Weak point identification and virtual sensing method and system for unmonitored area of high-temperature rotating equipment
Through the deep neural network prediction model, the service working condition data of high-temperature rotating equipment is analyzed, and the weaknesses and cumulative damage are identified in real time, which solves the problem that physical sensors cannot monitor equipment damage, and realizes real-time damage status perception and health of high-temperature rotating equipment Status evaluation, optimized operation and maintenance management.
Patent Information
- Application Number
- CN202410971304.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-18
- Publication Date
- 2024-11-08
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Existing physical sensors cannot monitor the damage accumulated during the start-steady-state-stop cycle of high-temperature rotating equipment in real time, resulting in the inability to effectively evaluate the health status of the equipment and identify the weaknesses in the unmonitorable areas, affecting the service safety and operation and maintenance management of the equipment.
By obtaining the service working condition data of high-temperature rotating equipment in real time for pre-processing, and using deep neural network prediction model to predict single-week damage data, identifying weaknesses and cumulative damage data, real-time virtual sensing of unmonitorable areas is achieved.
Real-time damage state perception of high-temperature rotating equipment is realized, the information of physical sensors is extended, and technical means to evaluate the health status of equipment and optimize operation and maintenance management is provided. On the basis of adding additional physical sensors, the equipment service process is realized. real-time perception of damage state in.
Smart Images

Figure CN118917187A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of digital twins and virtual sensing, and in particular to the identification of weaknesses in unmonitored areas of high-temperature rotating equipment, and virtual sensing methods and systems. Background Art
[0002] Large high-temperature rotating equipment such as steam turbine rotors are key components of thermal power turbine units. Their service environment is extremely harsh, and their operation mode is gradually becoming flexible under the background of peak load regulation. In order to ensure the safe and reliable service of high-temperature rotating equipment, engineers usually arrange a large number of measuring points inside it to monitor the operating status of the equipment in real time. However, existing physical sensors can only measure common status data such as temperature, pressure, speed and vibration. Such data will not show a degradation trend during the crack initiation stage of high-temperature rotating equipment. Therefore, it is impossible to evaluate the health status of high-temperature rotating equipment by using such sensor data.
[0003] Virtual sensing technology can effectively make up for the shortcomings of physical sensing. By combining mature physical models with machine learning intelligent algorithms, virtual sensing can expand and extend the limited data collected by physical sensors, thereby obtaining parameters that cannot be directly measured by physical sensing.
[0004] High-temperature rotating equipment will continue to accumulate damage during the start-steady-state-shutdown cycle. When the accumulated damage reaches the damage threshold, macro cracks will be generated and eventually break quickly. Therefore, how to realize real-time damage virtual sensing of high-temperature rotating equipment based on existing data and models is of great significance to ensure the service safety and operation and maintenance management optimization of high-temperature rotating equipment. Summary of the invention
[0005] The purpose of this application is to provide a method and system for identifying weaknesses in unmonitored areas of high-temperature rotating equipment and virtual sensing, which can realize real-time virtual sensing of damage of high-temperature rotating equipment, and ensure the service safety and operation and maintenance management optimization of high-temperature rotating equipment.
[0006] To achieve the above objectives, this application provides the following solutions:
[0007] In the first aspect, the present application provides a method for identifying weaknesses in unmonitorable areas of high-temperature rotating equipment and a virtual sensing method, including:
[0008] Acquire the service condition data of the target high-temperature rotating equipment in real time; the service condition data includes the condition parameters under each cycle; the condition parameters include: initial rotor temperature, speed-up rate, load-up rate, initial main steam temperature and main steam temperature rise rate; each completion of the service process of the target high-temperature rotating equipment from start-up to steady state to shutdown is recorded as one cycle;
[0009] Preprocessing the service condition data to obtain service condition processed data; the preprocessing includes: data cleaning, filtering and standardization;
[0010] Obtain the number of cycles of the target high-temperature rotating equipment at any time;
[0011] For any of the cycle times, the service condition processing data corresponding to the cycle times are input into the deep neural network prediction model, and single-cycle damage data are output; the deep neural network prediction model is obtained by training based on the service condition data of known single-cycle damage data; the deep neural network prediction model includes an input layer, a hidden layer and an output layer connected in sequence; the single-cycle damage data corresponds to the value of fatigue damage and creep damage on any component part of the target high-temperature rotating equipment;
[0012] For any of the cycles, a weak point is determined according to the single cycle damage data; the weak point is a component of the target high-temperature rotating equipment corresponding to the maximum damage value obtained after summing and comparing the fatigue damage and creep damage in the single cycle damage data;
[0013] Based on the linear damage accumulation criterion, the cumulative damage data of each weak point is determined according to the number of cycles at the current moment and the single-cycle damage data; the cumulative damage data is used to realize real-time damage status sensing to ensure the service safety of the target high-temperature rotating equipment.
[0014] Optionally, data cleaning can be performed by removing outliers or filling in missing values.
[0015] Optionally, a median filter or a Kalman filter is used for filtering.
[0016] Optionally, the calculation formula for the cumulative damage data is:
[0017]
[0018] Among them, D N is the cumulative damage data; N is the number of cycles of the high-temperature rotating equipment at the current moment; i is the sequence number of the cycle; d i is the single-week injury data; X i Processing data for service conditions corresponding to the cycle times.
[0019] Optionally, physical sensors are used to acquire the service condition data of the target high-temperature rotating equipment in real time; the physical sensors include: a temperature sensor, a pressure sensor and a rotation speed sensor.
[0020] In the second aspect, the present application provides a high-temperature rotating equipment unmonitorable area weakness identification and virtual sensing system, including:
[0021] A data acquisition module is used to acquire the service condition data of the target high-temperature rotating equipment in real time; the service condition data includes the operating condition parameters under each cycle; the operating condition parameters include: initial rotor temperature, speed increase rate, load increase rate, initial main steam temperature and main steam temperature rise rate; each completion of the service process of the target high-temperature rotating equipment from start-up to steady state to shutdown is recorded as one cycle;
[0022] A preprocessing module, used for preprocessing the service condition data to obtain service condition processed data; the preprocessing includes: data cleaning, filtering and standardization;
[0023] A cycle number acquisition module is used to obtain the cycle number of the target high-temperature rotating equipment at any time;
[0024] A single-cycle damage data determination module is used for inputting the service condition processing data corresponding to any of the cycle cycles into a deep neural network prediction model, and outputting the single-cycle damage data; the deep neural network prediction model is obtained by training based on the service condition data of known single-cycle damage data; the deep neural network prediction model includes an input layer, a hidden layer and an output layer connected in sequence; the single-cycle damage data corresponds to the fatigue damage and creep damage values on any component part of the target high-temperature rotating equipment;
[0025] A weak point location determination module is used to determine the weak point location for any of the cycle times according to the single cycle damage data; the weak point location is the component part of the target high-temperature rotating equipment corresponding to the maximum damage value after summing fatigue damage and creep damage and comparing the values in the single cycle damage data;
[0026] The cumulative damage data determination module is used to determine the cumulative damage data of each weak point location based on the linear damage accumulation criterion, according to the current cycle number and the single cycle damage data; the cumulative damage data is used to realize real-time damage status sensing to ensure the service safety of the target high-temperature rotating equipment.
[0027] According to the specific embodiments provided in this application, this application discloses the following technical effects:
[0028] The present application provides a method and system for identifying weaknesses in unmonitored areas of high-temperature rotating equipment and for virtual sensing, which collects service condition data of high-temperature rotating equipment in real time, and performs real-time virtual sensing of damage status through data cleaning, filtering and standardization. The real-time virtual sensing method for damage to weak points of high-temperature rotating equipment of the present application realizes the information extension of existing physical sensing of high-temperature rotating equipment, and indirectly obtains key state characteristic parameters that determine the service life of high-temperature rotating equipment in the crack initiation stage, providing a technical solution for the health status assessment and situation-based maintenance management of high-temperature rotating equipment in the early stage of service. The present application realizes the extension of data collected by limited physical sensors, and realizes real-time damage status perception of high-temperature rotating equipment during its service without adding additional physical sensors. BRIEF DESCRIPTION OF THE DRAWINGS
[0029] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0030] Figure 1 This is a flow chart of a method for identifying weaknesses in an unmonitorable area of high-temperature rotating equipment and a virtual sensing method in one embodiment of the present application;
[0031] Figure 2 A schematic diagram of a design process of a virtual sensing method and a weakness identification method for an unmonitorable area of a high-temperature rotating equipment provided in an embodiment of the present application;
[0032] Figure 3 It is a schematic diagram of the finite element simulation cloud diagram and weak point identification results of a two-dimensional axisymmetric steam turbine rotor;
[0033] Figure 4 This is a schematic diagram of the main steam temperature changes during 50 startup-steady-state-shutdown cycles of the steam turbine after data processing;
[0034] Figure 5 This is a schematic diagram of the real-time damage virtual sensing results of the weak points of the turbine rotor. DETAILED DESCRIPTION
[0035] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions 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. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0036] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.
[0037] In an exemplary embodiment, Figure 1 As shown, the weakness identification and virtual sensing method of the unmonitorable area of high-temperature rotating equipment includes:
[0038] Step 100: Acquire the service condition data of the target high-temperature rotating equipment in real time. The service condition data includes the operating condition parameters under each cycle; the operating condition parameters include: initial rotor temperature, speed increase rate, load increase rate, initial main steam temperature and main steam temperature rise rate; each time the target high-temperature rotating equipment completes a service process of start-up-steady state-shutdown, it is recorded as a cycle.
[0039] Step 200: pre-process the service condition data to obtain service condition processed data. The pre-processing includes: data cleaning, filtering and standardization.
[0040] Step 300: Obtain the number of cycles of the target high-temperature rotating equipment at any time.
[0041] Step 400: For any cycle, the service condition processing data corresponding to the cycle is input into the deep neural network prediction model, and the single cycle damage data is output. The deep neural network prediction model is trained based on the service condition data of the known single cycle damage data; the deep neural network prediction model includes an input layer, a hidden layer and an output layer connected in sequence; the single cycle damage data corresponds to the fatigue damage and creep damage values on any component of the target high temperature rotating equipment.
[0042] Step 500: For any cycle, determine the weak point according to the single cycle damage data. The weak point is the component part of the target high temperature rotating equipment corresponding to the maximum damage value after summing and comparing the fatigue damage and creep damage in the single cycle damage data.
[0043] Step 600: Based on the linear damage accumulation criterion, the cumulative damage data of each weak point is determined according to the cycle and single cycle damage data at the current moment. The cumulative damage data is used to realize the sensing of the real-time damage state to ensure the service safety of the target high-temperature rotating equipment.
[0044] That is, at any moment, the processing operations of step 400 to step 600 are performed to finally determine the accumulated damage data at that moment.
[0045] As an optional implementation, a method of eliminating outliers or filling missing values is used to clean the data, and a median filter or a Kalman filter is used to perform filtering.
[0046] In one embodiment, the calculation formula for the cumulative damage data is:
[0047]
[0048] Among them, D N is the cumulative damage data; N is the number of cycles of the high-temperature rotating equipment at the current moment; i is the sequence number of the cycle; d i is the single-week injury data; X i Processing data for service conditions corresponding to the cycle times.
[0049] As an optional implementation, physical sensors are used to acquire the service condition data of the target high-temperature rotating equipment in real time; the physical sensors include: a temperature sensor, a pressure sensor and a rotation speed sensor.
[0050] Specifically, in practical applications, the design ideas of the method mentioned in this application include the following steps:
[0051] S1: Accumulate simulation data of high-temperature rotating equipment under different working conditions.
[0052] Different working conditions refer to different combinations of physical information parameters obtained by physical sensing during the service of high-temperature rotating equipment; the simulation data of high-temperature rotating equipment includes single-cycle damage data of high-temperature rotating equipment under different working conditions calculated based on elastic-plastic finite element analysis.
[0053] S2: Identify weak points of high-temperature rotating equipment.
[0054] Based on the simulation results of high-temperature rotating equipment, the part (unit) with the largest single-cycle damage value is selected as the weak point.
[0055] S3: Construct a weak point damage prediction model based on deep neural network.
[0056] Based on the accumulation of simulation data of the identified weak points, a deep neural network prediction model is established to complete the mapping from operating parameter input to single-cycle damage output.
[0057] S4: Real-time collection of service condition data of high-temperature rotating equipment.
[0058] The collected real-time service condition data of the high-temperature rotating equipment is the physical information parameter data obtained by physical sensing in step S1.
[0059] S5: Collect data cleaning, filtering and standardization.
[0060] The data cleaning methods used include but are not limited to removing outliers, filling in missing values, etc., and the data filtering methods include but are not limited to median filtering, Kalman filtering, etc.
[0061] S6: Real-time damage status virtual sensing.
[0062] Real-time damage status virtual sensing is achieved by inputting the processed real-time service condition data into the constructed deep neural network model to obtain the single-cycle damage value of the weak point of the high-temperature rotating equipment, and obtaining the real-time cumulative damage value of the weak point under the current cycle through the linear damage accumulation criterion. Under the linear damage accumulation criterion, the corresponding cumulative damage data expression is:
[0063]
[0064] Where D N is the cumulative damage value, that is, the cumulative damage value of the weak point of the high-temperature rotating equipment when the high-temperature rotating equipment is started-steady-state-shutdown cycle N times; d i is the single-cycle damage data, that is, the damage value of the weak point of the high-temperature rotating equipment during the i-th start-steady-state-shutdown cycle of the high-temperature rotating equipment; X i The service condition processing data corresponding to the cycle period is the operating condition parameter data after data cleaning, filtering and standardization during the i-th start-steady-state-shutdown cycle of the high-temperature rotating equipment.
[0065] The present application also provides an application scenario, which applies the above-mentioned weakness identification and virtual sensing methods for unmonitored areas of high-temperature rotating equipment. Specifically: The weakness identification and virtual sensing methods for unmonitored areas of high-temperature rotating equipment provided in this embodiment can be applied in the scenario of evaluating the health status of high-temperature rotating equipment. The scenario of evaluating the health status of high-temperature rotating equipment includes a service condition data acquisition link, a data processing link, and an evaluation link; the service condition data enters the data processing link from the service condition data acquisition link, and the corresponding cumulative damage data is obtained by means of a deep neural network prediction model and a linear damage accumulation criterion, and enters the downstream evaluation link. The weakness identification and virtual sensing methods for unmonitored areas of high-temperature rotating equipment provided in this embodiment belong to the data processing link.
[0066] Take the turbine rotor of a 300MW thermal power unit as an example. Figure 2 shown.
[0067] S1: Accumulate simulation data of high-temperature rotating equipment under different working conditions.
[0068] According to the historical operation data of the steam turbine, the variation range of parameters under different operating conditions is summarized, as shown in Table 1.
[0069] Table 1 Variation range of parameters under different working conditions
[0070] parameter Range of variation unit Initial rotor temperature 20-215 ℃ Initial main steam temperature 320-400 ℃ Main steam temperature rise rate 0.7-2 ℃ / min Ramp-up rate 20-100 <![CDATA[r / min 2 ]]> Load increase rate 0.2-0.8 % / min
[0071] Assuming that the parameters of different working conditions obey Gaussian distribution, 50 groups of input samples are extracted by Latin hypercube sampling, and the output samples of single-cycle damage under the corresponding input samples are obtained by elastic-plastic finite element analysis.
[0072] S2: Identify weak points of high-temperature rotating equipment.
[0073] The steam turbine rotor simulation is performed using ABAQUS software. The simulation analysis results of the steam turbine rotor under the mean operating parameters are as follows: Figure 3 As shown in the figure, according to the simulation results of the steam turbine rotor under the mean operating condition parameters, the part (unit) with the largest single-cycle damage value is selected as the weak point. Finally, the single-cycle damage value of the U-shaped groove part of the steam turbine rotor reaches the maximum and is identified as the weak point of the steam turbine rotor.
[0074] S3: Construct a weak point damage prediction model based on deep neural network.
[0075] In this embodiment, a deep neural network model is built through the MATLAB platform, which includes an input layer, three hidden layers and an output layer. The number of neurons in the input and output layers is determined according to the dimensions of the input and output samples, and the optimal number of hidden layer neurons is determined to be 10 through trial and error.
[0076] S4: Collect real-time service condition data of high-temperature rotating equipment.
[0077] S5: Collect data cleaning, filtering and standardization.
[0078] In this embodiment, the collected data is cleaned, filtered and standardized through the MATLAB platform. The main steam temperature change during the 50 start-up-steady-state-shutdown cycles of the steam turbine after data processing is shown in Figure 1. Figure 4 .
[0079] S6: Real-time damage status virtual sensing.
[0080] The processed real-time service condition data is input into the constructed deep neural network model to obtain the single-cycle damage value of the U-shaped groove of the turbine rotor, and the real-time cumulative damage value of the weak point after 50 cycles is obtained by the linear damage accumulation criterion, which can be expressed as:
[0081]
[0082] Where D 50It is the cumulative damage value of the U-shaped groove of the steam turbine rotor when the steam turbine is cycled 50 times from start-up to steady state and from shutdown; is the damage value of the weak point of the steam turbine rotor predicted by the deep neural network model during the i-th start-up-steady-state-shutdown cycle of the steam turbine; is the operating parameter data of the steam turbine during the i-th start-up-steady-state-shutdown cycle after data cleaning, filtering, and standardization. Finally, the damage accumulation trajectory of the weak points of the steam turbine rotor obtained by damage virtual sensing is as follows: Figure 5 As shown in the figure, it can be seen that after 50 start-stop cycles of the steam turbine, the rotor weak point has accumulated a certain degree of damage, among which the fatigue damage accumulation of this part is greater than the creep damage accumulation. The real-time damage virtual sensing results of the weak point of the steam turbine rotor will provide important technical support for the health status assessment and condition-based maintenance management of the steam turbine in the early crack initiation stage.
[0083] Based on the same inventive concept, the embodiment of the present application also provides a high-temperature rotating equipment unmonitored area weakness identification and virtual sensing system for realizing the above-mentioned high-temperature rotating equipment unmonitored area weakness identification and virtual sensing method. The implementation scheme for solving the problem provided by the system is similar to the implementation scheme recorded in the above-mentioned method, so the specific limitations in the embodiment of the high-temperature rotating equipment unmonitored area weakness identification and virtual sensing system provided below can refer to the limitations of the high-temperature rotating equipment unmonitored area weakness identification and virtual sensing method in the above text, and will not be repeated here.
[0084] In an exemplary embodiment, a high-temperature rotating equipment unmonitorable area weakness identification and virtual sensing system is provided, comprising:
[0085] The data acquisition module is used to acquire the service condition data of the target high-temperature rotating equipment in real time; the service condition data includes the operating condition parameters under each cycle; the operating condition parameters include: initial rotor temperature, speed increase rate, load increase rate, initial main steam temperature and main steam temperature rise rate; each completion of the service process of start-up-steady state-shutdown of the target high-temperature rotating equipment is recorded as a cycle.
[0086] The preprocessing module is used to preprocess the service condition data to obtain service condition processed data; the preprocessing includes: data cleaning, filtering and standardization.
[0087] The cycle number acquisition module is used to obtain the cycle number of the target high-temperature rotating equipment at any time.
[0088] A single-cycle damage data determination module is used to input the service condition processing data corresponding to any of the cycle cycles into a deep neural network prediction model, and output single-cycle damage data; the deep neural network prediction model is trained based on the service condition data of known single-cycle damage data; the deep neural network prediction model includes an input layer, a hidden layer and an output layer connected in sequence; the single-cycle damage data corresponds to the numerical value of fatigue damage and creep damage on any component part of the target high-temperature rotating equipment.
[0089] A weak point location determination module is used to determine the weak point location for any of the cycle times according to the single cycle damage data; the weak point location is the component part of the target high-temperature rotating equipment corresponding to the maximum damage value after summing and comparing the fatigue damage and creep damage in the single cycle damage data.
[0090] The cumulative damage data determination module is used to determine the cumulative damage data of each weak point location based on the linear damage accumulation criterion, according to the current cycle number and the single cycle damage data; the cumulative damage data is used to realize real-time damage status sensing to ensure the service safety of the target high-temperature rotating equipment.
[0091] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0092] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.
Claims
1. A method for identifying weaknesses in unmonitorable areas of high-temperature rotating equipment and a virtual sensing method, characterized in that: The method for identifying weaknesses in unmonitorable areas of high-temperature rotating equipment and for virtual sensing includes: Acquire the service condition data of the target high-temperature rotating equipment in real time; the service condition data includes the condition parameters under each cycle; the condition parameters include: initial rotor temperature, speed-up rate, load-up rate, initial main steam temperature and main steam temperature rise rate; each completion of the service process of the target high-temperature rotating equipment from start-up to steady state to shutdown is recorded as one cycle; Preprocessing the service condition data to obtain service condition processed data; the preprocessing includes: data cleaning, filtering and standardization; Obtain the number of cycles of the target high-temperature rotating equipment at any time; For any of the cycle times, the service condition processing data corresponding to the cycle times are input into the deep neural network prediction model, and single-cycle damage data are output; the deep neural network prediction model is obtained by training based on the service condition data of known single-cycle damage data; the deep neural network prediction model includes an input layer, a hidden layer and an output layer connected in sequence; the single-cycle damage data corresponds to the value of fatigue damage and creep damage on any component part of the target high-temperature rotating equipment; For any of the cycles, a weak point is determined according to the single cycle damage data; the weak point is a component of the target high-temperature rotating equipment corresponding to the maximum damage value obtained after summing fatigue damage and creep damage and comparing the values in the single cycle damage data; Based on the linear damage accumulation criterion, the cumulative damage data of each weak point is determined according to the current cycle number and the single cycle damage data; the cumulative damage data is used to realize real-time damage status sensing to ensure the service safety of the target high-temperature rotating equipment.
2. The method for identifying weaknesses and virtual sensing in unmonitorable areas of high-temperature rotating equipment according to claim 1 is characterized in that: Data cleaning is performed by removing outliers or filling in missing values.
3. The method for identifying weaknesses and virtual sensing in unmonitorable areas of high-temperature rotating equipment according to claim 1 is characterized in that: The median filter or Kalman filter method is used for filtering.
4. The method for identifying weaknesses and virtual sensing in unmonitorable areas of high-temperature rotating equipment according to claim 1 is characterized in that: The calculation formula for the cumulative damage data is: Among them, D N is the cumulative damage data; N is the number of cycles of the high-temperature rotating equipment at the current moment; i is the sequence number of the cycle; d i is the single-week injury data; X i Processing data for service conditions corresponding to the cycle times.
5. The method for identifying weaknesses and virtual sensing in unmonitorable areas of high-temperature rotating equipment according to claim 1 is characterized in that: Physical sensors are used to obtain the service condition data of the target high-temperature rotating equipment in real time; the physical sensors include: temperature sensors, pressure sensors and rotation speed sensors.
6. Weakness identification and virtual sensing system for unmonitored areas of high-temperature rotating equipment, characterized in that: The high-temperature rotating equipment unmonitorable area weakness identification and virtual sensing system includes: A data acquisition module is used to acquire the service condition data of the target high-temperature rotating equipment in real time; the service condition data includes the operating condition parameters under each cycle; the operating condition parameters include: initial rotor temperature, speed increase rate, load increase rate, initial main steam temperature and main steam temperature rise rate; each completion of the service process of the target high-temperature rotating equipment from start-up to steady state to shutdown is recorded as one cycle; A preprocessing module, used for preprocessing the service condition data to obtain service condition processed data; the preprocessing includes: data cleaning, filtering and standardization; A cycle number acquisition module is used to obtain the cycle number of the target high-temperature rotating equipment at any time; A single-cycle damage data determination module is used for inputting the service condition processing data corresponding to any of the cycle cycles into a deep neural network prediction model, and outputting the single-cycle damage data; the deep neural network prediction model is obtained by training based on the service condition data of known single-cycle damage data; the deep neural network prediction model includes an input layer, a hidden layer and an output layer connected in sequence; the single-cycle damage data corresponds to the fatigue damage and creep damage values on any component part of the target high-temperature rotating equipment; A weak point location determination module is used to determine the weak point location for any of the cycle times according to the single cycle damage data; the weak point location is the component part of the target high-temperature rotating equipment corresponding to the maximum damage value after summing fatigue damage and creep damage and comparing the values in the single cycle damage data; The cumulative damage data determination module is used to determine the cumulative damage data of each weak point location based on the linear damage accumulation criterion, according to the current cycle number and the single cycle damage data; the cumulative damage data is used to realize real-time damage status sensing to ensure the service safety of the target high-temperature rotating equipment.
Citation Information
Patent Citations
Creep-fatigue-oxidation real-time damage accumulation evaluation method for material
CN111879636A
Constant velocity universal joint raceway damage calculation method based on neural network
CN116257938A
Steam turbine rotor monitoring method and system and storage medium
CN118167447A
