Generator stator temperature state monitoring method, device, equipment and medium

Abstract

The invention relates to the field of generator operation monitoring, and discloses a generator stator temperature state monitoring method, device, equipment and medium, which can determine a plurality of target operation parameters having high correlation with the stator temperature according to the real operation data of a generator, and determine the temperature state of the generator based on an electromagnetic field finite element model and a stator fluid temperature field finite element model. Determining simulation operation data of the generator under a plurality of simulation operation conditions, and training the to-be-trained model according to parameter values of target operation parameters in the simulation operation data and the stator temperature value to obtain a trained stator temperature prediction model, and monitoring the stator temperature state of the generator based on the trained stator temperature prediction model. According to the invention, the monitoring modes of the generator stator temperature state can be enriched, and diversification of the monitoring modes of the generator stator temperature state is realized.

Description

Technical Field

[0001] The present invention relates to the field of generator operation monitoring, and in particular, to a method, device, equipment and medium for monitoring the stator temperature state of a generator. Background Art

[0002] Large generators play a main role in power production and require high operating reliability.

[0003] Most serious faults of generators are caused by overheating, and the copper loss of the stator winding is one of the main losses of the generator. To ensure the safe operation of the generator, relevant technologies monitor the stator temperature of the generator.

[0004] However, in related technologies, the stator temperature of the generator is generally collected through sensors, and when it is determined that the stator temperature of the generator exceeds the limit, it is determined that the stator temperature of the generator is abnormal and an alarm is given. The monitoring method of the stator temperature state is relatively single. Summary of the Invention

[0005] The present invention provides a method, device, equipment and medium for monitoring the stator temperature state of a generator, so as to solve the defect that the monitoring method of the stator temperature state in related technologies is relatively single, enrich the monitoring method of the stator temperature state of the generator, and realize the diversification of the monitoring method of the stator temperature state of the generator.

[0006] In a first aspect, the present invention provides a method for monitoring the stator temperature state of a generator, including:

[0007] Obtaining the actual operation data, electromagnetic field finite element model and stator fluid temperature field finite element model of the generator;

[0008] According to the actual operation data of the generator, determining a plurality of target operation parameters whose correlation with the stator temperature exceeds a preset correlation threshold among a plurality of operation parameters;

[0009] Based on the electromagnetic field finite element model, the stator fluid temperature field finite element model, each of the target operation parameters and the stator temperature, respectively determining simulation operation data under a plurality of simulated operation conditions, and each of the simulation operation data includes the parameter value of each of the target operation parameters and the stator temperature value;

[0010] Training a model to be trained according to the parameter value of the target operation parameter and the stator temperature value in each of the simulation operation data, so as to obtain a trained stator temperature prediction model;

[0011] Monitoring the stator temperature state of the generator based on the trained stator temperature prediction model.

[0012] Optionally, the true operating data of the generator includes the operating data at multiple operating time points, and the operating data at each operating time point includes the parameter values of each operating parameter and the stator temperature value;

[0013] Determining, according to the true operating data of the generator, multiple target operating parameters with a relevance to the stator temperature exceeding a preset relevance threshold among multiple operating parameters, includes:

[0014] Calculating the relevance of each operating parameter to the stator temperature according to the parameter values of each operating parameter and the stator temperature value in the operating data at each operating time point;

[0015] Determining multiple target relevances exceeding the preset relevance threshold among the relevances of each operating parameter to the stator temperature;

[0016] Determining the operating parameter corresponding to each target relevance as the target operating parameter.

[0017] Optionally, at least two of the multiple target operating parameters include stator voltage, stator current, cooling water flow rate, cooling water temperature, gas flow rate, and gas temperature.

[0018] Optionally, determining the simulation operating data under multiple simulated operating conditions respectively based on the electromagnetic field finite element model, the stator fluid temperature field finite element model, each target operating parameter, and the stator temperature, includes:

[0019] Determining a corresponding solution equation based on the electromagnetic field finite element model, the stator fluid temperature field finite element model, each target operating parameter, and the stator temperature, and the solution equation includes each target operating parameter and the stator temperature;

[0020] For any one of the simulated operating conditions, obtaining the excitation source and boundary conditions corresponding to the simulated operating condition, and determining the simulation operating data under the simulated operating condition based on the electromagnetic field finite element model, the stator fluid temperature field finite element model, the solution equation, the excitation source, and the boundary conditions.

[0021] Optionally, the multiple simulated operating conditions include the normal operating common conditions, abnormal operating conditions, and various fault conditions of the generator.

[0022] Optionally, monitoring the stator temperature state of the generator based on the trained stator temperature prediction model, includes:

[0023] During the operation of the generator, obtaining the target stator temperature value detected by the sensor of the generator;

[0024] If it is determined that the target stator temperature value is not greater than a preset stator temperature threshold, the stator temperature state of the generator is monitored according to the stator temperature prediction model.

[0025] Optionally, the monitoring of the stator temperature state of the generator according to the stator temperature prediction model includes:

[0026] Obtain the operation parameter data of the generator at the operation time point to be monitored, where the operation parameter data includes the parameter values of each target operation parameter at the operation time point to be monitored;

[0027] Input the parameter values of each target operation parameter at the operation time point to be monitored into the stator temperature prediction model for stator temperature prediction, and obtain the stator temperature prediction value output by the stator temperature prediction model;

[0028] Calculate the temperature deviation between the stator temperature prediction value and the target stator temperature value;

[0029] Judge whether the temperature deviation is greater than a preset temperature deviation threshold, where the temperature deviation threshold is the temperature deviation of the generator under fault conditions and normal conditions calculated according to the finite element model;

[0030] If the temperature deviation is greater than the temperature deviation threshold, it is determined that the stator temperature state of the generator is abnormal;

[0031] If the temperature deviation is not greater than the temperature deviation threshold, it is determined that the stator temperature state of the generator is normal.

[0032] In a second aspect, the present invention provides a generator stator temperature state monitoring device, including:

[0033] An acquisition unit for acquiring the actual operation data of the generator, the electromagnetic field finite element model, and the stator fluid temperature field finite element model;

[0034] A parameter determination unit for determining, according to the actual operation data of the generator, a plurality of target operation parameters whose relevance to the stator temperature exceeds a preset relevance threshold among a plurality of operation parameters; wherein, the operation parameters are not the stator temperature;

[0035] A data determination unit for respectively determining the simulation operation data under a plurality of simulated operation conditions based on the electromagnetic field finite element model, the stator fluid temperature field finite element model, each target operation parameter, and the stator temperature, and each simulation operation data includes the parameter values of each target operation parameter and the stator temperature value;

[0036] A model training unit, configured to train a model to be trained according to the parameter values of the target operating parameters and the stator temperature values in each of the simulation operation data, so as to obtain a trained stator temperature prediction model;

[0037] A temperature monitoring unit, configured to monitor the stator temperature state of the generator based on the trained stator temperature prediction model.

[0038] In a third aspect, the present invention provides a computer device, including: a memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to execute the generator stator temperature state monitoring method according to the first aspect or any corresponding embodiment thereof.

[0039] In a fourth aspect, the present invention provides a computer-readable storage medium, on which computer instructions are stored. The computer instructions are used to cause a computer to execute the generator stator temperature state monitoring method according to the first aspect or any corresponding embodiment thereof.

[0040] The generator stator temperature state monitoring method, device, equipment and medium provided by the present invention can determine multiple target operating parameters with high correlation with the stator temperature according to the actual operating data of the generator, determine the simulation operation data of the generator under multiple simulated operating conditions based on the electromagnetic field finite element model and the stator fluid temperature field finite element model, train the model to be trained according to the parameter values of the target operating parameters and the stator temperature values in the simulation operation data, so as to obtain a trained stator temperature prediction model, and monitor the stator temperature state of the generator based on the trained stator temperature prediction model, enrich the monitoring methods of the stator temperature state of the generator, and realize the diversification of the monitoring methods of the stator temperature state of the generator. Description of the Drawings

[0041] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0042] Figure 1 It is a flowchart of a generator stator temperature state monitoring method provided by an embodiment of the present invention;

[0043] Figure 2 It is a structural schematic diagram of an electromagnetic field finite element model of a generator provided by an embodiment of the present invention;

[0044] Figure 3Schematic structural diagram of a finite element model for the fluid temperature field of a generator stator provided by an embodiment of the present invention;

[0045] Figure 4 Flowchart of another method for monitoring the temperature state of a generator stator provided by an embodiment of the present invention;

[0046] Figure 5 Schematic structural diagram of a device for monitoring the temperature state of a generator stator provided by an embodiment of the present invention;

[0047] Figure 6 Schematic structural diagram of a computer device provided by an embodiment of the present invention. Detailed implementation manners

[0048] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without making creative efforts shall fall within the protection scope of the present invention.

[0049] The following is combined with Figures 1 - 4 to describe the method for monitoring the temperature state of the generator stator of the present invention.

[0050] As Figure 1 shown, the first method for monitoring the temperature state of the generator stator is proposed in this embodiment, and the method may include the following steps:

[0051] S101. Obtain the actual operation data, electromagnetic field finite element model, and stator fluid temperature field finite element model of the generator.

[0052] Among them, the actual operation data is the operation data during the on-line operation of the generator. Specifically, the actual operation data may include the parameter values of the operation parameters at each time point during the long-term operation of the generator, and the operation parameters may be voltage, current, active power, reactive power, cooling water flow rate, inlet and outlet temperatures, hydrogen pressure, inlet and outlet temperatures, outlet temperature of the stator bar, and interlayer insulation temperature, etc.

[0053] Specifically, the electromagnetic field finite element model can be constructed in this embodiment based on Maxwell's equations, the boundary conditions of the electromagnetic field in the generator, the excitation current, material properties, geometric model, and the finite element method.

[0054] Specifically, the finite element model of the fluid temperature field of the generator stator can be constructed in this embodiment based on fluid dynamics and fluid-structure interaction theory, the cooling system of the generator, fluid properties, internal heat source distribution, geometric model, and the finite element method.

[0055] S102. Determine, based on the actual operation data of the generator, multiple target operation parameters from multiple operation parameters, where the correlation degree of each of the multiple target operation parameters with the stator temperature exceeds a preset correlation degree threshold.

[0056] Specifically, the multiple operation parameters may be pre-set operation parameters. In this embodiment, those skilled in the art can pre-select, according to experience, operation parameters with a relatively high correlation degree with the stator temperature and those that may have a correlation degree with the stator temperature.

[0057] Specifically, in this embodiment, based on the actual operation data of the generator, target operation parameters with a relatively high correlation degree with the stator temperature can be determined from the set multiple operation parameters.

[0058] Optionally, the actual operation data of the generator includes operation data at multiple operation time points, and the operation data at each operation time point includes the parameter value of each operation parameter and the stator temperature value. At this time, step S102 may include:

[0059] Calculate the correlation degree between each operation parameter and the stator temperature according to the parameter value of each operation parameter and the stator temperature value in the operation data at each operation time point;

[0060] Determine, from the correlation degrees between each operation parameter and the stator temperature, multiple target correlation degrees that exceed the preset correlation degree threshold;

[0061] Determine the operation parameter corresponding to each target correlation degree as a target operation parameter.

[0062] Among them, the operation time point may be a moment or a time period with a specific duration (such as 10 seconds or 20 seconds).

[0063] Specifically, when the operation time point is a moment, for any operation parameter, in this embodiment, the parameter value of the operation parameter at a certain operation time point during the operation of the generator can be collected as the parameter value of the operation parameter at this operation time point.

[0064] Specifically, when the operation time point is a specific duration, in this embodiment, the average parameter value of the operation parameter within this specific duration can be determined as the parameter value of the operation parameter at this operation time point, or the parameter value of the operation parameter can be collected every this specific duration, and the collected parameter value can be used as the parameter value of the operation parameter at this operation time point.

[0065] It can be understood that in this embodiment, the parameter values of each operation parameter collected and determined at multiple operation time points of the generator can be used as the above-mentioned actual operation data as a whole.

[0066] Among them, the preset correlation threshold can be set by technicians according to actual needs. For example, it can be set to 0.9, and the specific size of this embodiment is not limited.

[0067] Specifically, this embodiment can calculate the correlation between each operating parameter and the stator temperature according to the parameter values of each operating parameter and the stator temperature in the above-mentioned actual operating data, as well as according to the calculation formula of the Pearson correlation coefficient or the Spearman correlation coefficient. Among the calculated correlations, determine the correlations greater than the preset correlation threshold as the target correlations. Then, this embodiment can determine the operating parameters with the correlation with the stator temperature being the target correlation as the target operating parameters.

[0068] Optionally, at least two of the above-mentioned multiple target operating parameters include stator voltage, stator current, cooling water flow rate, cooling water temperature, gas flow rate, and gas temperature.

[0069] S103. Based on the electromagnetic field finite element model, the stator fluid temperature field finite element model, each target operating parameter, and the stator temperature, respectively determine the simulation operating data under multiple simulated operating conditions. Each simulation operating data includes the parameter value of each target operating parameter and the stator temperature value.

[0070] Specifically, the simulated operating conditions can correspond to the actual operating conditions of the generator. This embodiment can perform a simulation operation on the generator and collect the simulation operating data of the generator under the simulated operating conditions.

[0071] Optionally, the above-mentioned multiple simulated operating conditions include normal operating common conditions, abnormal operating conditions, and various fault conditions.

[0072] Specifically, this embodiment can perform a simulation operation on the generator under each simulated operating condition, and respectively collect and record the simulation operating data of the generator under each simulated operating condition.

[0073] Optionally, step S103 may include:

[0074] Based on the electromagnetic field finite element model, the stator fluid temperature field finite element model, each target operating parameter, and the stator temperature, determine the corresponding solution equations, and the solution equations include each target operating parameter and the stator temperature;

[0075] For any simulated operating condition, obtain the excitation source and boundary conditions corresponding to the simulated operating condition, and based on the electromagnetic field finite element model, the stator fluid temperature field finite element model, the solution equations, the excitation source, and the boundary conditions, determine the simulation operating data under the simulated operating condition.

[0076] It should be noted that in this embodiment, in a finite element analysis software, different simulated operating conditions can be reflected by defining certain conditions. Specifically, in this embodiment, in the finite element analysis software, the excitation source and boundary conditions can be assigned according to the simulated operating conditions to reflect the simulated operating conditions.

[0077] Among them, the excitation sources in the electromagnetic field finite element model include voltage, current, etc., and the excitation sources in the stator fluid temperature field finite element model include the magnitude of the heat source, etc. In this embodiment, the excitation source and boundary conditions can be assigned according to different simulated operating conditions, and in the finite element analysis software, the assigned excitation source, boundary conditions, the constructed electromagnetic field finite element model and stator fluid temperature field finite element model are combined for simulation analysis to determine the corresponding simulation operation data.

[0078] Among them, the simulation operation data under any simulated operating condition can include the parameter values of each target operating parameter and the stator temperature at multiple simulation operation time points under this simulated operating condition.

[0079] S104. Train the model to be trained according to the parameter values of the target operating parameter and the stator temperature value in each simulation operation data to obtain a trained stator temperature prediction model.

[0080] Among them, the model to be trained can be a model that initially has the performance of temperature prediction. The model type of the model to be trained can be a convolutional neural network model, or a backpropagation (BP) neural network model, or a support vector machine. This embodiment does not limit the model type of the model to be trained.

[0081] Specifically, in this embodiment, the simulation operation data under each simulated operating condition can be used to train the pre-trained stator temperature prediction model until a trained stator temperature prediction model is obtained.

[0082] It should be noted that the input data of the stator temperature prediction model can be the parameter values of each of the above target operating parameters at a certain time point during the actual operation or simulation operation of the generator, and the output data can be the predicted parameter values of the stator temperature at this time point.

[0083] Optionally, the stator temperature prediction model is used to predict the stator outlet water temperature, stator winding temperature, and / or stator interlayer insulation temperature of the generator.

[0084] Specifically, the stator temperature can specifically include the stator outlet water temperature, stator winding temperature, and / or stator interlayer insulation temperature. The stator temperature prediction model can be used to predict the parameter values of the stator outlet water temperature, stator winding temperature, and / or stator interlayer insulation temperature at a certain time point during the actual operation of the generator.

[0085] S105. Monitor the stator temperature status of the generator based on the trained stator temperature prediction model.

[0086] Specifically, in this embodiment, the trained stator temperature prediction model can be used to predict the stator temperature of the generator at each time point during the actual operation, and the stator temperature status can be monitored based on the predicted stator temperature.

[0087] Optionally, step S105 may include:

[0088] During the operation of the generator, obtain the target stator temperature value detected by the sensor of the generator;

[0089] If it is determined that the target stator temperature value is not greater than the preset stator temperature threshold, then monitor the stator temperature status of the generator according to the stator temperature prediction model.

[0090] Specifically, in this embodiment, during the operation of the generator, the stator temperature collected by the sensor can be obtained. Determine whether the stator temperature is greater than the stator temperature threshold. If it is determined that the stator temperature is greater than the stator temperature threshold, it can be directly determined that the stator temperature of the generator is abnormal and an alarm can be issued. If it is determined that the stator temperature is not greater than the stator temperature threshold, the stator temperature prediction model can be used to intervene and monitor the stator temperature status of the generator.

[0091] Optionally, the above-mentioned monitoring of the stator temperature status of the generator according to the stator temperature prediction model includes:

[0092] Obtain the operation parameter data of the generator at the time point to be monitored. The operation parameter data includes the parameter values of each target operation parameter at the time point to be monitored;

[0093] Input the parameter values of each target operation parameter at the time point to be monitored into the stator temperature prediction model for stator temperature prediction, and obtain the stator temperature prediction value output by the stator temperature prediction model;

[0094] Calculate the temperature deviation between the stator temperature prediction value and the target stator temperature value;

[0095] Judge whether the stator temperature status of the generator is abnormal according to the temperature deviation.

[0096] Among them, the stator temperature threshold can be a temperature threshold set by technicians according to the actual situation.

[0097] Specifically, in this embodiment, the stator temperature can be predicted through a stator temperature prediction model to obtain a predicted stator temperature value, and the temperature deviation between the predicted stator temperature value and the stator temperature collected by the sensor is calculated, that is, the absolute value of the difference between the two. Then, this embodiment can determine whether the stator temperature of the generator is abnormal according to this temperature deviation.

[0098] Optionally, determining whether the stator temperature state of the generator is abnormal according to the temperature deviation includes:

[0099] Determine whether the temperature deviation is greater than a preset temperature deviation threshold, and the temperature deviation threshold is the temperature deviation of the generator under fault conditions and normal conditions calculated according to the finite element model;

[0100] If the temperature deviation is greater than the temperature deviation threshold, it is determined that the stator temperature state of the generator is abnormal;

[0101] If the temperature deviation is not greater than the temperature deviation threshold, it is determined that the stator temperature state of the generator is normal.

[0102] Among them, the temperature deviation threshold can also be set by technicians according to the actual situation, and this embodiment does not make a limit.

[0103] Specifically, in this embodiment, when it is determined that the above temperature deviation is greater than the temperature deviation threshold, it is determined that the stator temperature state of the generator is abnormal and an alarm is issued. When it is determined that the above temperature deviation is not greater than the temperature deviation threshold, it is determined that the stator temperature state of the generator is normal.

[0104] It should be noted that this embodiment can compare the magnitudes and change trends of the online operation data and calculation data of the generator temperature, and make a comprehensive analysis for the operation and management personnel in a short time, so as to realize the early warning of the generator thermal fault.

[0105] This embodiment can combine the functions such as the stator temperature value collected by the sensor and parameter over-limit alarm in the related technology, and propose a function of comparing and judging the temperature prediction data by combining the online operation temperature data of the generator and the temperature prediction model of the stator. By judging whether the difference between the calculated value and the operation value of the stator temperature under different working conditions and the threshold, it is determined whether the generator has a temperature fault. This avoids the situation that when the generator has faults such as water blockage and air blockage, the generator does not reach full load operation and the stator temperature does not exceed the limit and no alarm is issued, and provides an early warning for the generator temperature fault to prevent the occurrence or expansion of accidents.

[0106] The generator stator temperature state monitoring method proposed in this embodiment can determine multiple target operating parameters with a high degree of correlation with the stator temperature according to the actual operating data of the generator. Based on the electromagnetic field finite element model and the stator fluid temperature field finite element model, the simulation operating data of the generator under multiple simulated operating conditions is determined. The model to be trained is trained according to the parameter values of the target operating parameters and the stator temperature values in the simulation operating data to obtain a trained stator temperature prediction model. The stator temperature state of the generator is monitored based on the trained stator temperature prediction model, enriching the monitoring methods for the stator temperature state of the generator and realizing the diversification of the monitoring methods for the stator temperature state of the generator.

[0107] In the related art, large generators play a main role in power production and have high requirements for operating reliability. The failure and outage of generators will not only damage the expensive generators themselves, causing huge direct economic losses, but also directly threaten the safety and reliable power supply of the entire power system, resulting in adverse social impacts. Most serious generator failures are caused by overheating, and local overheating often occurs in the early stage of the failure. Some components of the generator overheat due to reasons such as overloading operation, loose contact, or poor cooling effect. If not discovered and solved in time, it will accelerate the aging of the surrounding insulating materials and even lead to serious accidents. The copper loss of the stator winding is one of the main losses of the generator, and its cooling effect is crucial for the safe and stable operation of the generator and the health status of the winding insulation.

[0108] It should also be noted that the related art usually uses the equivalent thermal network method and the three-dimensional finite element method to monitor the stator temperature state of the generator, followed by the motor temperature prediction model based on parameter identification and the fingerprint coefficient method.

[0109] Among them, the equivalent thermal network method cannot take into account the specific temperature distribution inside the generator and can only reflect the average temperature. Especially when a fault occurs at a certain position inside the generator, the equivalent thermal network method cannot be considered for calculation. Although the finite element method has high calculation accuracy and can calculate the temperature distribution inside the generator in detail to find the position of the highest temperature point, the boundary treatment is difficult, the given boundary conditions have a great influence on the results, and it has high requirements for the computer configuration and a long calculation time. The motor temperature prediction model based on parameter identification and the fingerprint coefficient method are greatly affected by the temperature data of the training samples. The online operation historical data of generators are generally common normal operating conditions, and there is little or no operating data for the existence of operating faults such as changes in a small number of cooling media, blockage of internal water channels of the generator, blockage of cooling gases, and large disturbances in the power grid. If the operating conditions are not comprehensively covered, it will have a great impact on the calculation results of the temperature prediction model or the fingerprint coefficient. When the operating conditions of the generator change or a fault occurs, it cannot reflect the true temperature situation of the generator, and there are large errors in the prediction and judgment of the temperature.

[0110] In the related art, the sensors installed in turbo-generators are mainly temperature signal sensors, and the number of them exceeds 80% of all sensors. The sensor data is accessed to the data collection and distribution control system and management information system of the power plant, etc. In the stator temperature monitoring method in the related art, no further analysis and processing are performed on the temperature monitoring data and other related status data, and basically it can only play functions such as reflecting the current value of the monitoring component and parameter over-limit alarm. When a generator fails, in the face of a large amount of on-line monitoring data, it is very difficult for the operating personnel and management personnel to make a comprehensive analysis in a short time and accurately judge the current state of the generator on this basis. In addition, in the thermal fault monitoring method in the related art that judges faults by the sensor data exceeding the alarm threshold, it can only be detected when the fault is very serious, and it is difficult to realize the early warning of potential hazards and avoid the expansion of accidents.

[0111] Therefore, how to perform on-line monitoring and thermal fault diagnosis on the stator winding temperature of large units, and achieve early warning to prevent the occurrence or expansion of accidents, is of great significance.

[0112] Based on Figure 1 , this embodiment proposes a second method for monitoring the stator temperature state of a generator. In this method, the construction process of the electromagnetic field finite element model includes:

[0113] Determine the solution domain and basic assumptions of the electromagnetic field finite element model to be constructed;

[0114] Based on the solution domain and basic assumptions, construct the magnetic potential vector boundary value equation;

[0115] Among them, the solution domain includes stator teeth, stator yoke, upper and lower layer windings of the stator, stator slot wedges, air gap, rotor core, rotor slot wedges, exciting windings and damping windings;

[0116] Among them, the basic assumptions include:

[0117] The electromagnetic field in the generator is uniformly distributed along the axial direction;

[0118] Without considering the end region, the current density vector and the magnetic potential vector only have components in the axial direction;

[0119] The magnetic potential vector on the outer edges of the iron cores of the stator and rotor of the generator is equal to 0;

[0120] The resistivity and magnetic permeability of the materials of the generator are not affected by temperature.

[0121] The construction process of the stator fluid temperature field finite element model includes:

[0122] Determine the solid solution domain and fluid solution domain of the finite element model to be constructed;

[0123] The finite volume method is used to calculate the conservation relations in the solid solution domain and the fluid solution domain to obtain the mass conservation equation, the momentum conservation equation, and the energy conservation equation;

[0124] The fluid motion in the solid solution domain and the fluid solution domain is simulated by the standard turbulence equation to obtain the fluid motion simulation equation;

[0125] The mass conservation equation, the momentum conservation equation, the energy conservation equation, and the fluid motion simulation equation are taken as the finite element model of the stator fluid temperature field as a whole;

[0126] Among them, the solid solution domain includes the upper and lower layer windings of the stator, the main insulation, the interlayer insulation, the slot wedge, the under-wedge spacer, and the stator core;

[0127] The fluid solution domain includes the cooling water in the winding, the core slot wedge, the air gap, the rotor core, the rotor slot wedge, the field winding, and the damper winding.

[0128] Specifically, in this embodiment, the stator copper loss, iron loss, additional loss, etc. of the generator under normal operating conditions, leading power factor operation, and strong excitation operation of the generator can be calculated, and the loss is used as the heat source of the generator fluid and heat transfer model, and the cooling medium is used as the boundary condition of the fluid and heat transfer model to calculate the change of the cooling medium, the stator temperature value of the generator under faults such as water blockage and air blockage, and the model is verified with the online operation historical data of the generator to ensure the correctness of the above finite element model.

[0129] Such as Figure 2 For the electromagnetic field finite element model of the generator shown, the solution domain may include parts such as stator teeth, stator yoke, upper and lower layer windings of the stator, stator slot wedge, air gap, rotor core, rotor slot wedge, field winding, and damper winding.

[0130] When solving the electromagnetic field of the generator, the following basic assumptions are made to simplify the calculation:

[0131] It is assumed that the electromagnetic field in the generator is the same along the axial direction;

[0132] The end region is not considered, and the current density vector J and the magnetic potential vector A have only components in the axial direction;

[0133] The outer edge A of the iron cores of the stator and rotor of the generator is 0;

[0134] The resistivity and magnetic permeability of the materials used in the generator are not affected by temperature.

[0135] In the entire field domain Ω of the generator, the magnetic potential vector A Z Satisfies the following boundary value equation:

[0136]

[0137] where μ is the magnetic permeability; A Z is the component of the vector magnetic potential on the z-axis; J Z is the component of the current density on the z-axis; σ is the electrical conductivity; Γ is the first kind of boundary condition.

[0138] As Figure 3 shown in the finite element model of the stator fluid temperature field, specifically the finite element model of the fluid field and temperature field of the stator winding and core of a large generator with air-cooling and water-cooling structures. The solution domain includes the solid domain and the fluid domain. Among them, the solid domain can include the upper and lower layer windings of the stator, main insulation, interlayer insulation, slot wedges, under-wedge pads, and stator core, etc. The fluid domain can include parts such as the cooling water inside the winding, core slot wedges, air gap, rotor core, rotor slot wedges, field winding, and damping winding.

[0139] Specifically, the local fluid field and temperature field model of the generator stator is calculated using the finite volume method, satisfying the mass conservation equation, momentum conservation equation, and energy conservation equation.

[0140] Among them, the mass conservation equation is:

[0141]

[0142] where ρ is the fluid density; t is the time; V is the fluid velocity vector.

[0143] Among them, the momentum conservation equation is:

[0144]

[0145] where v r , v θ and v z are the relative velocity vectors along r, θ, and z; p is the static pressure acting on the fluid microelement; μ is the viscosity coefficient; S r , S θ and S z are the generalized source terms of the momentum conservation equation.

[0146] Among them, the energy conservation equation is:

[0147]

[0148] where t is the time; θ, r, and z are the circumferential, radial, and axial coordinate components in the solution domain under the cylindrical coordinate system; v θ , v r and v z are the components of the fluid velocity in the circumferential, radial, and axial directions; υ is the kinematic viscosity; υ T is the turbulent kinematic viscosity; p is the static pressure acting on the fluid microelement; T is the temperature; Pr is the Prandtl number; PrT is the turbulent Prandtl number; S T is the ratio of the heat generated by the heat source per unit volume to the specific heat capacity.

[0149] In the fluid region, the Reynolds number Re of the fluid > 2320. The stator solution domain does not involve rotational motion. The standard k-ε turbulence equation is selected to simulate the fluid motion in the solution domain. The fluid motion simulation equation is:

[0150]

[0151] In the formula, k is the turbulent kinetic energy; ε is the diffusion factor; ρ is the fluid density; V is the fluid velocity vector; μ t is the turbulent viscosity coefficient; G k is the turbulent production rate; G 1ε and G 2ε are constants; σ k and σ ε are the turbulent Prandtl constants.

[0152] Specifically, as Figure 4 shown, this embodiment can combine the online operation historical data of the generator and the data results obtained from the multi-physics field modeling of the generator, establish a large dataset with parameters such as the voltage, current, flow rate and temperature of the cooling water and gas of the generator as input parameters, and the stator outlet water temperature and interlayer insulation temperature and other parameters as output parameters. Based on the dataset, an artificial intelligence algorithm calculation model is established using artificial intelligence algorithms including but not limited to neural networks and genetic algorithms, and the generator stator temperature fingerprint model, that is, the stator temperature prediction model, is trained.

[0153] Among them, the online operation historical data of the generator may include the parameter values of operation parameters such as voltage, current, active power, reactive power, cooling medium temperature, bar outlet water temperature and interlayer insulation temperature. This embodiment can perform principal component analysis, correlation analysis, dimensionality reduction processing, etc. on the parameter values of each operation parameter in the online operation historical data of the generator to determine the target operation parameters that have a strong correlation with the stator temperature.

[0154] Among them, in the multi-physics field modeling calculation of the generator, it includes constructing an electromagnetic field model, a fluid field model and a temperature field model. And under various normal operation conditions and fault conditions such as multi-condition, variable cooling condition operation conditions, water blockage and gas blockage, the simulation operation data of the generator is determined.

[0155] Specifically, this embodiment can determine the non-linear input and output parameters and construct a large dataset based on the processed simulation operation data and the parameter values of the target operation parameters, and use the large dataset to train the artificial intelligence algorithm calculation model to obtain the trained generator stator temperature fingerprint model, that is, the stator temperature prediction model.

[0156] Among them, a neural network algorithm is used to train and model a large dataset. The operating data of the generator is used as the training, validation, and test samples of the neural network. Since the input sample values of the stator current, cooling gas, and cooling water flow and temperature and the target sample value of the stator bar outlet temperature are relatively scattered, the numerical span is large, and there are differences in the order of magnitude between the data of each dimension. To avoid training failure or large prediction errors caused by large differences in the order of magnitude of the input and output data, before neural network training and prediction, the sample data is normalized, and all data is normalized to numbers between [0, 1], that is, the data is centered and compressed:

[0157]

[0158] In the formula, i = 1, 2, …, m; j = 1, 2, …, p; n is the number of samples; p is the number of variables; x ij , x ij * are the sample values before and after standardization respectively, is the average value of the sample, s j is the standard deviation of the sample.

[0159] Assume that the output layer has m units, and the unit output vector is Z m =(z 1 , z 2 , …, z m ) T , and the target output vector is T m =(t 1 , t 2 , …, t m ) T , and the error between the network output and the target output is:

[0160]

[0161] Among them, ε δ is the error, and k is the unit serial number.

[0162] Specifically, in this embodiment, it can be determined whether the online operating data of the generator stator temperature exceeds the early warning limit. If it exceeds, it is determined as a generator temperature fault. If the stator temperature is not over the limit, the data online monitored by the generator sensor is used as the input parameter set and input into the stator temperature prediction model. The temperature value calculated by the stator temperature prediction model is compared with the temperature value online monitored by the generator operation. If the calculated temperature deviation from the actual operation is greater than or equal to the threshold value, it is considered that the generator temperature is faulty. If the temperature deviation is less than the threshold value, it is considered that the generator temperature is normal.

[0163] It should be noted that the stator temperature prediction model of this embodiment is built by training the model based on artificial intelligence algorithms. To avoid the singularity of training samples and the lack of data information under other working conditions and faults, a finite element model of multiple physical fields such as generator electromagnetics, fluid, and temperature is established to calculate the stator temperature of the generator under different working conditions, faults, and cooling conditions, and the online operation historical data under typical working conditions of the generator is used for verification and correction. Combining the historical data of the generator's online operation and the data under various working conditions calculated by the finite element model, a data set with voltage, current, cooling water, gas flow rate and temperature, etc. as input parameters and stator temperature as output parameters covering the full working conditions of the generator is established.

[0164] This embodiment can address the deficiencies in the generator temperature modeling technology and temperature monitoring means in the related art, and propose a new large generator stator thermal model (i.e., the above-mentioned finite element model) and thermal monitoring. This model can calculate the normal operation common working conditions, abnormal working conditions, and various faults of the generator, etc., to achieve full coverage of the generator's operating conditions. Through the calculation model, a large data set of input and output parameters of the generator temperature sensitivity parameters is obtained. Based on artificial intelligence algorithms, a stator temperature prediction model of the generator is built to calculate the standard temperature under the online operation conditions, and compare it with the size and change trend of the online operation temperature data to judge, so as to realize the online temperature monitoring and early warning of thermal faults of the generator, and ensure the safe operation of the unit.

[0165] The generator stator temperature state monitoring model proposed in this embodiment can cover the stator temperature change characteristics of the generator under various operating conditions and various faults. The database calculated and operated under full working conditions can provide a relatively comprehensive judgment basis for the monitoring and diagnosis of the generator's thermal faults. This embodiment takes into account both the parameter identification method and the finite element method of the generator. Among them, the online operation data verifies the multi-physical field data of the generator calculated by the finite element method. At the same time, the temperature data under full working conditions calculated by the finite element method is a supplement to the generator operation data.

[0166] As Figure 5 shown, this embodiment proposes a generator stator temperature state monitoring device, which may include:

[0167] An acquisition unit 501, configured to acquire the actual operation data of the generator, the electromagnetic field finite element model, and the stator fluid temperature field finite element model;

[0168] A parameter determination unit 502, configured to determine, according to the actual operation data of the generator, multiple target operation parameters whose correlation with the stator temperature exceeds a preset correlation threshold among multiple operation parameters; wherein, the operation parameters are not the stator temperature;

[0169] A data determination unit 503, configured to respectively determine simulation operation data under multiple simulated operation conditions based on an electromagnetic field finite element model, a stator fluid temperature field finite element model, each target operation parameter, and the stator temperature, where each simulation operation data includes the parameter value of each target operation parameter and the stator temperature value;

[0170] A model training unit 504, configured to train a model to be trained according to the parameter value of the target operation parameter and the stator temperature value in each simulation operation data, so as to obtain a trained stator temperature prediction model;

[0171] A temperature monitoring unit 505, configured to monitor the stator temperature state of the generator based on the trained stator temperature prediction model.

[0172] It should be noted that the processing procedures and the beneficial effects brought by the acquisition unit 501, the parameter determination unit 502, the data determination unit 503, the model training unit 504, and the temperature monitoring unit 505 can respectively refer to Figure 1 Steps S101 to S105 in, which will not be elaborated here.

[0173] Optionally, the actual operation data of the generator includes operation data at multiple operation time points, and the operation data at each operation time point includes the parameter value of each operation parameter and the stator temperature value;

[0174] The parameter determination unit 502 is further configured to:

[0175] Calculate the correlation degree between each operation parameter and the stator temperature according to the parameter value of each operation parameter and the stator temperature value in the operation data at each operation time point;

[0176] Determine multiple target correlation degrees that exceed a preset correlation degree threshold among the correlation degrees between each operation parameter and the stator temperature;

[0177] Determine the operation parameter corresponding to each target correlation degree as the target operation parameter.

[0178] Optionally, at least two of the above multiple target operation parameters include stator voltage, stator current, cooling water flow rate, cooling water temperature, gas flow rate, and gas temperature.

[0179] Optionally, the data determination unit 503 is further configured to:

[0180] Based on the electromagnetic field finite element model, the stator fluid temperature field finite element model, each target operation parameter, and the stator temperature, determine a corresponding solution equation, where the solution equation includes each target operation parameter and the stator temperature;

[0181] For any simulated operating condition, obtain the excitation source and boundary conditions corresponding to the simulated operating condition, and determine the simulation operation data under the simulated operating condition based on the electromagnetic field finite element model, the stator fluid temperature field finite element model, the solution equation, the excitation source, and the boundary conditions.

[0182] Optionally, among the multiple simulated operating conditions, there are the normal operating common conditions, abnormal operating conditions, and various fault conditions of the generator.

[0183] Optionally, the temperature monitoring unit 505 is further configured to:

[0184] During the operation of the generator, obtain the target stator temperature value detected by the sensors of the generator;

[0185] If it is determined that the target stator temperature value is not greater than the preset stator temperature threshold, then monitor the stator temperature state of the generator according to the stator temperature prediction model.

[0186] Optionally, the temperature monitoring unit 505 is further configured to:

[0187] Obtain the operation parameter data of the generator at the operation time point to be monitored, where the operation parameter data includes the parameter values of each target operation parameter at the operation time point to be monitored;

[0188] Input the parameter values of each target operation parameter at the operation time point to be monitored into the stator temperature prediction model for stator temperature prediction, and obtain the stator temperature prediction value output by the stator temperature prediction model;

[0189] Calculate the temperature deviation between the stator temperature prediction value and the target stator temperature value;

[0190] Judge whether the stator temperature state of the generator is abnormal according to the temperature deviation.

[0191] Optionally, the temperature monitoring unit 505 is further configured to:

[0192] Judge whether the temperature deviation is greater than the preset temperature deviation threshold, where the temperature deviation threshold is the temperature deviation of the generator calculated according to the finite element model under the fault condition and the normal condition;

[0193] If the temperature deviation is greater than the temperature deviation threshold, then determine that the stator temperature state of the generator is abnormal;

[0194] If the temperature deviation is not greater than the temperature deviation threshold, then determine that the stator temperature state of the generator is normal.

[0195] The generator stator temperature state monitoring device proposed in this embodiment can determine multiple target operating parameters with a high degree of correlation with the stator temperature based on the actual operating data of the generator. Based on the electromagnetic field finite element model and the stator fluid temperature field finite element model, the simulation operating data of the generator under multiple simulated operating conditions is determined. The model to be trained is trained according to the parameter values of the target operating parameters and the stator temperature values in the simulation operating data, so as to obtain a trained stator temperature prediction model. The stator temperature state of the generator is monitored based on the trained stator temperature prediction model, enriching the monitoring methods for the stator temperature state of the generator and realizing the diversification of the monitoring methods for the stator temperature state of the generator.

[0196] The generator stator temperature state monitoring device in this embodiment is presented in the form of functional units. Here, the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and a memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0197] The embodiment of the present invention also provides a computer device having the above Figure 5 shown generator stator temperature state monitoring device.

[0198] Please refer to Figure 6 , a schematic structural diagram of a computer device provided by an alternative embodiment of the present invention. The computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Each component communicates with each other using different buses and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed within the computer device, including instructions stored in the memory or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some alternative embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a set of blade servers, or a multi-processor system). Figure 6 Taking one processor 10 as an example in

[0199] The processor 10 can be a central processing unit, a network processor, or a combination thereof. Among them, the processor 10 can further include a hardware chip. The above hardware chip can be an application specific integrated circuit, a programmable logic device, or a combination thereof. The above programmable logic device can be a complex programmable logic device, a field programmable gate array, a general array logic, or any combination thereof.

[0200] Among them, the memory 20 stores instructions that can be executed by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiments.

[0201] The memory 20 may include a program storage area and a data storage area. Among them, the program storage area can store an operating system and application programs required for at least one function. The data storage area can store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some alternative embodiments, the memory 20 may optionally include a memory remotely provided with respect to the processor 10, and these remote memories can be connected to the computer device through a network. Examples of the above network include but are not limited to the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.

[0202] The memory 20 may include a volatile memory, such as a random access memory. The memory may also include a non-volatile memory, such as a flash memory, a hard disk, or a solid-state drive. The memory 20 may also include a combination of the above types of memories.

[0203] The computer device further includes a communication interface 30 for the computer device to communicate with other devices or communication networks.

[0204] An embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or be implemented as computer code that can be recorded on a storage medium, or be implemented as computer code originally stored in a remote storage medium or a non-temporary machine-readable storage medium and downloaded through a network and to be stored in a local storage medium, so that the method described herein can be processed by such software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid-state drive, etc.; further, the storage medium may also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code, and when the software or computer code is accessed and executed by the computer, the processor, or the hardware, the method shown in the above embodiments is implemented.

[0205] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for monitoring the temperature status of a generator stator, characterized in that: include: Obtain the actual operation data of the generator, the electromagnetic field finite element model and the stator fluid temperature field finite element model; According to the actual operating data of the generator, a plurality of target operating parameters whose correlation with the stator temperature exceeds a preset correlation threshold are determined from the plurality of operating parameters; Based on the electromagnetic field finite element model, the stator fluid temperature field finite element model, each of the target operating parameters and the stator temperature, respectively determine simulation operation data under a plurality of simulated operation conditions, each of the simulation operation data including a parameter value of each of the target operating parameters and a stator temperature value; Training the model to be trained according to the parameter value of the target operating parameter and the stator temperature value in each of the simulation operation data to obtain a trained stator temperature prediction model; The stator temperature state of the generator is monitored based on the trained stator temperature prediction model.

2. The method according to claim 1, characterized in that The real operation data of the generator includes operation data at multiple operation time points, and the operation data at each operation time point includes a parameter value of each operation parameter and a stator temperature value; The step of determining, based on the actual operating data of the generator, a plurality of target operating parameters whose correlation with the stator temperature exceeds a preset correlation threshold from among the plurality of operating parameters comprises: Calculating the correlation between each of the operating parameters and the stator temperature according to the parameter value of each of the operating parameters in the operating data at each of the operating time points and the stator temperature value; Among the correlations between each of the operating parameters and the stator temperature, a plurality of target correlations exceeding the preset correlation threshold are determined; The operating parameter corresponding to each target relevance is determined as the target operating parameter.

3. The method according to claim 2, characterized in that The multiple target operating parameters include at least two of stator voltage, stator current, cooling water flow, cooling water temperature, gas flow and gas temperature.

4. The method according to claim 1, characterized in that The step of determining the simulation operation data under a plurality of simulation operation conditions based on the electromagnetic field finite element model, the stator fluid temperature field finite element model, each of the target operation parameters and the stator temperature comprises: Determine a corresponding solution equation based on the electromagnetic field finite element model, the stator fluid temperature field finite element model, each of the target operating parameters and the stator temperature, wherein the solution equation includes each of the target operating parameters and the stator temperature; For any of the simulated operating conditions, obtain the excitation source and boundary conditions corresponding to the simulated operating condition, and determine the simulation operation data under the simulated operating condition based on the electromagnetic field finite element model, the stator fluid temperature field finite element model, the solved equations, the excitation source and the boundary conditions.

5. The method according to claim 4, characterized in that The multiple simulated operating conditions include normal operating conditions, unusual operating conditions and various fault conditions of the generator.

6. The method according to any one of claims 1 to 5, characterized in that The monitoring of the stator temperature state of the generator based on the trained stator temperature prediction model includes: When the generator is in operation, obtaining a target stator temperature value detected by a sensor of the generator; If it is determined that the target stator temperature value is not greater than a preset stator temperature threshold, the stator temperature state of the generator is monitored according to the stator temperature prediction model.

7. The method according to claim 6, characterized in that The step of monitoring the stator temperature state of the generator according to the stator temperature prediction model includes: Acquiring operating parameter data of the generator at the operating time point to be monitored, wherein the operating parameter data includes a parameter value of each target operating parameter at the operating time point to be monitored; Inputting the parameter value of each target operating parameter at the operating time point to be monitored into the stator temperature prediction model to perform stator temperature prediction, and obtaining a stator temperature prediction value output by the stator temperature prediction model; Calculating a temperature deviation between the predicted stator temperature value and the target stator temperature value; Determining whether the temperature deviation is greater than a preset temperature deviation threshold, wherein the temperature deviation threshold is a temperature deviation of the generator under a fault condition and a normal condition calculated according to a finite element model; If the temperature deviation is greater than the temperature deviation threshold, determining that the stator temperature state of the generator is abnormal; If the temperature deviation is not greater than the temperature deviation threshold, it is determined that the stator temperature state of the generator is normal.

8. A generator stator temperature status monitoring device, characterized in that: include: An acquisition unit, used for acquiring real operation data of the generator, an electromagnetic field finite element model and a stator fluid temperature field finite element model; A parameter determination unit, configured to determine, according to the actual operating data of the generator, a plurality of target operating parameters whose correlation with the stator temperature exceeds a preset correlation threshold value from among a plurality of operating parameters; wherein the operating parameter is not the stator temperature; A data determination unit, for determining simulation operation data under a plurality of simulated operation conditions respectively based on the electromagnetic field finite element model, the stator fluid temperature field finite element model, each of the target operation parameters and the stator temperature, wherein each of the simulation operation data includes a parameter value of each of the target operation parameters and a stator temperature value; A model training unit, used for training the to-be-trained model according to the parameter value of the target operating parameter and the stator temperature value in each of the simulation operation data, so as to obtain a trained stator temperature prediction model; The temperature monitoring unit is used to monitor the stator temperature state of the generator based on the trained stator temperature prediction model.

9. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the generator stator temperature status monitoring method according to any one of claims 1 to 7 by executing the computer instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the generator stator temperature state monitoring method according to any one of claims 1 to 7.

Images