A rapid reliability assessment method for CNC machine tools based on inverse power law model
By dividing CNC machine tools by motion axis and setting acceleration conditions, combining hierarchical analysis method and inverse power law model, the problems of long time period and insufficient accuracy of traditional CNC machine tools are solved, and faster, more accurate and reliable The evaluation of the overall reliability of CNC machine tools provides a scientific and reasonable evaluation.
Patent Information
- Application Number
- CN202210253280.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-03-15
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2042-03-15
AI Technical Summary
The reliability evaluation method of traditional CNC machine tools has a long time period and insufficient rapid evaluation accuracy, making it difficult to adapt to the iteration cycle of new products. There will be large deviations when using the entire electromechanical stress as the acceleration stress alone.
The CNC machine tool is divided into subsystems according to the motion axis, set up two accelerated operating conditions, collect the power of each motion axis motor, use the hierarchical analysis method to determine the weight of each subsystem, and calculate the corrected total of the entire machine based on the inverse power law model. Power, perform Weibull distribution model fitting and K-S test to solve the inverse power law model parameters, and realize the rapid reliability evaluation of CNC machine tools.
It improves the rapid accuracy of reliability evaluation, provides a more scientific and reasonable reliability evaluation of the whole machine, reduces the evaluation cycle, enhances the accuracy of the evaluation results, and conforms to the actual situation.
Smart Images

Figure CN114647906B_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of reliability engineering, in particular to the field of machine tool reliability assessment, and more particularly relates to a method for rapid reliability assessment of numerically controlled machine tools based on an inverse power law model. Background Art
[0002] As an advanced mechanical processing equipment, CNC machine tools are widely used in industrial production due to their high precision, high efficiency and high automation. They have become the mainstay of the manufacturing industry, and the technical level of their equipment is an important reflection of a country's comprehensive national strength. In recent years, with the continuous development of CNC technology, the reliability level of machine tools has become a more concerned issue in the industry, so reliability assessment technology is particularly important.
[0003] Traditional reliability assessment is based on on-site tracking and statistics to collect machine tool fault information, but the average failure-free time of various components of contemporary machine tools is often thousands or even tens of thousands of hours, and the machine tool product design process is relatively mature. This results in many faults being unable to be exposed in a short period of time, and long-term tracking and statistics cannot adapt to the iteration cycle of new products, which seriously restricts the improvement of the service performance of CNC equipment. Therefore, seeking a scientific and fast reliability assessment method has become an urgent need for the CNC machine tool industry.
[0004] The inverse power law model is one of the most commonly used accelerated life models and is often used for mechanical products with fatigue and wear as failure mechanisms. This model uses electrical stress as the accelerating stress and achieves the purpose of accelerating product failure by changing performance parameters such as voltage, current or power. According to existing literature research, the application of the inverse power law model to the key functional components and subsystems of CNC machine tools has achieved good results. However, since CNC machine tools are conventional mechatronic and hydraulic integrated products, the relationship between their subsystems at all levels is intricate. Using the electrical stress of the entire machine as the accelerating stress to solve the reliability index of the machine tool often results in large deviations.
[0005] The analytic hierarchy process (AHP) decomposes the relevant factors into several levels from top to bottom by establishing a hierarchical model, and constructs a comparison matrix using the paired comparison method to calculate the combined weight vector. The advantage of the AHP is that it makes decisions based on the thinking mode of decomposition, comparison, judgment and synthesis, and can quantify the impact of each factor on the result. Summary of the invention
[0006] The purpose of the present invention is to provide a CNC machine tool reliability rapid assessment method based on an inverse power law model to address the problems of long time period and insufficient rapid assessment accuracy of existing CNC machine tool reliability assessment technologies. This method is used to perform reliability assessment on CNC machine tools more quickly, accurately and in line with reality.
[0007] The technical solution to achieve the purpose of the present invention is: a method for rapid reliability assessment of CNC machine tools based on an inverse power law model, the method comprising the following steps:
[0008] Step 1, the entire CNC machine tool is divided into n subsystems according to the motion axis, and two acceleration operating conditions s2 and s3 are set according to the normal operating condition s1 of the equipment;
[0009] Step 2, respectively collect the power of the motor of each moving axis of the machine tool in the next working cycle of working conditions s1, s2, and s3, and use the second-order average power of the single axis to represent the single axis working condition load;
[0010] Step 3: Use the analytic hierarchy process (AHP) to determine the weight vector of the influence of the power of each motion axis of the CNC machine tool on the total power.
[0011] Step 4: Calculate the corrected total power v of the whole machine under each working condition based on the single-axis working condition load and the influence weight vector P ;
[0012] Step 5, carry out reliability field tracking statistical tests under working conditions s2 and s3 respectively, and record fault data by regular truncation;
[0013] Step 6, respectively calculating the fault data under working condition s2 and working condition s3 to obtain the corresponding empirical distribution function, and using the least squares method to preliminarily obtain the Weibull distribution model parameters;
[0014] Step 7, perform KS test on the Weibull distribution model obtained in step 6. If the test passes, it indicates that the fault distribution under the acceleration condition obeys the Weibull distribution. Otherwise, return to step 5 and execute again until the Weibull distribution model passes the test.
[0015] Step 8, calculating the characteristic life of the Weibull distribution model;
[0016] Step 9, solving the estimated values of the inverse power law model parameters according to the characteristic life and working load under different acceleration conditions;
[0017] Step 10, using the inverse power law model to calculate the reliability index of the CNC machine tool under the normal operating condition s1.
[0018] Furthermore, the calculation formula for the single-axis load in step 2 is:
[0019]
[0020] Where, ν Pi is the power of the i-th motion axis subsystem, i = 1, 2, ..., n; N is the number of sampling points; P ij (t) is the power of the jth sampling point of the i-th motion axis subsystem.
[0021] Furthermore, the influence weight vector of the power of each motion axis of the CNC machine tool on the total power is determined by using the analytic hierarchy process AHP in step 3. The specific process includes:
[0022] Step 3-1, compare the importance of each motion axis of the CNC machine tool in pairs, and construct a quantitative judgment matrix A according to Table 1 below. The dimension of the matrix is equal to the number n of motion axes of the machine tool;
[0023] Table 1 Importance meaning
[0024]
[0025] Step 3-2: Normalize the elements of the judgment matrix A by column, then sum and normalize by row to obtain the eigenvector corresponding to the maximum eigenvalue of the judgment matrix A. Then calculate λ max :
[0026] A·W=λ max ·W (2)
[0027] Where W i is the influence weight of the power of the i-th motion axis subsystem on the total power, i = 1, 2, ..., n;
[0028] Step 3-3, calculate the consistency index CR by combining the following Table 2 and formulas (3) and (4). If the value of the consistency index CR is less than 0.1, the vector obtained in step 3-2 is the influence weight vector. Otherwise, return to step 3-1 to reconstruct the judgment matrix A.
[0029]
[0030]
[0031] Table 2. Comparison table of consistency indicators
[0032]
[0033] Furthermore, in step 4, the corrected total power v of the whole machine under each working condition is calculated based on the single-axis working condition load and the influence weight vector. P , the calculation formula is:
[0034]
[0035] Furthermore, in step 6, the corresponding empirical distribution function is calculated based on the fault data, and the least squares method is used to preliminarily obtain the parameters of the Weibull distribution model, which specifically includes:
[0036] Step 6-1, pre-process the failure time by sorting, remove the truncated time, and then use the approximate median rank estimate to calculate the empirical distribution function according to formula (6);
[0037]
[0038] Step 6-2, calculate x i ,y i :
[0039] x i = ln t i (7)
[0040]
[0041] Then, the least square method y=Ax+B is used for fitting to obtain A and B;
[0042] Step 6-3, calculate the scale parameter θ and shape parameter m of the Weibull distribution, the calculation formula is:
[0043] m=A (9)
[0044]
[0045] Furthermore, in step 7, a KS test is performed on the Weibull distribution model, specifically including:
[0046] Step 7-1, sort the fault data in ascending order, and then calculate the F0(t i ) value, and calculate the empirical distribution function F corresponding to each fault data n (t i )value;
[0047] Step 7-2, calculate the F0(t i ) and F n (t i ), and the largest absolute value is used as the test statistic D n Observed value of
[0048] Step 7-3, determine whether the following conditions are met:
[0049]
[0050] If satisfied, it means that the Weibull distribution model passes the test, otherwise it fails;
[0051] Where D n,α is the critical value of the test statistic, F n (t i ) is the empirical distribution function with a sample size of n.
[0052] Table 3D n,α Comparison Table
[0053]
[0054]
[0055] Furthermore, the calculation formula of the characteristic life of the Weibull distribution model in step 8 is:
[0056] ξ=t(e -1 ) (12)
[0057] Where ξ is the characteristic life, t(e -1 ) is the reliability e -1 Reliable life.
[0058] Furthermore, the solution formula for the estimated value of the inverse power law model parameters in step 9 is:
[0059] lnξ=a+blnν (13)
[0060] Where a and b are the parameters of the inverse power law model, are constants, and their estimated values are ν is the load, which can be power, current or vibration.
[0061] Furthermore, the step 10 uses the inverse power law model to calculate the reliability index of the CNC machine tool under the conventional operating condition s1, specifically including:
[0062] Step 10-1, calculating the scale parameter θ and shape parameter m of the Weibull model of the failure time distribution of the CNC machine tool under the normal operating condition s1;
[0063]
[0064]
[0065] Where m1 and m2 are the shape parameters of the Weibull model fitting under the two acceleration operating conditions s2 and s3 respectively;
[0066] Step 10-2, calculate the reliability R(t), failure rate function h(t) and mean time between failures MTBF:
[0067]
[0068]
[0069]
[0070] Where E[T] is the expected value of the failure time; Γ(x) is the gamma function,
[0071] Compared with the prior art, the present invention has the following significant advantages:
[0072] 1) The weight of each level of motion axis subsystem to the whole system is obtained by using the hierarchical analysis method. The stress of each level of subsystem is weighted to obtain the corrected stress. The corrected stress is used for model calculation, which improves the accuracy of the acceleration model.
[0073] 2) Reliability assessment provides an important basis for the comprehensive evaluation of CNC machine tools and the weak links in the traceability design. By combining the accelerated stresses of each motion axis subsystem through the hierarchical analysis method and using the inverse power law model to perform accelerated reliability assessment of the entire CNC machine tool, the assessment results can be scientific and reasonable and more in line with actual conditions.
[0074] 3) Compared with the existing CNC machine tool reliability evaluation method, the cycle is long; the rapid evaluation method in the industry is only aimed at functional components, and there is a lack of rapid evaluation methods for complete machine tools. The present invention sets two different accelerated operating conditions and uses the mechanical product acceleration model with fatigue and wear as the failure mechanism - the inverse power law model to solve a series of reliability indicators under conventional conditions, thereby achieving rapid evaluation of reliability.
[0075] 4) When solving the reliability index under normal working conditions based on the inverse power law model, the corrected whole machine power is used as the load of the inverse power law model, and the impact of the amplified working conditions on the life of each motion axis is considered, making the evaluation result more accurate.
[0076] The present invention is further described in detail below in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0077] Figure 1 The present invention is a flowchart of a method for rapid reliability assessment of CNC machine tools based on an inverse power law model. DETAILED DESCRIPTION
[0078] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0079] In one embodiment, in combination Figure 1 , a rapid reliability assessment method for CNC machine tools based on the inverse power law model is provided, and the data of a certain model of three-axis CNC milling machine is analyzed as an example.
[0080] Step 1, divide the entire CNC machine tool into n subsystems according to the motion axis; set two acceleration operating conditions s2 and s3 according to the normal operating condition s1 of the equipment;
[0081] In this embodiment, it is determined that a certain model of three-axis CNC milling machine includes a total of 4 motion axis subsystems, namely: spindle, feed axis X, feed axis Y, feed axis Z. The normal operation condition s1 of the equipment and the two acceleration operation conditions s2 and s3 correspond to the following Table 4 respectively;
[0082] Table 4 Cutting parameters of three different working conditions
[0083]
[0084]
[0085] Step 2, respectively collect the power of the motor of each moving axis of the machine tool in the next working cycle of working conditions s1, s2, and s3, and use the second-order average power of the single axis to represent the single axis working condition load;
[0086] In this embodiment, the motor power of each motion axis is sampled under three different working conditions, and is calculated using formula (1) to obtain Table 5;
[0087] Table 5 Second-order average power of each motion axis motor under three different working conditions
[0088]
[0089] Step 3: Determine the weight vector of the influence of the power of each motion axis of the CNC machine tool on the total power
[0090] In this embodiment, firstly, the importance of two-by-two motion axes is compared according to Table 1 to construct a quantized judgment matrix A;
[0091]
[0092] Normalize the elements of the judgment matrix A by column, then sum and normalize them by row to obtain the eigenvector corresponding to the maximum eigenvalue of the judgment matrix A. The maximum eigenvalue λ is obtained by formula (2): max =4.17.
[0093] According to formula (3) and (4), the consistency index comparison table 2 is used for inspection. If the test is passed, the weight vector of each motion axis is
[0094] Step 3: Use formula (5) to calculate the corrected total power ν of the whole machine under each working condition: P As shown in Table 6 below;
[0095] Table 6 Corrected total power of the whole machine under three different working conditions
[0096]
[0097] Step 5: Conduct reliability field tracking statistical tests under working conditions s2 and s3 respectively. The 4000h scheduled truncated record failure data is shown in Table VII.
[0098] Table 7 Timing truncation fault data table
[0099]
[0100] Step 6, respectively calculating the fault data under working condition s2 and working condition s3 to obtain the empirical distribution function, and using the least squares method to preliminarily obtain the Weibull distribution model parameters;
[0101] The fault time in Table 7 is sorted and the empirical distribution function is calculated using the approximate median rank estimate according to formula (6); then, x is calculated using formulas (7) and (8). i ,y i The values of are shown in Table 8. The data in Table 8 are then fitted by the least squares method y = Ax + B. Finally, the scale parameter θ and shape parameter m of the Weibull distribution are calculated using formulas (9) and (10). See Table 9.
[0102] Table 8x i ,y i Data Sheet
[0103]
[0104] Table 9 Scale parameter θ and shape parameter m
[0105]
[0106] Step 7, respectively use KS test on the above Weibull distribution model, if the conditions are met, the fitted model is the fault distribution model under the acceleration condition;
[0107] In this embodiment, the F0(t i ) and the empirical distribution function F n (t i ) to compare and calculate the test statistic D n The values are 0.0745 and 0.1284 respectively. n With critical value D 20,0.05 , D 30,0.05 After comparison, if formula (11) is satisfied, the null hypothesis, i.e. the Weibull distribution model obtained in step 6, is accepted.
[0108] Step 8, for the Weibull distribution model that has passed the test, calculate its characteristic life using formula (12);
[0109] In this embodiment, the characteristic life of the Weibull distribution model corresponding to the working condition s2 and the working condition s3 is calculated to be ξ2=1204.775 and ξ3=792.435 respectively.
[0110] Step 9: According to the characteristic life and working load under different acceleration conditions, solve the estimated value of the inverse power law model parameters according to formula (13):
[0111] In this embodiment, the estimated value of the inverse power law model parameter is calculated using formula (13):
[0112] Step 10, using the inverse power law model to calculate the reliability index of the CNC machine tool under the normal operating condition s1.
[0113] In this embodiment, the scale parameter θ=2156.9 and the shape parameter m=4.0865 of the Weibull model of the failure time distribution of the CNC machine tool under the normal operating condition s1 are calculated using formulas (14) and (15); the reliability R(t), the failure rate function h(t) and the mean time between failures MTBF are calculated respectively according to formulas (16), (17) and (18), as shown in Table 10.
[0114] Table 10 Equipment reliability indicators under normal operating conditions
[0115]
[0116] The rapid reliability evaluation method for CNC machine tools based on the inverse power law model proposed in the present invention can perform reliability evaluation on CNC machine tools, and is faster, more accurate and more practical than existing reliability evaluation methods for CNC machine tools.
[0117] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.
Claims
1. A method for rapid reliability assessment of CNC machine tools based on an inverse power law model, characterized in that: The method comprises the following steps: Step 1, the entire CNC machine tool is divided into n subsystems according to the motion axis, and two acceleration operating conditions s2 and s3 are set according to the normal operating condition s1 of the equipment; Step 2, respectively collect the power of the motor of each moving axis of the machine tool in the next working cycle of working conditions s1, s2, and s3, and use the second-order average power of the single axis to represent the single axis working load; the calculation formula of the single axis working load is: In the formula, v Pi is the power of the i-th motion axis subsystem, i = 1, 2, ..., n; N is the number of sampling points; P ij (t) is the power of the jth sampling point of the i-th motion axis subsystem; Step 3: Use the analytic hierarchy process (AHP) to determine the weight vector of the influence of the power of each motion axis of the CNC machine tool on the total power. Step 4: Calculate the corrected total power v of the whole machine under each working condition based on the single-axis working condition load and the influence weight vector P ; Step 5, carry out reliability field tracking statistical tests under working conditions s2 and s3 respectively, and record fault data by regular truncation; Step 6, respectively calculating the fault data under working condition s2 and working condition s3 to obtain the corresponding empirical distribution function, and using the least squares method to preliminarily obtain the Weibull distribution model parameters; Step 7, perform KS test on the Weibull distribution model obtained in step 6. If the test passes, it indicates that the fault distribution under the acceleration condition obeys the Weibull distribution. Otherwise, return to step 5 and execute again until the Weibull distribution model passes the test. Step 8, calculating the characteristic life of the Weibull distribution model; Step 9, solving the estimated values of the inverse power law model parameters according to the characteristic life and working load under different acceleration conditions; the solution formula for the estimated values of the inverse power law model parameters is: lnξ=a+blnv (2) Where a and b are the parameters of the inverse power law model, are constants, and their estimated values are ν is the load, take the total power v of the whole machine in step 4 P ; Step 10, using the inverse power law model to calculate the reliability index of the CNC machine tool under the normal operating condition s1.
2. The method for rapid reliability assessment of CNC machine tools based on the inverse power law model according to claim 1 is characterized in that: Step 3 uses the analytic hierarchy process (AHP) to determine the influence weight vector of the power of each motion axis of the CNC machine tool on the total power. The specific process includes: Step 3-1, compare the importance of each motion axis of the CNC machine tool in pairs, and construct a quantitative judgment matrix A according to Table 1 below. The dimension of the matrix is equal to the number n of motion axes of the machine tool; Table 1 Importance meaning Step 3-2: Normalize the elements of the judgment matrix A by column, then sum and normalize by row to obtain the eigenvector corresponding to the maximum eigenvalue of the judgment matrix A. Then calculate λ max : Where W i is the influence weight of the power of the i-th motion axis subsystem on the total power, i = 1, 2, ..., n; Step 3-3, calculate the consistency index CR by combining the following Table 2 and formulas (4) and (5). If the value of the consistency index CR is less than 0.1, the vector obtained in step 3-2 is the influence weight vector. Otherwise, return to step 3-1 to reconstruct the judgment matrix A. Table 2. Comparison table of consistency indicators 。 3. The method for rapid reliability assessment of CNC machine tools based on the inverse power law model according to claim 2 is characterized in that: Step 4: Based on the uniaxial load and the influence weight vector, the corrected total power v of the whole machine under each working condition is calculated. P , the calculation formula is:
4. The method for rapid reliability assessment of CNC machine tools based on the inverse power law model according to claim 3 is characterized in that: In step 6, the corresponding empirical distribution function is calculated based on the fault data, and the least squares method is used to preliminarily obtain the parameters of the Weibull distribution model, which specifically includes: Step 6-1, pre-process the failure time by sorting, remove the truncated time, and then use the approximate median rank estimate to calculate the empirical distribution function according to formula (7); Step 6-2, calculate x i ,y i : x i =lnt i (8) Then, the least square method y=Ax+B is used for fitting to obtain A and B; Step 6-3, calculate the scale parameter θ and shape parameter m of the Weibull distribution, the calculation formula is: m=A(10) 5. The method for rapid reliability assessment of CNC machine tools based on the inverse power law model according to claim 4 is characterized in that: In step 7, the KS test is performed on the Weibull distribution model, which specifically includes: Step 7-1, sort the fault data in ascending order, and then calculate the F0(t i ) value, and calculate the empirical distribution function F corresponding to each fault data n (t i )value; Step 7-2, calculate the F0(t i ) and F n (t i ), and the largest absolute value is used as the test statistic D n Observed value of Step 7-3, determine whether the following conditions are met: If satisfied, it means that the Weibull distribution model passes the test, otherwise it fails; Where D n,α is the critical value of the test statistic, F n (t i ) is the empirical distribution function with a sample size of n.
6. The method for rapid reliability assessment of CNC machine tools based on the inverse power law model according to claim 5 is characterized in that: The calculation formula for the characteristic life of the Weibull distribution model in step 8 is: ξ=t(e -1 ) (13) Where ξ is the characteristic life, t(e -1 ) is the reliability e -1 reliable lifespan.
7. The method for rapid reliability assessment of CNC machine tools based on the inverse power law model according to claim 6 is characterized in that: Step 10 uses the inverse power law model to calculate the reliability index of the CNC machine tool under the normal operating condition s1, specifically including: Step 10-1, calculating the scale parameter θ and shape parameter m of the Weibull model of the failure time distribution of the CNC machine tool under the normal operating condition s1; Where m1 and m2 are the shape parameters of the Weibull model fitting under the two acceleration operating conditions s2 and s3 respectively; Step 10-2, calculate the reliability R(t), failure rate function h(t) and mean time between failures MTBF: Where E[T] is the expected value of the failure time; Γ(x) is the gamma function,
Citation Information
Patent Citations
Precision evaluation method and service life prediction method for numerically controlled machine tool based on meta-action unit and integrated subjective-objective weight
CN110209119A
Numerical control machine tool linear feeding system reliability distribution method based on FAHP
CN112116213A
