Automatic diagnosis method for vibration fault of steam turbine unit under all working conditions and computer readable medium

Abstract

The present invention relates to a method for automatic diagnosis of vibration faults of a steam turbine unit under all working conditions and a computer-readable medium, wherein the method for automatic diagnosis of faults comprises: step 1: establishing a database of all working conditions; step 2: real-time monitoring of vibration data and operating parameter data of the steam turbine unit during operation; step 3: if the difference between the vibration data and the data in the database of all working conditions exceeds a preset threshold, then executing step 4; step 4: sequentially determining the position where the vibration abnormality occurs, the state of the steam turbine unit when the vibration abnormality occurs, and the characteristic parameters of the position under the current state; step 5: according to the characteristic parameters determined in step 4, comparing the parameter values ​​collected in real time with the parameter values ​​in the database of all working conditions to obtain the fault diagnosis result. Compared with the prior art, the present invention has the advantages of being able to effectively realize real-time monitoring of vibration parameters and operating parameters of the steam turbine unit under all working conditions, and combining the operating parameters for joint diagnosis when a vibration fault occurs.

Description

Technical Field

[0001] The invention relates to the technical field of steam turbine unit fault diagnosis, and in particular to a method for automatic diagnosis of full-operating-condition vibration faults of steam turbine units and a computer-readable medium. Background Art

[0002] Steam turbine is one of the most important equipment in power generation enterprises. Whether it can operate safely and stably affects the safety and economy of the entire generator set. For steam turbine, vibration failure is the most important equipment failure. Being able to handle steam turbine vibration failure in a timely and reasonable manner is an important guarantee for the safe and stable production of power plants.

[0003] At present, some power plants only analyze the vibration failure of steam turbines by focusing on the vibration failure itself, while ignoring the deep-seated causes of the vibration. For example, the most common vibration failure component of steam turbines is the 1-fold frequency component, which indicates that there is a mass imbalance problem in the steam turbine. The most direct way to solve the mass imbalance problem is dynamic balancing. Therefore, some power plants directly perform dynamic balancing without specifically analyzing the causes of the mass imbalance. The temporary results may be ideal, but such problems are likely to occur again later.

[0004] As for the storage and management of vibration information, since the steam turbine transmits vibration signals all the time, the amount of information is large, and the vibration data in the power plant monitoring and control equipment will be automatically deleted every fixed period, so the vibration data and other operating parameters of the unit during the run-up, constant speed, load process and various working conditions cannot be well saved, which is not conducive to the later vibration analysis. Even if some units have saved data for a long time, they have not been sorted out and summarized and used, resulting in a waste of resources and not conducive to the analysis of vibration faults.

[0005] Therefore, there is an urgent need for a vibration fault analysis method. This vibration fault diagnosis method analyzes the deep-seated causes behind the vibration based on the established database under various working conditions, solves the fundamental problems, and improves the economy, safety, and management level of the power plant. It realizes the integration of rotating machinery vibration supervision, fault diagnosis, and information management, and improves the comprehensive management level of rotating equipment in power plants. Summary of the invention

[0006] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and to provide a method for automatic diagnosis of vibration faults of steam turbine units under all operating conditions and a computer-readable medium that can effectively realize real-time monitoring of vibration faults of steam turbine units under all operating conditions.

[0007] The purpose of the present invention can be achieved by the following technical solutions:

[0008] A method for automatically diagnosing vibration faults of a steam turbine unit under all working conditions, the method comprising:

[0009] Step 1: Establish a full working condition database;

[0010] Step 2: Real-time monitoring of vibration data and operating parameter data of the steam turbine unit during operation;

[0011] Step 3: If the difference between the vibration data and the data in the full working condition database exceeds the preset threshold, execute step 4;

[0012] Step 4: sequentially determine the location where the abnormal vibration occurs, the state of the steam turbine unit when the abnormal vibration occurs, and the characteristic parameters of the location in the current state;

[0013] Step 5: According to the characteristic parameters determined in step 4, the parameter values ​​collected in real time are compared with the parameter values ​​in the full working condition database to obtain the fault diagnosis results.

[0014] Preferably, the full operating condition database includes normal vibration data and normal range values ​​of operating parameters of the steam turbine unit in the start-up and shutdown states, constant speed state and load state.

[0015] More preferably, the start-stop state includes a warm-up state and an over-critical state; the constant speed state is specifically a constant speed state of 3000 r / min; and the load state includes a load state at several ambient temperatures.

[0016] Preferably, the step 3 further comprises:

[0017] If the difference between the vibration data and the data in the full working condition database does not exceed the preset threshold, and the operating parameter data is not within the normal range, an alarm will be issued.

[0018] Preferably, the step 4 is specifically:

[0019] Abnormal vibration occurs at locations including high and medium pressure cylinders, low pressure cylinders and generators;

[0020] The status of the steam turbine unit when abnormal vibration occurs includes the unit start-up and shutdown status, constant speed status and load status;

[0021] The characteristic parameters include start-stop characteristic parameters, high and medium pressure cylinder rotor characteristic parameters after load, low pressure rotor characteristic parameters after load and generator rotor characteristic parameters.

[0022] More preferably, the start-stop characteristic parameters include cylinder temperature, expansion difference, high-pressure cylinder expansion, shaft seal temperature and vacuum; the high and medium pressure cylinder rotor characteristic parameters after load include gap voltage, main steam temperature, pressure, flow, valve opening and bearing temperature; the low-pressure rotor characteristic parameters after load include vacuum, shaft seal temperature and bearing temperature; the generator rotor characteristic parameters include sealing bearing oil temperature, hydrogen pressure, hydrogen temperature, excitation current and excitation voltage.

[0023] Preferably, the step 5 is specifically:

[0024] Step 5-1: Determine whether the deviation between each characteristic parameter and a preset threshold is greater than a preset deviation threshold. If so, determine the cause of the fault directly according to the fault parameter. Otherwise, execute step 5-2.

[0025] Step 5-2: Calculate the correlation coefficient matrix between vibration and characteristic parameters, and determine whether the correlation coefficient exceeds the standard. If so, directly determine the cause of the fault based on the correlation coefficient. Otherwise, execute step 5-3.

[0026] Step 5-3: Determine whether the vibration is parameter-independent vibration, and judge whether the vibration is periodic vibration. If so, diagnose it as a random friction fault, otherwise, execute step 5-4;

[0027] Step 5-4: Determine that the rotor has an internal defect and perform manual diagnosis.

[0028] More preferably, the correlation coefficient is specifically the Pearson correlation coefficient.

[0029] More preferably, the calculation method of the Pearson correlation coefficient is:

[0030]

[0031]

[0032] Among them, Y i X is the variable data within the analysis period; i ={Δt,2Δt,3Δt,...,n·Δt}, Y i ={y1,y2,y3,...,y n}, Δt is the sampling interval; μ X is the average value of the process quantity during the analysis period, σ(X) is the fluctuation range of the process quantity during the analysis period; μ Y is the average value of the vibration during the analysis period, σ(Y) is the fluctuation amplitude of the vibration during the analysis period;

[0033]

[0034]

[0035]

[0036]

[0037] r xy ' is the correlation coefficient between vibration and characteristic parameters, r xy is the correlation between vibration and process parameters, calculated as:

[0038] r xy =2|r' xy |-1

[0039] τ is a time delay parameter, which is used to reflect the time difference between the changing trends of two variables, τ≤n-2.

[0040] A computer-readable medium stores the automatic diagnosis method for vibration faults of a steam turbine unit under all operating conditions as described in any one of the above items.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] Effectively realize real-time monitoring of vibration faults of steam turbine units under all operating conditions: The automatic vibration fault diagnosis method in the present invention is based on a parameter-fused vibration fault analysis method for steam turbine units under all operating conditions, and establishes various databases including run-up, constant speed, load processes and various operating conditions, recording the normal operating range of various parameters of the unit under different states; the vibration data and operating parameters of the unit can be monitored in real time, and for abnormal vibrations, the cause of the fault can be determined based on the full operating condition database according to characteristic parameters and correlation coefficients, and the cause of the fault can be deeply excavated, effectively realizing real-time monitoring of vibration parameters and operating parameters of steam turbine machines under all operating conditions, and when a vibration fault occurs, combined diagnosis is made in combination with the operating parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a schematic flow chart of the method for automatically diagnosing vibration faults of steam turbine units under all operating conditions in the present invention. DETAILED DESCRIPTION

[0044] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0045] A method for automatic diagnosis of vibration faults of steam turbine units under all working conditions, the process is as follows: Figure 1 As shown, including:

[0046] Step 1: Establish a full working condition database;

[0047] The full operating condition database includes normal vibration data and normal range values ​​of operating parameters of the steam turbine unit in the start-up and shutdown states, constant speed states and load states; the start-up and shutdown states include warm-up state and over-critical state; the constant speed state is specifically a constant speed state of 3000r / min; the load state includes load states at several ambient temperatures;

[0048] Step 2: Real-time monitoring of vibration data and operating parameter data of the steam turbine unit during operation;

[0049] Step 3: If the difference between the vibration data and the data in the full working condition database exceeds the preset threshold, execute step 4; if the difference between the vibration data and the data in the full working condition database does not exceed the preset threshold, and the operating parameter data is not within the normal range, an alarm is issued;

[0050] In this embodiment, an interval band is defined within 10% of the normal operating range of each parameter to determine whether the operating parameter deviates from the normal operating interval band. If it deviates, an alarm prompt is given, otherwise, each parameter value is displayed in real time;

[0051] Step 4: sequentially determine the location where the abnormal vibration occurs, the state of the steam turbine unit when the abnormal vibration occurs, and the characteristic parameters of the location in the current state;

[0052] Abnormal vibration occurs at locations including high and medium pressure cylinders, low pressure cylinders and generators;

[0053] The status of the steam turbine unit when abnormal vibration occurs includes the unit start-up and shutdown status, constant speed status and load status;

[0054] The characteristic parameters include start-stop characteristic parameters, high and medium pressure cylinder rotor characteristic parameters after load, low pressure rotor characteristic parameters after load and generator rotor characteristic parameters;

[0055] The start-stop characteristic parameters include cylinder temperature, expansion difference, high-pressure cylinder expansion, shaft seal temperature and vacuum; the high- and medium-pressure cylinder rotor characteristic parameters after load include gap voltage, main steam temperature, pressure, flow, valve opening and bearing temperature; the low-pressure rotor characteristic parameters after load include vacuum, shaft seal temperature and bearing temperature; the generator rotor characteristic parameters include sealing bearing oil temperature, hydrogen pressure, hydrogen temperature, excitation current and excitation voltage;

[0056] Step 5: According to the characteristic parameters determined in step 4, the parameter values ​​collected in real time are compared with the parameter values ​​in the full working condition database to obtain the fault diagnosis results;

[0057] Step 5-1: Determine whether the deviation between each characteristic parameter and a preset threshold is greater than a preset deviation threshold. If so, determine the cause of the fault directly according to the fault parameter. Otherwise, execute step 5-2.

[0058] This embodiment determines whether the characteristic parameter deviates from the threshold value recorded in the database by more than 20%. If it exceeds 20%, the root cause of the fault can be determined based on the characteristic parameter. Otherwise, execute step 5-2;

[0059] Step 5-2: Calculate the correlation coefficient matrix between vibration and characteristic parameters, and determine whether the correlation coefficient exceeds the standard. If so, directly determine the cause of the fault based on the correlation coefficient. Otherwise, execute step 5-3.

[0060] Step 5-3: Determine whether the vibration is parameter-independent vibration, and judge whether the vibration is periodic vibration. If so, diagnose it as a random friction fault, otherwise, execute step 5-4;

[0061] Step 5-4: Determine that the rotor has an internal defect and perform manual diagnosis.

[0062] The correlation coefficient in this embodiment is specifically the Pearson correlation coefficient, and the calculation method is:

[0063] The variable data of the analysis period is Y i , the sampling interval is Δt (in s);

[0064] Note X i ={Δt,2Δt,3Δt,...,n·Δt}, Y i ={y1,y2,y3,...,y n};

[0065] The calculation method of Pearson correlation coefficient is:

[0066]

[0067]

[0068] μ X is the average value of the process quantity during the analysis period; σ(X) is the fluctuation range of the process quantity during the analysis period; μ Y is the average value of vibration in a period of time; σ(Y) is the fluctuation amplitude of vibration in the analysis period, and the calculation methods are:

[0069]

[0070]

[0071]

[0072]

[0073] r xy ' is the correlation coefficient between vibration and characteristic parameters. The correlation characteristic between vibration and characteristic parameters is defined as:

[0074] r xy =2|r' xy |-1

[0075] When performing correlation analysis on vibration data and process parameters, it is necessary to consider both the correlation coefficient and the time delay between the two, otherwise it will cause a large error in the extraction of correlation features. Taking the excitation current as an example, when the excitation current changes, the vibration changes rapidly, and there is no lag between the two, indicating that the vibration is mainly caused by electrical defects; if the vibration changes gradually after the excitation current changes, and it takes a period of time to stabilize, the vibration change lags behind the excitation current, indicating that the vibration is mainly caused by thermal deformation.

[0076] In order to calculate the correlation coefficient between the vibration signal and the process parameter and reflect the lead or lag phenomenon between the two when conducting correlation analysis between the vibration signal and the process parameter, the traditional correlation coefficient calculation formula is improved by adding a delay. The calculation method of the correlation coefficient between the two variables with time lag (lead) is as follows:

[0077]

[0078]

[0079] Among them, xy ' is the correlation coefficient of the inter-band lag between the vibration and process parameters. τ is the time delay parameter, which is used to reflect the time difference between the changing trends of the two variables, τ≤n-2.

[0080] In general, the independent variable changes first, and the dependent variable changes after no lag or a period of lag. The maximum lag time is generally 60 minutes. In correlation analysis, the relevant features in the lag time interval of 0-60 minutes are calculated respectively, and the maximum value is taken as the final relevant feature, and the lag time corresponding to the maximum correlation coefficient can be obtained at the same time.

[0081] The present embodiment also relates to a computer-readable medium, in which any one of the above-mentioned methods for automatic diagnosis of vibration faults of a steam turbine unit under all operating conditions is stored.

[0082] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for automatic diagnosis of vibration faults of steam turbine units under all operating conditions, characterized in that: The automatic fault diagnosis method comprises: Step 1: Establish a full working condition database; Step 2: Real-time monitoring of vibration data and operating parameter data of the steam turbine unit during operation; Step 3: If the difference between the vibration data and the data in the full working condition database exceeds the preset threshold, execute step 4; Step 4: sequentially determine the location where the abnormal vibration occurs, the state of the steam turbine unit when the abnormal vibration occurs, and the characteristic parameters of the location in the current state; Step 5: According to the characteristic parameters determined in step 4, the parameter values ​​collected in real time are compared with the parameter values ​​in the full working condition database to obtain the fault diagnosis results; The step 5 is specifically as follows: Step 5-1: Determine whether the deviation between each characteristic parameter and a preset threshold is greater than a preset deviation threshold. If so, determine the cause of the fault directly according to the fault parameter. Otherwise, execute step 5-2. Step 5-2: Calculate the correlation coefficient matrix between vibration and characteristic parameters, and determine whether the correlation coefficient exceeds the standard. If so, directly determine the cause of the fault based on the correlation coefficient. Otherwise, execute step 5-3. Step 5-3: Determine whether the vibration is parameter-independent vibration, and judge whether the vibration is periodic vibration. If so, diagnose it as a random friction fault, otherwise, execute step 5-4; Step 5-4: Determine that it is an internal defect of the rotor and perform manual diagnosis; The correlation coefficient is specifically the Pearson correlation coefficient; The calculation method of the Pearson correlation coefficient is: in, To analyze the variable data within the time period; , , is the sampling interval; is the average value of the process quantity during the analysis period, The fluctuation range of the process quantity during the analysis period; is the average value of the vibration during the analysis period, The fluctuation amplitude of vibration during the analysis period; is the correlation coefficient between vibration and characteristic parameters, is the correlation between vibration and characteristic parameters, calculated as: It is a time delay parameter, which is used to reflect the time difference between the vibration and characteristic parameter change trends. .

2. The method for automatic diagnosis of vibration faults of steam turbine units under all operating conditions according to claim 1, characterized in that: The full operating condition database includes normal vibration data and normal range values ​​of operating parameters of the steam turbine unit in the start-up and shutdown states, the constant speed state and the load state.

3. The method for automatic diagnosis of vibration faults of steam turbine units under all operating conditions according to claim 2, characterized in that: The start-stop state includes a warm-up state and an over-critical state; the constant speed state is specifically a constant speed state of 3000r / min; and the load state includes a load state under several ambient temperatures.

4. The method for automatic diagnosis of vibration faults of steam turbine units under all operating conditions according to claim 1, characterized in that: The step 3 also includes: If the difference between the vibration data and the data in the full working condition database does not exceed the preset threshold, and the operating parameter data is not within the normal range, an alarm will be issued.

5. The method for automatic diagnosis of vibration faults of steam turbine units under all operating conditions according to claim 1, characterized in that: The step 4 is specifically as follows: Abnormal vibration occurs at locations including high and medium pressure cylinders, low pressure cylinders and generators; The status of the steam turbine unit when abnormal vibration occurs includes the unit start-up and shutdown status, constant speed status and load status; The characteristic parameters include start-stop characteristic parameters, high and medium pressure cylinder rotor characteristic parameters after load, low pressure rotor characteristic parameters after load and generator rotor characteristic parameters.

6. The method for automatic diagnosis of vibration faults of steam turbine units under all operating conditions according to claim 5, characterized in that: The start-stop characteristic parameters include cylinder temperature, expansion difference, high-pressure cylinder expansion, shaft seal temperature and vacuum; the high- and medium-pressure cylinder rotor characteristic parameters after load include gap voltage, main steam temperature, pressure, flow, valve opening and bearing temperature; the low-pressure rotor characteristic parameters after load include vacuum, shaft seal temperature and bearing temperature; the generator rotor characteristic parameters include sealing bearing oil temperature, hydrogen pressure, hydrogen temperature, excitation current and excitation voltage.

7. A computer-readable medium, characterized in that The computer-readable medium stores the automatic diagnosis method for vibration faults of a steam turbine unit under all operating conditions as described in any one of claims 1 to 6.

Images