Turbine differential expansion online monitoring method based on strain measurement

Abstract

The invention provides a steam turbine differential expansion online monitoring method based on strain measurement, and aims to solve the problems that a direct measurement mode involved in the prior art has relatively high requirements on the space size of a measured position, the area size of a measured surface and the actual structure of a unit; and for a large-capacity steam turbine generator unit and a steam turbine low-pressure cylinder with an obvious Poisson effect, the defect that the measuring range of the sensor does not meet the requirement exists. The invention relates to the field of steam turbine differential expansion monitoring, and the method comprises a cylinder support structure strain measurement scheme, a steam turbine differential expansion prediction model, and a steam turbine key position differential expansion monitoring and over-limit early warning method. And differential expansion monitoring and over-limit early warning of the key position of the steam turbine are realized in a manner of combining a strain gauge measurement result with an axial clearance prediction model based on a deep neural network.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of steam turbine expansion difference monitoring, and particularly relates to a steam turbine expansion difference online monitoring method based on strain measurement. BACKGROUND

[0002] Under the working conditions of steam turbine startup, shutdown or deep peak regulation, due to the inconsistent speed of temperature change of the cylinder and the rotor, the axial deformation difference between the cylinder and the rotor occurs, that is, the expansion difference. When the axial expansion deformation of the rotor is greater than that of the cylinder, it is positive expansion difference, and vice versa, it is negative expansion difference. The expansion difference exceeding the limit may cause the rubbing of the moving and static parts of the steam turbine, the increase of the unit vibration, and even the serious accidents such as blade falling and large shaft bending, which seriously affects the safe operation of the steam turbine generator unit. Therefore, the steam turbine expansion difference monitoring is an important project of the steam turbine operation regulation and protection, and the accurate and reliable evaluation method is the key of the expansion difference monitoring.

[0003] At present, the measurement of the steam turbine expansion difference mainly relies on the large-diameter eddy current displacement sensor arranged at the position of the thrust bearing of the steam turbine to realize the measurement. This direct measurement method has high requirements for the space size of the measured position, the area size of the measured surface and the actual structure of the unit, and for the large-capacity steam turbine generator unit and the low-pressure cylinder of the steam turbine with obvious Poisson effect, the sensor range does not meet the requirements. SUMMARY

[0004] In view of the defects of the existing steam turbine expansion difference measurement method, the present application provides a steam turbine expansion difference online monitoring method based on strain measurement. The method realizes the expansion difference monitoring and over-limit early warning of the key position of the steam turbine by combining the strain gauge measurement results with the axial gap prediction model based on deep neural network.

[0005] To achieve the purpose of the present application, the technical scheme adopted is: A steam turbine expansion difference online monitoring method based on strain measurement, including a cylinder support structure strain measurement scheme, a steam turbine expansion difference prediction model, and a steam turbine key position expansion difference monitoring and over-limit early warning method.

[0006] Among them, the cylinder support structure strain measurement scheme uses strain measurement equipment to monitor the strain of the key area of the cylinder outer front and rear support structure in real time; the strain measurement equipment includes a strain gauge, a strain meter and a computer.

[0007] Preferably, the strain gauge in the above strain measurement equipment is a three-axis 45° strain rosette.

[0008] The steam turbine expansion difference prediction model includes the following steps: (1) Determine the boundary conditions of temperature and force of the turbine cylinder and rotor under various operating conditions, and establish a three-dimensional finite element thermal-structure coupling analysis model of the cylinder and rotor; the boundary conditions need to determine the temperature boundary value space of each surface of the turbine cylinder and rotor under all operating conditions such as cold start, warm start, hot start, 15-100% MCR operating condition, normal shutdown and other common conditions and the rotor speed value range ; The subscript 2 of the subscript 2 is the working medium temperature and surface heat transfer coefficient of the temperature boundary, and the subscript m is the number of corresponding regions where the cylinder and rotor may have rubbing.

[0009] (2) Use Latin hypercube sampling method to sample on the operating boundary of the turbine cylinder and rotor, and establish a sample composed of boundary conditions of different operating conditions; the sample size of the boundary condition sample set needs to be determined , each sample is composed of elements, of which elements are temperature boundaries and 1 element is force boundary. Divide the value range of each element into equal probability zones, then randomly take one number from each equal probability zone of different elements that have not been taken out, and repeat times to form the boundary condition sample set .

[0010] (3) Complete the finite element analysis of the cylinder and rotor under all boundary conditions in the sample, extract the Mises equivalent stress of the cylinder support arm as input, and the expansion difference of the interface end surface of the cylinder and the rotor as output, and establish a data set with input and output mapping relationship; the position of the stress of the support structure extracted in the finite element analysis result should correspond to the position of the strain gauge in the strain measurement, and the interface end surface of the cylinder and the rotor should include the axial end surface of the stages of static blades and dynamic blades, front and rear steam seal teeth and shaft shoulder which are prone to rubbing.

[0011] (4) Establish a turbine cylinder and rotor expansion difference prediction model based on deep neural network, and complete the training and testing of the model using the data set; the deep neural network is a multi-layer fully connected neural network, which is composed of an input layer containing neurons, hidden layers and output layers, wherein is the number of strain gauges pasted at the support arm, is the number of interface end surfaces of the cylinder and the rotor, The number of layers of the hidden layer, the number of neurons in each layer are determined, and is usually not less than 64; the data set is divided into a training set and a test set according to a ratio of 7:3, in the process of training the network, the total iteration number is set to 300 rounds, first, the optimizer is set to Adam, and the initial learning rate is set to 0.001, then the learning rate is reduced to 1 / 5 of the original when the training number reaches 100, 180, 230 and 260 steps; after the training of the model is completed, if the prediction accuracy of the difference of the model on the test set is less than 98%, the number of neurons in the hidden layer, the total iteration number, the initial learning rate and other hyperparameters are modified, and the model training is performed again until the accuracy meets the requirements.

[0012] The method for monitoring and over-limit early warning of the difference of expansion of the key position of the steam turbine is to convert the strain measured by the cylinder support structure into stress and input the steam turbine difference of expansion prediction model, that is, to obtain the difference of expansion monitoring result of the key position; different difference of expansion limit values are determined according to the ambient temperature and the running condition, and a flexible difference of expansion over-limit early warning scheme is established.

[0013] The above technical scheme has the following beneficial effects: The application provides an online monitoring method for the difference of expansion of a steam turbine based on strain measurement, which is used for monitoring and over-limit early warning of the difference of expansion of a key position of the steam turbine.Unlike the traditional direct measurement method at the thrust bearing, the application indirectly measures by arranging strain gauges at the cylinder support position, is not limited by the space size of the measured position, the area size of the measured surface and the actual structure of the unit, can realize the difference of expansion monitoring of multiple axial end surfaces of the cylinder and the rotor, the static blades and the dynamic blades of each stage, the front and rear steam seal teeth and the shaft shoulder and the like which are prone to rubbing, has the advantages of convenient installation, rich monitoring data and the like, and can effectively guarantee the safety and stability of the unit in the variable working condition operation process. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 Fig. 1 is a schematic diagram of the arrangement of strain gauges of a low-pressure inner cylinder floor type support; Figure 2 Fig. 2 is a flow chart of the online monitoring method for the difference of expansion of the steam turbine based on strain measurement provided by the application; Figure 3 Fig. 3 is a schematic diagram of the strain gauge measurement data processing mode of the application; Figure 4 Fig. 4 is a strain gauge measurement signal diagram in embodiment 1 of the application; Figure 5 Fig. 5 is a difference of expansion output signal diagram in embodiment 1 of the application. DETAILED DESCRIPTION

[0015] The present invention will be further described in detail below with reference to specific examples. The following description is one application of the present invention, but it is not limited thereto; those skilled in the art can modify the parameters according to specific circumstances.

[0016] Example 1 like Figure 1 The diagram shows a strain gauge arrangement for a low-pressure inner cylinder floor-mounted support. A triaxial 45° strain rosette is arranged on the end face and belly of each support arm, and connected to a computer via a strain gauge. This embodiment combines... Figure 1 and Figure 2 The specific implementation steps are as follows: (1) Considering common operating conditions such as cold start, warm start, hot start, 15%~100% MCR, and normal shutdown, determine the temperature boundary value space of each surface of the low-pressure inner cylinder and rotor under these operating conditions. Rotor speed range The temperature and force boundary conditions of the cylinder and rotor under various operating conditions are determined, and a three-dimensional finite element thermo-mechanical coupling analysis model of the cylinder and rotor is established. In this embodiment, m =12, , .

[0017] (2) Initialize the sample size of the boundary condition sample set In the temperature boundary value space Rotor speed range Latin hypercube sampling was performed to establish a boundary condition sample set. .

[0018] (3) Complete Finite element analysis of the cylinder and rotor under all boundary conditions is performed. The Mises equivalent stress at the location corresponding to the strain rosette position on the low-pressure cylinder support arm in the example is extracted to form the input tensor. The expansion difference between the stationary and moving blades, and between the front and rear steam seal teeth and the shaft shoulder, is extracted to form the output tensor. Establish a dataset with input-output mapping relationships. Table 1 shows the data for a sample in the mapping dataset.

[0019] Table 1 shows the data of a sample in the input-output mapping dataset of this embodiment.

[0020] (4) A prediction model for the differential expansion of the turbine cylinder rotor is established based on a multi-layer fully connected neural network, and the model structure parameters are set as shown in Table 2. The dataset is... The network was divided into training and test sets in a 7:3 ratio. During network training, the total number of iterations was set to 300. The optimizer was initially set to Adam with an initial learning rate of 0.001. After training reached 100, 180, 230, and 260 steps, the learning rate was reduced to 1 / 5 of its original value. After training, if the model's error prediction accuracy on the test set was less than 98%, the model's structural parameters, total number of iterations, initial learning rate, and other hyperparameters were modified, and the model was retrained until the required accuracy was achieved. For example: if the model's accuracy on both the training and test sets is below 90%, you can try increasing the number of hidden layers, increasing the number of neurons in the hidden layers, reducing the Dropout in the hidden layers, increasing the total number of iterations, or increasing the initial learning rate. If the model's accuracy on the training set is above 98%, but its accuracy on the test set is below 95%, you can try reducing the number of hidden layers and neurons, or increasing the Dropout in the hidden layers. If the model's accuracy on the training set is below 60% compared to its accuracy on the test set, you need to check whether the correspondence between the input tensors and output tensors in the dataset is correct.

[0021] Table 2 shows the model structure parameters of the turbine cylinder rotor expansion difference prediction model in this embodiment.

[0022] (5) Deploy the turbine cylinder rotor expansion difference prediction model on the computer; compile the strain data measured by the cylinder support arm for each sampling period of the strain gauge according to... Figure 3 The process is performed as shown to obtain the equivalent stress tensor corresponding to the finite element analysis. The data is then input into the turbine expansion difference prediction model to obtain the expansion difference monitoring results at key locations. .

[0023] (6) Determine different upper limits for expansion difference based on ambient temperature and operating conditions. and lower limit value When the expansion difference monitoring results of the end faces of the stationary blade and the moving blade, the front and rear steam seal teeth and the shaft shoulder, etc. If the differential expansion is outside the specified limits, an alarm will be triggered, and the turbine's operating parameters will be adjusted. Specifically, if the differential expansion monitoring result exceeds the upper alarm limit or falls below the lower alarm limit, on-site technicians must closely monitor the trend of the differential expansion monitoring results and data from other sensors, such as the shaft vibration sensor, during turbine operation. If the monitoring data remains stable, technicians can adjust the start-up or load change rate by changing the opening of the main steam regulating valve; if the monitoring data continues to increase and quickly exceeds the shutdown limit, and the rotor shaft vibration becomes increasingly severe, technicians need to consider an emergency shutdown of the turbine. Figure 4The embodiment shown gives the measurement results of the strain gauge in a period of time, the measurement results are input into a steam turbine cylinder rotor expansion difference prediction model, and expansion difference monitoring results are obtained Figure 5 The expansion difference monitoring results are shown. Figure 5 Taking the results as an example, the expansion difference monitoring results change smoothly and are all located in the expansion difference limit value range, indicating that the expansion coordination of the steam turbine rotor and stator is good and the steam turbine runs stably.

[0024] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application but not limit the present application. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the technical solutions of the present application, and all should be covered in the scope of the claims of the present application.

Claims

1. A method for on-line monitoring of differential expansion in a steam turbine based on strain measurements, characterized by, The application relates to a steam turbine cylinder and rotor expansion difference prediction method. The steam cylinder support structure strain measurement scheme comprises the following steps: real-time monitoring of strains of key regions of front and rear support structures outside a steam cylinder by using a strain measurement device; The steam turbine expansion difference prediction model is established, comprising the following steps: (1) determining boundary conditions of temperatures and forces of the steam turbine cylinder and rotor under various operating conditions, and establishing a three-dimensional finite element thermal-elastic coupling analysis model of the cylinder and rotor; (2) using a Latin hypercube sampling method to sample on the operating boundaries of the steam turbine cylinder and rotor, and establishing a sample composed of boundary conditions of different operating conditions; (3) completing finite element analysis of the cylinder and rotor under all boundary conditions in the sample, extracting Mises equivalent stresses of cylinder support arms as inputs, and extracting expansion differences of each junction end surface of the cylinder and rotor as outputs, and establishing a data set with input and output mapping relationships; (4) establishing a steam turbine cylinder and rotor expansion difference prediction model based on a deep neural network, and completing training and testing of the model by using the data set; The expansion difference monitoring and overrun early warning method of the key positions: the strains measured by the steam cylinder support structure are converted into stresses, and the steam turbine expansion difference prediction model is inputted, so that the expansion difference monitoring results of the key positions are obtained; different expansion difference limit values are determined according to environmental temperatures and operating conditions, and a flexible expansion difference overrun early warning scheme is established.

2. The strain measurement based online monitoring of expansion difference in steam turbines method according to claim 1, characterized in that: The strain measurement device comprises strain gauges, strain meters and a computer.

3. The strain measurement based on-line monitoring of differential expansion in steam turbines method according to claim 2, characterized in that: The strain gauges are triaxial 45-degree strain flowers.

4. The strain measurement based online monitoring of expansion difference in steam turbines method as claimed in claim 1 wherein: The determination of the boundary condition in step (1) is according to cold start, warm start, hot start, 15-100% MCR working condition, normal shutdown working condition, and determination of the temperature boundary value space of each surface of the steam turbine cylinder and rotor in all working conditions and the rotor speed value range ; the subscript 2 of the is the working medium temperature of the temperature boundary and the surface heat transfer coefficient; the subscript m is the corresponding area number of the cylinder and rotor internal collision.

5. The strain measurement based online monitoring of expansion difference in steam turbines method as claimed in claim 1, wherein: The sample composed of the boundary conditions in step (2) is a sample of a sample set of boundary conditions, each sample being composed of elements, wherein one element is a temperature boundary and one element is a force boundary; the value range of each element is evenly divided into equal probability zones, and then one number is randomly taken from the equal probability zones of different elements which have not been taken out, and this is repeated times to form a sample set of boundary conditions .

6. The strain measurement based online monitoring of expansion difference in steam turbines method as claimed in claim 1 wherein: The positions of the support structure stresses extracted from the results of the finite element analysis in step (3) should correspond to the positions of the strain gauges in the strain measurement; the junction end surfaces of the cylinder and the rotor should include axial end surfaces of the stages of static blades and dynamic blades, front and rear steam seal teeth and shaft shoulders and other end surfaces prone to collision and abrasion.

7. The strain measurement based online monitoring of expansion difference in steam turbines method as claimed in claim 1, wherein: The deep neural network in step (4) is a multilayer fully connected neural network.

8. The strain measurement based online monitoring of expansion difference in steam turbines method according to claim 7, characterized in that: The multi-layer fully connected neural network consists of an input layer comprising neurons, a hidden layer comprising neurons, and an output layer comprising neurons.

9. The strain measurement based online monitoring of expansion difference in steam turbines method according to claim 8, characterized in that: The The number of strain gauges attached to the support arm; The number of end faces at the junction of the cylinder and the rotor; The number of hidden layers and the number of neurons in each layer are determined by the number of layers, and should not be less than 64.

10. The strain measurement based online monitoring of expansion difference in steam turbines method as claimed in claim 1 wherein: Step (4) involves dividing the dataset into a training set and a test set in a 7:3 ratio. During network training, the total number of iterations is set to 300 rounds. The optimizer is set to Adam, and the initial learning rate is set to 0.

001. Then, when the number of training steps reaches 100, 180, 230, and 260, the learning rate is reduced to 1 / 5 of the original rate. After the model training is completed, if the model's error prediction accuracy on the test set is less than 98%, the number of hidden layer neurons is modified. The model is retrained until the accuracy meets the requirements, including the total number of iterations, the initial learning rate, and other hyperparameters.

Images