Heavy-duty gas turbine intelligent modeling method based on Mamba large model and Transform network

Abstract

The invention discloses a heavy duty gas turbine intelligent modeling method based on a Mama large model and a Transform network, and belongs to the technical field of electric power engineering, and the method comprises the following steps: S1, building a coupling system model structure according to steam-water flow dynamic characteristics in a peak shaving process of a heavy duty gas turbine, and determining an input variable and an output variable; s2, constructing a peak shaving process model of the heavy duty gas turbine on the basis of a Mamb + Transform multi-mode large model; and S3, training and verifying the peak shaving process model of the heavy duty gas turbine by adopting historical operation data of the heavy duty gas turbine, and completing modeling. According to the method, the characteristic of large span of variable data in the peak shaving process of the heavy duty gas turbine is fully considered, and the generalization ability of the established model is high; the model effectively fuses the characteristics of Mama and Transform, can efficiently process a multivariate long-time sequence, can quickly identify and obtain an accurate industrial control-oriented heavy duty gas turbine peak regulation process model, is beneficial to improving the deep peak regulation capability of a unit, and provides support for high-proportion new energy power consumption.

Description

Technical Field

[0001] The invention relates to the technical field of electric power industry, and in particular to an intelligent modeling method for a heavy-duty gas turbine based on a Mamba large model and a Transformer network. Background Art

[0002] The high penetration of large-scale random wind power generation and photovoltaic power generation in the power grid has made the peak-shaving and frequency-regulating needs of my country's power system increasingly prominent. In this context, the flexible operation of thermal power generating units under all working conditions and the rapid load reduction to the island operation mode with factory power supply have become an inevitable development trend. Heavy-duty gas turbines are highly favored in the peak-shaving market due to their advantages such as high thermal efficiency, fast start and stop, fast load response speed, and high cleanliness.

[0003] However, the premise of rapid start and stop and variable load optimization control of heavy-duty gas turbines is the accurate acquisition of multivariable models of the power generation process. Therefore, developing intelligent modeling methods for heavy-duty gas turbines to promote the participation of heavy-duty gas turbines in peak load regulation is a practical and effective way to improve the safe and stable operation level of the power grid. In order to promote the participation of heavy-duty gas turbines in power grid peak load regulation, it is urgent to build a modeling method that can quickly obtain an accurate heavy-duty gas turbine peak load regulation process model, and this model can serve the design of industrial process controllers. Traditional modeling methods cannot fully consider the large span characteristics of the variable data of the heavy-duty gas turbine peak load regulation process, and cannot effectively integrate multi-source data.

[0004] Based on this, the present invention designs a multivariable sequence intelligent modeling method by integrating Mamba and Transformer large models on the basis of analyzing the dynamic characteristics of the peak-shaving process of heavy-duty gas turbines, so as to improve the deep peak-shaving capability of heavy-duty gas turbines from the perspective of modeling. Summary of the invention

[0005] The purpose of the present invention is to provide a heavy-duty gas turbine intelligent modeling method based on the Mamba large model and the Transformer network to solve the problems in the background technology.

[0006] To achieve the above object, the present invention provides a heavy-duty gas turbine intelligent modeling method based on the Mamba large model and the Transformer network, comprising the following steps:

[0007] S1. According to the dynamic characteristics of the steam-water process in the peak regulation process of heavy-duty gas turbines, the coupling system model structure is constructed to determine the input variables and output variables;

[0008] S2, obtain the multivariable mixed data sequence of the heavy-duty gas turbine peak-shaving process and perform data processing, and build a heavy-duty gas turbine peak-shaving process model based on the Mamba+Transformer multimodal large model;

[0009] S3. Use the historical operating data of heavy-duty gas turbines to train and verify the peak-shaving process model of heavy-duty gas turbines to complete the modeling.

[0010] Preferably, in S1, the input variables are the air intake volume and the air intake guide vane opening during the power generation process of the heavy-duty gas turbine, and the output variables are the outlet pressure of the compressor, the rotor speed, the gas turbine power and the exhaust temperature.

[0011] Preferably, the specific steps of S2 are as follows:

[0012] S21, obtaining a multivariate mixed data sequence of a heavy-duty gas turbine peak-shaving process, performing data construction, pooling operations, and multi-layer perception processing to obtain multi-sequence input data;

[0013] The data construction process designs a unique pre-training mode based on multiple sequences of different dimensions and time scales, so that the model can distinguish the temporal and spatial dependencies between multiple sequences;

[0014] The pooling operation maps the gradually longer input sequence into a fixed-length vector. Adding the pooling operation to the multimodal large model can compress the gradually longer multivariate input sequence to obtain a fixed-length vector that is convenient for subsequent module learning.

[0015] The multi-layer perceptron consists of two linear transformation layers and a nonlinear activation function. The first linear transformation layer increases the dimension of the input sequence so that the network can learn complex tasks. The second linear transformation layer restores the sequence dimension to the same size as the input sequence. Through residual connections and layer normalization, the stable training of the model is ensured. The nonlinear activation function helps the multimodal large model capture complex patterns in long-term multi-sequence input data.

[0016] S22, performing Mamba module processing, Transformer module processing, and normalization processing on the result of S21, and completing the construction of the Mamba+Transformer multimodal large model;

[0017] S23. Expand the input sequence dimension of the Mamba+Transformer multimodal large model to 2, and the output sequence dimension to 4, and complete the construction of the heavy-duty gas turbine peak-shaving process model.

[0018] Preferably, in S22, the processing process of the Mamba module is as follows:

[0019] 1) The state space model is used to process multi-sequence input data, which can be expressed as:

[0020]

[0021] yk =Cx k ;

[0022] Among them, x k is the state variable at the kth moment, x k-1 is the state variable at the k-1th moment, u k The input vector for the model, y k Output vector for the model;

[0023] C is the parameter matrix of the state space model;

[0024]

[0025] Where e is the base of the exponential function, Δ is the increment, and I is the identity matrix;

[0026] 2) Perform convolution kernel calculation on the processed multi-sequence input data as follows:

[0027]

[0028] in, is the convolution kernel, N is the number of rows, is the transpose of the parameter matrix; for The Vandermonde matrix of

[0029] is the extended matrix;

[0030] The purpose is to generate All power forms of , the state vector is mapped to a high-dimensional feature space through power expansion for convolution calculation;

[0031] 3) Output the predicted value, expressed as:

[0032]

[0033] Among them, u is the model input vector and y is the model output vector.

[0034] Preferably, in S22, the specific process of the Transformer model processing is: learning the sequence using a sinusoidal position encoding mode, which is expressed as:

[0035]

[0036] Among them, pos is the position of the current semantics in the sequence, i is the dimension of semantic embedding, and d model is the dimension of the input sequence.

[0037] Preferably, in S22, the sequence processed by the Transformer module enters the Mamba module for further processing to accelerate the learning efficiency of complex sequences, and then the output value is normalized to output the model prediction value. The normalization operation of the prediction value before output is conducive to stabilizing and standardizing the data and ensuring the stable transmission of data between network layers.

[0038] Preferably, in S23, the air intake volume and the opening of the air intake guide vanes during the power generation process of the heavy-duty gas turbine are used as model inputs, and the outlet pressure of the compressor, the rotor speed, the power of the gas turbine and the exhaust temperature are used as model outputs.

[0039] Preferably, in S3, step S2 is used for training and verification, the model prediction value is output and compared with the real data in the historical operation data, and the root mean square error, modeling error, and modeling accuracy are calculated to evaluate the model.

[0040] Preferably, the calculation formula of the root mean square error is:

[0041]

[0042] The calculation formula of modeling error is:

[0043] E(k)=|y(k)-r(k)|;

[0044] The calculation formula for modeling accuracy is:

[0045]

[0046] Among them, RMSE(k) is the root mean square error, E(k) is the modeling error, R 2 is the modeling accuracy, N represents the number of arrays, r(k) is the output variable of the heavy-duty gas turbine peak-shaving process model, and y(k) is the model prediction value.

[0047] Therefore, the heavy-duty gas turbine intelligent modeling method based on the Mamba large model and the Transformer network of the present invention has the following beneficial effects:

[0048] (1) The method protected by the present invention deeply explores the deep peak-shaving capability of heavy-duty gas turbines to promote the construction of a high-proportion new energy power system, fully considers the dynamic characteristics of the steam-water process in the peak-shaving process of heavy-duty gas turbines, and combines the Mamba+Transformer multimodal large model that can quickly and effectively capture the long-term and short-term dependencies of multivariate long-term series to design a multivariate power generation process intelligent modeling method. This method fully considers the large span characteristics of the variable data in the heavy-duty gas turbine peak-shaving process, and the constructed model has the advantage of strong generalization ability.

[0049] (2) The Mamba+Transformer large model designed in this method has a stable selective state space model architecture and strong generalization ability, and can efficiently process multivariate long time series; in addition, the model can quickly identify and obtain accurate heavy-duty gas turbine peak-shaving processes for industrial control, while taking into account the coupling between the variables in the heavy-duty gas turbine peak-shaving process and accurately capturing the mapping relationship, which helps to improve the peak-shaving capacity of heavy-duty gas turbines, improve the deep peak-shaving capacity of the unit, and provide support for the high proportion of new energy power consumption.

[0050] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 A simplified structural schematic diagram of a heavy-duty gas turbine peak-shaving process model according to an embodiment of the present invention;

[0052] Figure 2 This is a flowchart of the Mamba+Transformer multimodal modeling according to an embodiment of the present invention. DETAILED DESCRIPTION

[0053] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0054] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments.

[0055] Example

[0056] This embodiment provides a heavy-duty gas turbine intelligent modeling method based on the Mamba large model and the Transformer network, based on 20,000 groups and 12,000 groups of historical operation data of the actual peak-shaving process of a 300MW F-class heavy-duty gas turbine, the data including the intake volume, intake guide vane opening, compressor outlet pressure, rotor speed, gas turbine power, and exhaust temperature, including the following steps:

[0057] S1. As the power grid's requirements for the depth and speed of heavy-duty gas turbine peak load regulation become higher and higher, the operating load area of ​​heavy-duty gas turbines is becoming wider and wider. As a result, the dynamic characteristics of the unit are becoming more and more obvious. The values ​​of various variables such as air intake, air intake guide vane opening, compressor outlet pressure, gas turbine power, and exhaust temperature vary greatly. The heat resistance of the blades inside the unit and the stable and safe operation of various components face great challenges. Therefore, according to the dynamic characteristics of the steam-water process during the peak load regulation of heavy-duty gas turbines, a simplified structure of the two-input and four-output coupled system model is obtained, such as Figure 1As shown, the input variables are determined to be the intake air volume and intake guide vane opening during the power generation process of the heavy-duty gas turbine, and the output variables are the compressor outlet pressure, rotor speed, gas turbine power and exhaust temperature;

[0058] S2. Obtain the multivariable mixed data sequence of the heavy-duty gas turbine peak-shaving process and perform data processing, and build a heavy-duty gas turbine peak-shaving process model based on the Mamba+Transformer multimodal large model; the details are as follows:

[0059] S21, obtaining a multivariate mixed data sequence of a heavy-duty gas turbine peak-shaving process, performing data construction, pooling operations, and multi-layer perception processing to obtain multi-sequence input data;

[0060] S22. Perform Mamba module processing, Transformer module processing, and normalization processing on the result of S21 to complete the construction of the Mamba+Transformer multimodal large model; integrating Mamba and Transformer to construct the Mamba+Transformer multimodal large model has practical significance for learning the large-span and long-term multivariate series of the heavy gas turbine peak-shaving process.

[0061] The Mamba+Transformer multimodal model includes the Mamba and Transformer models. The sequence from top to bottom passes through the Mamba module and then enters the Transformer module to complete the model establishment of the heavy-duty gas turbine peak-shaving process. The ratio of the Mamba module to the Transformer module in the Mamba+Transformer multimodal model is 7:1. Figure 2 shown.

[0062] The underlying architecture of the Mamba large model is a selective state space model, which is good at processing long time series quickly and does not require additional storage space. The processing process of the Mamba module is as follows:

[0063] 1) The state space model is used to process multi-sequence input data, which can be expressed as:

[0064]

[0065] y k =Cx k ;

[0066] Among them, x k is the state variable at the kth moment, x k-1 is the state variable at the k-1th moment, u k The input vector for the model, y k Output vector for the model;

[0067] C is the parameter matrix of the state space model;

[0068]

[0069]

[0070] Where e is the base of the exponential function, Δ is the increment, and I is the identity matrix;

[0071] 2) Perform convolution kernel calculation on the processed multi-sequence input data as follows:

[0072]

[0073] in, is the convolution kernel, N is the number of rows, is the transpose of the parameter matrix; for The Vandermonde matrix of

[0074] is the extended matrix;

[0075] The purpose is to generate All power forms of , the state vector is mapped to a high-dimensional feature space through power expansion for convolution calculation;

[0076] 3) Output the predicted value, expressed as:

[0077]

[0078] Among them, u is the model input vector and y is the model output vector.

[0079] The Transformer model consists of an attention mechanism and a feedforward neural network, with superior multi-task parallel computing performance and strong generalization ability. The specific process of the Transformer model is: the sequence is learned using a sinusoidal position encoding mode, expressed as:

[0080]

[0081] Among them, pos is the position of the current semantics in the sequence, i is the dimension of semantic embedding, and d model is the dimension of the input sequence.

[0082] The sequence processed by the Transformer module enters the Mamba module for further processing to accelerate the learning efficiency of complex sequences. The output value is then normalized and the model prediction value is output. The normalization operation of the prediction value before output is conducive to stabilizing and standardizing the data and ensuring the stable transmission of data between network layers.

[0083] S23. Expand the input sequence dimension of the Mamba+Transformer multimodal large model to 2, and use the air intake volume and air intake guide vane opening during the power generation process of the heavy-duty gas turbine as the model input; expand the output sequence dimension to 4, and use the compressor outlet pressure, rotor speed, gas turbine power and exhaust temperature as the model output; complete the construction of the heavy-duty gas turbine peak-shaving process model.

[0084] S3. Use 20,000 groups and 12,000 groups of historical operation data of the actual peak-shaving process of a 300MW F-class heavy-duty gas turbine to train and verify the heavy-duty gas turbine peak-shaving process model according to step S2, output the model prediction value and compare it with the actual value of the unit to evaluate the performance of the modeling method.

[0085] Combine all loop model predictions and unit true values ​​to calculate the root mean square error RMSE, modeling error E, and modeling accuracy R 2 The modeling time T can be obtained by adding a timer to the model, and RMSE, E and R 2 The calculation of is as follows:

[0086]

[0087] E(k)=|y(k)-r(k)|;

[0088]

[0089] Where N represents the number of arrays, r(k) is the output variable of the heavy-duty gas turbine peak-shaving process model, namely the actual value of the compressor outlet pressure, rotor speed, gas turbine power and exhaust temperature, and y(k) is the model prediction value.

[0090] Therefore, the present invention provides a heavy-duty gas turbine intelligent modeling method based on the Mamba large model and the Transformer network, deeply explores the deep peak-shaving capability of the heavy-duty gas turbine to promote the construction of a high-proportion new energy power system, fully considers the dynamic characteristics of the steam-water process in the peak-shaving process of the heavy-duty gas turbine, and combines the Mamba+Transformer multimodal large model that can quickly and effectively capture the long-term and short-term dependencies of multivariable long-term series to design a multivariable power generation process intelligent modeling method. This method fully considers the characteristics of the large span of each variable data in the heavy-duty gas turbine peak-shaving process, and the constructed model has the advantage of strong generalization ability.

[0091] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A heavy-duty gas turbine intelligent modeling method based on the Mamba large model and the Transformer network, characterized in that: The following steps are involved: S1. According to the dynamic characteristics of the steam-water process in the peak regulation process of heavy-duty gas turbines, the coupling system model structure is constructed to determine the input variables and output variables; S2, obtain the multivariable mixed data sequence of the heavy-duty gas turbine peak-shaving process and perform data processing, and build a heavy-duty gas turbine peak-shaving process model based on the Mamba+Transformer multimodal large model; S3. Use the historical operating data of heavy-duty gas turbines to train and verify the peak-shaving process model of heavy-duty gas turbines to complete the modeling.

2. According to claim 1, a heavy-duty gas turbine intelligent modeling method based on the Mamba large model and the Transformer network is characterized in that: In S1, the input variables are the intake air volume and intake guide vane opening during the power generation process of the heavy-duty gas turbine, and the output variables are the outlet pressure of the compressor, the rotor speed, the engine power and the exhaust temperature.

3. The heavy-duty gas turbine intelligent modeling method based on the Mamba large model and the Transformer network according to claim 1 is characterized in that: The specific steps of S2 are as follows: S21, obtaining a multivariate mixed data sequence of a heavy-duty gas turbine peak-shaving process, performing data construction, pooling operations, and multi-layer perception processing to obtain multi-sequence input data; S22, performing Mamba module processing, Transformer module processing, and normalization processing on the result of S21, and completing the construction of the Mamba+Transformer multimodal large model; S23. Expand the input sequence dimension of the Mamba+Transformer multimodal large model to 2, and the output sequence dimension to 4 to obtain the heavy-duty gas turbine peak-shaving process model.

4. The heavy-duty gas turbine intelligent modeling method based on the Mamba large model and the Transformer network according to claim 3 is characterized by: In S22, the processing process of the Mamba module is as follows: 1) The state space model is used to process multi-sequence input data, which can be expressed as: y k =Cx k ; Among them, x k is the state variable at the kth moment, x k-1 is the state variable at the k-1th moment, u k The input vector for the model, y k Output vector for the model; C is the parameter matrix of the state space model; Where e is the base of the exponential function, Δ is the increment, and I is the identity matrix; 2) Perform convolution kernel calculation on the processed multi-sequence input data as follows: in, is the convolution kernel, N is the number of rows, is the transpose of the parameter matrix, for The Vandermonde matrix of is the extended matrix; 3) Output the predicted value, expressed as: Among them, u is the model input vector and y is the model output vector.

5. The heavy-duty gas turbine intelligent modeling method based on the Mamba large model and the Transformer network according to claim 4 is characterized in that: In S22, the specific process of the Transformer model processing is: learning the sequence using the sinusoidal position encoding mode, which is expressed as: Among them, pos is the position of the current semantics in the sequence, i is the dimension of semantic embedding, and d model is the dimension of the input sequence.

6. The heavy-duty gas turbine intelligent modeling method based on the Mamba large model and the Transformer network according to claim 5 is characterized in that: In S22, the sequence processed by the Transformer module enters the Mamba module for further processing, and then the output value is normalized to output the model prediction value.

7. The heavy-duty gas turbine intelligent modeling method based on the Mamba large model and the Transformer network according to claim 6 is characterized by: In S23, the intake volume and intake guide vane opening during the power generation process of the heavy-duty gas turbine are used as model inputs, and the outlet pressure of the compressor, the rotor speed, the engine power and the exhaust temperature are used as model outputs.

8. The heavy-duty gas turbine intelligent modeling method based on the Mamba large model and the Transformer network according to claim 1 is characterized by: In S3, step S2 is used for training and verification, the model prediction value is output and compared with the real data in the historical operation data, and the root mean square error, modeling error, and modeling accuracy are calculated to evaluate the model.

9. The heavy-duty gas turbine intelligent modeling method based on the Mamba large model and the Transformer network according to claim 8 is characterized in that: The calculation formula of the root mean square error is: The calculation formula of modeling error is: E(k)=|y(k)-r(k)|; The calculation formula for modeling accuracy is: Among them, RMSE(k) is the root mean square error, E(k) is the modeling error, R 2 is the modeling accuracy, N represents the number of arrays, r(k) is the output variable of the heavy-duty gas turbine peak-shaving process model, and y(k) is the model prediction value.

Images