits. With the advent of the big data era, real-time monitoring of servo valve malfunctions can be achieved by collecting key data from the valves.

[0003] The invention application with application number CN202010840213.0 discloses "a method and system for fault diagnosis of electro-hydraulic servo valves based on transfer learning", which includes the following steps: acquiring an electro-hydraulic servo valve fault database, including no-load flow data and fault types under corresponding current; preprocessing the electro-hydraulic servo valve data to establish a training set; using the training set, establishing a fault diagnosis model based on a transfer learning framework of Mahalanobis metric; processing the sample data of the electro-hydraulic servo valve to be tested, and using the established fault model to identify faults in the electro-hydraulic servo valve data.

[0004] The invention application with application number CN201611161353.5 discloses "a fault diagnosis method for an electro-hydraulic servo valve", which includes the following steps: 1) Parameter acquisition: Acquiring servo valve parameters including valve left chamber pressure, valve right chamber pressure, valve inlet flow rate, valve outlet flow rate, valve body temperature, and valve current; 2) Data processing: Processing the acquired servo valve parameters using singular value decomposition and cosine analysis; 3) Fault diagnosis: Training a fast and accurate neural network model using prepared parameters, and performing fault diagnosis based on the network model; 4) Result output and processing: Outputting and inverse normalizing the model, displaying and processing the diagnostic results.

[0005] The invention application with application number CN201910823504.6 discloses a "method and device for diagnosing faults in electro-hydraulic servo valves, a storage medium, and an electronic device". The method includes: collecting data from several electro-hydraulic servo valves; dividing the acquired characteristic curves into an upper characteristic curve and a lower characteristic curve with the flow rate value equal to zero as the boundary; calculating the flow rate difference between the upper and lower characteristic curves corresponding to the same current value; verifying whether each flow rate difference conforms to a normal distribution using the Kolmogorov-Smirnov method; if it conforms to a normal distribution, calculating the mean and standard deviation of the flow rate difference between the upper and lower characteristic curves of each electro-hydraulic servo valve; and performing statistical tests on each electro-hydraulic servo valve based on the statistical F-test to determine whether each electro-hydraulic servo valve has malfunctioned. Summary of the Invention

[0006] This invention provides an intelligent monitoring method for servo valve faults, the specific technical solution of which is as follows:

[0007] A method for intelligent monitoring of servo valve faults, characterized by establishing real-time monitoring of whether the servo valve is operating safely through the following steps:

[0008] S1: Based on the historical data distribution characteristics of the servo amplifier input voltage and the servo valve main valve spool position, set the envelope curve function;

[0009] S2: Establish a judgment model based on the set envelope curve function;

[0010] S3: The real-time collected data is sent to the judgment model according to the set collection cycle, and the judgment result of the judgment model is used to determine whether the servo valve is in a safe working state.

[0011] According to the present invention, a servo valve fault intelligent monitoring method is characterized in that:

[0012] The envelope curve function is:

[0013]

[0014] in,

[0015] x: Input voltage;

[0016] y: Main valve spool position;

[0017] a: The semi-major axis of the elliptic function;

[0018] b: The minor semi-axis of the elliptic function;

[0019] Note: The input voltage and valve core position are converted digital values ​​without units. The servo valve used in this case has a range of -16000 to +16000.

[0020] According to the present invention, a servo valve fault intelligent monitoring method is characterized in that:

[0021] The specific judgment model in step S2 is as follows:

[0022]

[0023] According to the present invention, a servo valve fault intelligent monitoring method is characterized in that:

[0024] The aforementioned intelligent monitoring of servo valve faults also includes real-time monitoring of the servo valve's zero-point drift status.

[0025] The real-time monitoring of the servo valve's zero-point drift status specifically includes:

[0026] SS1: Determine the theoretical zero-point position of the servo valve based on historical data of the servo amplifier input voltage and the position of the servo valve main valve spool;

[0027] SS2: Real-time acquisition of the servo amplifier input voltage and the servo valve main valve core position according to the set acquisition cycle, and determination of the real-time zero point position of the servo valve based on the acquisition data within the cycle;

[0028] SS3: Compare the real-time zero position with the theoretical zero position. If there is no deviation between the two, it is determined that no zero drift has occurred; otherwise, it is determined that zero drift has occurred.

[0029] According to the present invention, a servo valve fault intelligent monitoring method is characterized in that:

[0030] In the envelope curve function, a and b are determined through the following steps:

[0031] S21: Determine the value of a based on the historical maximum and minimum values ​​of the servo amplifier input voltage and the slope of the historical data distribution fitting;

[0032] S22: Determine the value of b based on three factors: the determined value of a, the envelope curve function, and the requirements for hysteresis calculation.

[0033] According to the present invention, a servo valve fault intelligent monitoring method is characterized in that:

[0034] If the real-time data calculation results within 5 consecutive acquisition cycles meet the judgment model, the servo valve is determined to be in a safe working state; otherwise, the servo valve is determined to be faulty.

[0035] According to the present invention, a servo valve fault intelligent monitoring method is characterized in that:

[0036] The slope based on historical data distribution fitting is specifically as follows:

[0037] A least-squares-based linear fit is performed on the historical data distribution of the servo amplifier input voltage and the servo valve main valve spool position, and the slope is determined based on the fit.

[0038] According to the present invention, a servo valve fault intelligent monitoring method is characterized in that:

[0039] The established real-time monitoring of whether the servo valve is operating safely and the established real-time monitoring of the servo valve's zero-point drift status are operated in parallel.

[0040] According to the present invention, a servo valve fault intelligent monitoring method is characterized in that:

[0041] The theoretical zero point position and the real-time zero point position are both determined based on the weighted average of the data values ​​in their respective data spaces.

[0042] According to the present invention, a servo valve fault intelligent monitoring method is characterized in that:

[0043] The real-time collected data is first converted into a numerical expression in the rectangular coordinate system of the envelope curve function through linear rotation, and then the corresponding judgment or comparison is performed.

[0044] The rotation angle of the linear rotation is determined based on the slope of the historical data distribution fit.

[0045] According to the present invention, a servo valve fault intelligent monitoring method is characterized in that:

[0046] The value of 'a' is determined as follows:

[0047]

[0048] in,

[0049] u l The minimum input voltage value in historical data;

[0050] u r The maximum input voltage value in historical data;

[0051] θ: The angle between the fitted slopes.

[0052] According to the present invention, a servo valve fault intelligent monitoring method is characterized in that:

[0053] The slope based on historical data distribution fitting is specifically as follows:

[0054] A least-squares-based linear fit is performed on the historical data distribution of the servo amplifier input voltage and the servo valve main valve spool position, and the slope is determined based on the fit.

[0055] This invention discloses an intelligent fault monitoring method for servo valves. By analyzing historical data of the servo amplifier input voltage and the servo valve main valve spool position, an envelope curve function is designed for real-time fault monitoring of the servo valve. Simultaneously, the actual zero-point coordinates of the servo valve can be obtained. On one hand, if a data point repeatedly falls outside the envelope curve region within a continuous time period, a fault is considered to have occurred, and an alarm is triggered. Simultaneously, a fault-tolerance mechanism is introduced into the monitoring model; if only a single data point exceeds the envelope range, it is considered normal and no alarm is triggered. On the other hand, by using a time-shifting sliding window, the data center over a period of time is calculated and compared with the center zero-point coordinates obtained from historical data to obtain the current zero-point offset of the servo valve. Using this invention, it is possible to monitor in real-time whether the servo valve's data points are within the envelope curve range and to detect zero-point drift, thereby achieving real-time fault monitoring of the servo valve. This invention is a prerequisite for fault monitoring of hydraulic servo systems and also forms the foundation for intelligent operation and maintenance of mechanical equipment containing hydraulic servo systems. Attached Figure Description

[0056] Figure 1 This is a schematic diagram of the steps of the present invention;

[0057] Figure 2 This is a schematic diagram of the steps for monitoring the zero-point drift of the servo valve in this invention;

[0058] Figure 3 This is a schematic diagram illustrating the steps for determining the values ​​of a and b in the envelope curve function of this invention.

[0059] Figure 4 This is a schematic diagram of the intelligent monitoring process for servo valve faults according to an embodiment of the present invention;

[0060] Figure 5 This is a schematic diagram of the signal control principle of the hydraulic servo system in an embodiment of the present invention;

[0061] Figure 6 This is a schematic diagram illustrating the coordinate transformation principle of step 6 in this embodiment of the invention;

[0062] Figure 7 This is a schematic diagram of the servo data distribution and envelope curve of the DS side of the leveling machine in an embodiment of the present invention;

[0063] Figure 8 This is a schematic diagram of the servo data distribution and envelope curve of the WS side of the leveling machine in an embodiment of the present invention;

[0064] Figure 9 This is a simulation result diagram of the DS-side servo data of the leveling machine in Embodiment 1 of the present invention;

[0065] Figure 10 This is a simulation result diagram of the servo data on the WS side of the leveling machine in Embodiment 1 of the present invention;

[0066] Figure 11 This is a simulation result diagram of the DS-side servo data of the leveling machine in Embodiment 2 of the present invention;

[0067] Figure 12 This is a simulation result diagram of the servo data of the leveling machine on the WS side in Implementation Example 2 of the present invention. Detailed Implementation

[0068] The following is a further detailed description of a servo valve fault intelligent monitoring method of the present invention, based on the accompanying drawings and specific embodiments.

[0069] A method for intelligent monitoring of servo valve faults, such as Figure 1 As shown, real-time monitoring of the safe operation of the servo valve is established through the following steps:

[0070] S1: Based on the historical data distribution characteristics of the servo amplifier input voltage and the servo valve main valve spool position, set the envelope curve function;

[0071] S2: Establish a judgment model based on the set envelope curve function;

[0072] S3: The real-time collected data is sent to the judgment model according to the set collection cycle, and the judgment result of the judgment model is used to determine whether the servo valve is in a safe working state.

[0073] in,

[0074] The envelope curve function is:

[0075]

[0076] in,

[0077] x: Input voltage;

[0078] y: Main valve spool position;

[0079] a: The semi-major axis of the elliptic function;

[0080] b: The minor semi-axis of the elliptic function.

[0081] in,

[0082] The specific judgment model in step S2 is as follows:

[0083]

[0084] in,

[0085] The aforementioned intelligent monitoring of servo valve faults also includes real-time monitoring of the servo valve's zero-point drift status.

[0086] The aforementioned real-time monitoring of the servo valve's zero-point drift status, such as Figure 2 As shown, specifically:

[0087] SS1: Determine the theoretical zero-point position of the servo valve based on historical data of the servo amplifier input voltage and the position of the servo valve main valve spool;

[0088] SS2: Real-time acquisition of the servo amplifier input voltage and the servo valve main valve core position according to the set acquisition cycle, and determination of the real-time zero point position of the servo valve based on the acquisition data within the cycle;

[0089] SS3: Compare the real-time zero position with the theoretical zero position. If there is no deviation between the two, it is determined that no zero drift has occurred; otherwise, it is determined that zero drift has occurred.

[0090] Among them, such as Figure 3 As shown,

[0091] In the envelope curve function, a and b are determined through the following steps:

[0092] S21: Determine the value of a based on the historical maximum and minimum values ​​of the servo amplifier input voltage and the slope of the historical data distribution fitting;

[0093] S22: Determine the value of b based on three factors: the determined value of a, the envelope curve function, and the requirements for hysteresis calculation.

[0094] in,

[0095] If the real-time data calculation results within 5 consecutive acquisition cycles meet the judgment model, the servo valve is determined to be in a safe working state; otherwise, the servo valve is determined to be faulty.

[0096] in,

[0097] The slope based on historical data distribution fitting is specifically as follows:

[0098] A least-squares-based linear fit is performed on the historical data distribution of the servo amplifier input voltage and the servo valve main valve spool position, and the slope is determined based on the fit.

[0099] in,

[0100] The established real-time monitoring of whether the servo valve is operating safely and the established real-time monitoring of the servo valve's zero-point drift status are operated in parallel.

[0101] in,

[0102] The theoretical zero point position and the real-time zero point position are both determined based on the weighted average of the data values ​​in their respective data spaces.

[0103] in,

[0104] The real-time collected data is first converted into a numerical expression in the rectangular coordinate system of the envelope curve function through linear rotation, and then the corresponding judgment or comparison is performed.

[0105] The rotation angle of the linear rotation is determined based on the slope of the historical data distribution fit.

[0106] in,

[0107] The value of 'a' is determined as follows:

[0108]

[0109] in,

[0110] u l The minimum input voltage value in historical data;

[0111] u r The maximum input voltage value in historical data;

[0112] θ: The angle between the fitted slopes.

[0113] Working process, principle and implementation examples

[0114] The following reference Figure 4-8 Explanation of the working process principle:

[0115] First, using the servo amplifier input voltage as the horizontal axis and the servo valve main valve core position as the vertical axis, an envelope curve function is designed based on the historical distribution of its data points, serving as the "safety domain" of the data distribution.

[0116] Secondly, by using the coordinates of the historical data of the input voltage of the servo amplifier and the position of the main valve core of the servo valve, the data center point can be determined as the actual zero point of the servo valve.

[0117] Finally, by determining whether the real-time data points of the input voltage and the main valve spool position are within the envelope function region, it is determined whether the servo valve is in normal working condition. Simultaneously, by using a time-shifting sliding window to obtain the data center over a period of time and comparing it with the center zero-point coordinates obtained from historical data, the current zero-point offset of the servo valve can be obtained, thereby achieving servo valve fault monitoring.

[0118] To achieve the above technical effects, the specific process and principles are explained through the following steps:

[0119] Step 1: Based on the historical distribution of the servo amplifier input voltage and the servo valve main valve spool position data, design an envelope curve function as the "safety region" of the data distribution. The specific envelope function is shown in formula (1).

[0120]

[0121] In the formula, x represents the input voltage, y represents the position of the main valve core, and a and b represent the major and minor axes of the envelope region, respectively. Note: The input voltage and valve core position are converted digital values ​​without units. The servo valve range used in this case is -16000 to +16000.

[0122] Step 2: The envelope curve function mentioned in Step 1 is located in the new coordinate system X1O1Y1, such as... Figure 6 As shown, the coordinate system X1OY1 belongs to the new rectangular coordinate system of the "fault monitoring and judgment area", and UOV is the rectangular coordinate system where the original data is located.

[0123] Step 3: The parameters a and b of the envelope curve function described in Step 1 are determined based on the distribution of historical data to ensure that normal data points are within the envelope region.

[0124] Step 4: According to formulas (2) and (3), in the coordinate system UOV, the data center point of the servo amplifier input voltage and the servo valve main valve core position data can be obtained as the actual zero point of the servo valve.

[0125] u o =ave(u1, u2, ..., u n ), n∈N + (2)

[0126] v o =ave(v1, v2, ..., v n ), n∈N + (3)

[0127] In the formula, u o v represents the actual zero point x-coordinate of the servo valve. o This represents the actual zero point ordinate of the servo valve, while ave represents the average of the x or y coordinates of all data points.

[0128] Step 5: The coordinate transformation relationship between coordinate system X1O1Y1 and coordinate system UOV described in Step 2 is as follows: Figure 6 As shown, the change in the origin translation can be obtained through step 4, i.e. (u o v o The rotation angle θ of the coordinate system change is determined by the slope k of the straight line fitted based on historical data points using a linear regression method, as shown in formula (4).

[0129] θ = arctank (4)

[0130] Step 6: In the envelope curve function described in Steps 1 and 2, the determination of parameters a and b is crucial, such as... Figure 3 and 4 As shown, in the coordinate system UOV, the two points with the smallest and largest x-coordinates are first selected from the historical data point distribution, and their corresponding x-coordinates are u and u, respectively. l with u r The parameter a can be calculated using formula (5).

[0131]

[0132] The parameter b is determined according to the requirements of hysteresis calculation and satisfies constraint condition (6).

[0133]

[0134] τ represents the hysteresis loop, and empirically, τ ≤ 5%. U1 and U2 are the two voltage values ​​corresponding to the envelope function under the same main valve spool position (i.e., the same ordinate). Figure 5 As shown. U n It is the rated voltage, which is the difference between the voltage at the center zero point and the maximum range.

[0135] Step 7: Substitute the real-time data points of the servo amplifier input voltage and the servo valve main valve core position into formula (1) (Note: the real-time data points need to be converted to the monitoring coordinate system) to determine whether the data points are within the envelope region, i.e., whether they conform to formula (7). If they do, the data points are within the envelope region; if they do not, the data points are outside the envelope region.

[0136]

[0137] Step 8: If the judgment result described in Step 4 is that all data points are within the envelope area, then the working state of the servo valve can be determined to be normal. Conversely, if all data points are outside the envelope area, then the working state of the servo valve can be determined to be abnormal. However, in order to reduce the false alarm rate of the monitoring model, a fault tolerance mechanism is introduced. The number of data points that are continuously outside the envelope area is set to N. When the formula (8) is satisfied, then the servo valve can be determined to be in a normal working state. Conversely, the working state of the servo valve can be determined to be abnormal.

[0138] N≤5 (8)

[0139] Step 9: The process described in steps 6, 7, and 8 involves a coordinate transformation, requiring the conversion of the real-time data point coordinates of the current servo amplifier input voltage and the servo valve main valve spool position into coordinate values ​​in the Cartesian coordinate system of the envelope curve function. The specific process is as follows... Figure 6 As shown.

[0140] This invention can be used to achieve the following:

[0141] (1) Real-time fault monitoring of servo valves.

[0142] (2) Monitor the zero-point drift of the servo valve.

[0143] (3) Effectively avoid equipment failures caused by servo valve failures, thereby reducing economic losses.

[0144] Example 1:

[0145] Taking the hydraulic servo system of the leveling machine in a hot-dip galvanizing production line of a steel plant as an example, this invention illustrates an intelligent fault monitoring method for servo valves. The specific flowchart is as follows: Figure 4 As shown, the signal control principle is as follows: Figure 5 As shown, the programming software used is Visual Studio Code, and the programming language is Python. The specific steps include the following:

[0146] Simulation was performed on the servo valves of the hydraulic servo system of the leveling machine in a steel plant's hot-dip galvanizing production line. First, historical data on the input voltage and main valve core position of the servo valves on both the WS and DS sides of the leveling machine were acquired. Then, following steps 1 to 6, based on the distribution of their historical coordinate data, such as... Figure 7 and Figure 8 As shown, the envelope curve function is determined, as shown in formula (6).

[0147]

[0148] Then, according to formulas (2) and (3), the actual zero point (3747, 3718) of the servo valve of the hydraulic servo system of the leveling machine in the hot-dip galvanizing production line of the steel plant was calculated. Finally, based on the real-time data of the input voltage and main valve core position of the servo valves on both sides of the leveling machine WS and DS, real-time data was displayed according to steps 7 to 9, and the results are as follows. Figure 9 and Figure 10 As shown in the figure. The results show that the input voltage of the servo valves on the WS and DS sides of the leveling machine and the coordinates of the main valve core position data are within the safe range. At the same time, the zero point of the data center coincides with the actual zero point. Therefore, it can be determined that the servo valve is in normal working condition.

[0149] Example 2:

[0150] Taking the hydraulic servo system of the leveling machine in a hot-dip galvanizing production line of a steel plant as an example, this invention illustrates an intelligent monitoring method for servo valve faults. The specific flowchart is as follows: Figure 4 As shown, the signal control principle is as follows: Figure 5As shown, the programming software used is Visual Studio Code, and the programming language used is Python.

[0151] In this embodiment, the servo valves of the hydraulic servo system of the leveling machine in a steel plant's hot-dip galvanizing production line are monitored. First, historical data on the input voltage and main valve core position of the servo valves on both the WS and DS sides of the leveling machine are acquired. Then, following steps 1 to 6, based on the distribution of their historical coordinate data, such as... Figure 7 and Figure 8 As shown, the envelope curve function is determined, as shown in formula (9).

[0152] Then, according to formulas (2) and (3), the actual zero point (3736, 3705) of the servo valve of the hydraulic servo system of the leveling machine in the hot-dip galvanizing production line of the steel plant was calculated. Finally, based on the real-time data of the input voltage and main valve core position of the servo valves on both sides of the leveling machine WS and DS, the operation was carried out according to steps 7 to 9, and the results are as follows. Figure 11 and Figure 12 As shown in the figure. The results show that the input voltage of the servo valves on the WS and DS sides of the leveling machine and the main valve core position data are outside the safe range for 6 consecutive coordinate points. At the same time, the zero point of the data center is offset from the actual zero point. Therefore, it can be determined that the servo valve is malfunctioning.

[0153] This invention discloses an intelligent fault monitoring method for servo valves. By analyzing historical data of the servo amplifier input voltage and the servo valve main valve spool position, an envelope curve function is designed for real-time fault monitoring of the servo valve. Simultaneously, the actual zero-point coordinates of the servo valve can be obtained. On one hand, if a data point repeatedly falls outside the envelope curve region within a continuous time period, a fault is considered to have occurred, and an alarm is triggered. Simultaneously, a fault-tolerance mechanism is introduced into the monitoring model; if only a single data point exceeds the envelope range, it is considered normal and no alarm is triggered. On the other hand, by using a time-shifting sliding window, the data center over a period of time is calculated and compared with the center zero-point coordinates obtained from historical data to obtain the current zero-point offset of the servo valve. Using this invention, it is possible to monitor in real-time whether the servo valve's data points are within the envelope curve range and to detect zero-point drift, thereby achieving real-time fault monitoring of the servo valve. This invention is a prerequisite for fault monitoring of hydraulic servo systems and also forms the foundation for intelligent operation and maintenance of mechanical equipment containing hydraulic servo systems.

Claims

1. A method for intelligent monitoring of servo valve faults, characterized in that: Establish real-time monitoring of the safe operation of the servo valve by following these steps: S1: Based on the historical data distribution characteristics of the servo amplifier input voltage and the servo valve main valve spool position, set the envelope curve function; S2: Establish a judgment model based on the set envelope curve function; S3: The real-time collected data is sent to the judgment model according to the set collection cycle, and the judgment result of the judgment model is used to determine whether the servo valve is in a safe working state. The envelope curve function is: ， in, x Input voltage; y : Main valve spool position; a: The semi-major axis of the elliptic function; b: The minor semi-axis of the elliptic function.

2. The intelligent monitoring method for servo valve faults according to claim 1, characterized in that: The specific judgment model in step S2 is as follows: 。 3. The intelligent monitoring method for servo valve faults according to claim 1, characterized in that: The aforementioned intelligent monitoring of servo valve faults also includes real-time monitoring of the servo valve's zero-point drift status. The real-time monitoring of the servo valve's zero-point drift status specifically includes: SS1: Determine the theoretical zero-point position of the servo valve based on historical data of the servo amplifier input voltage and the position of the servo valve main valve spool; SS2: Real-time acquisition of the servo amplifier input voltage and the servo valve main valve core position according to the set acquisition cycle, and determination of the real-time zero point position of the servo valve based on the acquisition data within the cycle; SS3: Compare the real-time zero position with the theoretical zero position. If there is no deviation between the two, it is determined that no zero drift has occurred; otherwise, it is determined that zero drift has occurred.

4. The intelligent fault monitoring method for a servo valve according to claim 1, characterized in that: In the envelope curve function, a and b are determined through the following steps: S21: Determine the value of a based on the historical maximum and minimum values ​​of the servo amplifier input voltage and the slope of the historical data distribution fitting; S22: Determine the value of b based on three factors: the determined value of a, the envelope curve function, and the requirements for hysteresis calculation.

5. The intelligent monitoring method for servo valve faults according to claim 2, characterized in that: If the real-time data calculation results within 5 consecutive acquisition cycles meet the judgment model, the servo valve is determined to be in a safe working state; otherwise, the servo valve is determined to be faulty.

6. The intelligent fault monitoring method for a servo valve according to claim 4, characterized in that: The slope based on historical data distribution fitting is specifically as follows: A least-squares-based linear fit is performed on the historical data distribution of the servo amplifier input voltage and the servo valve main valve spool position, and the slope is determined based on the fit.

7. The intelligent monitoring method for servo valve faults according to claim 3, characterized in that: The established real-time monitoring of whether the servo valve is operating safely and the established real-time monitoring of the servo valve's zero-point drift status are operated in parallel.

8. The intelligent monitoring method for servo valve faults according to claim 3, characterized in that: The theoretical zero point position and the real-time zero point position are both determined based on the weighted average of the data values ​​in their respective data spaces.

9. A servo valve fault intelligent monitoring method according to claim 1 or 3, characterized in that: The real-time collected data is first converted into a numerical expression in the rectangular coordinate system of the envelope curve function through linear rotation, and then the corresponding judgment or comparison is performed. The rotation angle of the linear rotation is determined based on the slope of the historical data distribution fit.

10. The intelligent monitoring method for servo valve faults according to claim 4, characterized in that: The value of 'a' is determined as follows: ， in, u l The minimum input voltage value in historical data; u r The maximum input voltage value in historical data; : The angle between the fitted slopes.

11. The intelligent monitoring method for servo valve faults according to claim 9, characterized in that: The slope based on historical data distribution fitting is specifically as follows: A least-squares-based linear fit is performed on the historical data distribution of the servo amplifier input voltage and the servo valve main valve spool position, and the slope is determined based on the fit.

Images