A method and system for automatically evaluating the stability performance of a hydro-generator set

Abstract

A method and system for automatically evaluating the stability performance of a hydro-turbine generator set, which layers the stability performance of the hydro-turbine generator set from top to bottom, obtains the corresponding indicators of the branch performance, stator core mechanical performance, main shaft sealing water stop performance, and hydraulic performance of the fourth layer, and calculates layer by layer from the fourth layer according to the weighted average model, to obtain the stability performance evaluation index of the hydro-turbine generator set and the equipment evaluation index of each layer, and realizes the visual display of the performance index in the system. The present invention can realize the automatic evaluation of the stability performance of the hydro-turbine generator set and form a rich expert knowledge base according to the different equipment structures of the hydro-turbine generator set, and combines the neural network algorithm in the field of big data analysis to timely reflect the performance changes of the equipment during normal operation and the long-term change trend of the health status of the hydro-turbine generator set, so as to ensure the accuracy of the evaluation results.

Description

Technical Field

[0001] The invention relates to a hydroelectric generator set, in particular to a method and a system for automatically evaluating the stability performance of a hydroelectric generator set. Background Art

[0002] At present, the stability performance evaluation and abnormal monitoring of hydro-turbine generator sets mainly rely on the threshold alarm monitoring of online monitoring data. The operation and maintenance personnel analyze and evaluate the equipment operation status based on the alarm information of the online monitoring system. This evaluation method has the following limitations:

[0003] (1) It can only reflect the abnormal problems of local equipment and has a single evaluation index, but cannot reflect the overall quantitative performance index of the hydro-generator unit.

[0004] (2) Monitoring only the alarm information cannot timely reflect the performance changes of the equipment during normal operation and the long-term trend of the health status of the turbine generator set.

[0005] (3) It is impossible to provide personnel in different positions with the information that they are most concerned about. Personnel in different positions have different concerns about the operating status of the hydro-turbine generator set. Technical personnel are concerned about the performance of the equipment under their jurisdiction, while supervisors are concerned about the overall performance of the hydro-turbine generator set. Summary of the invention

[0006] In order to solve the above technical problems, the present invention provides a method and system for automatically evaluating the stability performance of a hydro-turbine generator set, which obtains the stability performance index of the hydro-turbine generator set and the evaluation index of each level of equipment, and realizes the visual display of the performance index in the system.

[0007] In order to achieve the above technical purpose, the technical solution adopted is: a method for automatically evaluating the stability performance of a hydro-generator set, comprising the following steps:

[0008] Step 1: Layer the stability performance of the hydro-generator set from top to bottom;

[0009] The top layer is the stability performance F0 of the hydro-generator set; the second layer is the mechanical performance F11 of the generator and the performance F12 of the turbine; the third layer is the mechanical performance F21 of the stator core, the performance F22 of the upper guide bearing, the performance F23 of the lower guide bearing, and the thrust bearing performance F24 included in the mechanical performance F11 of the generator, and the performance F25 of the water-guide bearing, the water-stopping performance F26 of the main shaft seal, and the hydraulic performance F27 included in the turbine performance F12; the fourth layer is the vibration performance F31 of the upper frame, the swing performance F32 of the upper guide bearing, and the performance F33 of the upper guide shoe included in the upper guide bearing performance F22, the vibration performance F34 of the lower frame, the swing performance F35 of the lower guide bearing, and the performance F36 of the lower guide shoe included in the lower guide bearing performance F23, the thrust shoe performance F37 included in the thrust bearing performance F24, and the top cover vibration performance F38, the swing performance F39 of the water-guide bearing, and the performance F310 of the water-guide shoe included in the water-guide bearing performance F25;

[0010] Step 2, obtain the corresponding data of each branch performance of the fourth layer, the corresponding data of the mechanical performance of the stator core, the corresponding data of the main shaft seal water-stopping performance, and the corresponding original data of the hydraulic performance through communication, extract the characteristic value according to the original data, build the algorithm model corresponding to the performance of each branch of the fourth layer, the mechanical performance of the stator core, the water-stopping performance of the main shaft, and the hydraulic performance, and obtain the initial performance score corresponding to each performance according to the algorithm model and the operation data corresponding to the performance, discretize the initial performance score, correct the points with large deviations from the performance score of the unit operation status, and obtain the expert knowledge sample library corresponding to the performance of each branch of the fourth layer, the mechanical performance of the stator core, the water-stopping performance of the main shaft, and the hydraulic performance, and use the neural network algorithm to train the expert knowledge sample library corresponding to the performance of each branch of the fourth layer, the mechanical performance of the stator core, the water-stopping performance of the main shaft, and the hydraulic performance, and obtain the corresponding indicators of the performance of each branch of the fourth layer, the mechanical performance of the stator core, the water-stopping performance of the main shaft, and the hydraulic performance;

[0011] Step 3: Starting from the fourth layer, from the bottom up, the branch performance is used as the evaluation basis for the upper layer performance. According to the weighted average model, the initial performance score of each layer of equipment is calculated layer by layer to obtain the initial performance score. The initial performance score is discretized, and the points with large deviations from the unit operation status performance score are corrected to obtain the expert knowledge sample library of the equipment performance of each layer. The expert knowledge sample library of each performance of each layer of equipment is trained using a neural network algorithm to obtain the stability performance evaluation index of the hydro-turbine generator set and the evaluation index of each layer of equipment.

[0012] The corresponding data of each branch performance on the fourth layer, the corresponding data of the mechanical performance of the stator core, the corresponding data of the main shaft seal water-stopping performance and the corresponding data of the hydraulic performance are the unit bearing bushing temperature data in the computer monitoring system of the whole plant, the start and stop time data of the turbine top cover drainage pump, and the unit operation vibration swing and pressure pulsation data of the unit stability online monitoring system.

[0013] The bearing temperature data of the unit includes the bearing temperature data of the upper guide, lower guide, water guide and thrust bearing of the generator set.

[0014] The unit operation vibration swing data includes the peak-to-peak value of the main shaft swing in the +X and -Y directions at the shaft collars of the upper guide, lower guide and water guide bearings of the generator, the peak-to-peak value of the frame vibration in the +X and -Y directions of the upper and lower frames of the generator, the 100Hz component amplitude of the vibration of the generator stator core in the +X and -Y directions, and the peak-to-peak value of the pressure pulsation at the outlet of the pressure measuring pipeline at the lower cavity of the top cover, the tailwater pipe outlet and the tailwater pipe entrance.

[0015] The data extraction features include:

[0016] Upper frame vibration performance characteristic value: Upper frame +X direction vibration peak value A 上机架+x , Upper frame - Y-axis vibration peak value A 上机架-y ;

[0017] Upper guide bearing swing performance characteristic value: Upper guide + X-axis swing peak value A 上导+x , Upward guide - Y-direction swing peak-to-peak value A 上导-y ;

[0018] Upper guide bearing performance characteristic value: Maximum upper guide bearing bearing temperature T 上导max , the average temperature of the upper guide bearing bushing T 上导avg , Standard deviation value of upper guide bearing bush temperature T 上导sd ;

[0019] Lower frame vibration performance characteristic value: Lower frame +X direction vibration peak value A 下机架+x , Lower frame - Y-axis vibration peak value A 下机架-y ;

[0020] Lower guide bearing performance characteristic value: Maximum temperature of lower guide bearing bearing T 下导max , the average temperature of the lower guide bearing bushing T 下导avg , Standard deviation value of lower guide bearing bush temperature T 下导sd ;

[0021] Lower guide bearing swing performance characteristic value: Lower guide + X-axis swing peak value A 下导+x , Downward guide - Y direction swing peak value A 下导-y ;

[0022] Thrust bearing performance characteristic value: Maximum thrust bearing temperature T 推力max , Average thrust bearing temperature T 推力avg , thrust bearing bush temperature standard deviation value T 推力sd ;

[0023] Top cover vibration performance characteristic value: Top cover + X-axis vibration peak value A 顶盖+x , Top cover - Y direction vibration peak value A 顶盖-y , Peak-to-peak value of top cover Z-direction vibration A 顶盖z ;

[0024] Water-guided bearing swing performance characteristic value: Water-guided + X-axis swing peak value A 水导+x , Water guide - Y direction swing peak value A 水导-y ;

[0025] Water-guided bearing performance characteristic value: Maximum water-guided bearing bearing temperature T 水导max , Average water-guided bearing bushing temperature T 水导avg , water-guided bearing bush temperature standard deviation value T 水导sd ;

[0026] Mechanical performance characteristic value of stator core: Stator core + X-axis vibration 100Hz component amplitude A 铁芯+x , Stator core - Y-axis vibration 100Hz component amplitude A 铁芯-y ;

[0027] Main shaft seal water-stopping performance characteristic value: Top cover drainage pump start-up time T 1 , The last stop time of the top cover T 2 ;

[0028] Hydraulic performance characteristic value: Peak-to-peak value of pressure pulsation at the tailwater inlet P 尾水进人门 , peak value of tailwater outlet pressure pulsation P 尾水管出口 .

[0029] A system for automatically evaluating the stability performance of a hydro-generator set, comprising:

[0030] The layered display module layers the stability performance of the hydro-generator units from top to bottom;

[0031] The top layer is the stability performance F0 of the hydro-generator set; the second layer is the mechanical performance F11 of the generator and the performance F12 of the turbine; the third layer is the mechanical performance F21 of the stator core, the performance F22 of the upper guide bearing, the performance F23 of the lower guide bearing, and the thrust bearing performance F24 included in the mechanical performance F11 of the generator, and the performance F25 of the water-guide bearing, the water-stopping performance F26 of the main shaft seal, and the hydraulic performance F27 included in the turbine performance F12; the fourth layer is the vibration performance F31 of the upper frame, the swing performance F32 of the upper guide bearing, and the performance F33 of the upper guide shoe included in the upper guide bearing performance F22, the vibration performance F34 of the lower frame, the swing performance F35 of the lower guide bearing, and the performance F36 of the lower guide shoe included in the lower guide bearing performance F23, the thrust shoe performance F37 included in the thrust bearing performance F24, and the top cover vibration performance F38, the swing performance F39 of the water-guide bearing, and the performance F310 of the water-guide shoe included in the water-guide bearing performance F25;

[0032] The bottom performance index calculation module obtains the corresponding data of each branch performance of the fourth layer, the corresponding data of the mechanical performance of the stator core, the corresponding data of the main shaft seal water-stopping performance, and the corresponding data of the hydraulic performance through communication, extracts characteristic values ​​according to the data, builds an algorithm model corresponding to each branch performance of the fourth layer, the mechanical performance of the stator core, the main shaft seal water-stopping performance, and the hydraulic performance, obtains the corresponding initial performance score according to the algorithm model and the operating data corresponding to the performance, discretizes the initial performance score, corrects the points with large deviations from the unit operation status performance score, obtains the expert knowledge sample library corresponding to each branch performance of the fourth layer, the mechanical performance of the stator core, the main shaft seal water-stopping performance, and the hydraulic performance, uses the neural network algorithm to train the expert knowledge sample library corresponding to each branch performance of the fourth layer, the mechanical performance of the stator core, the main shaft seal water-stopping performance, and the hydraulic performance, and obtains the corresponding indicators of each branch performance of the fourth layer, the mechanical performance of the stator core, the main shaft seal water-stopping performance, and the hydraulic performance;

[0033] The automatic evaluation module, starting from the fourth layer from the bottom up, uses the branch performance as the evaluation basis for the upper layer performance, calculates layer by layer according to the weighted average model, obtains the initial performance score of the equipment performance at each layer, discretizes the initial performance score, corrects the points with large deviations from the unit operation status performance score, and obtains the expert knowledge sample library of the equipment performance at each layer. The expert knowledge sample library of each performance of each layer of equipment is trained using a neural network algorithm to obtain the stability performance evaluation index of the hydro-turbine generator set and the evaluation index of each layer of equipment.

[0034] The system displays the performance evaluation indicators of each layer with a score of 0-100, among which 0-50 points indicate that the equipment is severely damaged and requires a long time to deal with (more than 15 days); 51-60 points indicate that the equipment is slightly damaged and requires a certain amount of time to deal with (such as 2-15 days); 61-80 points indicate that the equipment has an alarm and needs to be temporarily shut down and dealt with in a short time (within 1 day); 81-100 means that the equipment can operate stably for a long time, among which 81-90 points indicate that the equipment has an abnormal alarm but can still operate for a long time, and 91-100 points indicate that the equipment operates normally, is in good working condition, and can operate stably for a long time.

[0035] When the performance score of a certain layer of equipment is lower than 80 points, the score in the system will turn yellow. When the score is lower than 60 points, the score will turn red, and relevant warning information will be pushed to the system users.

[0036] The beneficial effects of the present invention are as follows: compared with the prior art, the present invention provides a method and system for automatically evaluating the stability performance of a hydro-turbine generator set, which can be based on the different equipment structures of the hydro-turbine generator set, and can be classified and implemented according to local conditions to achieve automatic evaluation of the stability performance of the hydro-turbine generator set, and form a rich expert knowledge base, combined with the neural network algorithm in the field of big data analysis, to ensure the accuracy of the evaluation results. The performance score of the hydro-turbine generator set equipment can provide the operation and maintenance personnel with a basis for judging the macroscopic state of the unit. When the score is reduced, the system will issue an early warning in time to prompt the operation and maintenance personnel to check the cause of the equipment failure and ensure the safe and stable operation of the equipment. At the same time, the present invention can timely reflect the performance changes of the equipment during normal operation and the long-term change trend of the health status of the hydro-turbine generator set, and provide the most direct information to personnel in different positions. The staff in different positions have different concerns about the operating status of the hydro-turbine generator set. The technical personnel pay attention to the performance of the equipment under their jurisdiction, and the supervisor pays attention to the overall performance of the hydro-turbine generator set. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 A diagram showing a system model of the present invention;

[0038] Figure 2 It is a performance indicator trend analysis diagram of the present invention. DETAILED DESCRIPTION

[0039] The preferred embodiments of the invention are given below in conjunction with the accompanying drawings to explain the technical solution of the present invention in detail. Here, the corresponding drawings are given to explain the present invention in detail. It should be particularly noted that the preferred embodiments described here are only used to illustrate and explain the present invention, and are not used to limit or restrict the present invention.

[0040] A method for automatically evaluating the stability performance of a hydro-generator set comprises the following steps:

[0041] Step 1: Layer the stability performance of the hydro-generator set from top to bottom;

[0042] The top layer is the stability performance F0 of the hydro-generator set; the second layer is the mechanical performance F11 of the generator and the performance F12 of the turbine; the third layer is the mechanical performance F21 of the stator core, the performance F22 of the upper guide bearing, the performance F23 of the lower guide bearing, and the thrust bearing performance F24 included in the mechanical performance F11 of the generator, and the performance F25 of the water-guide bearing, the water-stopping performance F26 of the main shaft seal, and the hydraulic performance F27 included in the turbine performance F12; the fourth layer is the vibration performance F31 of the upper frame, the swing performance F32 of the upper guide bearing, and the performance F33 of the upper guide shoe included in the upper guide bearing performance F22, the vibration performance F34 of the lower frame, the swing performance F35 of the lower guide bearing, and the performance F36 of the lower guide shoe included in the lower guide bearing performance F23, the thrust shoe performance F37 included in the thrust bearing performance F24, and the top cover vibration performance F38, the swing performance F39 of the water-guide bearing, and the performance F310 of the water-guide shoe included in the water-guide bearing performance F25;

[0043] Step 2, obtain the corresponding data of each branch performance of the fourth layer, the corresponding data of the mechanical performance of the stator core, the corresponding data of the main shaft sealing water-stopping performance, and the corresponding original data of the hydraulic performance through communication, extract characteristic values ​​according to the original data, each branch performance of the fourth layer, the mechanical performance of the stator core, the main shaft sealing water-stopping performance, and the hydraulic performance have their own characteristic values, build an algorithm model corresponding to each branch performance of the fourth layer, the mechanical performance of the stator core, the main shaft sealing water-stopping performance, and the hydraulic performance, and obtain the initial performance score corresponding to each performance according to the algorithm model and the original operation data corresponding to the performance, discretize the initial performance score, correct the points with large deviations from the unit operation status performance score, and obtain the expert knowledge sample library corresponding to each branch performance of the fourth layer, the mechanical performance of the stator core, the main shaft sealing water-stopping performance, and the hydraulic performance, and use the neural network algorithm to train the expert knowledge sample library corresponding to each branch performance of the fourth layer, the mechanical performance of the stator core, the main shaft sealing water-stopping performance, and the hydraulic performance, and obtain the corresponding indicators of each branch performance of the fourth layer, the mechanical performance of the stator core, the main shaft sealing water-stopping performance, and the hydraulic performance;

[0044] Step 3: Starting from the fourth layer, from the bottom up, the branch performance is used as the evaluation basis for the upper layer performance. According to the weighted average model, the initial performance score of each layer of equipment is calculated layer by layer to obtain the initial performance score. The initial performance score is discretized, and the points with large deviations from the unit operation status performance score are corrected to obtain the expert knowledge sample library of the equipment performance of each layer. The expert knowledge sample library of each performance of each layer of equipment is trained using a neural network algorithm to obtain the stability performance evaluation index of the hydro-turbine generator set and the evaluation index of each layer of equipment.

[0045] like Figure 1As shown in the figure, the upper frame vibration performance F31, the upper guide bearing swing performance F32, and the upper guide bearing performance F33 are used as the evaluation basis for the upper guide bearing performance F22. According to the weighted average model, the initial performance score of the upper guide bearing performance F22 is calculated layer by layer to obtain the initial performance score of the upper guide bearing performance F22. The initial performance score of the upper guide bearing performance F22 is discretized, and the points with large deviations from the unit operation status performance score are corrected to obtain the expert knowledge sample library of the upper guide bearing performance F22. The expert knowledge sample library of the upper guide bearing performance F22 is trained using a neural network to obtain the evaluation index of the upper guide bearing performance F22. Similarly, the lower frame vibration performance F34, the lower guide bearing swing performance F32, and the upper guide bearing performance F33 are used to evaluate the upper guide bearing performance F22. The performance F35 and the lower guide bearing performance F36 are used as the evaluation basis for the lower guide bearing performance F23, and the evaluation index of the lower guide bearing performance F23 is obtained. By analogy, the stability performance of the hydro-generator set is evaluated step by step. The mechanical performance F11 of the generator is evaluated based on the stator core mechanical performance F21, the upper guide bearing performance F22, the lower guide bearing performance F23, and the thrust bearing performance F24. The performance of the turbine is evaluated based on the guide bearing performance F25, the main shaft seal water-stop performance F26, and the hydraulic performance F27. According to step 3, the evaluation indexes of each layer of equipment can be obtained, and finally the stability performance F0 evaluation index of the hydro-generator set is obtained.

[0046] The corresponding data of each branch performance on the fourth layer, the corresponding data of the mechanical performance of the stator core, the corresponding data of the main shaft seal water-stopping performance and the corresponding data of the hydraulic performance are the unit bearing bushing temperature data in the computer monitoring system of the whole plant, the start and stop time data of the turbine top cover drainage pump, and the unit operation vibration swing and pressure pulsation data of the unit stability online monitoring system.

[0047] The bearing temperature data of the unit includes the bearing temperature data of the upper guide, lower guide, water guide and thrust bearing of the generator set.

[0048] The unit operation vibration swing data includes the peak-to-peak value of the main shaft swing in the +X and -Y directions at the shaft collars of the upper guide, lower guide and water guide bearings of the generator, the peak-to-peak value of the frame vibration in the +X and -Y directions of the upper and lower frames of the generator, the 100Hz component amplitude of the vibration of the generator stator core in the +X and -Y directions, and the peak-to-peak value of the pressure pulsation at the outlet of the pressure measuring pipeline at the lower cavity of the top cover, the tailwater pipe outlet and the tailwater pipe entrance.

[0049] The data extraction characteristic values ​​include: upper frame vibration performance characteristic value, upper guide bearing performance characteristic value, lower frame vibration performance characteristic value, lower guide bearing performance characteristic value, lower guide bearing swing performance characteristic value, thrust bearing performance characteristic value, top cover vibration performance characteristic value, water guide bearing swing performance characteristic value and water guide bearing performance characteristic value. Upper frame vibration performance characteristic value: upper frame +X direction vibration peak value A 上机架+x , Upper frame - Y-axis vibration peak value A 上机架-y .

[0050] Upper guide bearing swing performance characteristic value: Upper guide + X-axis swing peak value A 上导+x , Upward guide - Y-direction swing peak-to-peak value A 上导-y .

[0051] Upper guide bearing performance characteristic value: Maximum upper guide bearing bearing temperature T 上导max , the average temperature of the upper guide bearing bushing T 上导avg , Standard deviation value of upper guide bearing bush temperature T 上导sd .

[0052] Lower frame vibration performance characteristic value: Lower frame +X direction vibration peak value A 下机架+x , Lower frame - Y-axis vibration peak value A 下机架-y .

[0053] Lower guide bearing performance characteristic value: Maximum temperature of lower guide bearing bearing T 下导max , the average temperature of the lower guide bearing bushing T 下导avg , Standard deviation value of lower guide bearing bush temperature T 下导sd .

[0054] Lower guide bearing swing performance characteristic value: Lower guide + X-axis swing peak value A 下导+x , Downward guide - Y direction swing peak value A 下导-y .

[0055] Thrust bearing performance characteristic value: Maximum thrust bearing temperature T 推力max , Average thrust bearing temperature T 推力avg , thrust bearing bush temperature standard deviation value T 推力sd .

[0056] Top cover vibration performance characteristic value: Top cover + X-axis vibration peak value A 顶盖+x , Top cover - Y direction vibration peak value A 顶盖-y , Peak-to-peak value of top cover Z-direction vibration A 顶盖z .

[0057] Water-guided bearing swing performance characteristic value: Water-guided + X-axis swing peak value A 水导+x , Water guide - Y direction swing peak value A 水导-y .

[0058] Water-guided bearing performance characteristic value: Maximum water-guided bearing bearing temperature T 水导max , Average water-guided bearing bushing temperature T 水导avg , water-guided bearing bush temperature standard deviation value T 水导sd .

[0059] Mechanical performance characteristic value of stator core: Stator core + X-axis vibration 100Hz component amplitude A 铁芯+x , Stator core - Y-axis vibration 100Hz component amplitude A铁芯-y .

[0060] Main shaft seal water-stopping performance characteristic value: Top cover drainage pump start-up time T 1 , The last stop time of the top cover T 2 . Hydraulic performance characteristic value: peak-to-peak value of pressure pulsation at the tailwater inlet 尾水进人门 , peak value of tailwater outlet pressure pulsation P 尾水管出口 .

[0061] A system for automatically evaluating the stability performance of a hydro-generator set, comprising:

[0062] The layered display module layers the stability performance of the hydro-generator units from top to bottom;

[0063] The top layer is the stability performance F0 of the hydro-generator set; the second layer is the mechanical performance F11 of the generator and the performance F12 of the turbine; the third layer is the mechanical performance F21 of the stator core, the performance F22 of the upper guide bearing, the performance F23 of the lower guide bearing, and the thrust bearing performance F24 included in the mechanical performance F11 of the generator, and the performance F25 of the water-guide bearing, the water-stopping performance F26 of the main shaft seal, and the hydraulic performance F27 included in the turbine performance F12; the fourth layer is the vibration performance F31 of the upper frame, the swing performance F32 of the upper guide bearing, and the performance F33 of the upper guide shoe included in the upper guide bearing performance F22, the vibration performance F34 of the lower frame, the swing performance F35 of the lower guide bearing, and the performance F36 of the lower guide shoe included in the lower guide bearing performance F23, the thrust shoe performance F37 included in the thrust bearing performance F24, and the top cover vibration performance F38, the swing performance F39 of the water-guide bearing, and the performance F310 of the water-guide shoe included in the water-guide bearing performance F25;

[0064] The bottom performance index calculation module obtains the corresponding data of each branch performance of the fourth layer, the corresponding data of the mechanical performance of the stator core, the corresponding data of the main shaft seal water-stopping performance, and the corresponding data of the hydraulic performance through communication, extracts characteristic values ​​according to the data, builds an algorithm model corresponding to each branch performance of the fourth layer, the mechanical performance of the stator core, the main shaft seal water-stopping performance, and the hydraulic performance, obtains the corresponding initial performance score according to the algorithm model and the operating data corresponding to the performance, discretizes the initial performance score, corrects the points with large deviations from the unit operation status performance score, obtains the expert knowledge sample library corresponding to each branch performance of the fourth layer, the mechanical performance of the stator core, the main shaft seal water-stopping performance, and the hydraulic performance, uses the neural network algorithm to train the expert knowledge sample library corresponding to each branch performance of the fourth layer, the mechanical performance of the stator core, the main shaft seal water-stopping performance, and the hydraulic performance, and obtains the corresponding indicators of each branch performance of the fourth layer, the mechanical performance of the stator core, the main shaft seal water-stopping performance, and the hydraulic performance;

[0065] The automatic evaluation module, starting from the fourth layer from the bottom up, uses the branch performance as the evaluation basis for the upper layer performance, calculates layer by layer according to the weighted average model, obtains the initial performance score of the equipment performance at each layer, discretizes the initial performance score, corrects the points with large deviations from the unit operation status performance score, and obtains the expert knowledge sample library of the equipment performance at each layer. The expert knowledge sample library of each performance of each layer of equipment is trained using a neural network algorithm to obtain the stability performance index of the hydro-turbine generator set and the evaluation index of each layer of equipment.

[0066] The system displays the performance evaluation results of each layer with a score of 0-100, among which 0-50 points indicate that the equipment is severely damaged and requires a long time to deal with (more than 15 days); 51-60 points indicate that the equipment is slightly damaged and requires a certain amount of time to deal with (such as 2-15 days); 61-80 points indicate that the equipment has an alarm and needs to be temporarily shut down and dealt with in a short time (within 1 day); 81-100 means that the equipment can operate stably for a long time, among which 81-90 points indicate that the equipment has an abnormal alarm but can still operate for a long time, and 91-100 points indicate that the equipment operates normally, is in good working condition, and can operate stably for a long time.

[0067] When the performance score of a certain layer of equipment is lower than 80 points, the score in the system will turn yellow. When the score is lower than 60 points, the score will turn red, and relevant warning information will be pushed to the system users.

[0068] The algorithm model of the stator core mechanical property F21 described in step 2 is:

[0069] Let x = Max(A 铁芯+x ,A 铁芯-y ), the mechanical performance rating of the stator core is:

[0070]

[0071] The algorithm model of the upper frame vibration performance F31 and the lower frame vibration performance F34 described in step 2 is:

[0072] make or The algorithm model of the upper frame vibration performance F31 and the lower frame vibration performance F34 is:

[0073]

[0074] The upper frame vibration performance F31 and the lower frame vibration performance F34 use the same algorithm model.

[0075] The algorithm model of the upper guide bearing swing performance F32 and the lower guide bearing swing performance F35 described in step 2 is:

[0076] make or The algorithm model of the upper guide bearing swing performance F32 and the lower guide bearing swing performance F35 is:

[0077]

[0078] The upper guide bearing swing performance F32 and the lower guide bearing swing performance F35 use the same algorithm model.

[0079] The algorithm model of the water-guided bearing swing performance F39 described in step 2 is:

[0080] make The algorithm model of water-guided bearing swing performance F39 is:

[0081]

[0082] The algorithm model of the top cover vibration performance F38 described in step 2 is:

[0083] make The algorithm model of the top cover vibration performance F38 is:

[0084]

[0085] The algorithm model of the upper guide pad performance F33, the lower guide pad performance F36, the thrust pad performance F37, and the water guide pad performance F310 described in step 2 is:

[0086] Collect the temperature of the i-th bearing of the upper guide, lower guide, thrust or water guide bearing in real time, and let the maximum value of the bearing temperature of the upper guide, lower guide, thrust and water guide bearing in the same period be T max , average value T avg , standard deviation value T sd As the eigenvalue, T i represents the temperature of the i-th bearing pad of the upper guide, lower guide, thrust or water guide bearing, then

[0087] T max =Max(T 1 ,T 2 ,...,T n )

[0088]

[0089]

[0090]

[0091] The upper guide bearing performance F33, the lower guide bearing performance F36, the thrust bearing performance F37 and the water guide bearing performance F310 use the same algorithm model.

[0092] The algorithm model of the spindle seal water-stopping performance F26 described in step 2 is:

[0093] Let x be the time interval between the top cover drainage pump’s stop and start, then x = T 1 -T 2 , the algorithm model of the spindle seal water-stopping performance F26 is:

[0094]

[0095] The algorithm model of the hydraulic performance F27 described in step 2 is:

[0096] Let x = Max(P 尾水进人门 ,P 尾水管出口 ), hydraulic performance F27 score is:

[0097]

[0098] The algorithm model described above is used to obtain an initial performance score, which is discretized and corrected to form an expert knowledge sample library. The characteristic values ​​or operating data in these samples should cover all operating conditions of the equipment to ensure the reliability of the samples. Due to the different operating rules of different equipment, the neural network algorithm is used to train the equipment evaluation samples, and the accuracy of the sample data is continuously iterated and optimized based on the training results, so that the underlying equipment can finally obtain accurate and reliable performance indicators.

[0099] The neural network algorithm uses the independent variable x value in the algorithm model continuously collected or calculated in real time during the operation of the unit as the input condition, and f(x) as the output of the model, and gradually iteratively optimizes the sample data and neural network model parameters. Among them, the hidden layer uses the S-type function as the activation function. The S-type function expression

[0100]

[0101] Where: α is the slope parameter of the S-type function.

[0102] According to the error signal back propagation mechanism of BP neural network, assuming that the input learning sample is X = [x 1 ,…x p ,…,x P ], the output is obtained after the pth sample is input Then, the mean square error E of the sample p Calculate according to the following formula:

[0103]

[0104] In the formula, is the expected output, and m is the number of nodes in the output layer.

[0105] When the mean square error E p When the value is less than 1%, the final model parameters can be obtained. The optimized model can realize the automatic evaluation of equipment performance during the operation of the hydro-generator set.

[0106] According to the performance evaluation indicators of the lowest-level equipment, performance evaluation samples are formulated for the upper-level unit equipment from bottom to top, and then the neural network learning algorithm is used for training iterations, and the stability performance indicators of the hydro-turbine generator set are obtained step by step.

[0107] Example 1

[0108] On November 6, 2021, the operating status data of a certain unit is shown in Table 1. Based on the bottom-level data, the computer imports the automatic evaluation system of the stability performance of the hydropower generator set, and the performance score results of each level of equipment are shown in Table 1:

[0109] Table 1 Unit operating status and performance rating results

[0110]

[0111]

[0112]

[0113] Table 2 Model weight coefficient table

[0114]

[0115] When the performance score of a certain level of equipment is lower than 80 points, the score in the system will turn yellow. When the score is lower than 60 points, the score will turn red, and relevant warning information will be pushed to the system users. The water-stopping performance of the main shaft seal is 79%, indicating that the water leakage of the main shaft seal is too large and needs to be strengthened. The performance of the water guide shoe is 79%, indicating that the temperature of the water guide shoe exceeds the alarm value and needs to be strengthened. When the turbine performance is 88%, it means that the overall operating performance of the turbine is good and can run for a long time without restrictions. When the turbine performance is lower than 80%, it indicates that more attention needs to be paid, and the reasons for the reduction in turbine performance should be checked and repaired at the right time.

[0116] The above are only preferred embodiments of the present invention and are not intended to limit or restrict the present invention. For researchers or technicians in this field, the present invention may have various changes and variations. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection declared by the present invention.

Claims

1. A method for automatically evaluating the stability performance of a hydro-generator set, characterized in that: The following steps are involved: Step 1: Layer the stability performance of the hydro-generator set from top to bottom; The top layer is the stability performance F0 of the hydro-generator set; The second layer is the generator mechanical performance F11 and the turbine performance F12; the third layer is the stator core mechanical performance F21, upper guide bearing performance F22, lower guide bearing performance F23, thrust bearing performance F24 included in the generator mechanical performance F11, and the water guide bearing performance F25, main shaft seal water stop performance F26, hydraulic performance F27 included in the turbine performance F12; the fourth layer is the upper frame vibration performance F31, upper guide bearing swing performance F32, upper guide shoe performance F33 included in the upper guide bearing performance F22, the lower frame vibration performance F34, lower guide bearing swing performance F35, lower guide shoe performance F36 included in the lower guide bearing performance F23, the thrust shoe performance F37 included in the thrust bearing performance F24, the top cover vibration performance F38, water guide bearing swing performance F39, water guide shoe performance F310 included in the water guide bearing performance F25; Step 2, obtain the corresponding data of each branch performance of the fourth layer, the corresponding data of the mechanical performance of the stator core, the corresponding data of the main shaft sealing water-stopping performance, and the corresponding original data of the hydraulic performance through communication, extract characteristic values ​​according to the original data, build an algorithm model corresponding to each branch performance of the fourth layer, the mechanical performance of the stator core, the main shaft sealing water-stopping performance, and the hydraulic performance, and obtain the initial performance score corresponding to each performance according to the algorithm model and the operating data corresponding to the performance, discretize the initial performance score, and obtain the expert knowledge sample library corresponding to each branch performance of the fourth layer, the mechanical performance of the stator core, the main shaft sealing water-stopping performance, and the hydraulic performance, and use the neural network algorithm to train the expert knowledge sample library corresponding to each branch performance of the fourth layer, the mechanical performance of the stator core, the main shaft sealing water-stopping performance, and the hydraulic performance, and obtain the corresponding indicators of each branch performance of the fourth layer, the mechanical performance of the stator core, the main shaft sealing water-stopping performance, and the hydraulic performance; The algorithm model of the stator core mechanical property F21 is: Let x = Max(A 铁芯+x ,A 铁芯-y ), the mechanical performance rating of the stator core is: The algorithm model of the upper frame vibration performance F31 and the lower frame vibration performance F34 is: make or The scores of upper frame vibration performance F31 and lower frame vibration performance F34 are: The upper frame vibration performance F31 and the lower frame vibration performance F34 use the same algorithm model; The algorithm model of the upper guide bearing swing performance F32 and the lower guide bearing swing performance F35 is: make or The scores for the upper guide bearing swing performance F32 and the lower guide bearing swing performance F35 are: The upper guide bearing swing performance F32 and the lower guide bearing swing performance F35 use the same algorithm model; The algorithm model of the water-guided bearing swing performance F39 is: make The scores of the water-guided bearing swing performance F39 are: The algorithm model of the top cover vibration performance F38 is: make The ratings for the top cover vibration performance F38 are: The algorithm model of the upper guide pad performance F33, the lower guide pad performance F36, the thrust pad performance F37, and the water guide pad performance F310 is: Collect the temperature of the i-th bearing of the upper guide, lower guide, thrust or water guide bearing in real time, and let the maximum value of the bearing temperature of the upper guide, lower guide, thrust and water guide bearing in the same period be T max , average value T avg , standard deviation value T sd As the eigenvalue, T i represents the temperature of the i-th bearing pad of the upper guide, lower guide, thrust or water guide bearing, then T max =Max(T1,T2,...,T n ) The upper guide pad performance F33, lower guide pad performance F36, thrust pad performance F37, and water guide pad performance F310 use the same algorithm model; The algorithm model of the spindle seal water-stopping performance F26 is: Let x be the interval between the stop and start of the top cover drainage pump, then x = T1-T2, and the score of the main shaft seal water-stop performance F26 is: The algorithm model of the hydraulic performance F27 is: Let x = Max(P 尾水进人门 ,P 尾水管出口 ), hydraulic performance F27 score is: Step 3: Starting from the fourth layer, from the bottom up, the branch performance is used as the evaluation basis for the upper layer performance. The weighted average model is used to calculate each layer to obtain the initial performance score of the equipment performance at each layer. The initial performance score is discretized to obtain the expert knowledge sample library of the equipment performance at each layer. The neural network algorithm is used to train the expert knowledge sample library of each performance of the equipment at each layer to obtain the stability performance evaluation index of the hydro-turbine generator set and the evaluation index of the equipment at each layer.

2. The method for automatically evaluating the stability performance of a hydro-generator set according to claim 1, characterized in that: The corresponding data of each branch performance on the fourth layer, the corresponding data of the mechanical performance of the stator core, the corresponding data of the main shaft seal water-stopping performance and the corresponding data of the hydraulic performance are the unit bearing bushing temperature data in the computer monitoring system of the whole plant, the start and stop time data of the turbine top cover drainage pump, and the unit operation vibration swing and pressure pulsation data of the unit stability online monitoring system.

3. The method for automatically evaluating the stability performance of a hydro-generator set according to claim 2, characterized in that: The bearing temperature data of the unit includes the bearing temperature data of the upper guide, lower guide, water guide and thrust bearing of the generator set.

4. A method for automatically evaluating the stability performance of a hydro-generator set as claimed in claim 2, characterized in that: The unit operation vibration swing and pressure pulsation data include the peak-to-peak value of the main shaft swing in the +X and -Y directions at the shaft collar of the upper guide, lower guide and water guide bearings of the generator, the peak-to-peak value of the frame vibration in the +X and -Y directions of the upper and lower frames of the generator, the 100Hz component amplitude of the vibration of the generator stator core in the +X and -Y directions, and the peak-to-peak value of the pressure pulsation at the outlet of the pressure measuring pipeline at the lower cavity of the top cover, the tailwater pipe outlet and the tailwater pipe entrance.

5. The method for automatically evaluating the stability performance of a hydro-generator set according to claim 1, characterized in that: The feature values ​​extracted from the original data include: Upper frame vibration performance characteristic value: Upper frame +X direction vibration peak value A 上机架+x , Upper frame - Y-axis vibration peak value A 上机架-y ; Upper guide bearing swing performance characteristic value: Upper guide + X-axis swing peak value A 上导+x , Upward guide - Y-direction swing peak-to-peak value A 上导-y ; Upper guide bearing performance characteristic value: Maximum upper guide bearing bearing temperature T 上导max , the average temperature of the upper guide bearing bushing T 上导avg , Standard deviation value of upper guide bearing bush temperature T 上导sd ; Lower frame vibration performance characteristic value: Lower frame +X direction vibration peak value A 下机架+x , Lower frame - Y-axis vibration peak value A 下机架-y ; Lower guide bearing performance characteristic value: Maximum temperature of lower guide bearing bearing T 下导max , the average temperature of the lower guide bearing bushing T 下导avg , Standard deviation value of lower guide bearing bush temperature T 下导sd ; Lower guide bearing swing performance characteristic value: Lower guide + X-axis swing peak value A 下导+x , Downward guide - Y direction swing peak value A 下导-y ; Thrust bearing performance characteristic value: Maximum thrust bearing temperature T 推力max , Average thrust bearing temperature T 推力avg , thrust bearing bush temperature standard deviation value T 推力sd ; Top cover vibration performance characteristic value: Top cover + X-axis vibration peak value A 顶盖+x , Top cover - Y direction vibration peak value A 顶盖-y , Peak-to-peak value of top cover Z-direction vibration A 顶盖z ; Water-guided bearing swing performance characteristic value: Water-guided + X-axis swing peak value A 水导+x , Water guide - Y direction swing peak value A 水导-y ; Water-guided bearing performance characteristic value: Maximum water-guided bearing bearing temperature T 水导max , Average water-guided bearing bushing temperature T 水导avg , water-guided bearing bush temperature standard deviation value T 水导sd ; Mechanical performance characteristic value of stator core: Stator core + X-axis vibration 100Hz component amplitude A 铁芯+x , Stator core - Y-axis vibration 100Hz component amplitude A 铁芯-y ; Main shaft seal water-stopping performance characteristic values: top cover drainage pump start time T1, top cover last stop time T2; Hydraulic performance characteristic value: Peak-to-peak value of pressure pulsation at the tailwater inlet P 尾水进人门 , peak value of tailwater outlet pressure pulsation P 尾水管出口 .

6. A system for automatically evaluating the stability performance of a hydro-generator set, characterized in that: include: The layered display module layers the stability performance of the hydro-generator units from top to bottom; The top layer is the stability performance F0 of the hydro-generator set; The second layer is the generator mechanical performance F11 and the turbine performance F12; the third layer is the stator core mechanical performance F21, upper guide bearing performance F22, lower guide bearing performance F23, thrust bearing performance F24 included in the generator mechanical performance F11, and the water guide bearing performance F25, main shaft seal water stop performance F26, hydraulic performance F27 included in the turbine performance F12; the fourth layer is the upper frame vibration performance F31, upper guide bearing swing performance F32, upper guide shoe performance F33 included in the upper guide bearing performance F22, the lower frame vibration performance F34, lower guide bearing swing performance F35, lower guide shoe performance F36 included in the lower guide bearing performance F23, the thrust shoe performance F37 included in the thrust bearing performance F24, the top cover vibration performance F38, water guide bearing swing performance F39, water guide shoe performance F310 included in the water guide bearing performance F25; The bottom performance index calculation module obtains the corresponding data of each branch performance of the fourth layer, the corresponding data of the mechanical performance of the stator core, the corresponding data of the main shaft sealing water-stopping performance, and the corresponding data of the hydraulic performance through communication, extracts characteristic values ​​according to the data, builds the algorithm model corresponding to each branch performance of the fourth layer, the mechanical performance of the stator core, the main shaft sealing water-stopping performance, and the hydraulic performance, obtains the corresponding initial performance score according to the algorithm model and the operating data corresponding to the performance, discretizes the initial performance score, obtains the expert knowledge sample library corresponding to each branch performance of the fourth layer, the mechanical performance of the stator core, the main shaft sealing water-stopping performance, and the hydraulic performance, uses the neural network algorithm to train the expert knowledge sample library corresponding to each branch performance of the fourth layer, the mechanical performance of the stator core, the main shaft sealing water-stopping performance, and the hydraulic performance, and obtains the corresponding indicators of each branch performance of the fourth layer, the mechanical performance of the stator core, the main shaft sealing water-stopping performance, and the hydraulic performance; The automatic evaluation module, starting from the fourth layer from the bottom up, uses the branch performance as the evaluation basis for the upper layer performance, calculates layer by layer according to the weighted average model, obtains the initial performance score of the equipment performance of each layer, discretizes the initial performance score, and obtains the expert knowledge sample library of the equipment performance of each layer. The expert knowledge sample library of each performance of each layer of equipment is trained using the neural network algorithm to obtain the stability performance evaluation index of the hydro-turbine generator set and the evaluation index of each layer of equipment.

7. The automatic evaluation system for stability performance of a hydro-generator set according to claim 6, characterized in that: The system displays the performance evaluation indicators of each layer with a score of 0-100, among which 0-50 points indicate serious damage to the equipment; 51-60 points indicate slight damage to the equipment; 61-80 points indicate that the equipment has an alarm and needs to be temporarily shut down; 81-100 points indicate that the equipment can operate stably for a long time.

8. The automatic evaluation system for stability performance of a hydro-generator set according to claim 7, characterized in that: When the performance score of a certain layer of equipment is lower than 80 points, the score in the system will turn yellow. When the score is lower than 60 points, the score will turn red, and relevant warning information will be pushed to the system users.

Images