An intelligent early warning method for turbine top cover failure

Abstract

The present invention belongs to the field of intelligent operation and maintenance of hydro-generator sets. Aiming at the fault of the turbine top cover, an intelligent early warning method for the fault of the turbine top cover is proposed to achieve the purpose of predicting the fault of the top cover in advance. By arranging two swing measuring points in the X and Y directions of the water guide bearing of the turbine, two vibration measuring points in the horizontal X and Y directions of the top cover, one vibration measuring point in the vertical Z direction of the top cover, and a pressure pulsation measuring point in the bladeless area, the signals of four physical quantities, namely, the swing of the water guide bearing, the horizontal vibration of the top cover, the vertical vibration of the top cover, and the pressure pulsation of the bladeless area of ​​the top cover, are obtained. After being analyzed and processed by the intelligent fault early warning platform, accurate and efficient early warning of the top cover fault is achieved.

Description

Technical Field

[0001] The invention belongs to the field of intelligent operation and maintenance of hydraulic machinery, and particularly proposes an intelligent fault early warning method for a turbine top cover component. Background Art

[0002] Whether the abnormal operation of hydropower units can be discovered in time and the unit failure can be predicted in advance is of great significance to maintaining the safety and stability of hydropower stations and pumped storage power stations. At the same time, the development of intelligent unit fault warning and the gradual implementation of condition-based maintenance to improve the operating stability of hydropower units are also the general trend of the hydropower station industry. Based on the structural characteristics of the unit, data analysis based on the fusion of multiple monitoring quantities can realize fault warning of unit components, providing an effective means to further improve the intelligence level of hydropower units.

[0003] At present, there have been many research results in the field of intelligent operation and maintenance of hydropower units. The Chinese patent with announcement number CN202110462387.2 introduces a method for early warning of the operating data trend of hydropower units, including collecting the operating power data of the hydropower units, predicting the power trend based on the historical power data, setting the low power warning value, etc. This patent can realize the abnormal early warning of the operating power and perform trend prediction; the invention patent with announcement number CN201810573391.4 publishes a five-in-one hydraulic machinery fault diagnosis method. This patent obtains the signals of five physical quantities of flow, water pressure, current, vibration, and temperature at the upper frame, lower frame, guide bearing, volute, and tailwater pipe of the hydraulic machinery, and uses the DS evidence theory to realize efficient judgment of turbine faults. This patent realizes a fault diagnosis method for hydraulic machinery based on multi-physical quantity fusion probability analysis.

[0004] Most of the current hydraulic machinery fault diagnosis systems set alarm values ​​for single or multiple monitored physical quantity signals to achieve the function of conventional fault alarm. Relevant research on unique fault warnings for specific components of hydropower units is relatively rare. In order to achieve the fault warning of the turbine top cover, this application, based on an in-depth analysis of the specific characteristics of the specific fault of the top cover, selects relevant monitored physical quantity signals according to its symptom characteristics, sets fault warning setting values, and other steps to achieve the goal of top cover fault warning. Summary of the invention

[0005] In order to realize the fault warning function of hydraulic machinery, this application proposes an intelligent warning method for turbine top cover fault. According to the characteristics of turbine top cover fault, by monitoring four physical quantity signals, namely water guide swing, top cover horizontal vibration, top cover vertical vibration, and top cover bladeless area pressure pulsation, the purpose of turbine top cover fault warning is achieved by integrating turbine structural mechanism and logical reasoning, and the specific fault category can be located.

[0006] The technical solution adopted by the present invention is:

[0007] Step 1: At the water-guided bearing, an eddy current sensor is arranged in the X and Y directions to measure the displacement and obtain the swing signal; an acceleration sensor is arranged in the horizontal X and Y directions and the vertical Z direction of the top cover to measure the acceleration and obtain the vibration signal; a piezoelectric sensor is arranged above the bladeless area of ​​the top cover to obtain the pressure pulsation signal of the bladeless area of ​​the top cover, and the monitoring data number of each time node and the data information of each time node are uploaded to the database of the early warning system;

[0008] Step 2: Take the historical data of the last 30 days for the swing values ​​of the water guide in the X and Y directions and calculate the average value, which is used as the swing setting value of the water guide in the X and Y directions. The algorithm formula for calculating the swing setting value is:

[0009]

[0010] Where N represents the sampling times of the water conductance swing signal value within 30 days, X si , Y si They represent the i-th sampling value of the water guide X-direction swing and the water guide Y-direction swing, respectively. s ,Y s They are the set values ​​of water guide swing in X and Y directions respectively;

[0011] Step 3: Take the historical data of the latest 30 days for the acceleration values ​​of the top cover horizontal vibration signal in the X and Y directions and calculate the average value respectively, which is used as the setting value of the top cover horizontal vibration signal in the X and Y directions. The algorithm formula for calculating the acceleration setting value of the top cover horizontal vibration signal is:

[0012]

[0013] Where N is the number of times the acceleration value of the top cover horizontal vibration signal is sampled within 30 days, X DSi , Y DSi Respectively represent the i-th sampling value of the acceleration value of the top cover horizontal vibration signal in the X direction and the Y direction; DS , Y DS They represent the acceleration setting values ​​of the vibration signal in the horizontal X and Y directions of the top cover respectively;

[0014] Step 4: Take the historical data of the last 30 days for the vertical vibration acceleration value of the top cover in the Z direction and calculate the average value as the Z direction setting value of the vertical vibration signal of the top cover. The algorithm formula for calculating the acceleration setting value of the vertical vibration signal of the top cover is:

[0015]

[0016] Where N is the number of times the acceleration value of the vertical vibration signal of the roof is sampled within 30 days, and Z is DsiRepresents the i-th sampling value of the acceleration value of the vertical vibration signal of the top cover in the Z direction, Z Ds It is the acceleration setting value of the vertical vibration signal of the top cover;

[0017] Step 5: Take the historical data of the last 30 days for the pressure pulsation monitoring value of the top cover without leaves and calculate the average value. The algorithm formula for calculating the top cover pressure pulsation setting value is:

[0018]

[0019] Where N is the number of times the acceleration value of the vertical vibration signal of the roof is sampled within 30 days, P Dsi represents the i-th sampling value of the acceleration value of the vertical vibration signal of the top cover in the Z direction, P Ds It is the top cover pressure pulsation setting value;

[0020] Step 6: Define event A as the comparison between the swing value of the water guide in the X and Y directions and the set value in the X and Y directions: when the swing value of the water guide in the X direction is continuously greater than the set value in the X direction by more than 10% for 6 hours, and the swing value of the water guide in the Y direction is continuously greater than the set value in the Y direction by more than 10% for 6 hours, event A is TRUE, otherwise event A is FALSE;

[0021] Step 7: Define event B as the comparison between the acceleration values ​​of the top cover horizontal vibration in the X and Y directions and the set values ​​in the X and Y directions: when the acceleration value of the top cover horizontal vibration in the X direction is continuously greater than the set value in the X direction by more than 10% for 6 hours, and the acceleration value of the top cover horizontal vibration in the Y direction is continuously greater than the set value in the Y direction by more than 10% for 6 hours, event B is TRUE, otherwise event B is FALSE;

[0022] Step 8: Define event C as the comparison between the vertical vibration acceleration value of the top cover and the set value: when the vertical vibration acceleration value of the top cover is continuously greater than the vertical setting value by more than 10% for 6 hours, event C is TRUE, otherwise event C is FALSE;

[0023] Step 9: Define event D as the comparison between the pressure pulsation monitoring value of the top cover without leaves and the pressure pulsation setting value: when the pressure pulsation monitoring value of the top cover without leaves is greater than the pressure pulsation setting value, event D is TRUE, otherwise event D is FALSE;

[0024] Step 10: When event B is true, a fault warning of excessive hydraulic unbalance force is issued; when events A and B are both true, a fault warning of loose top cover handle and bolts is issued; when event C is true, a fault warning of insufficient top cover axial stiffness is issued; when event D is true, a fault warning of static and dynamic interference of top cover is issued;

[0025] Step 11: Using MySQL database as a platform, establish an equipment information management database to save historical fault warning information and related data, conduct statistical analysis on the warning records, fault records, fault types, and maintenance plans of the turbine, and use the attribute parameters of different faults as index identification numbers and store them in the equipment management information database;

[0026] Step 12: When a fault warning occurs, use the unit downtime for maintenance to repair the top cover related components. The maintenance process includes:

[0027] When a fault warning of excessive hydraulic unbalance force occurs, stop the machine to check whether the natural air supply system is working normally, and repair the natural air supply system; when a fault warning of loose top cover handles and bolts occurs, stop the machine to check whether the top cover handles and bolts are loose, and tighten all top cover handles and bolts; when a fault warning of insufficient axial stiffness of the top cover occurs, stop the machine to check whether the top cover structure is damaged or worn, and weld and reinforce the worn parts and parts prone to wear; when a fault warning of dynamic and static interference of the top cover occurs, stop the machine to check the flow components of the impeller, guide vanes, and wear-resistant plates, and make slight adjustments to the clearance of the flow components.

[0028] In the above-mentioned intelligent early warning method for turbine top cover fault, in the step 2, the water guide X direction and Y direction swing setting values ​​are respectively recorded as X s ,Y s ,but

[0029]

[0030] Where N represents the number of sampling times of the water conductance swing signal value in one month, X si , Y si They represent the i-th sampling value of the water guide X-direction swing and the water guide Y-direction swing, respectively. s ,Y s They are the set values ​​of the water guide swing in the X and Y directions respectively.

[0031] In the above-mentioned intelligent early warning method for turbine top cover fault, in step 3, the acceleration setting values ​​of the top cover horizontal X and Y directions of vibration signal are recorded as X DS , Y DS ,but:

[0032]

[0033]

[0034] Where N is the number of times the acceleration value of the top cover horizontal vibration signal is sampled within one month, and X DSi , Y DSiRespectively represent the i-th sampling value of the acceleration value of the top cover horizontal vibration signal in the X direction and the Y direction; DS , Y DS They represent the acceleration setting values ​​of the vibration signal in the horizontal X and Y directions of the top cover respectively.

[0035] In the above-mentioned intelligent early warning method for turbine top cover fault, in step 4, the acceleration setting value of the top cover vertical vibration signal is recorded as Z Ds ,but:

[0036]

[0037] Where N is the number of times the acceleration value of the vertical vibration signal of the roof is sampled within one month, and Z is Dsi Represents the i-th sampling value of the acceleration value of the vertical vibration signal of the top cover in the Z direction, Z Ds is the acceleration value of the vertical vibration signal of the top cover.

[0038] In the above-mentioned intelligent early warning method for turbine top cover fault, in step 5, the top cover pressure pulsation setting value is recorded as P Ds ,but:

[0039]

[0040] Where N is the number of times the acceleration value of the vertical vibration signal of the top cover is sampled within one month, and P Dsi represents the i-th sampling value of the acceleration value of the vertical vibration signal of the top cover in the Z direction, P Ds It is the setting value of top cover pressure pulsation.

[0041] In the above-mentioned intelligent early warning method for turbine top cover failure, in step 12, the contents of the inspection and treatment measures are as follows:

[0042] When a fault warning of excessive hydraulic unbalance force occurs, stop the machine to check whether the natural air supply system is working normally, and repair the natural air supply system; when a fault warning of loose top cover handles and bolts occurs, stop the machine to check whether the top cover handles and bolts are loose, and tighten all top cover handles and bolts; when a fault warning of insufficient axial stiffness of the top cover occurs, stop the machine to check whether the top cover structure is damaged or worn, and weld and reinforce the worn parts and parts prone to wear; when a fault warning of excessive dynamic and static interference of the top cover occurs, stop the machine to check the flow components of the impeller, guide vanes, and wear-resistant plates, and make slight adjustments to the gaps of the flow components by grinding and repair welding.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] 1. The installation method of the measuring point sensor of the present invention is convenient, and the early warning result is intuitive and accurate. Through the monitoring data of the water guide bearing measuring point, the top cover horizontal vibration measuring point, the top cover vertical vibration measuring point, and the pressure pulsation measuring point, after the inference machine analyzes, a fault early warning signal is issued;

[0045] 2. The eddy current velocity sensor and piezoelectric acceleration sensor used in the present invention are easy to install on site, and are particularly suitable for use on large structures;

[0046] 3. The present invention can provide a practical and simple top cover fault early warning method for the intelligent fault operation and maintenance system of the hydropower unit. The method of the present invention can intuitively monitor the real-time status of the top cover and provide powerful quantitative data when the unit is running. It has a good application prospect and provides a reliable test method and technical means for the safe operation of the hydro-generator. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is a diagram of the measuring point arrangement of the present invention;

[0048] Figure 2 This is the fault warning logic diagram of the present invention.

[0049] Figure 1 Markings: 1- water guide bearing, 2- horizontal vibration monitoring position of top cover, 3- vertical vibration monitoring position of top cover, 4- bladeless area of ​​top cover DETAILED DESCRIPTION

[0050] Specific implementation method 1: Figure 1 and Figure 2 The method for intelligent early warning of turbine top cover fault shown comprises the following steps:

[0051] 1) At the turbine water guide bearing 1, an eddy current sensor is arranged in the X and Y directions to measure the displacement and obtain the swing signal; at the top cover horizontal vibration monitoring position 2, an acceleration sensor is arranged in the X and Y directions, and at the top cover vertical vibration monitoring position 3, an acceleration sensor is arranged in the Z direction to measure the acceleration and obtain the vibration signal; above the top cover bladeless area 4, a piezoelectric sensor is arranged to obtain the pressure pulsation signal of the top cover bladeless area;

[0052] 2) Take one month’s worth of historical data for the water conduction swing values ​​in the X and Y directions and calculate the average values, which will be used as the water conduction swing setting values ​​in the X and Y directions;

[0053] 3) Take one month of historical data for the acceleration values ​​of the top cover horizontal vibration signal in the X and Y directions and calculate the average values ​​respectively, which are used as the setting values ​​of the top cover horizontal vibration signal in the X and Y directions;

[0054] 4) Take one month’s historical data of the vertical vibration acceleration value of the top cover in the Z direction and calculate the average value as the Z-direction setting value of the vertical vibration signal of the top cover;

[0055] 5) Take one month’s historical data of the pressure pulsation value of the top cover without leaves and calculate the average value as the pressure pulsation setting value;

[0056] 6) Define event A as the comparison between the X-direction and Y-direction swing values ​​of the water guide bearing and the X-direction and Y-direction setting values. When the X-direction swing value of the water guide bearing is continuously greater than the X-direction setting value by more than 10% for 6 hours and the Y-direction swing value of the water guide bearing is continuously greater than the Y-direction setting value by more than 10% for 6 hours, event A is TRUE, otherwise event A is FALSE;

[0057] 7) Define event B as the comparison between the acceleration values ​​of the top cover horizontal vibration in the X and Y directions and the set values ​​in the X and Y directions. When the acceleration value of the top cover horizontal vibration in the X direction is continuously greater than the set value in the X direction by more than 10% for 6 hours and the acceleration value of the top cover horizontal vibration in the Y direction is continuously greater than the set value in the Y direction by more than 10% for 6 hours, event B is TRUE, otherwise event B is FALSE;

[0058] 8) Define event C as the comparison between the vertical vibration acceleration value of the top cover and the set value. When the vertical vibration acceleration value of the top cover is continuously greater than the vertical setting value by more than 10% for 6 hours, event C is TRUE, otherwise event C is FALSE;

[0059] 9) Define event D as the comparison between the pressure pulsation monitoring value of the top cover without leaves and the pressure pulsation setting value. When the pressure pulsation monitoring value of the top cover without leaves is greater than the pressure pulsation setting value, event D is TRUE, otherwise event D is FALSE;

[0060] 10) When event B is true, a fault warning of excessive hydraulic imbalance force is issued; when events A and B are both true, a fault warning of loose top cover handle and bolts is issued; when event C is true, a fault warning of insufficient top cover axial stiffness is issued; when event D is true, a fault warning of static and dynamic interference of top cover is issued;

[0061] 11) Using MySQL database as a platform, establish an equipment information management database to store historical fault warning information and related data. Perform statistical analysis on the warning records, fault records, fault types, and maintenance plans of the turbines, and use the attribute parameters of different faults as index identification numbers and store them in the equipment management information database;

[0062] Specific implementation method 2: Figure 1 and Figure 2 As shown, this embodiment further limits step 2) described in the specific embodiment 1. In this embodiment, the calculation steps of the water conductance swing setting value in step 2) are as follows:

[0063] Specifically, the swing values ​​of the water guide in the X and Y directions are collected and divided into time nodes at equal intervals of every 30 minutes, so there are 1440 time node data in each direction for 30 days, and the data are stored in the system database according to the numbering of the time nodes from far to near, and the setting values ​​of the X and Y directions are calculated. The calculation formula is:

[0064]

[0065] Where N represents the number of sampling times of the water-guided bearing swing signal value in one month, which is equal to 1440, and X si , Y si They represent the i-th sampling value of the swing of the water-guided bearing in the X direction and the Y direction, respectively. s ,Y s They are the set values ​​of water guide swing in X and Y directions respectively;

[0066] This implementation method introduces the average value of the historical data of the unit's water guide bearing as the fault alarm setting value. By comparing the current unit's real-time monitoring data with its own historical average level, the real-time fault status of the unit's water guide bearing can be obtained.

[0067] Specific implementation method three: Figure 1 and Figure 2 As shown, this implementation is to further limit step 3) described in the specific implementation mode 1. In this implementation mode, the calculation steps of the top cover horizontal vibration setting value in step 3) are as follows:

[0068] Take one month of historical data for the acceleration values ​​of the top cover horizontal vibration signal in the X and Y directions and calculate the average values ​​respectively, which are used as the X and Y setting values ​​of the top cover horizontal vibration signal;

[0069] Specifically, the acceleration values ​​of the horizontal vibration signal of the top cover in the X and Y directions are collected and divided into time nodes at equal intervals of every 30 minutes. There are 1440 time node data in each direction for 30 days, and the data are stored in the system database according to the numbering of the time nodes from far to near, and the setting values ​​of the X and Y directions are calculated. The calculation formula is:

[0070]

[0071] Where N is the number of times the acceleration value of the top cover horizontal vibration signal is sampled within one month, and X DSi , Y DSi Respectively represent the i-th sampling value of the acceleration value of the top cover horizontal vibration signal in the X direction and the Y direction; DS , Y DS They represent the acceleration setting values ​​of the vibration signal in the horizontal X and Y directions of the top cover respectively;

[0072] This implementation method introduces the average value of the historical data of the horizontal vibration of the unit's top cover as the fault alarm setting value. By comparing the current real-time monitoring data of the unit with its own historical average level, it can be concluded whether the horizontal vibration level of the unit's top cover is abnormal.

[0073] Specific implementation method four: Figure 1 and Figure 2 As shown, this embodiment further limits step 4) described in the specific embodiment 1. In this embodiment, the calculation steps of the top cover vertical vibration setting value in step 4) are as follows:

[0074] Take one month's historical data of the vertical vibration acceleration value of the top cover in the Z direction and calculate the average value, which is used as the Z-direction setting value of the vertical vibration signal of the top cover;

[0075] Specifically, the acceleration value of the vertical vibration signal of the top cover is collected and divided into time nodes at equal intervals of every 30 minutes, so there are 1440 time node data in each direction for 30 days, and the data are stored in the system database according to the numbering of the time nodes from far to near, and the set value is calculated. The calculation formula is:

[0076]

[0077] Where N is the number of times the acceleration value of the vertical vibration signal of the roof is sampled within one month, and Z is Dsi Represents the i-th sampling value of the acceleration value of the vertical vibration signal of the top cover in the Z direction, Z Ds It is the acceleration setting value of the vertical vibration signal of the top cover;

[0078] This implementation method introduces the average value of the historical data of the vertical vibration of the unit top cover as the fault alarm setting value. By comparing the current real-time monitoring data of the unit with its own historical average level, it can be concluded whether the vertical vibration level of the unit top cover is abnormal.

[0079] Specific implementation method five: Figure 1 and Figure 2 As shown, this embodiment further limits step 5) described in the specific embodiment 1. In this embodiment, the calculation steps of the pressure pulsation setting value of the top cover without leaves in step 5) are as follows:

[0080] Specifically, the pressure pulsation value of the top cover leafless area is collected and the time nodes are divided into equal intervals of every 30 minutes, so there are 1440 time node data in each direction for 30 days, and the data are stored in the system database according to the numbering of the time nodes from far to near, and the set value is calculated. The calculation formula is:

[0081]

[0082] Where N is the number of times the top cover pressure pulsation signal is sampled within a month, PDsi represents the i-th sampling value of the top cover pressure pulsation signal, P Ds It is the top cover pressure pulsation setting value;

[0083] This implementation method introduces the average historical data of the vertical vibration of the unit top cover as the fault alarm setting value. By comparing the current real-time monitoring data of the unit with its own historical average level, it can be determined whether the pressure pulsation level of the unit top cover is abnormal.

[0084] Specific implementation method 6: This implementation method is an example of step 6) described in specific implementation method 1:

[0085] Specifically, for example, the value of event A in step 6) is first assumed to be FALSE, and the set value of the water guide bearing X-direction swing value in the past 30 days and 1440 monitoring values ​​is calculated to be 130um, and the set value of the Y-direction swing value in the past 30 days and 1440 monitoring values ​​is calculated to be 138um; then in the next 6 hours, the X-direction and Y-direction swing values ​​of the unit are collected in real time at a time node of every 30 minutes, and saved to the system database. If the 12 monitoring values ​​of the X-direction and Y-direction are respectively greater than 143um and 151.8um, event A is set to TRUE and saved to the system;

[0086] Specific implementation method 7: This implementation method is an example of step 7) described in specific implementation method 1:

[0087] Specifically, for example, the default value of event B in step 7) is FALSE, and the set value is calculated to be 130um / s for the 1440 monitoring values ​​of the horizontal vibration of the top cover in the X direction for the past 30 days. 2 The Y direction acceleration value is calculated from 1440 monitoring values ​​in the past 30 days and the set value is 138um / s 2 ; Then for the next 6 hours, the vibration acceleration values ​​of the unit in the X and Y directions are collected in real time at a time node of every 30 minutes and saved in the system database. If the 12 monitoring values ​​in the X and Y directions are both greater than 143um / s 2 、151.8um / s 2 , then set event B to TRUE and save it to the system;

[0088] Specific implementation eight: This implementation is an example of step 8) described in specific implementation one:

[0089] Specifically, for example, the default value of event C in step 8) is FALSE, and the set value is calculated to be 130um / s for 1440 monitoring values ​​in the vertical Z direction of the top cover in the past 30 days. 2; Then for the next 6 hours, the Z-axis vibration acceleration value of the unit is collected in real time every 30 minutes and saved in the system database. If all 12 monitoring values ​​in the Z-axis are greater than 143um / s 2 , then set event C to TRUE and save it to the system;

[0090] Specific implementation method 9: This implementation method is an example of step 9) described in specific implementation method 1:

[0091] Specifically, for example, the event D value in step 9) is first assumed to be FALSE, and the set value is calculated to be 130Pa for the 1440 monitoring values ​​of the pressure pulsation value of the bladeless area of ​​the top cover in the past 30 days; then the pressure pulsation value of the bladeless area of ​​the unit is collected in real time at a time node of every 30 minutes for the next 6 hours, and saved to the system database. If the 12 monitoring values ​​in the Z direction are all greater than 143Pa, the event D is set to TRUE and saved to the system;

[0092] Specific implementation method ten: This implementation method is an application description of step 10) described in specific implementation method one:

[0093] Specifically, the system has a built-in logic inference engine program, which can read the Boolean value of the event at any time under the operator's command. When the value of the read event B is true, a fault warning of excessive hydraulic imbalance force is issued; when the values ​​of the read events A and B are both true, a fault warning of loose top cover handle and bolt is issued; when the value of the read event C is true, a fault warning of insufficient top cover axial stiffness is issued; when the value of the read event D is true, a fault warning of static and dynamic interference of the top cover is issued;

[0094] Specific implementation eleven: This implementation is an application description of step 11) described in specific implementation one:

[0095] Specifically, a first data table of the database is established, which is sorted from far to near in time, and the time node sequence is used as the first column of the database. The data of the measuring points are collected once every thirty minutes and stored in the first data table synchronously. The second column of the first data table stores the X-direction swing monitoring value of the water guide bearing according to the time node, the third column stores the Y-direction swing monitoring value of the water guide bearing according to the time node, the fourth column stores the X-direction monitoring value of the horizontal vibration of the top cover according to the time node, the fifth column stores the Y-direction detection value of the horizontal vibration of the top cover according to the time node, the sixth column stores the vertical vibration direction monitoring value of the top cover according to the time node, the seventh column stores the vertical vibration direction monitoring value of the top cover according to the time node, and the eighth column stores the pressure pulsation monitoring value of the leafless area of ​​the top cover according to the time node;

[0096] Specifically, a second data table of the database is established, sorted by time from far to near, and the Boolean values ​​of event A, event B, event C, and event D are stored. The first column of the second data table is the time node when the inference engine starts running, the second column is the Boolean value of event A, the third column is the Boolean value of event B, the fourth column is the Boolean value of event C, and the fifth column is the Boolean value of event D;

[0097] Specifically, a third data table of the database is established, sorted by time from far to near, and fault warning results are stored. A value of TRUE indicates that a fault warning is issued, and a value of FALSE indicates that no fault warning is issued. The first column of the third data table is the start-up time node of the inference engine, the second column is the fault warning result of excessive hydraulic unbalanced force, the third column is the fault warning result of loose top cover handle and bolts, the fourth column is the fault warning result of insufficient top cover axial stiffness, and the fifth column is the fault warning result of static-dynamic interference of the top cover;

[0098] This implementation method establishes three data tables in the power station database to respectively store the monitoring data described in the specific implementation method one, the Boolean values ​​of events A, B, C, and D, and the fault warning results, thereby achieving clear and unified management of the system for turbine top cover faults and related monitoring data.

Claims

1. An intelligent early warning method for turbine top cover failure, Features include Follow these steps: Step 1: at the water guide bearing (1) of the turbine, an eddy current sensor is arranged in the X and Y directions to measure the displacement and obtain a swing signal; an acceleration sensor is arranged in the horizontal (2) X and Y directions and in the vertical (3) Z direction of the top cover to measure the acceleration and obtain a vibration signal; a piezoelectric sensor is arranged above the bladeless area of ​​the top cover (4) to obtain a pressure pulsation signal of the bladeless area of ​​the top cover; Step 2: Take the historical data of the latest month for the swing values ​​of the water-guided bearing in the X and Y directions and calculate the average values, which are used as the setting values ​​of the swing values ​​of the water-guided bearing in the X and Y directions; Step 3: Take the historical data of the latest month for the acceleration values ​​of the top cover horizontal vibration signal in the X and Y directions and calculate the average values ​​respectively, which are used as the setting values ​​of the top cover horizontal vibration signal in the X and Y directions; Step 4: Take the historical data of the last 30 days for the vertical vibration acceleration value of the top cover in the Z direction and calculate the average value as the setting value of the vertical vibration signal of the top cover in the Z direction; Step 5: Take the historical data of the pressure pulsation monitoring value of the top cover without leaves in the last 30 days to calculate the average value as the pressure pulsation setting value; Step 6: Define event A as the comparison between the swing value of the water guide in the X and Y directions and the set value in the X and Y directions: when the swing value of the water guide in the X direction is continuously greater than the set value in the X direction by more than 10% for 6 hours, and the swing value of the water guide in the Y direction is continuously greater than the set value in the Y direction by more than 10% for 6 hours, event A is TRUE, otherwise event A is FALSE; Step 7: Define event B as the comparison between the acceleration values ​​of the top cover horizontal vibration in the X and Y directions and the set values ​​in the X and Y directions: when the acceleration value of the top cover horizontal vibration in the X direction is continuously greater than the set value in the X direction by more than 10% for 6 hours, and the acceleration value of the top cover horizontal vibration in the Y direction is continuously greater than the set value in the Y direction by more than 10% for 6 hours, event B is TRUE, otherwise event B is FALSE; Step 8: Define event C as the comparison between the vertical vibration acceleration value of the top cover and the set value: when the vertical vibration acceleration value of the top cover is continuously greater than the vertical setting value by more than 10% for 6 hours, event C is TRUE, otherwise event C is FALSE; Step 9: Define event D as the comparison between the pressure pulsation monitoring value of the top cover without leaves and the pressure pulsation setting value: when the pressure pulsation monitoring value of the top cover without leaves is greater than the pressure pulsation setting value, event D is TRUE, otherwise event D is FALSE; Step 10: When event B is true, a fault warning of excessive hydraulic unbalance force is issued; when events A and B are both true, a fault warning of loose top cover handle and bolts is issued; when event C is true, a fault warning of insufficient top cover axial stiffness is issued; when event D is true, a fault warning of static and dynamic interference of top cover is issued; Step 11: Using MySQL database as a platform, establish an equipment information management database to save historical fault warning information and related data, conduct statistical analysis on the warning records, fault records, fault types, and maintenance plans of the turbine, and use the attribute parameters of different faults as index identification numbers and store them in the equipment management information database; Step 12: When a fault warning occurs, use the unit downtime for maintenance to repair the top cover related components.

2. According to claim 1, a method for intelligent early warning of turbine top cover failure, Its characteristics are: In the step 2, the water conduction X direction and Y direction swing setting values ​​are respectively recorded as X s ,Y s ,but Where N represents the number of sampling times of the water conductance swing signal value in one month, X si , Y si They represent the i-th sampling value of the water guide X-direction swing and the water guide Y-direction swing, respectively. s ,Y s They are the set values ​​of the water guide swing in the X and Y directions respectively.

3. According to claim 1, a method for intelligent early warning of turbine top cover failure, Its characteristics are: In step 3, the acceleration setting values ​​of the vibration signal in the horizontal X and Y directions of the top cover are respectively recorded as X DS , Y DS ,but: Where N represents the number of times the acceleration value of the top cover horizontal vibration signal is sampled within one month, and X DSi , Y DSi Respectively represent the i-th sampling value of the acceleration value of the top cover horizontal vibration signal in the X direction and the Y direction; DS , Y DS They represent the acceleration setting values ​​of the vibration signal in the horizontal X and Y directions of the top cover respectively.

4. According to claim 1, a method for intelligent early warning of turbine top cover failure, Its characteristics are: In step 4, the acceleration setting value of the vertical vibration signal of the top cover is recorded as Z Ds ,but: Where N is the number of times the acceleration value of the vertical vibration signal of the roof is sampled within one month, and Z is Dsi Represents the i-th sampling value of the acceleration value of the vertical vibration signal of the top cover in the Z direction, Z Ds is the acceleration value of the vertical vibration signal of the top cover.

5. According to claim 1, a method for intelligent early warning of turbine top cover failure, Its characteristics are: In step 5, the top cover pressure pulsation setting value is recorded as P Ds ,but: Where N is the number of times the acceleration value of the vertical vibration signal of the roof is sampled within one month, and P Dsi represents the i-th sampling value of the acceleration value of the vertical vibration signal of the top cover in the Z direction, P Ds It is the setting value of top cover pressure pulsation.

6. According to claim 1, a method for intelligent early warning of turbine top cover failure, Its characteristics are: In step 12, the contents of the inspection and treatment measures are as follows: When a fault warning of excessive hydraulic unbalance force occurs, stop the machine to check whether the natural air supply system is working normally, and repair the natural air supply system; when a fault warning of loose top cover handles and bolts occurs, stop the machine to check whether the top cover handles and bolts are loose, and tighten all top cover handles and bolts; when a fault warning of insufficient axial stiffness of the top cover occurs, stop the machine to check whether the top cover structure is damaged or worn, and weld and reinforce the worn parts and parts prone to wear; when a fault warning of excessive dynamic and static interference of the top cover occurs, stop the machine to check the flow components of the impeller, guide vanes, and wear-resistant plates, and make slight adjustments to the gaps of the flow components by grinding and repair welding.

Images