A hydraulic system oil path fault early warning method and system

Abstract

The present application relates to the technical field of data processing, more particularly, the present application relates to a kind of hydraulic system oil circuit fault early warning method and system, comprising: collecting parameter data in hydraulic system oil circuit, the parameter data includes pressure in oil circuit;Pressure prediction value at any time is predicted using improved exponential smoothing method, and according to the size of pressure prediction value at current time, fault early warning is carried out, wherein the initial smoothing coefficient of the pressure actual value collected at each time is modified in the improved exponential smoothing method.The present application not only considers the inherent characteristics (such as mutation degree and fluctuation degree) of pressure data when calculating reliability, but also introduces the environmental purity index of hydraulic oil, more comprehensively evaluates the reliability of data, can more accurately reflect the change trend of pressure data by dynamically adjusting the smoothing coefficient of pressure actual value at each time, reduces the influence of prediction deviation and data noise, so as to improve the accuracy of fault early warning.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing. More particularly, the present application relates to a hydraulic system oil circuit fault early warning method and system. BACKGROUND

[0002] The hydraulic system oil circuit in the hydraulic device is the core part of realizing hydraulic energy transmission and control, mainly composed of a hydraulic pump, a hydraulic valve, a hydraulic cylinder (or a hydraulic motor), a pipeline and an oil tank, etc. When working, the hydraulic pump converts the mechanical energy output by the motor into hydraulic energy, and delivers the pressure oil to each execution element through the pipeline. Under the adjustment of the control element (such as the reversing valve, the overflow valve, the throttle valve, etc.), the pressure oil flows according to the predetermined path, realizes the movement control of the execution element, and thus completes various actions such as pushing, pulling and rotating. The hydraulic oil circuit can be composed of an open or closed system according to the use requirements, has the characteristics of large output force, fast response and smooth operation, and is widely used in engineering machinery, industrial automation equipment, aerospace and shipbuilding fields, etc.

[0003] In the hydraulic system oil circuit fault early warning, it is usually judged whether the pressure prediction value at any time is in the normal pressure interval. When the pressure prediction value is outside the normal pressure interval, the fault early warning is triggered. The pressure prediction value needs to be predicted by some algorithm, for example: exponential smoothing method, which depends on the fixed smoothing coefficient to weight the historical actual value. However, in actual application, the parameter data collected at different time points may have noise interference or abnormal fluctuations (such as sensor error, oil pollution interference, etc.). If the same smoothing coefficient is used to weight all historical actual values, it will inevitably lead to giving too high weight to low credibility data, and thus affecting the accuracy of the prediction model. SUMMARY

[0004] The present application provides a hydraulic system oil circuit fault early warning method and system, which aims to solve the problem that in the related art, the same smoothing coefficient is used to weight all historical actual values, which will inevitably lead to giving too high weight to low credibility data, and will affect the accuracy of the prediction model.

[0005] In a first aspect, the present application provides a hydraulic system oil circuit fault early warning method, comprising: collecting parameter data in the hydraulic system oil circuit, the parameter data including pressure in the oil circuit; predicting a pressure prediction value at any time using an improved exponential smoothing method, and performing fault early warning according to the size of the pressure prediction value at the current time, wherein the improved exponential smoothing method is to correct an initial smoothing coefficient of the collected pressure actual value at each time to obtain a final smoothing coefficient of the collected pressure actual value at each time, and the final smoothing coefficient is positively correlated with the reliability of the pressure actual value at the time and the reliability of the pressure actual value in the target period corresponding to the time, wherein the target period corresponding to any time is composed of the time and a preset number of previous times; calculating the reliability of the collected pressure actual value at any time, which is negatively correlated with the fluctuation degree of the pressure actual value in the target period corresponding to the time, and positively correlated with the environmental purity index of the hydraulic oil, which reflects the cleanliness of the hydraulic oil. Through the improved exponential smoothing method, dynamic smoothing coefficient correction, and comprehensive consideration of multiple factors such as pressure mutation degree, fluctuation degree, and hydraulic oil environmental purity index, the accuracy of pressure prediction is significantly improved.

[0006] Further, the parameter data further includes the number and size of particles in the hydraulic oil, and the kinematic viscosity of the hydraulic oil; and the environmental purity index of the hydraulic oil is calculated based on the number and size of particles in the hydraulic oil, and the kinematic viscosity of the hydraulic oil.

[0007] Further, the environmental purity index of the hydraulic oil is calculated, comprising: the environmental purity index of the hydraulic oil is positively correlated with the number and size of particles in the hydraulic oil, and positively correlated with the difference between the kinematic viscosity of the hydraulic oil and the standard operating viscosity, wherein the number and size of particles in the hydraulic oil, and the kinematic viscosity of the hydraulic oil all represent the cleanliness of the hydraulic oil. By integrating particle characteristics and viscosity deviation, the environmental state of the hydraulic oil is quantified, the judgment accuracy of pollution risk is improved, and key variables are provided for subsequent data reliability evaluation and prediction modeling.

[0008] Further, the reliability of the collected pressure actual value at any time is also positively correlated with the mutation degree of the collected pressure actual value at the time, and the mutation degree of the pressure actual value is represented by the absolute value of the difference between the pressure actual value and the mean value of all pressure actual values in the target period corresponding to the time.

[0009] Further, the initial smoothing coefficient of the pressure actual value collected at each time is corrected, including: if the difference between the credibility of the pressure actual value collected at the time and the credibility of the pressure actual value in the target period corresponding to the time is greater than a preset threshold, the initial smoothing coefficient of the pressure actual value at the time is increased to obtain a final smoothing coefficient; if the difference between the credibility of the pressure actual value collected at the time and the credibility of the pressure actual value in the target period corresponding to the time is less than a preset threshold, the initial smoothing coefficient of the pressure actual value at the time is reduced to obtain a final smoothing coefficient. The smoothing coefficient at each time is adaptively adjusted, the response capability of the model to high-quality data is enhanced, the accuracy and stability of the prediction result are improved, and the misjudgment or warning error caused by data noise is reduced.

[0010] Further, obtaining the final smoothing coefficient further includes: assigning an adjustment factor to the difference for controlling the influence degree of the credibility on the final smoothing coefficient, wherein the adjustment factor is 0.3.

[0011] Further, the fault warning is performed according to the size of the pressure prediction value, including: if the pressure prediction value at any time is outside the normal pressure interval, a warning is performed. The warning is automatically triggered based on the deviation of the prediction value from the normal pressure interval, without manual intervention.

[0012] Further, the fluctuation degree of the pressure actual value in the target period corresponding to the time includes: the fluctuation degree is characterized by the variance of all pressure actual values in the target period corresponding to the time. The variance as a statistical quantity can reflect the overall dispersion and fluctuation range of the pressure values in the period, and is more stable than the mutation determination of a single value.

[0013] Further, the method for obtaining the number and size of particles in the hydraulic oil includes: the size and number of particles in the hydraulic oil in the oil circuit are monitored in real time by using a contamination degree sensor.

[0014] The second aspect of the application also provides a fault warning system for a hydraulic system oil circuit, including a processor and a memory, the memory stores a computer program, and the processor executes the computer program to realize the fault warning method for the hydraulic system oil circuit as described in any one of the above.

[0015] Beneficial effects: the method predicts the pressure data in the hydraulic system oil circuit by improved exponential smoothing method, and combines the pressure prediction value at the current time for fault warning. The improved exponential smoothing method obtains the final smoothing coefficient which is positively related to the pressure actual value reliability and the pressure reliability difference in the target period by correcting the initial smoothing coefficient, and when calculating the reliability, not only considers the internal characteristics of the pressure data (such as mutation degree and fluctuation degree), but also introduces the environmental purity index of hydraulic oil, more comprehensively evaluates the reliability of data, and through dynamically adjusting the smoothing coefficient of pressure actual value at each time, can more accurately reflect the change trend of pressure data, reduce the influence of prediction deviation and data noise, so as to improve the accuracy of fault warning. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 is a flow chart illustrating the acquisition of pressure prediction value at any time according to an embodiment of the present application. DETAILED DESCRIPTION

[0017] The specific embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0018] Step S101: collecting parameter data in the hydraulic system oil circuit.

[0019] In one embodiment, the parameter data in the hydraulic system oil circuit is collected by using a sensor, wherein the parameter data includes the pressure in the oil circuit, the impurities (particles) in the hydraulic oil and the kinematic viscosity of the hydraulic oil. Specifically, a high-precision pressure sensor is used to monitor the pressure value at a specified position in the hydraulic oil circuit in real time, a contamination sensor is used to monitor the particles in the hydraulic oil in the detected area in the oil circuit in real time, and the size and number of particles can be obtained by using the contamination sensor; then a vibration type viscosity sensor is used to monitor the viscosity of the hydraulic oil in the oil circuit in real time. And the collected parameter data is normalized, the dimension is unified, and the interference of dimension difference on subsequent data analysis is eliminated. It should be noted that the number and size of particles in the hydraulic oil, as well as the kinematic viscosity of the hydraulic oil, are used to represent the cleanliness of the hydraulic oil.

[0020] S102: calculating the environmental purity index of the hydraulic oil in the oil circuit.

[0021] In one embodiment, when calculating the predicted value at any given time using the exponential smoothing method, the environmental purity index of the hydraulic oil in the hydraulic system circuit (which reflects the cleanliness of the hydraulic oil) is not taken into account. This leads to errors in the predicted value calculated at that time. The reason is that the environmental purity index of the hydraulic oil in the hydraulic system circuit at any given time will affect the actual pressure value at that time. The smaller the environmental purity index of the hydraulic oil, the higher the purity of the hydraulic oil and the less affected it is by external factors. Therefore, the actual pressure value at that time is closer to the true value, and the higher the accuracy of the obtained actual pressure value. The greater the confidence in the actual pressure value at that time, the larger the smoothing coefficient should be for the actual value at that time. Therefore, when calculating the predicted value, the smoothing coefficient of the actual pressure value should be larger, thereby improving the accuracy of the calculated predicted value.

[0022] In one embodiment, the environmental purity index is related to the size and number of particles detected in the hydraulic oil at the current moment. The larger the size and the greater the number of particles, the lower the environmental purity index of the hydraulic oil at the current moment. It is also related to the viscosity of the hydraulic oil at the current moment. The higher the viscosity, the greater the resistance to the movement of particles in the oil, and the easier it is for the particles to settle or remain at the bottom of the detection chamber, which makes it impossible for the laser beam to effectively irradiate the particles, resulting in missed detection. Therefore, when the viscosity is higher and the number of particles detected is greater, the environmental purity index at the current moment is lower.

[0023] In one embodiment, the environmental purity index of the hydraulic oil in the oil circuit at any given time is calculated using the following formula: In the formula, Indicates the first The environmental purity index of the hydraulic oil in the oil circuit at all times. Indicates the first The total size of all particles is constantly monitored. Indicates the first The total number of all particles is constantly detected. Indicates the first The kinematic viscosity of the hydraulic oil is constantly monitored. This indicates the standard kinematic viscosity of the hydraulic oil used in the oil circuit.

[0024] in, The value of the product reflects the severity of pollution; a larger product indicates larger particle size and greater quantity, resulting in a higher pollution risk and a lower environmental purity index. The value indirectly reflects the risk of missing particles in the hydraulic oil: the greater the viscosity deviates from the standard value, the higher the risk of missed detection even if particles are detected. Although the viscosity is relatively small, the actual risk of contamination is still considered higher (because large particles that are not detected may exist if the viscosity is too high or too low).

[0025] Step S103: Calculate the reliability of the actual pressure value collected at any given time.

[0026] In one embodiment, a lower environmental purity index for the hydraulic oil indicates more severe particulate contamination and a potential deviation from the standard value in kinematic viscosity. This directly or indirectly affects the actual pressure measured in the hydraulic system. For pressure sensors, ambient interference can cause errors in the measured pressure values. For example, a sudden change in pressure can lead to inaccurate readings. Using inaccurate pressure values ​​for prediction will affect the accuracy of the prediction results. Therefore, the degree of change in the actual pressure value within a given time period can be calculated. A greater degree of change indicates a higher likelihood that the actual pressure value is noise, resulting in lower reliability. A higher degree of change in the actual pressure value at any given moment necessitates a higher reliability of the actual pressure value when using exponential smoothing to calculate the predicted value. This requires a higher smoothing coefficient to reduce the impact of noise and improve prediction accuracy. It should be noted that determining the target time period corresponding to any given moment includes using the time period consisting of that moment and a predetermined number of preceding moments as the target time period. In this embodiment, the preset quantity is 10. In other embodiments, the preset quantity can be 12 or 18, etc., and can be adjusted according to the specific implementation scenario.

[0027] In one embodiment, the reliability of the actual pressure value collected at any given time is calculated. This reliability is negatively correlated with the fluctuation of the actual pressure value within the corresponding target time period and positively correlated with the environmental purity index of the hydraulic oil at that time. The fluctuation is the variance of all actual pressure values ​​within the corresponding target time period. Specifically, the product of the reciprocal of the variance of all actual pressure values ​​within the corresponding target time period and the environmental purity index of the hydraulic oil at that time is calculated, and the normalized product is used as the reliability of the actual pressure value collected at that time.

[0028] In another embodiment, the reliability of the actual pressure value collected at any given time is not only negatively correlated with the variance of all actual pressure values ​​within the target time period corresponding to that time, and positively correlated with the environmental purity index of the hydraulic oil at that time, but also positively correlated with the degree of abrupt change in the actual pressure value collected at that time. The degree of abrupt change in the actual pressure value is characterized as the absolute value of the difference between the actual pressure value and the mean of all actual pressure values ​​within the target time period corresponding to that time. The calculation formula is as follows: In the formula, Indicates the first [number]th [time period] within the target time period The reliability of constantly collecting actual pressure values. Indicates the first [number]th [time period] within the target time period The actual pressure value collected at all times. Indicates the first The time corresponds to the average of the actual pressure values ​​collected at all times within the target time period. Indicates the first The variance of all actual pressure values ​​collected within the target time period corresponds to each moment. Indicates the first [number]th [time period] within the target time period The environmental purity index of the hydraulic oil at all times.

[0029] in, Reflects the first The degree of abrupt change in the actual pressure value collected at any given time; the larger this value, the more significant the change. The greater the deviation of the actual pressure value collected at any given time from the overall distance, the more... The lower the reliability of the actual pressure values ​​collected at all times.

[0030] S104: Predict the pressure value at any given time using the improved exponential smoothing method.

[0031] In one embodiment, the improved exponential smoothing method modifies the initial smoothing coefficient of the actual pressure values ​​collected at each time step in the algorithm (the exponential smoothing method is a single-step exponential smoothing method). Specifically, it modifies the coefficient using the reliability of the actual pressure values ​​collected at each time step to obtain the final smoothing coefficient of the pressure values ​​collected at each time step. Then, the predicted pressure values ​​at each time step are predicted using the final smoothing coefficient. The empirical value of the initial smoothing coefficient is 0.3. The exponential smoothing method is calculated based on a weighted average of the actual values ​​and historical smoothed values. The exponential smoothing method is prior art and will not be elaborated upon here.

[0032] In one embodiment, the final smoothing coefficient of the actual pressure values ​​collected at each time point is calculated using the following formula: In the formula, Indicates the first The final smoothing coefficient at time step, Indicates the first The initial smoothing coefficient at time 1. Indicates the first The reliability of constantly collected pressure values. Indicates the first The mean of the reliability of all actual pressure values ​​within the target time period at any given time. This represents the adjustment factor, which controls the degree of influence of confidence on the final smoothing coefficient. In this embodiment, The value is 0.3. In other embodiments, The value can be 0.4 or 0.45, etc., and can be adjusted according to the specific implementation situation. When This indicates that the reliability of the actual pressure value collected at that moment is at an average level, and therefore no correction to the smoothing value is needed; when This indicates that the current collected actual pressure value is reliable, thus increasing the dependence on the actual pressure value and consequently increasing the final smoothing coefficient of the actual pressure value; when This indicates that the currently collected actual pressure value is unreliable. Therefore, we reduce the reliance on the actual value, thereby reducing the final smoothing coefficient of the actual pressure value. This completes the initial smoothing coefficient correction for the actual pressure values ​​at each time point.

[0033] S105: Provide fault warning based on the magnitude of the predicted pressure value.

[0034] In one embodiment, the pressure prediction value at any given time can be obtained using the improved exponential smoothing method, and a fault warning is issued based on the magnitude of the pressure prediction value. Specifically, if the pressure prediction value at any given time is outside the normal pressure range, a warning is issued; if the pressure prediction value at any given time is within the normal pressure range, no warning is issued, and monitoring continues. The normal pressure range can be determined based on... Criteria are used to determine this, for example: the historical average stress level is... The standard deviation is The normal pressure range can then be set as follows: .

[0035] The present invention also provides a fault early warning system for hydraulic system oil circuits. The system includes a processor and a memory, the memory storing computer program instructions. When the processor executes the computer program instructions, it implements a fault early warning method for hydraulic system oil circuits according to the first aspect of the present invention.

[0036] The system also includes other components well known to those skilled in the art, such as communication buses and communication interfaces, the settings and functions of which are known in the art and therefore will not be described in detail here.

[0037] In this invention, the aforementioned memory can be any tangible medium containing or storing a program that can be used or combined with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium can be any suitable magnetic or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc., or any other medium that can be used to store desired information and can be accessed by an application, module, or both. Any such computer storage medium can be part of a device or accessible to or connected to a device. Any application or module described in this invention can be implemented using computer-readable / executable instructions stored or otherwise maintained on such a computer-readable medium.

[0038] The embodiments described above are merely examples of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention.

Claims

1. A hydraulic system oil passage failure early warning method characterized by, The method comprises the following steps: Collecting parameter data in the hydraulic system oil circuit, the parameter data including pressure in the oil circuit; Using the improved exponential smoothing method to predict the pressure prediction value at any time, and performing fault early warning according to the size of the pressure prediction value at the current time, wherein the improved exponential smoothing method is to correct the initial smoothing coefficient of the collected pressure actual value at each time to obtain the final smoothing coefficient of the collected pressure actual value at each time, and the final smoothing coefficient is positively correlated with the difference between the reliability of the collected pressure actual value at the current time and the reliability of the pressure actual value in the target period corresponding to the current time, and the target period corresponding to any time is composed of the current time and a preset number of previous times; Calculating the reliability of the collected pressure actual value at any time, which is negatively correlated with the fluctuation degree of the pressure actual value in the target period corresponding to the current time, and is positively correlated with the environmental purity index of the hydraulic oil at the current time, the environmental purity index of the hydraulic oil reflecting the cleanliness of the hydraulic oil; The parameter data further includes the number and size of particles in the hydraulic oil, and the kinematic viscosity of the hydraulic oil; the environmental purity index of the hydraulic oil is calculated based on the number and size of particles in the hydraulic oil and the kinematic viscosity of the hydraulic oil; The environmental purity index of the hydraulic oil is calculated, including that the environmental purity index of the hydraulic oil is positively correlated with the number and size of particles in the hydraulic oil, and the difference between the kinematic viscosity of the hydraulic oil and the standard operating viscosity is positively correlated, wherein the number and size of particles in the hydraulic oil and the kinematic viscosity of the hydraulic oil all represent the cleanliness of the hydraulic oil; The product of the inverse of the variance of all pressure actual values in the target period corresponding to the current time and the environmental purity index of the hydraulic oil at the current time is calculated, and the normalized product is taken as the reliability of the collected pressure actual value at the current time.

2. The hydraulic system oil passage failure early warning method according to claim 1, characterized by, The reliability of the collected pressure actual value at any time is also positively correlated with the mutation degree of the collected pressure actual value at the current time, and the mutation degree of the pressure actual value is represented by the absolute value of the difference between the pressure actual value and the mean value of all pressure actual values in the target period corresponding to the current time.

3. The hydraulic system oil passage failure early warning method according to claim 2, characterized by, The initial smoothing coefficient of the collected pressure actual value at each time is corrected, including: If the difference between the reliability of the collected pressure actual value at the current time and the reliability of the pressure actual value in the target period corresponding to the current time is greater than a preset threshold, the initial smoothing coefficient of the pressure actual value at the current time is increased to obtain the final smoothing coefficient; If the difference between the reliability of the collected pressure actual value at the current time and the reliability of the pressure actual value in the target period corresponding to the current time is less than a preset threshold, the initial smoothing coefficient of the pressure actual value at the current time is decreased to obtain the final smoothing coefficient.

4. The hydraulic system oil passage failure early warning method according to claim 3, characterized by, The final smoothing coefficient is also obtained, including: An adjustment factor is given to the difference between the reliability of the collected pressure actual value at the current time and the reliability of the pressure actual value in the target period corresponding to the current time, which is used to control the influence degree of the reliability on the final smoothing coefficient, wherein the adjustment factor is 0.

3.

5. The hydraulic system oil passage failure early warning method according to claim 1, characterized by, The fault early warning according to the size of the pressure prediction value, including: If the pressure prediction value at any time is outside the normal pressure interval, early warning is performed.

6. The hydraulic system oil passage failure early warning method according to claim 1, characterized by, The fluctuation degree of the pressure actual value in the target period corresponding to the current time, including: The fluctuation degree is characterized by the variance of all actual pressure values in the target period corresponding to the moment.

7. The hydraulic system oil passage failure early warning method according to claim 1, characterized by, Method for obtaining the number and size of particles in hydraulic oil, comprising: The size and number of particles in the hydraulic oil in the oil circuit are monitored in real time by using a contamination degree sensor.

8. A hydraulic system oil path failure early warning system comprising a processor and a memory, wherein, The memory stores a computer program, and the processor executes the computer program to realize the fault early warning method for the hydraulic system oil circuit according to any one of claims 1-7.

Images