Fault monitoring system and method for turbine blade processing machine tool

Abstract

The invention relates to the technical field of fault processing, in particular to a turbine blade processing machine tool fault monitoring system and method. In the monitoring period, operating parameters of a turbine blade machining machine tool are collected; according to the collected operation parameters of the turbine blade processing machine tool, constructing and training to obtain a machine tool fault identification model; operating parameters of a turbine blade machining machine tool are collected in real time and input into the machine tool fault recognition model, and the fault recognition model outputs a real-time machine tool fault type according to the obtained operating parameters; if the output real-time machine tool fault type shows that a specific fault exists, fault information is sent to a manager of the steam turbine blade machining machine tool, and the manager carries out fault removal on the steam turbine blade machining machine tool according to the fault information; compared with a traditional fault diagnosis method in the prior art, the method has the advantage that the adaptability and accuracy of fault identification are improved.

Description

Technical Field

[0001] The invention relates to the technical field of fault processing, and in particular to a fault monitoring system and method for a steam turbine blade processing machine tool. Background Art

[0002] Steam turbine blades are one of the key components of a steam turbine. Their main function is to convert the thermal energy and kinetic energy of steam into mechanical energy. The blades are usually installed on the rotor of the steam turbine. As steam passes through the blades, the blades are impacted by the steam and push the rotor to rotate, thereby driving the generator or mechanical equipment. The processing of steam turbine blades is a highly precise and complex process, which usually requires the use of specialized machine tools. These machine tools can ensure that the accuracy, shape and surface quality of the blades meet the design requirements during the production process, thereby meeting the working requirements of the steam turbine in high temperature, high pressure and high speed environments.

[0003] However, after long-term use and high-load operation, machine tools will be affected by various factors such as mechanical wear, material aging, temperature changes, vibration and shock, and are prone to various faults, such as wear of parts, failure of bearings, failure of electrical systems, poor lubrication and corrosion, etc. These faults cause the machine tools to reduce or lose their intended functions and reduce production capacity. Therefore, how to improve fault detection capabilities and reduce production delays caused by late detection of faults has become an issue that needs to be urgently addressed.

[0004] For example, a Chinese patent with application number 202310983557.0 and application date of August 7, 2023, and the invention name is "A method for fault monitoring of CNC machine tools". Its technical solution is: a method for fault monitoring of CNC machine tools, comprising the following steps: the equipment operates by collecting sensor data of the CNC machine tool through a state perception model, and performs data preprocessing on the sensor data through a lightweight computing model to predict the equipment operation, and obtain the machine tool operation data. When the CNC machine tool is running, the operating status characteristic information is sent to the fault diagnosis module through the fault detection module, and the fault diagnosis module The operating status characteristic information is received, and the operating status characteristic information is compared with the operating status characteristic information in a fault database to generate a fault diagnosis result. The fault prediction module predicts the fault of the operating status characteristic according to relevant models and algorithms, and analyzes possible faulty components, fault causes, fault types, etc. Whether the equipment needs to be repaired is determined through the fault prediction analysis, and corresponding control instructions are sent to the CNC system of the CNC machine tool according to the fault diagnosis result. The fault information is displayed by sound and light through the alarm module, and the fault characteristic information, the fault diagnosis result and the control instruction are sent to the operator of the CNC machine tool.

[0005] The above patent realizes fault detection through a fault detection module and a fault diagnosis module. However, in the above patent, after obtaining the machine tool related data in the fault detection module, the fault diagnosis module only compares it with the fault database to determine whether a fault occurs. This monitoring method will make it impossible to obtain the fault situation from the database when a new fault that has never occurred before occurs, thereby causing a misjudgment of the machine tool fault. Therefore, the fault monitoring method adopted in the above patent has a large error in fault judgment and a low accuracy. Summary of the invention

[0006] In order to solve the problems existing in the prior art, the present invention provides a turbine blade processing machine tool fault monitoring system and method, which uses a machine tool fault identification model constructed based on deep learning to make more accurate judgments on machine tool faults and output more accurate fault results.

[0007] In order to achieve the above technical effects, the technical solution of the present invention is as follows: A fault monitoring system for a steam turbine blade processing machine tool, comprising: An operating parameter collection module is used to collect operating parameters of a steam turbine blade processing machine tool; The recognition model building module builds and trains a machine tool fault recognition model based on the operating parameters of the steam turbine blade processing machine tool collected by the operating parameter collection module; A fault identification module is used to input the operating parameters of the steam turbine blade processing machine tool into the machine tool fault identification model in the identification model building module, thereby outputting the real-time machine tool fault type; The information push module sends fault information to the management personnel of the steam turbine blade processing machine tool according to the machine tool fault type output by the fault identification module.

[0008] Furthermore, in the recognition model building module, the specific method steps for building and training a machine tool fault recognition model are as follows: Step a: within the monitoring period, determine the abnormal operation time period of the steam turbine blade processing machine tool, and record the fault type corresponding to the time period; Step b: within the monitoring period, confirm the normal operation time period of the steam turbine blade processing machine tool, and record the fault type corresponding to the time period; Step c: Establish a database to store the operating parameters and the fault types corresponding to the operating parameters, then establish a fault identification model through deep learning, use the database to train the fault identification model, and finally obtain a trained machine tool fault model; In step a, the specific method of determining the abnormal operation time period of the steam turbine blade processing machine tool is: the time point when each fault starts to occur is the target time T, then the abnormal operation time period of the steam turbine blade processing machine tool is: [T i, T i +ΔT], where T i represents the i-th target time, ΔT is the preset time length; The operating parameters of the turbine blade processing machine tool collected in the operating parameter acquisition module are simultaneously input into the PLC and CNC to determine the fault type. The fault type corresponding to the operating parameters output by the PLC and CNC is used as the comparison result, and the comparison result is compared with the fault type corresponding to the operating parameters output by the machine tool fault identification model. When the comparison result is different from the fault type corresponding to the operating parameters output by the machine tool fault identification model, the comparison result is input into the machine tool fault identification model, and deep learning training is performed again to back-propagate and update the machine tool fault identification model.

[0009] Furthermore, in the information push module, the specific method and steps for sending fault information to the staff of the steam turbine blade processing machine tool are as follows: Step a1: First, set m information sending frequency levels and m management personnel number levels from low to high; Step b1: when a fault occurs and the total number of managers is at the ath manager number level, the information sending frequency level is ma, where m is greater than a; the higher the information sending frequency level, the higher the information sending frequency; The fault information includes each fault type, the fault level corresponding to each fault type and the location where the fault occurs; the fault levels include level one fault, level two fault and level three fault.

[0010] A method for monitoring faults of a steam turbine blade processing machine tool comprises the following specific method steps: Step 1: During the monitoring period, collect the operating parameters of the steam turbine blade processing machine tool; Step 2: Based on the collected operating parameters of the steam turbine blade processing machine tool, a machine tool fault identification model is constructed and trained; Step 3: The operating parameters of the steam turbine blade processing machine tool are collected in real time and input into the machine tool fault identification model. The fault identification model outputs the real-time machine tool fault type according to the obtained operating parameters; Step 4: If the output real-time machine tool fault type shows that there is a specific fault, the fault information is sent to the manager of the steam turbine blade processing machine tool, and the manager troubleshoots the steam turbine blade processing machine tool based on the fault information.

[0011] Furthermore, the operating parameters include an abnormal operating time period of the steam turbine blade processing machine tool and a normal operating time period of the steam turbine blade processing machine tool.

[0012] Furthermore, in step 2, the specific method steps for constructing and training a machine tool fault identification model are as follows: Step a: within the monitoring period, determine the abnormal operation time period of the steam turbine blade processing machine tool, and record the fault type corresponding to the time period; Step b: within the monitoring period, confirm the normal operation time period of the steam turbine blade processing machine tool, and record the fault type corresponding to the time period; Step c: Establish a database to store the operating parameters and the fault types corresponding to the operating parameters, then establish a fault identification model through deep learning, use the database to train the fault identification model, and finally obtain the trained machine tool fault model.

[0013] Furthermore, in step a, the specific method of determining the abnormal operation time period of the steam turbine blade processing machine tool is: the time point when each fault starts to occur is the target time T, then the abnormal operation time period of the steam turbine blade processing machine tool is: [T i , T i +ΔT], where T i represents the i-th target time, and ΔT is the preset time length.

[0014] Furthermore, in the step b, the intersection of the normal operation time period of the steam turbine blade processing machine tool and the abnormal operation time period of the steam turbine blade processing machine tool is empty.

[0015] Furthermore, in the step three, the collected operating parameters are simultaneously input into the PLC and CNC to determine the fault type, the fault type corresponding to the operating parameters output by the PLC and CNC is used as a comparison result, and the comparison result is compared with the fault type corresponding to the operating parameters output by the machine tool fault identification model. When the comparison result is different from the fault type corresponding to the operating parameters output by the machine tool fault identification model, the comparison result is input into the machine tool fault identification model, and deep learning training is performed again to back-propagate and update the machine tool fault identification model.

[0016] Furthermore, in step 4, the specific method and steps for sending fault information to the staff of the steam turbine blade processing machine tool are as follows: Step a1: First, set m information sending frequency levels and m management personnel quantity levels from low to high; Step b1: When a fault occurs and the total number of managers is at the ath manager number level, the information sending frequency level is ma, where m is greater than a; the higher the information sending frequency level, the higher the information sending frequency; the fault information includes each fault type, the fault level corresponding to each fault type and the location where the fault occurs; the fault levels include level one fault, level two fault and level three fault.

[0017] According to the above technical solution, the beneficial effects of the present invention are as follows: 1. The method adopted in the present invention utilizes deep learning to construct a machine tool fault identification model. Since deep learning has the ability to process high-dimensional data and complex nonlinear relationships, it can identify fault modes more accurately than traditional analysis methods. Compared with traditional fault diagnosis methods in the prior art, which often require direct comparison with fault data stored in a database or rely on manually set rules and expert knowledge, the deep learning model can automatically learn through a data-driven method to improve the adaptability and accuracy of fault identification.

[0018] 2. When constructing a machine tool fault identification model, the method of the present invention determines the normal operating time period of the turbine blade processing machine tool, provides a benchmark for the machine tool fault identification model, enables the model to identify the normal operating status of the machine tool, helps the model learn the characteristics under normal operation, and avoids mistaking normal fluctuations for faults, thereby reducing the false alarm rate of the machine tool fault identification model.

[0019] 3. The method adopted by the present invention sends fault information to management personnel, and divides the fault levels corresponding to the fault types into primary faults, secondary faults and tertiary faults, thereby achieving the technical effect of optimizing resource allocation, ensuring rapid response to key issues and reducing risks.

[0020] 4. The system adopted by the present invention sends fault information to management personnel through an information push module, thereby ensuring that relevant personnel can understand the fault details in time and handle them quickly, avoiding production delays caused by late discovery of fault problems. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a system schematic diagram of the present invention.

[0022] Figure 2 It is a schematic flow chart of the method of the present invention. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments.

[0024] Example 1 like Figure 1 As shown, a fault monitoring system for a steam turbine blade processing machine tool comprises: An operating parameter acquisition module is used to acquire operating parameters of a steam turbine blade processing machine tool; using sensors commonly used in the prior art to acquire operating parameters of a steam turbine blade processing machine tool; The recognition model building module builds and trains a machine tool fault recognition model based on the operating parameters of the steam turbine blade processing machine tool collected by the operating parameter collection module; A fault identification module is used to input the operating parameters of the steam turbine blade processing machine tool into the machine tool fault identification model in the identification model building module, thereby outputting the real-time machine tool fault type; The information push module sends fault information to the management personnel of the steam turbine blade processing machine tool according to the machine tool fault type output by the fault identification module.

[0025] In the recognition model building module, the specific steps for building and training a machine tool fault recognition model are as follows: Step a: within the monitoring period, determine the abnormal operation time period of the steam turbine blade processing machine tool, and record the fault type corresponding to the time period; Step b: within the monitoring period, confirm the normal operation time period of the steam turbine blade processing machine tool, and record the fault type corresponding to the time period; Step c: Establish a database to store the operating parameters and the fault types corresponding to the operating parameters, then establish a fault identification model through deep learning, use the database to train the fault identification model, and finally obtain a trained machine tool fault model; In step a, the specific method of determining the abnormal operation time period of the steam turbine blade processing machine tool is: the time point when each fault starts to occur is the target time T, then the abnormal operation time period of the steam turbine blade processing machine tool is: [T i , T i +ΔT], where T i represents the i-th target time, ΔT is the preset time length; The operating parameters of the turbine blade processing machine tool collected in the operating parameter acquisition module are simultaneously input into the PLC and CNC to determine the fault type, and the fault type corresponding to the operating parameters output by the PLC and CNC is used as a comparison result, and the comparison result is compared with the fault type corresponding to the operating parameters output by the machine tool fault identification model. When the comparison result is different from the fault type corresponding to the operating parameter output by the machine tool fault identification model, the comparison result is input into the machine tool fault identification model, and deep learning training is performed again to back-propagate and update the machine tool fault identification model; PLC and CNC are both existing systems in the field, and the comparison results of the fault types corresponding to the operating parameters output by the PLC and CNC are used as a comparison reference, and the comparison results are input into the machine tool fault identification model for training, thereby improving the accuracy of the model and ensuring its compatibility with existing mature technologies.

[0026] In the information push module, the specific method and steps for sending fault information to the staff of the steam turbine blade processing machine tool are as follows: Step a1: First, set m information sending frequency levels and m management personnel quantity levels from low to high; Step b1: When a fault occurs and the total number of managers is at the ath level, the information sending frequency level is ma, where m is greater than a; the higher the information sending frequency level, the higher the information sending frequency; according to the different numbers of managers, the information sending frequency is dynamically adjusted to avoid excessive concentration of information or repeated sending to the same person when the number of managers is large. Therefore, when the number of managers increases, the information sending frequency decreases, and when the number of managers is small, the sending frequency needs to be increased to ensure that the fault information can be processed in time.

[0027] The fault information includes each fault type, the fault level corresponding to each fault type, and the location where the fault occurs; the fault levels include level 1 fault, level 2 fault, and level 3 fault. Level 1 fault is a major fault that affects safety and equipment, so it is handled with the highest priority. Level 2 faults and level 3 faults can have different response times and maintenance strategies, thereby optimizing resource allocation and ensuring that key issues are responded to quickly, thereby reducing the operating risk of turbine blade processing machines; among them, level 1 faults and level 2 faults are highlighted on the digital twin display interface, so that managers can focus on level 1 faults and level 2 faults instead of being distracted by minor level 3 faults. This visual reminder enables managers to quickly take corresponding troubleshooting measures and reduces the possibility of equipment damage and production interruptions.

[0028] Example 2 like Figure 2 As shown, a method for monitoring faults of a steam turbine blade processing machine tool comprises the following specific method steps: Step 1: During the monitoring period, the operating parameters of the steam turbine blade processing machine tool are collected by using sensors commonly used in the prior art; Step 2: Based on the collected operating parameters of the steam turbine blade processing machine tool, a machine tool fault identification model is constructed and trained; Step 3: collect the operating parameters of the steam turbine blade processing machine tool in real time and input them into the machine tool fault identification model. The fault identification model outputs the real-time machine tool fault type according to the obtained operating parameters; Step 4: If the real-time machine tool fault type output shows that there is a specific fault, the fault information is sent to the manager of the turbine blade processing machine tool, and the manager troubleshoots the turbine blade processing machine tool based on the fault information; the fault type is the fault condition of the turbine blade processing machine tool that is well known to technical personnel in this field.

[0029] The operation parameters include an abnormal operation time period of the steam turbine blade processing machine tool and a normal operation time period of the steam turbine blade processing machine tool.

[0030] In step 2, the specific method steps for constructing and training a machine tool fault identification model are as follows: Step a: within the monitoring period, determine the abnormal operation time period of the steam turbine blade processing machine tool, and record the fault type corresponding to the time period; Step b: within the monitoring period, confirm the normal operation time period of the steam turbine blade processing machine tool, and record the fault type corresponding to the time period; Step c: Establish a database to store the operating parameters and the fault types corresponding to the operating parameters, then establish a fault identification model through deep learning, use the database to train the fault identification model, and finally obtain the trained machine tool fault model.

[0031] In step a, the specific method of determining the abnormal operation time period of the steam turbine blade processing machine tool is: the time point when each fault starts to occur is the target time T, then the abnormal operation time period of the steam turbine blade processing machine tool is: [T i , T i +ΔT], where T i represents the i-th target time, and ΔT is the preset time length.

[0032] In step b, the intersection of the normal operation time period of the steam turbine blade processing machine tool and the abnormal operation time period of the steam turbine blade processing machine tool is empty.

[0033] In step three, the collected operating parameters are simultaneously input into PLC and CNC to determine the fault type, the fault type corresponding to the operating parameters output by PLC and CNC is used as a comparison result, and the comparison result is compared with the fault type corresponding to the operating parameters output by the machine tool fault identification model. When the comparison result is different from the fault type corresponding to the operating parameter output by the machine tool fault identification model, the comparison result is input into the machine tool fault identification model, and deep learning training is performed again to back-propagate and update the machine tool fault identification model; PLC and CNC are both existing systems in the field, and the comparison results of the fault types corresponding to the operating parameters output by PLC and CNC are used as a comparison reference, and the comparison results are input into the machine tool fault identification model for training, which improves the accuracy of the model and ensures its compatibility with existing mature technologies.

[0034] In step 4, the specific method and steps for sending fault information to the staff of the steam turbine blade processing machine tool are as follows: Step a1: First, set m information sending frequency levels and m management personnel quantity levels from low to high; Step b1: When a fault occurs and the total number of managers is at the ath level, the information sending frequency level is ma, where m is greater than a; the higher the information sending frequency level, the higher the information sending frequency; according to the different numbers of managers, the information sending frequency is dynamically adjusted to avoid excessive concentration of information or repeated sending to the same person when the number of managers is large. Therefore, when the number of managers increases, the information sending frequency decreases, and when the number of managers is small, the sending frequency needs to be increased to ensure that the fault information can be processed in time.

[0035] The fault information includes each fault type, the fault level corresponding to each fault type, and the location where the fault occurs; the fault levels include level 1 fault, level 2 fault, and level 3 fault. Level 1 fault is a major fault that affects safety and equipment, so it is handled with the highest priority. Level 2 faults and level 3 faults can have different response times and maintenance strategies, thereby optimizing resource allocation and ensuring that key issues are responded to quickly, thereby reducing the operating risk of turbine blade processing machines; among them, level 1 faults and level 2 faults are highlighted on the digital twin display interface, so that managers can focus on level 1 faults and level 2 faults instead of being distracted by minor level 3 faults. This visual reminder enables managers to quickly take corresponding troubleshooting measures and reduces the possibility of equipment damage and production interruptions.

[0036] Example 3 like Figure 1 As shown, a fault monitoring system for a steam turbine blade processing machine tool comprises: An operating parameter acquisition module is used to acquire operating parameters of a steam turbine blade processing machine tool; using sensors commonly used in the prior art to acquire operating parameters of a steam turbine blade processing machine tool; The recognition model building module builds and trains a machine tool fault recognition model based on the operating parameters of the steam turbine blade processing machine tool collected by the operating parameter collection module; A fault identification module is used to input the operating parameters of the steam turbine blade processing machine tool into the machine tool fault identification model in the identification model building module, thereby outputting the real-time machine tool fault type; The information push module sends fault information to the management personnel of the steam turbine blade processing machine tool according to the machine tool fault type output by the fault identification module.

[0037] In the recognition model building module, the specific steps for building and training a machine tool fault recognition model are as follows: Step a: within the monitoring period, determine the abnormal operation time period of the steam turbine blade processing machine tool, and record the fault type corresponding to the time period; Step b: within the monitoring period, confirm the normal operation time period of the steam turbine blade processing machine tool, and record the fault type corresponding to the time period; Step c: Establish a database to store the operating parameters and the fault types corresponding to the operating parameters, then establish a fault identification model through deep learning, use the database to train the fault identification model, and finally obtain a trained machine tool fault model; In step a, the specific method of determining the abnormal operation time period of the steam turbine blade processing machine tool is: the time point when each fault starts to occur is the target time T, then the abnormal operation time period of the steam turbine blade processing machine tool is: [T i , T i +ΔT], where T i represents the i-th target time, ΔT is the preset time length; The operating parameters of the turbine blade processing machine tool collected in the operating parameter acquisition module are simultaneously input into the PLC and CNC to determine the fault type, and the fault type corresponding to the operating parameters output by the PLC and CNC is used as a comparison result, and the comparison result is compared with the fault type corresponding to the operating parameters output by the machine tool fault identification model. When the comparison result is different from the fault type corresponding to the operating parameter output by the machine tool fault identification model, the comparison result is input into the machine tool fault identification model, and deep learning training is performed again to back-propagate and update the machine tool fault identification model; PLC and CNC are both existing systems in the field, and the comparison results of the fault types corresponding to the operating parameters output by the PLC and CNC are used as a comparison reference, and the comparison results are input into the machine tool fault identification model for training, thereby improving the accuracy of the model and ensuring its compatibility with existing mature technologies.

[0038] In the information push module, the specific method and steps for sending fault information to the staff of the steam turbine blade processing machine tool are as follows: Step a1: First, set m information sending frequency levels and m management personnel number levels from low to high; Step b1: When a fault occurs and the total number of managers is at the ath level, the information sending frequency level is ma, where m is greater than a; the higher the information sending frequency level, the higher the information sending frequency; according to the different numbers of managers, the information sending frequency is dynamically adjusted to avoid excessive concentration of information or repeated sending to the same person when the number of managers is large. Therefore, when the number of managers increases, the information sending frequency decreases, and when the number of managers is small, the sending frequency needs to be increased to ensure that the fault information can be processed in time.

[0039] The fault information includes each fault type, the fault level corresponding to each fault type, and the location where the fault occurs; the fault levels include level 1 fault, level 2 fault, and level 3 fault. Level 1 fault is a major fault that affects safety and equipment, so it is handled with the highest priority. Level 2 faults and level 3 faults can have different response times and maintenance strategies, thereby optimizing resource allocation and ensuring that key issues are responded to quickly, thereby reducing the operating risk of turbine blade processing machines; among them, level 1 faults and level 2 faults are highlighted on the digital twin display interface, so that managers can focus on level 1 faults and level 2 faults instead of being distracted by minor level 3 faults. This visual reminder enables managers to quickly take corresponding troubleshooting measures and reduces the possibility of equipment damage and production interruptions.

[0040] like Figure 2 As shown, a method for monitoring faults of a steam turbine blade processing machine tool comprises the following specific method steps: Step 1: During the monitoring period, the operating parameters of the steam turbine blade processing machine tool are collected by using sensors commonly used in the prior art; Step 2: Based on the collected operating parameters of the steam turbine blade processing machine tool, a machine tool fault identification model is constructed and trained; Step 3: The operating parameters of the steam turbine blade processing machine tool are collected in real time and input into the machine tool fault identification model. The fault identification model outputs the real-time machine tool fault type according to the obtained operating parameters; Step 4: If the real-time machine tool fault type output shows that there is a specific fault, the fault information is sent to the manager of the turbine blade processing machine tool, and the manager troubleshoots the turbine blade processing machine tool based on the fault information; the fault type is the fault condition of the turbine blade processing machine tool that is well known to technical personnel in this field.

[0041] The operation parameters include an abnormal operation time period of the steam turbine blade processing machine tool and a normal operation time period of the steam turbine blade processing machine tool.

[0042] In step 2, the specific method steps for constructing and training a machine tool fault identification model are as follows: Step a: within the monitoring period, determine the abnormal operation time period of the steam turbine blade processing machine tool, and record the fault type corresponding to the time period; Step b: within the monitoring period, confirm the normal operation time period of the steam turbine blade processing machine tool, and record the fault type corresponding to the time period; Step c: Establish a database to store the operating parameters and the fault types corresponding to the operating parameters, then establish a fault identification model through deep learning, use the database to train the fault identification model, and finally obtain the trained machine tool fault model.

[0043] In step a, the specific method of determining the abnormal operation time period of the steam turbine blade processing machine tool is: the time point when each fault starts to occur is the target time T, then the abnormal operation time period of the steam turbine blade processing machine tool is: [T i , T i +ΔT], where T i represents the i-th target time, and ΔT is the preset time length.

[0044] In step b, the intersection of the normal operation time period of the steam turbine blade processing machine tool and the abnormal operation time period of the steam turbine blade processing machine tool is empty.

[0045] In step three, the collected operating parameters are simultaneously input into PLC and CNC to determine the fault type, the fault type corresponding to the operating parameters output by PLC and CNC is used as a comparison result, and the comparison result is compared with the fault type corresponding to the operating parameters output by the machine tool fault identification model. When the comparison result is different from the fault type corresponding to the operating parameter output by the machine tool fault identification model, the comparison result is input into the machine tool fault identification model, and deep learning training is performed again to back-propagate and update the machine tool fault identification model; PLC and CNC are both existing systems in the field, and the comparison results of the fault types corresponding to the operating parameters output by PLC and CNC are used as a comparison reference, and the comparison results are input into the machine tool fault identification model for training, which improves the accuracy of the model and ensures its compatibility with existing mature technologies.

[0046] In step 4, the specific method and steps for sending fault information to the staff of the steam turbine blade processing machine tool are as follows: Step a1: First, set m information sending frequency levels and m management personnel number levels from low to high; Step b1: When a fault occurs and the total number of managers is at the ath level, the information sending frequency level is ma, where m is greater than a; the higher the information sending frequency level, the higher the information sending frequency; according to the different numbers of managers, the information sending frequency is dynamically adjusted to avoid excessive concentration of information or repeated sending to the same person when the number of managers is large. Therefore, when the number of managers increases, the information sending frequency decreases, and when the number of managers is small, the sending frequency needs to be increased to ensure that the fault information can be processed in time.

[0047] The fault information includes each fault type, the fault level corresponding to each fault type, and the location where the fault occurs; the fault levels include level 1 fault, level 2 fault, and level 3 fault. Level 1 fault is a major fault that affects safety and equipment, so it is handled with the highest priority. Level 2 faults and level 3 faults can have different response times and maintenance strategies, thereby optimizing resource allocation and ensuring that key issues are responded to quickly, thereby reducing the operating risk of turbine blade processing machines; among them, level 1 faults and level 2 faults are highlighted on the digital twin display interface, so that managers can focus on level 1 faults and level 2 faults instead of being distracted by minor level 3 faults. This visual reminder enables managers to quickly take corresponding troubleshooting measures and reduces the possibility of equipment damage and production interruptions.

[0048] The above description is a detailed description of the preferred feasible embodiment of the present application, but the embodiment is not intended to limit the patent application scope of the present application. All equivalent changes or modified changes made under the technical spirit suggested by the present application should fall within the patent scope covered by the present application.

Claims

1. A fault monitoring system for a steam turbine blade processing machine tool, characterized in that: include: An operating parameter collection module is used to collect operating parameters of a steam turbine blade processing machine tool; The recognition model building module builds and trains a machine tool fault recognition model based on the operating parameters of the steam turbine blade processing machine tool collected by the operating parameter collection module; A fault identification module is used to input the operating parameters of the steam turbine blade processing machine tool into the machine tool fault identification model in the identification model building module, thereby outputting the real-time machine tool fault type; The information push module sends fault information to the management personnel of the steam turbine blade processing machine tool according to the machine tool fault type output by the fault identification module.

2. A turbine blade processing machine tool fault monitoring system according to claim 1, characterized in that: In the recognition model building module, the specific steps for building and training a machine tool fault recognition model are as follows: Step a: within the monitoring period, determine the abnormal operation time period of the steam turbine blade processing machine tool, and record the fault type corresponding to the time period; Step b: within the monitoring period, confirm the normal operation time period of the steam turbine blade processing machine tool, and record the fault type corresponding to the time period; Step c: Establish a database to store the operating parameters and the fault types corresponding to the operating parameters, then establish a fault identification model through deep learning, use the database to train the fault identification model, and finally obtain a trained machine tool fault model; In step a, the specific method of determining the abnormal operation time period of the steam turbine blade processing machine tool is: the time point when each fault starts to occur is the target time T, then the abnormal operation time period of the steam turbine blade processing machine tool is: [T i , T i +ΔT], where T i represents the i-th target time, ΔT is the preset time length; The operating parameters of the turbine blade processing machine tool collected in the operating parameter acquisition module are simultaneously input into the PLC and CNC to determine the fault type. The fault type corresponding to the operating parameters output by the PLC and CNC is used as the comparison result, and the comparison result is compared with the fault type corresponding to the operating parameters output by the machine tool fault identification model. When the comparison result is different from the fault type corresponding to the operating parameters output by the machine tool fault identification model, the comparison result is input into the machine tool fault identification model, and deep learning training is performed again to back-propagate and update the machine tool fault identification model.

3. A turbine blade processing machine tool fault monitoring system according to claim 1, characterized in that: In the information push module, the specific method and steps for sending fault information to the staff of the steam turbine blade processing machine tool are as follows: Step a1: First, set m information sending frequency levels and m management personnel number levels from low to high; Step b1: when a fault occurs and the total number of managers is at the ath manager number level, the information sending frequency level is ma, where m is greater than a; the higher the information sending frequency level, the higher the information sending frequency; The fault information includes each fault type, the fault level corresponding to each fault type and the location where the fault occurs; the fault levels include level one fault, level two fault and level three fault.

4. A method for monitoring faults of a steam turbine blade processing machine tool, characterized in that: The specific method steps include the following: Step 1: During the monitoring period, collect the operating parameters of the steam turbine blade processing machine tool; Step 2: Based on the collected operating parameters of the steam turbine blade processing machine tool, a machine tool fault identification model is constructed and trained; Step 3: collect the operating parameters of the steam turbine blade processing machine tool in real time and input them into the machine tool fault identification model. The fault identification model outputs the real-time machine tool fault type according to the obtained operating parameters; Step 4: If the output real-time machine tool fault type shows that there is a specific fault, the fault information is sent to the manager of the steam turbine blade processing machine tool, and the manager troubleshoots the steam turbine blade processing machine tool based on the fault information.

5. A method for monitoring faults of a steam turbine blade machining machine tool according to claim 4, characterized in that: The operating parameters include an abnormal operating time period of the steam turbine blade machining machine tool and a normal operating time period of the steam turbine blade machining machine tool.

6. A method for monitoring faults of a steam turbine blade machining machine tool according to claim 4, characterized in that: In step 2, the specific method steps for constructing and training a machine tool fault identification model are as follows: Step a: within the monitoring period, determine the abnormal operation time period of the steam turbine blade processing machine tool, and record the fault type corresponding to the time period; Step b: within the monitoring period, confirm the normal operation time period of the steam turbine blade processing machine tool, and record the fault type corresponding to the time period; Step c: Establish a database to store the operating parameters and the fault types corresponding to the operating parameters, then establish a fault identification model through deep learning, use the database to train the fault identification model, and finally obtain the trained machine tool fault model.

7. A method for monitoring faults of a steam turbine blade machining machine tool according to claim 6, characterized in that: In step a, the specific method of determining the abnormal operation time period of the steam turbine blade processing machine tool is: the time point when each fault starts to occur is the target time T, then the abnormal operation time period of the steam turbine blade processing machine tool is: [T i , T i +ΔT], where T i represents the i-th target time, and ΔT is the preset time length.

8. A method for monitoring faults of a steam turbine blade machining machine tool according to claim 6, characterized in that: In the step b, the intersection of the normal operation time period of the steam turbine blade processing machine tool and the abnormal operation time period of the steam turbine blade processing machine tool is empty.

9. A method for monitoring faults of a steam turbine blade machining machine tool according to claim 4, characterized in that: In the step three, the collected operating parameters are simultaneously input into the PLC and CNC to determine the fault type, the fault type corresponding to the operating parameters output by the PLC and CNC is used as a comparison result, and the comparison result is compared with the fault type corresponding to the operating parameters output by the machine tool fault identification model. When the comparison result is different from the fault type corresponding to the operating parameters output by the machine tool fault identification model, the comparison result is input into the machine tool fault identification model, and deep learning training is performed again to back-propagate and update the machine tool fault identification model.

10. A method for monitoring faults of a steam turbine blade machining machine tool according to claim 4, characterized in that: In step 4, the specific method and steps for sending fault information to the staff of the steam turbine blade processing machine tool are as follows: Step a1: First, set m information sending frequency levels and m management personnel number levels from low to high; Step b1: When a fault occurs and the total number of managers is at the ath manager number level, the information sending frequency level is ma, where m is greater than a; the higher the information sending frequency level, the higher the information sending frequency; the fault information includes each fault type, the fault level corresponding to each fault type and the location where the fault occurs; the fault levels include level one fault, level two fault and level three fault.

Images