Turbine rotor wheel groove machining process management method based on adaptive monitoring system
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HARBIN TURBINE
- Filing Date
- 2026-03-24
- Publication Date
- 2026-06-12
Smart Images
Figure CN122194865A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of steam turbine rotor control technology. Background Technology
[0002] The machining process of steam turbine rotors, especially the machining of key features such as wheel grooves, typically faces the following technical challenges:
[0003] First, turbine rotors are often made of high-strength, high-temperature resistant alloy materials, which are both hard and tough. A key machining component is the wheel groove, which can have complex shapes, such as fir tree-shaped grooves, requiring multiple cutting tools in combination. The tools employ full-edge cutting, with the forming tool milling inside the workpiece, making it impossible to visually observe the tool's condition.
[0004] Secondly, traditional production management relies on the experience of on-site technicians and operators, making it impossible to adjust tool cutting parameters in real time based on actual processing conditions. When the actual processing differs from the theoretical conditions, abnormal situations such as tool breakage and chipping can easily occur, leading to processing interruptions, workpiece scrap, and increased production costs.
[0005] Furthermore, in the existing processing procedures, the processing equipment requires dedicated personnel to monitor it in real time, resulting in high labor intensity and low efficiency. Tool life management and processing records largely rely on manual paper records, making it difficult to accurately trace tool usage information, processing characteristics, and equipment status. When processing anomalies occur, there is a lack of effective data support to quickly locate the root cause of the problem, leading to low efficiency in anomaly handling and an inability to provide early warnings of potential quality risks.
[0006] In summary, existing rotor groove machining methods suffer from poor machining stability and low efficiency. Summary of the Invention
[0007] The purpose of this invention is to solve the problems of poor processing stability and low efficiency in existing rotor wheel groove processing methods, and to propose a process control method for turbine rotor wheel groove processing based on an adaptive monitoring system.
[0008] A method for controlling the machining process of turbine rotor wheel grooves based on an adaptive monitoring system, characterized in that the method includes the following:
[0009] Step 1: Set up the rotor groove machining process. Each machining process is used to perform roughing to fine machining on all the grooves of the rotor, and bind the corresponding tool information to each machining process.
[0010] Step 2: Use the bound tool to perform the nth machining operation on the rotor. Control the tool to process according to the preset feed rate reference value. During the machining process, the power and vibration signals of the tool are collected in real time. When the tool power is detected to be less than the preset reference power or the vibration signal exceeds the preset vibration range, the tool feed rate is increased to the preset upper limit value to improve machining efficiency. When the tool power is detected to be greater than the preset reference power, the machining is stopped and a tool abnormality alarm is triggered. The initial value of n is 1.
[0011] Step 3: Detect whether the nth machining operation has been completed. If so, process all power or vibration signals collected in the nth machining operation. Input the processed dataset into the pre-trained model and output the remaining tool life. When the remaining tool life is less than the set life threshold, issue a tool life alarm.
[0012] Step 4: Determine if n is equal to the set rotor wheel groove processing procedure. If not, set n = n + 1 and execute step 2. If yes, complete the processing of the turbine rotor wheel groove.
[0013] Preferably, step 2 further includes:
[0014] The processing time is recorded in real time, and a tool detection alarm is triggered when the processing time reaches the preset time.
[0015] The beneficial effects of this invention are:
[0016] This invention detects tool power and vibration signals. When the power is too low or the vibration is too high, it may mean that the cutting is lightly loaded or chattering. The feed rate is actively increased to improve processing efficiency. When the power is too high, processing is paused to prevent tool breakage or workpiece damage. After each process is completed, the remaining tool life is predicted and an early warning is given to avoid rotor groove size errors or scrap due to sudden tool failure.
[0017] This invention enables precise control of the machining process for each groove, ensuring that every machining step meets the process requirements. Taking the machining of steam turbine rotors as an example, it can monitor various parameters during the machining process in real time, guaranteeing high-quality machining of the rotor grooves, improving the overall product quality and reliability of the industry, and reducing after-sales costs and production losses caused by quality problems.
[0018] This invention utilizes advanced sensors and data acquisition systems to acquire key data such as tool monitoring status, actual machine tool operating parameters, and tool life utilization rate in real time and accurately. Real-time monitoring of the tool monitoring status allows for proactive tool replacement, preventing production interruptions. Simultaneously, based on data such as actual machine tool operating parameters and tool life utilization rate, reverse optimization of the process can be performed to improve tool utilization efficiency, reduce production costs, and increase overall production efficiency.
[0019] This invention can automatically adjust processing parameters based on real-time collected data, improving processing accuracy and efficiency. Combined with remote manual operation and maintenance monitoring of the processing status, it enables less-attentive operation in the production and manufacturing of heavy equipment, creating a new intelligent production model of one person operating multiple machines and centralized control, greatly improving labor productivity and reducing labor costs.
[0020] Digital manufacturing control systems deeply integrate information technology with manufacturing technology, enabling intelligent, information-based, and visualized production processes. Enterprises can monitor production progress, quality status, and equipment operating conditions in real time, allowing for timely decision-making and adjustments. Simultaneously, it facilitates information sharing and collaboration between enterprises and upstream and downstream companies in the supply chain, optimizing resource allocation across the entire industrial chain, driving the industry towards high-end and intelligent development, and enhancing its overall competitiveness. Attached Figure Description
[0021] Figure 1 A flowchart illustrating the process control method for the machining of turbine rotor wheel grooves based on an adaptive monitoring system;
[0022] Figure 2 This is a real-time power acquisition curve.
[0023] Figure 3 This is a schematic diagram illustrating real-time parameter acquisition during the processing.
[0024] Figure 4 A graph showing the acquisition of multiple parameters;
[0025] Figure 5 This is a screenshot of the centralized control platform interface.
[0026] Figure 6 Display graphs for multiple parameters;
[0027] Figure 7 This is a diagram illustrating the tool application analysis process.
[0028] Figure 8 This is a graph showing the alarm times.
[0029] Figure 9 This is a schematic diagram illustrating the control principle. Detailed Implementation
[0030] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0031] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but this is not intended to limit the scope of the invention.
[0032] Example 1:
[0033] A method for controlling the machining process of turbine rotor wheel grooves based on an adaptive monitoring system, the method comprising the following:
[0034] Step 1: Set up the rotor groove machining process to machine the rotor groove from rough to fine, and bind the corresponding tool information to each machining process;
[0035] Step 2: Use the bound tool to perform the nth machining operation on the rotor. Control the tool to process according to the preset feed rate reference value. During the machining process, the power or vibration signal of the tool is collected in real time. When the power of the tool is detected to be less than the preset reference power or the vibration signal exceeds the preset vibration range, the tool feed rate is increased to the preset upper limit value to improve machining efficiency. When the power of the tool is detected to be greater than the preset reference power, the machining is stopped and a tool abnormality alarm is triggered. The initial value of n is 1.
[0036] Step 3: Detect whether the nth processing operation has been completed. If so, process the power or vibration signal collected in the nth processing operation, input the processed dataset into the pre-trained model, and output the remaining tool life. When the remaining tool life is less than the set life threshold, issue a tool life alarm.
[0037] Step 4: Determine if n is equal to the set rotor wheel groove processing procedure. If not, set n = n + 1 and execute step 2. If yes, complete the processing of the turbine rotor wheel groove.
[0038] Specifically, the machining process of the rotor wheel grooves generally involves four machining steps. First, a first cutting tool is used for rough machining of all the grooves. Then, a second cutting tool is used for the first fine machining, followed by a third cutting tool for the second fine machining, and finally, a fourth cutting tool is used for the final fine machining. Therefore, n can be 4.
[0039] In step 3, after each groove is machined, all power or vibration signals collected during the machining of the groove are processed to predict the remaining tool life.
[0040] Further specifying, step 2 also includes:
[0041] When the processing time reaches the preset time, a tool detection alarm will be triggered.
[0042] Further specifying, step 2 also includes:
[0043] The number of processed rotor grooves is collected in real time and displayed.
[0044] Further specifying, step 2 also includes:
[0045] It displays the number of processing steps and the type of cutting tool in real time.
[0046] Further specifying, step 2 also includes:
[0047] Real-time display of power signal, vibration signal and feed rate.
[0048] Further specifying, the power or vibration signal collected during the nth processing step is processed as follows:
[0049] The power or vibration signals collected during the nth processing step are used to form a dataset. The dataset is then subjected to data cleaning, outlier removal, and feature extraction in sequence.
[0050] This embodiment achieves the following functions:
[0051] 1. Process Parameter Acquisition Module
[0052] Adopting a refined construction strategy of "one policy per type, one policy per machine," we conduct in-depth and comprehensive research and analysis on the unique attributes of rotor machining, the inherent attributes of the equipment itself, and the specific attributes of the process flow. We comprehensively develop and optimize the CNC functions of heavy-duty machine tools, establishing a stable and efficient communication connection with the machine tool's CNC system to record various data of key components in real time and accurately during machining, achieving closed-loop control of the entire machining process. By systematically reading and monitoring various key information during machining, we achieve full traceability of the machining process. Using the management concept of open-loop monitoring and closed-loop control, we promptly identify workpiece machining deformation problems and assess the impact of factors such as tool stress on machining accuracy. Through program development, we achieve cyclic recording of machining features, providing data support for subsequent quality analysis and process improvement.
[0053] 2. Processing and Equipment Data Acquisition and Monitoring Module
[0054] Multimodal sensors integrating vibration, sound, and power are installed at key equipment locations to monitor the real-time operating status of critical components, exhibiting significant differences between normal operation and abnormal conditions. This data is simultaneously fed back to the machine tool's own CNC system to determine the equipment's position, speed, and other information, achieving synchronous acquisition of high- and low-frequency data. Simultaneously, the system precisely and continuously monitors the flow and pressure of the coolant during processing, mitigating potential quality risks to the cutting tools and workpiece due to coolant fluctuations. The system also collects and closely monitors various tool status parameters in real-time, employing scientific mechanistic analysis and advanced signal fusion technology, including local lightweight data caching and initial filtering. Edge-based data cleaning removes noise interference. Time synchronization alignment integrates different signals such as vibration and power, achieving multi-source data fusion analysis. The system accurately quantifies tool wear and vibration during processing, automatically setting reasonable threshold ranges based on the actual processing conditions to identify abnormal states.
[0055] 3. Intelligent Adaptive Processing Module
[0056] Intelligent control of the machining process is achieved through data analysis. The system's self-learning function collects "healthy" data samples through repeated machining under positively given process parameters. Multi-dimensional sensor signals, including spindle power, current, vibration, acoustic emission, displacement, and temperature, are integrated to construct a high-dimensional machining state feature space. A synchronous sampling and time alignment mechanism ensures that different physical quantities are correlated under a unified time reference. Statistical learning or deep representation learning methods are used to establish a time-frequency domain distribution model of power, vibration, and other signals during normal machining, forming a "digital twin baseline." A "continuous over-limit judgment" mechanism is introduced, synchronously collecting multi-channel signals such as spindle power and three-dimensional vibration at a high-frequency sampling rate. Sliding window filtering is applied to suppress high-frequency noise and retain effective transient characteristics. Simultaneously, thresholds are automatically switched based on normal machining stages to achieve dynamic threshold boundary setting. Tool status is assessed, and machining parameters are automatically adjusted and optimized under normal and abnormal conditions given positive process parameters. Under normal machining conditions, when the system detects that the tool has detached from the workpiece (e.g., power suddenly drops below the baseline, vibration amplitude changes abruptly), the system control equipment automatically increases the feed rate and shortens the non-cutting time. At the moment of contact, a stepped deceleration algorithm is triggered to achieve a gradual, step-by-step deceleration, avoiding impact loads that could damage the tool or cause surface defects on the workpiece. After the tool detaches from the workpiece surface, the system automatically increases the tool feed rate. The system incorporates multiple judgment methods. For special processes such as full-edge machining of forming tools like rotor grooves, it automatically optimizes parameters when abnormal wear or vibration occurs. In extreme abnormal situations, the system automatically stops the feed and issues an alarm. Simultaneously, under normal machining conditions, the system performs dynamic performance evaluation of the equipment to predict machining risks in advance and reduce unnecessary downtime.
[0057] 4. Centralized Control Platform Module
[0058] A centralized control platform is constructed to seamlessly integrate various information collected by the analysis and monitoring system with the workshop data acquisition, analysis, and monitoring system within the turbine company's internal industrial Internet of Things (IoT) framework. Users can monitor tool wear, machining program execution progress, and the real-time status of machine tools and workpieces in a multi-dimensional manner, providing real-time and intuitive information. Countdown prompts and checkpoints are included to remind operators to inspect tool and machining quality. The system can provide early warnings and alerts based on real-time data analysis, offering decision support for operators. The platform supports remote operator intervention for reverse control of equipment, adjusting key parameters such as machining magnification. Simultaneously, the centralized control platform module possesses powerful data recording and analysis capabilities, providing data support for equipment maintenance, troubleshooting, and process optimization.
[0059] 5. Full data monitoring and management
[0060] Establish a comprehensive monitoring system covering every step from tool change to machining completion. Based on the prepared process technology documents, batch import process data in advance. During machining, monitor and record the machine tool's operating status and tool usage status in real time, locating machining characteristics based on the actual indexing values collected from the machine tool. Achieve precise one-to-one binding of various machining process information, including workpiece, features, tools, and parameters, ensuring the traceability of the entire machining process. Exported data can be used to analyze the machining process, optimize process parameters, and realize online workflow recording functionality for on-site data.
[0061] 6. Tool Application Analysis
[0062] Based on the collected data, various machining states are quantified. Through post-processing analysis of the exported data, further applications beyond real-time monitoring are derived. In cyclic machining processes with similar characteristics at the same level, machining data is comprehensively and deeply analyzed to perform a comprehensive analysis of tool status. By comparing the actual usage information of different tools, the manufacturer, production batch, furnace batch number, and coating characteristics of the tools are traced back, providing a reference for tool selection and process cutting experiments.
[0063] 7. Traceability of Abnormal Processes
[0064] An alarm data query linkage mechanism is introduced to accurately locate and collect relevant monitoring data and equipment processing parameters when abnormal processing occurs. The time of the anomaly, along with related data and trends, is recorded, providing a preliminary basis for problem tracing. Based on the alarm's time point, real-time operating data of various machine tools, key process parameter settings, and actual tool life are traced back to assist technicians in virtual reconstruction and simulation of the on-site operating status, pinpointing the root cause of the problem. This provides technical support and decision-making basis for optimizing on-site production management processes and adjusting and improving process parameters. Furthermore, this function can predict and warn of potential quality risks through in-depth mining and analysis of historical data, possessing powerful data visualization capabilities.
[0065] Figure 2 This demonstrates the consistency of various parameters during stable processing, and helps explain the monitoring and traceability functions of the process parameter acquisition module.
[0066] Figure 3 It presents the key locations and methods for collecting equipment status during the processing, and reflects the hardware layout and data collection status of the processing and equipment data acquisition and monitoring modules.
[0067] Figure 4 This demonstrates that the intelligent adaptive processing module achieves intelligent control of the processing process through data analysis, and displays the distribution and calculation of the collected parameters.
[0068] Figure 5 The interface layout and functions of the centralized control platform are presented in an intuitive way, demonstrating the multi-dimensional information display and operation functions provided by the centralized control platform modules to users.
[0069] Figure 6 This demonstrates the data collection process and methods in the full-data monitoring and management module, explaining how to achieve accurate binding and traceability of processing information.
[0070] Figure 7 This section presents the process and methods for analyzing machining data in the tool application analysis module, demonstrating how data post-processing can lead to more applications.
[0071] Figure 8 This is used to explain how alarm data is recorded and displayed in the abnormal process tracing module, assisting technicians in tracing and analyzing problems.
[0072] While the invention has been described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the invention. Therefore, it should be understood that many modifications can be made to the exemplary embodiments, and other arrangements can be designed without departing from the spirit and scope of the invention as defined by the appended claims. It should be understood that different dependent claims and features described herein can be combined in ways different from those described in the original claims. It is also understood that features described in conjunction with individual embodiments can be used in other described embodiments.
Claims
1. A method for controlling the machining process of turbine rotor wheel grooves based on an adaptive monitoring system, characterized in that, The method includes the following: Step 1: Set up the rotor groove machining process. Each machining process is used to perform roughing to fine machining on all the grooves of the rotor, and bind the corresponding tool information to each machining process. Step 2: Use the bound tool to perform the nth machining operation on the rotor. Control the tool to process according to the preset feed rate reference value. During the machining process, the power and vibration signals of the tool are collected in real time. When the tool power is detected to be less than the preset reference power or the vibration signal exceeds the preset vibration range, the tool feed rate is increased to the preset upper limit value to improve machining efficiency. When the tool power is detected to be greater than the preset reference power, the machining is stopped and a tool abnormality alarm is triggered. The initial value of n is 1. Step 3: Detect whether the nth machining operation has been completed. If so, process all power or vibration signals collected in the nth machining operation. Input the processed dataset into the pre-trained model and output the remaining tool life. When the remaining tool life is less than the set life threshold, issue a tool life alarm. Step 4: Determine if n is equal to the set rotor wheel groove processing procedure. If not, set n = n + 1 and execute step 2. If yes, complete the processing of the turbine rotor wheel groove.
2. The method for controlling the machining process of turbine rotor wheel grooves based on an adaptive monitoring system according to claim 1, characterized in that, Step 2 also includes: The processing time is recorded in real time, and a tool detection alarm is triggered when the processing time reaches the preset time.
3. The method for controlling the machining process of turbine rotor wheel grooves based on an adaptive monitoring system according to claim 1, characterized in that, Step 2 also includes: The number of processed rotor grooves is collected in real time and displayed.
4. The method for controlling the machining process of turbine rotor wheel grooves based on an adaptive monitoring system according to claim 1, characterized in that, Step 2 also includes: It displays the number of processing steps and the type of cutting tool in real time.
5. The method for controlling the machining process of turbine rotor wheel grooves based on an adaptive monitoring system according to claim 1, characterized in that, Step 2 also includes: Real-time display of power signal, vibration signal and feed rate.
6. The method for controlling the machining process of turbine rotor wheel grooves based on an adaptive monitoring system according to claim 1, characterized in that, The power or vibration signal collected during the nth processing step is processed as follows: The power or vibration signals collected during the nth processing step are used to form a dataset. The dataset is then subjected to data cleaning, outlier removal, and feature extraction in sequence.