A method for analyzing faults of a numerically controlled machine tool and a terminal

Abstract

A method and a terminal for analyzing faults of a numerical control machine tool, which obtain the fault data of the machine tool; based on the fault data of the machine tool, statistically obtain the number of fault occurrences and the fault occurrence duration within a preset time period, and calculate the mean time between failures; construct a multiple evaluation model according to the machine tool fault type, the machine tool fault level, and the mean time between failures; train the multiple evaluation model with the goal of maximizing the fault severity rate and the optimal weight coefficient; and use the trained multiple evaluation model to analyze the faults of the numerical control machine tool. The present invention calculates the mean time between failures based on the fault data of the machine tool, constructs a multiple evaluation model with the above data as input, trains the model with the goal of maximizing the fault severity rate and the optimal weight coefficient, and finally uses the trained model to analyze the faults of the numerical control machine tool, generate the safety factor of the numerical control machine tool, so as to discover potential problems and reduce the downtime and productivity losses caused by faults.

Description

Technical Field

[0001] The present invention relates to the field of fault analysis, and particularly to a method and a terminal for analyzing faults of a numerically controlled machine tool. Background Art

[0002] In modern manufacturing, numerically controlled machine tools (CNC machine tools), as one of the core production equipment, are widely used in fields such as precision machining and automated production. With the increasing complexity of production processes and the continuous improvement of production efficiency requirements, the stability and fault warning of numerically controlled machine tools become particularly important. During the operation of the machine tool, various faults may occur, including mechanical faults, electrical faults, control system faults, etc. These faults often have a direct impact on production progress, product quality, and equipment service life, and may even lead to the shutdown of the production line and huge economic losses. Therefore, timely and accurately diagnosing the fault types and severity of numerically controlled machine tools and taking appropriate maintenance or preventive measures are crucial for improving production efficiency, reducing maintenance costs, and extending equipment life.

[0003] However, the existing numerically controlled machine tool fault analysis technologies have the following problems: (1) Poor fault diagnosis accuracy: Traditional fault diagnosis methods mainly rely on manual experience and intuitive judgment, and it is often difficult to adapt to the situation of a large variety and complexity of machine tool faults. Experience-based diagnosis methods are not only easily affected by the personal abilities of operators, but also often difficult to find the root cause of problems in a timely manner when facing sudden faults, resulting in low maintenance efficiency. (2) Strong limitations of fault analysis methods: Many current fault analysis methods mainly focus on the processing of single fault types, ignoring the complexity of different fault types, levels, and their mutual influences. This makes it difficult for single fault analysis methods to meet the changing fault diagnosis requirements in a complex industrial environment and unable to provide comprehensive and accurate fault prediction. Summary of the Invention

[0004] The technical problem to be solved by the present invention is: to provide a method and a terminal for analyzing faults of a numerically controlled machine tool to solve the problem of difficult fault analysis of numerically controlled machine tools.

[0005] To solve the above technical problem, the technical solution adopted by the present invention is:

[0006] A method for analyzing faults of a numerically controlled machine tool includes the steps of:

[0007] S1. Obtain machine tool fault data;

[0008] S2. Based on the machine tool fault data, statistically obtain the number of fault occurrences and the fault occurrence duration within a preset time period, and calculate the mean time between failures.

[0009] S3. Construct a multiple evaluation model based on the machine tool failure type, machine tool failure level, and mean time between failures; train the multiple evaluation model with the goal of maximizing the failure severity rate and the optimal weight coefficient;

[0010] The multiple evaluation model is expressed as:

[0011] ;

[0012] In the formula, is the value of maximizing the failure severity rate; j is the index of the decision-making unit to be evaluated; is the weight coefficient; is the weight coefficient of the j-th decision-making unit to be evaluated; 、 are slack variables, where is the excessive input variable, is the insufficient output variable; represents the input vector, including at least the failure type ratio, failure level ratio, and mean time between failures; represents the output vector, including at least the total number of failures and the total cost of failure repair within the preset duration;

[0013] The constraint conditions of the multiple evaluation model are as follows:

[0014] ;

[0015] ;

[0016] ;

[0017] ;

[0018] In the formula, represents the first input vector, represents the first output vector;

[0019] S4. Use the trained multiple evaluation model to perform failure analysis on the numerically controlled machine tool.

[0020] To solve the above technical problems, another technical solution adopted by the present invention is:

[0021] A numerically controlled machine tool failure analysis terminal, characterized in that it includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the following steps are completed:

[0022] S1. Obtain machine tool failure data;

[0023] S2. Based on the machine tool fault data, count the number of fault occurrences and the fault occurrence duration within a preset time period, and calculate the mean time between failures (MTBF).

[0024] S3. Construct a multiple evaluation model according to the machine tool fault type, machine tool fault level, and mean time between failures (MTBF); train the multiple evaluation model with the goal of maximizing the fault severity rate and the optimal weight coefficient.

[0025] The multiple evaluation model is expressed as:

[0026] ;

[0027] In the formula, is the value of maximizing the fault severity rate; j is the index of the decision-making unit to be evaluated; is the weight coefficient; is the weight coefficient of the jth decision-making unit to be evaluated; , are slack variables, where is the excessive input variable, is the insufficient output variable; represents the input vector, including at least the fault type ratio, fault level ratio, and mean time between failures (MTBF); represents the output vector, including at least the total number of faults and the total cost of fault repair within a preset time period;

[0028] The constraint conditions of the multiple evaluation model are as follows:

[0029] ;

[0030] ;

[0031] ;

[0032] ;

[0033] In the formula, represents the first input vector, represents the first output vector;

[0034] S4. Use the trained multiple evaluation model to perform fault analysis on the CNC machine tool.

[0035] The beneficial effects of the present invention are as follows: It provides a method and a terminal for analyzing faults of a numerically controlled machine tool, calculates the mean time between failures (MTBF) based on the machine tool fault data, constructs a multiple evaluation model with the above data as input, trains the model with the goal of maximizing the fault severity rate and the optimal weight coefficient, and finally uses the trained model to analyze the faults of the numerically controlled machine tool, generates the safety factor of the numerically controlled machine tool, so as to discover potential problems and reduce the downtime and productivity losses caused by faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a flowchart of a method for analyzing faults of a numerically controlled machine tool in an embodiment of the present invention;

[0037] Figure 2 It is a schematic diagram of model construction in a method for analyzing faults of a numerically controlled machine tool in an embodiment of the present invention;

[0038] Figure 3 It is a schematic diagram of a terminal for analyzing faults of a numerically controlled machine tool in an embodiment of the present invention;

[0039] LABEL DESCRIPTION:

[0040] 1. A terminal for analyzing faults of a numerically controlled machine tool; 2. A memory; 3. A processor. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0041] To describe in detail the technical content, the achieved objectives and the effects of the present invention, the following is described in conjunction with the embodiments and with reference to the accompanying drawings.

[0042] Please refer to Figure 1 and Figure 2 , a method for analyzing faults of a numerically controlled machine tool, including the steps of:

[0043] S1. Obtain the machine tool fault data;

[0044] S2. Based on the machine tool fault data, count the number of fault occurrences and the fault occurrence duration within a preset time period, and calculate the mean time between failures;

[0045] S3. Construct a multiple evaluation model according to the machine tool fault type, the machine tool fault level and the mean time between failures; with the goal of maximizing the fault severity rate and the optimal weight coefficient, train the multiple evaluation model;

[0046] S4. Use the trained multiple evaluation model to analyze the faults of the numerically controlled machine tool.

[0047] It should be noted that the Mean Time Between Failures (MTBF) is a reliability indicator for measuring a product (especially electrical products). The unit is "hours". It reflects the time quality of the product and represents the ability of the product to maintain its functions within a specified time. Specifically, it refers to the average working time between two adjacent failures, also known as the mean time between failures.

[0048] As can be seen from the above description, a method for fault analysis based on numerically controlled machine tool fault data is provided, which can effectively improve the accuracy and efficiency of machine tool fault analysis. By obtaining fault data and counting the number of faults and the fault duration, combined with the calculated Mean Time Between Failures (MTBF), it can lay a foundation for further fault prediction and analysis. This method can accurately reflect the operating state of the machine tool within the preset duration, avoiding the problem of over-reliance on empirical judgment in traditional methods, thereby improving the scientificity and reliability of data-driven decision-making. Especially in a complex production environment, where the types and severities of numerically controlled machine tool faults vary, this method can comprehensively consider the fault type, fault level, and the working time of the machine tool to ensure that each decision-making unit can reflect the true state of the machine tool during evaluation. By accurately modeling the fault occurrence pattern of the machine tool, potential problems can be identified in advance, reducing sudden faults, and thus improving the operating stability and production efficiency of the production line.

[0049] In some embodiments, the step S3 further includes the steps:

[0050] The multiple evaluation model is expressed as:

[0051] ;

[0052] In the formula, is the maximum fault severity value; j is the index of the decision-making unit to be evaluated; is the weight coefficient; is the weight coefficient of the jth decision-making unit to be evaluated; , are slack variables, where is the excess input variable, is the insufficient output variable; represents the input vector, which includes at least the proportion of fault types, the proportion of fault levels, and the mean time between failures; represents the output vector, which includes at least the total number of faults and the total cost of fault repair within the preset duration;

[0053] Among them, there are differences in input variables, output variables, weight coefficients, and slack variables among the decision-making units to be evaluated. The input variables may vary due to differences in the proportion of failure types, the proportion of failure levels, and the mean time between failures; the output variables are affected by the total number of failures and the total cost of failure repairs within a preset time period, and the values of each decision-making unit may be different. In addition, the weight coefficients also vary according to the input-output conditions of the unit, reflecting their relative importance in the overall evaluation. The slack variables further reflect the performance of the unit in resource utilization and failure management, representing excessive input, that is, whether there is resource waste, while represents insufficient output, reflecting whether the ideal failure management effect has not been achieved. These factors jointly determine the performance of each unit in the optimization calculation and ultimately affect the calculated value of the maximum failure severity .

[0054] The constraint conditions of the multiple evaluation model are as follows:

[0055] ;

[0056] ;

[0057] ;

[0058] ;

[0059] In the formula, represents the first input vector, represents the first output vector.

[0060] As can be seen from the above description, by constructing a multiple evaluation model, the systematicness and flexibility of fault analysis are further enhanced. The core of the model is to obtain the optimal machine tool fault evaluation results by maximizing the fault severity and optimizing the weight coefficients. By linearly combining the inputs and outputs of each decision-making unit (i.e., the machine tool state) in multiple dimensions such as fault type, fault level, and time between failures, the working efficiency and potential risks of each machine tool can be more accurately quantified. In addition, the introduction of constraint conditions ensures that the complexity and diversity of faults are taken into account during the evaluation process, and the optimal weight coefficients and slack variables are obtained through optimization solving, making the model more in line with the fault occurrence law in the actual production environment. This method can not only provide accurate fault analysis reports for equipment maintenance personnel, but also help enterprises predict the occurrence of faults in advance while reducing downtime and productivity losses caused by equipment failures, thereby realizing refined and intelligent equipment management.

[0061] In some embodiments, the step S3 further includes the steps of:

[0062] Obtain the weight coefficients through the said constraint conditions λ and slack variables 、 The optimal solutions of,. According to the optimal solutions, if θ = 1, and and , then the decision-making unit j is considered effective.

[0063] As can be seen from the above description, in this example, it is proposed to obtain the optimal solution by solving linear programming, further improving the accuracy and operability of the fault analysis method. Through the solution of the constraint conditions, the optimal weight coefficients and slack variables can be obtained, thus ensuring that the evaluation results of each decision-making unit are the most real and accurate. The introduction of this method avoids the uncertainty of manually adjusting parameters in the traditional method and improves the automation degree of fault analysis. During the solution process, the fault severity rate is maximized through a multi-evaluation model, making the fault analysis results more in line with the actual operating conditions of the machine tool and ensuring the effectiveness of the decision-making results. Especially in the production environment, if the decision-making unit is determined to be effective (θ = 1 and the slack variable is 0), it means that the machine tool is operating in the best state, which can greatly improve production efficiency and reduce fault downtime. Therefore, this method has significant advantages in production equipment management, fault warning, and resource optimization, providing a strong guarantee for the production safety and economic benefits of enterprises. In addition, if θ is less than 1, the decision-making unit is considered ineffective at this time.

[0064] In some embodiments, the step S3 further includes:

[0065] The proportion of fault types is to calculate the proportion of each machine tool fault type in the total number of faults;

[0066] The proportion of fault levels is to calculate the proportion of each machine tool fault level in the total number of faults.

[0067] In some embodiments, the step S2 further includes the steps of:

[0068] Obtain the equipment operation duration within a preset duration according to the number of fault occurrences and the fault occurrence duration within the preset duration; Based on the equipment operation duration, calculate the mean time between failures.

[0069] From the above description, it can be seen that a detailed calculation method for the proportion of fault types and fault levels is provided, making the fault analysis of CNC machine tools more detailed and accurate. By accurately calculating the proportion of each machine tool fault type and level, the specific circumstances of the fault can be deeply analyzed to help enterprise managers identify the most common or most dangerous fault modes. Specifically, through the proportional analysis of various types of faults, it is possible to better understand the impact of different types of faults on machine tool operation and guide the priority of equipment maintenance and repair work based on actual data. The proportional analysis of fault levels helps to quantify the severity of faults and provide enterprises with targeted preventive measures, thereby avoiding minor faults from turning into serious problems and reducing the risks and losses caused by equipment failures in the production process. Specifically, fault levels include general, severe, and particularly severe.

[0070] Please refer to Figure 3 A numerical control machine tool fault analysis terminal 1 includes a memory 2, a processor 3, and a computer program stored in the memory 2 and executable on the processor 3. When the processor 3 executes the computer program, the steps in the numerical control machine tool fault analysis method are completed. That is, an execution carrier in the numerical control machine tool fault analysis method is provided.

[0071] The present invention provides a CNC machine tool fault analysis method and terminal, which are mainly used to comprehensively analyze machine tool faults, thereby discovering potential problems and reducing downtime and productivity losses caused by faults. The following is a specific description in conjunction with an embodiment:

[0072] Please refer to Figures 1 to 2 , Embodiment 1 of the present invention is:

[0073] A method for analyzing a fault of a numerically controlled machine tool comprises the following steps:

[0074] S1. Obtain machine tool fault data;

[0075] S2. Based on the machine tool fault data, the number of fault occurrences and the fault occurrence duration within a preset time are obtained by statistics, and the average trouble-free working time is calculated; specifically, the machine tool fault data statistics process is as follows: when a fault occurs, data is reported and the time when the fault is triggered is recorded; a complete alarm duration is obtained, that is, the difference between the alarm record trigger time and the alarm fault release time, and the complete number of fault occurrences within the preset time interval is calculated regularly;

[0076] S3. Construct a multiple evaluation model based on the machine tool fault type, machine tool fault level and mean trouble-free working time; train the multiple evaluation model with the goal of maximizing the fault severity rate and the optimal weight coefficient;

[0077] S4. Use the trained multiple evaluation model to perform fault analysis on CNC machine tools.

[0078] That is, in this embodiment, by obtaining the fault data, counting the number of faults and the fault duration, and combining with the calculated mean time between failures (MTBF), a foundation can be laid for further fault prediction and analysis. This method can accurately reflect the operating state of the machine tool within the preset duration, avoiding the problem of over-reliance on empirical judgment in traditional methods, thereby improving the scientificity and reliability of data-driven decision-making.

[0079] Embodiment 2 of the present invention is as follows:

[0080] Based on Embodiment 1, step S3 further includes the steps:

[0081] The multiple evaluation model is expressed as:

[0082] ;

[0083] In the formula, is the maximum fault severity value; j is the index of the decision-making unit to be evaluated; is the weight coefficient; is the weight coefficient of the jth decision-making unit to be evaluated; , are slack variables, where is the excessive input variable, is the insufficient output variable; represents the input vector, including at least the fault type ratio, the fault level ratio, and the mean time between failures; represents the output vector, including at least the total number of faults and the total cost of fault repair within the preset duration; the fault type ratio is the ratio of each machine tool fault type to the total number of faults; the fault level ratio is the ratio of each machine tool fault level to the total number of faults.

[0084] The constraint conditions of the multiple evaluation model are as follows:

[0085] ;

[0086] In the formula, represents the first input vector; this constraint means that the input of the decision-making unit i to be evaluated can be represented by a linear combination of the inputs of all other decision-making units, and non-negative slack variables can be added to represent the excess of the input. This means finding an effective frontier such that the input of the evaluation unit can be replaced by the inputs of other units while allowing waste of the input.

[0087] ;

[0088] In the formula, represents the first output vector; this constraint indicates that the output of the decision-making unit to be evaluated can also be represented by a linear combination of the outputs of all other decision-making units, and the non-negative slack variable can be reduced to represent the shortage of output. This indicates that on the effective frontier, the output of the evaluation unit can be replaced by the outputs of other units, taking into account the shortage of output.

[0089] ;

[0090] .

[0091] The weight coefficients and the slack variables 、 are obtained through the constraint conditions. According to the optimal solution, if and and , then the decision-making unit j is considered effective.

[0092] The purpose of these constraint conditions is to find a set of optimal weights and the slack variables 、 such that the efficiency index θ of the decision-making unit i to be evaluated reaches the maximum value. If and and , then the decision-making unit j is considered effective, that is, it achieves the maximum possible output at the current input level or uses the minimum input to achieve the current output level. If θ < 1, then the decision-making unit i is ineffective and can be adjusted by adjusting the input and output to achieve higher efficiency. By solving the linear programming, adjustments will be made according to the objective function and constraint conditions during the solution process until the optimal solution is found, which can make the efficiency score of the decision-making unit to be evaluated and the optimal weight coefficients λ. These weight coefficients reflect the relative importance of each decision-making unit in evaluating the severity of CNC machine tool failures, so as to analyze the importance of different fault types, levels, MTBF, etc. for maintenance costs.

[0093] Please refer to Figure 3 , Example 3 of the present invention is:

[0094] A CNC machine tool fault analysis terminal 1, including a memory 2, a processor 3, and a computer program stored on the memory 2 and operable on the processor. When the processor 3 executes the computer program, the steps in the CNC machine tool fault analysis method are completed. That is, an execution carrier in the CNC machine tool fault analysis method is provided.

[0095] In summary, for the CNC machine tool fault analysis method and terminal provided by the present invention, its comprehensive beneficial effects are mainly reflected in the following aspects:

[0096] (1) Improve the accuracy of fault analysis: By comprehensively collecting and analyzing the machine tool fault data, especially the detailed statistics of the number of fault occurrences, duration, fault types, and levels, the present invention can more accurately reflect the fault occurrence patterns of the machine tool during actual production. This data-driven analysis method avoids the limitations of traditional experience-based methods and improves the accuracy and scientific nature of fault diagnosis.

[0097] (2) Multi-dimensional fault assessment: By constructing multiple evaluation models, the present invention comprehensively considers multiple factors such as the fault types, levels, and fault-free working time of the machine tool, ensuring the comprehensiveness and objectivity of the fault analysis results. The model can not only evaluate the operating efficiency of each decision-making unit but also deeply analyze the impact of different fault modes on the machine tool performance, helping enterprises more accurately identify potential risks.

[0098] (3) Optimize equipment management decisions: The present invention obtains the optimal fault assessment results by solving linear programming to optimize the weight coefficients and slack variables. This enables equipment management personnel to allocate resources, maintain equipment, and issue fault warnings more scientifically. By identifying faults in advance and optimizing the equipment operation configuration, the downtime and productivity losses caused by sudden faults can be significantly reduced, thereby improving production efficiency and reducing operating costs.

[0099] The above are only embodiments of the present invention and do not limit the patent scope of the present invention. Any equivalent transformation made using the specification and drawings of the present invention, or directly or indirectly applied in related technical fields, shall be included in the patent protection scope of the present invention by the same token.

Claims

1. A method for analyzing a fault of a numerically controlled machine tool, characterized in that: Includes steps: S1. Obtain machine tool fault data; S2. Based on the machine tool fault data, obtain the number of faults occurring within a preset time and the duration of the faults, and calculate the average trouble-free working time; S3, constructing a multiple evaluation model according to the machine tool fault type, machine tool fault level and mean trouble-free working time; training the multiple evaluation model with the goal of maximizing the fault severity rate and the optimal weight coefficient; The multiple evaluation model is expressed as: ； In the formula, is the maximum fault severity value; j is the index of the decision-making unit being evaluated; is the weight coefficient; is the weight coefficient of the jth decision-making unit being evaluated; , is the slack variable, where For excessive input variables, For insufficient output variables; represents an input vector, including at least the fault type ratio, fault level ratio and mean time between failures; represents an output vector, which includes at least the total number of faults and the total cost of fault repair within a preset time; The constraints of the multiple evaluation model are as follows: ； ； ； ； In the formula, represents the first input vector, represents the first output vector; S4. Utilizing the trained multiple evaluation model to perform fault analysis on the CNC machine tool.

2. A method for analyzing a fault of a numerically controlled machine tool according to claim 1, characterized in that: The step S3 further comprises the steps of: The weight coefficient is obtained by the constraint condition and slack variables , The optimal solution of , according to the optimal solution, if ,and and , then the decision unit j is considered effective.

3. A method for analyzing a fault of a numerically controlled machine tool according to claim 1, characterized in that: The step S3 further comprises: The fault type ratio is the ratio of each machine tool fault type to the total number of faults; The fault level ratio is calculated by calculating the ratio of each machine tool fault level to the total number of faults.

4. A method for analyzing a fault of a numerically controlled machine tool according to claim 1, characterized in that: The step S2 further comprises the steps of: According to the number of faults occurring within a preset time period and the duration of the faults occurring, the equipment operation time within the preset time period is obtained; based on the equipment operation time, the average trouble-free working time is calculated.

5. A CNC machine tool fault analysis terminal, characterized in that: The invention comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the following steps are performed: S1. Obtain machine tool fault data; S2. Based on the machine tool fault data, obtain the number of faults occurring within a preset time and the duration of the faults, and calculate the average trouble-free working time; S3, constructing a multiple evaluation model according to the machine tool fault type, machine tool fault level and mean trouble-free working time; training the multiple evaluation model with the goal of maximizing the fault severity rate and the optimal weight coefficient; The multiple evaluation model is expressed as: ； In the formula, is the maximum fault severity value; j is the index of the decision-making unit being evaluated; is the weight coefficient; is the weight coefficient of the jth decision-making unit being evaluated; , is the slack variable, where For excessive input variables, For insufficient output variables; represents an input vector, including at least the fault type ratio, fault level ratio and mean time between failures; represents an output vector, which includes at least the total number of faults and the total cost of fault repair within a preset time; The constraints of the multiple evaluation model are as follows: ； ； ； ； In the formula, represents the first input vector, represents the first output vector; S4. Utilizing the trained multiple evaluation model to perform fault analysis on the CNC machine tool.

6. A CNC machine tool fault analysis terminal according to claim 5, characterized in that: The step S3 further comprises the steps of: The weight coefficient is obtained by the constraint condition and slack variables , The optimal solution of , according to the optimal solution, if ,and and , then the decision unit j is considered effective.

7. The CNC machine tool fault analysis terminal according to claim 5, characterized in that: The step S3 further comprises: The fault type ratio is the ratio of each machine tool fault type to the total number of faults; The fault level ratio is calculated by calculating the ratio of each machine tool fault level to the total number of faults.

8. The CNC machine tool fault analysis terminal according to claim 5, characterized in that: The step S2 further comprises the steps of: According to the number of faults occurring within a preset time period and the duration of the faults occurring, the equipment operation time within the preset time period is obtained; based on the equipment operation time, the average trouble-free working time is calculated.

Images