Dynamic simulation method for multi-stage compressor of compressed air energy storage system

Abstract

The invention relates to the technical field of novel energy, in particular to a multi-stage compressor dynamic simulation method for a compressed air energy storage system, and the method comprises the steps: building a compressed air energy storage system compression side static thermodynamic model frame according to a multi-stage compressor static thermodynamic model and an interstage heat exchanger static thermodynamic model; a pre-trained LSTM neural network is embedded into the compressed air energy storage system compression side static thermodynamic model framework, real-time variable working condition nonlinear characteristic simulation in the system dynamic process is carried out after embedding is completed, and the pre-trained LSTM neural network is obtained by training a preset LSTM neural network through a compressor unit characteristic data set. Therefore, the problem that in the background technology, it is difficult to accurately simulate and optimize the dynamic performance of the multi-stage compressor of the compressed air energy storage system under the variable working conditions is solved, and the performance prediction accuracy of the compressed air energy storage system under the variable working conditions is improved through the dynamic simulation method.

Description

Technical Field

[0001] The present application relates to the field of new energy technologies, and in particular to a dynamic simulation method for a multi-stage compressor of a compressed air energy storage system. Background Art

[0002] With the proposal of the "dual carbon" goal, the large-scale access of renewable energy represented by wind power and photovoltaics to the power grid has brought significant volatility and uncertainty, bringing new challenges to the stable and efficient operation of the new power system. Energy storage technology can store excess electricity when there is a surplus of new energy, and release the stored electricity for load use during peak electricity consumption, effectively smoothing the fluctuations of new energy, which is an effective way to improve the power supply quality of the new power system. In recent years, compressed air energy storage technology, as one of the large-scale energy storage technologies that can be commercially used, is not restricted by geographical conditions, is green, safe and efficient, and has become a hot area of ​​common concern in industry and academia. The construction of related demonstration projects and commercial power stations is also gradually advancing.

[0003] At present, compressed air energy storage power stations are mainly used for new energy consumption and volatility smoothing on the power supply side, in order to improve the external characteristics of renewable energy power stations such as wind power and photovoltaic power, and reduce their impact and influence on the power grid. On the power grid side, in addition to undertaking peak load regulation tasks, they also need to undertake auxiliary services such as frequency regulation and phase regulation. In these application scenarios, since the compression side of the compressed air energy storage system needs to track the fluctuating power output instructions, this causes the operating conditions of the multi-stage compressor units to often deviate from the rated conditions. When the compressor deviates from the rated operating conditions, its pressure ratio and isentropic efficiency will change with the mass flow rate and speed, and there is a mutual coupling relationship between the pipeline network characteristics and different compressors, which further aggravates the variable operating conditions. The variable operating characteristics caused by the coupling of multiple physical fields and multiple components are difficult to accurately characterize, which makes it difficult to optimize the dynamic performance of the compression side of the compressed air energy storage system, which needs to be solved urgently. Summary of the invention

[0004] The present application provides a dynamic simulation method for a multi-stage compressor of a compressed air energy storage system to solve the problem in the background technology that it is difficult to accurately simulate and optimize the dynamic performance of the multi-stage compressor of the compressed air energy storage system under variable operating conditions. The performance prediction accuracy of the compressed air energy storage system under variable operating conditions is improved through the dynamic simulation method.

[0005] The first aspect of the present application provides a method for dynamic simulation of a multi-stage compressor of a compressed air energy storage system, comprising the following steps:

[0006] Construct static thermodynamic model of multi-stage compressor and static thermodynamic model of inter-stage heat exchanger;

[0007] Constructing a compression-side static thermodynamic model framework of a compressed air energy storage system according to the static thermodynamic model of the multi-stage compressor and the static thermodynamic model of the inter-stage heat exchanger;

[0008] A pre-trained LSTM (Long Short-Term Memory) neural network is embedded into the static thermodynamic model framework of the compression side of the compressed air energy storage system, and after the embedding is completed, a real-time variable operating condition nonlinear characteristic simulation is performed in the dynamic process of the system, wherein the pre-trained LSTM neural network is obtained by training a preset LSTM neural network using a compressor unit characteristic data set.

[0009] According to one embodiment of the present application, after performing the real-time variable working condition nonlinear characteristic simulation, the method further includes:

[0010] Acquire multiple sets of time series data of mass flow rate, speed and pressure, set the initial time constant parameter estimation value and the corresponding error covariance matrix, and obtain the prediction data at the current moment based on the system model and the time constant parameter estimation value at the previous moment;

[0011] Calculating a Kalman gain according to the time series acquisition data and the prediction data, and updating an estimated value of a time constant parameter and an error covariance matrix based on the Kalman gain;

[0012] If the estimated value of the updated time constant parameter does not meet the preset convergence condition, the step of obtaining multiple sets of time-series acquisition data of mass flow rate, rotation speed and pressure is re-executed until the estimated value of the updated time constant parameter meets the preset convergence condition, and an inertia link is constructed according to the converged estimated value of the time constant parameter, and the inertia link is embedded in the static thermodynamic model framework of the compression side of the compressed air energy storage system to perform simulation fitting of the dynamic process.

[0013] According to one embodiment of the present application, before the pre-trained LSTM neural network is embedded into the static thermodynamic model framework of the compressed air energy storage system compression side, it also includes:

[0014] Collecting the pressure ratio, inlet and outlet temperatures, mass flow rate and speed of the multi-stage compressor under various working conditions, and constructing a compressor unit characteristic data set according to the pressure ratio, inlet and outlet temperatures, mass flow rate and speed of the multi-stage compressor under various working conditions;

[0015] Preprocessing the compressor unit characteristic data set to obtain a preprocessed compressor unit characteristic data set;

[0016] Dividing the preprocessed compressor unit characteristic data set into a training set and a validation set according to a preset division ratio;

[0017] The preset LSTM neural network is trained using the training set, and after the training is completed, the preset LSTM neural network is verified using the verification set to obtain a verification result, and if the verification result does not meet the preset verification condition, the network structure and / or hyperparameters of the preset LSTM neural network are optimized and adjusted until the verification result of the preset LSTM neural network meets the preset verification condition, thereby obtaining the pre-trained LSTM neural network.

[0018] According to one embodiment of the present application, the construction of a static thermodynamic model of a multi-stage compressor and a static thermodynamic model of an inter-stage heat exchanger includes:

[0019] Calculating the pressure ratio, outlet temperature, and isentropic efficiency of the multi-stage compressor, and calling a first preset physical property library to construct a static thermodynamic model of the multi-stage compressor;

[0020] Based on the compressor interstage heat exchanger equation and calling the second preset physical property library, a static thermodynamic model of the interstage heat exchanger is constructed.

[0021] According to one embodiment of the present application, the compressor interstage heat exchanger equation is:

[0022]

[0023] in, is the outlet air temperature of the heat exchanger, is the inlet air temperature of the heat exchanger, ΔT is the temperature change, is the outlet heat transfer medium temperature, is the inlet temperature of the heat exchange medium, is the derivative of temperature change, τ r is the heat transfer inertia time constant, ε r is the energy efficiency coefficient of the heat exchanger.

[0024] According to the dynamic simulation method of the multi-stage compressor of the compressed air energy storage system of the embodiment of the present application, the static thermodynamic model framework of the compression side of the compressed air energy storage system is constructed according to the static thermodynamic model of the multi-stage compressor and the static thermodynamic model of the inter-stage heat exchanger; the pre-trained LSTM neural network is embedded in the static thermodynamic model framework of the compression side of the compressed air energy storage system, and after the embedding is completed, the real-time variable working condition nonlinear characteristic simulation of the system dynamic process is performed, wherein the pre-trained LSTM neural network is obtained by training the preset LSTM neural network with the compressor unit characteristic data set. Thus, the problem that it is difficult to accurately simulate and optimize the dynamic performance of the multi-stage compressor of the compressed air energy storage system under variable working conditions in the background technology is solved, and the performance prediction accuracy of the compressed air energy storage system under variable working conditions is improved through the dynamic simulation method.

[0025] A second aspect of the present application provides a multi-stage compressor dynamic simulation device for a compressed air energy storage system, comprising:

[0026] The first building module is used to build a static thermodynamic model of a multi-stage compressor and a static thermodynamic model of an inter-stage heat exchanger;

[0027] A second construction module is used to construct a static thermodynamic model framework of the compression side of the compressed air energy storage system according to the static thermodynamic model of the multi-stage compressor and the static thermodynamic model of the inter-stage heat exchanger;

[0028] The simulation module is used to embed a pre-trained LSTM neural network into the static thermodynamic model framework of the compression side of the compressed air energy storage system, and after the embedding is completed, simulate the real-time variable working condition nonlinear characteristics of the system in the dynamic process, wherein the pre-trained LSTM neural network is obtained by training a preset LSTM neural network with a compressor unit characteristic data set.

[0029] According to an embodiment of the present application, after performing the real-time variable operating condition nonlinear characteristic simulation, the simulation module is further used to:

[0030] Acquire multiple sets of time series data of mass flow rate, speed and pressure, set the initial time constant parameter estimation value and the corresponding error covariance matrix, and obtain the prediction data at the current moment based on the system model and the time constant parameter estimation value at the previous moment;

[0031] Calculating a Kalman gain according to the time series acquisition data and the prediction data, and updating an estimated value of a time constant parameter and an error covariance matrix based on the Kalman gain;

[0032] If the estimated value of the updated time constant parameter does not meet the preset convergence condition, the step of obtaining multiple sets of time-series acquisition data of mass flow rate, rotation speed and pressure is re-executed until the estimated value of the updated time constant parameter meets the preset convergence condition, and an inertia link is constructed according to the converged estimated value of the time constant parameter, and the inertia link is embedded in the static thermodynamic model framework of the compression side of the compressed air energy storage system to perform simulation fitting of the dynamic process.

[0033] According to one embodiment of the present application, before the pre-trained LSTM neural network is embedded into the static thermodynamic model framework of the compressed air energy storage system compression side, the simulation module is further used to:

[0034] Collecting the pressure ratio, inlet and outlet temperatures, mass flow rate and speed of the multi-stage compressor under various working conditions, and constructing a compressor unit characteristic data set according to the pressure ratio, inlet and outlet temperatures, mass flow rate and speed of the multi-stage compressor under various working conditions;

[0035] Preprocessing the compressor unit characteristic data set to obtain a preprocessed compressor unit characteristic data set;

[0036] Dividing the preprocessed compressor unit characteristic data set into a training set and a validation set according to a preset division ratio;

[0037] The preset LSTM neural network is trained using the training set, and after the training is completed, the preset LSTM neural network is verified using the verification set to obtain a verification result, and if the verification result does not meet the preset verification condition, the network structure and / or hyperparameters of the preset LSTM neural network are optimized and adjusted until the verification result of the preset LSTM neural network meets the preset verification condition, thereby obtaining the pre-trained LSTM neural network.

[0038] According to one embodiment of the present application, the first building module is used to:

[0039] Calculating the pressure ratio, outlet temperature, and isentropic efficiency of the multi-stage compressor, and calling a first preset physical property library to construct a static thermodynamic model of the multi-stage compressor;

[0040] Based on the compressor interstage heat exchanger equation and calling the second preset physical property library, a static thermodynamic model of the interstage heat exchanger is constructed.

[0041] According to one embodiment of the present application, the compressor interstage heat exchanger equation is:

[0042]

[0043] in, is the outlet air temperature of the heat exchanger, is the inlet air temperature of the heat exchanger, ΔT is the temperature change, is the outlet heat transfer medium temperature, is the inlet temperature of the heat exchange medium, is the derivative of temperature change, τ r is the heat transfer inertia time constant, ε r is the energy efficiency coefficient of the heat exchanger.

[0044] According to the multi-stage compressor dynamic simulation device of the compressed air energy storage system of the embodiment of the present application, a static thermodynamic model framework of the compression side of the compressed air energy storage system is constructed according to the static thermodynamic model of the multi-stage compressor and the static thermodynamic model of the inter-stage heat exchanger; the pre-trained LSTM neural network is embedded in the static thermodynamic model framework of the compression side of the compressed air energy storage system, and after the embedding is completed, the real-time variable working condition nonlinear characteristic simulation of the system dynamic process is performed, wherein the pre-trained LSTM neural network is obtained by training the preset LSTM neural network with the compressor unit characteristic data set. Thus, the problem that it is difficult to accurately simulate and optimize the dynamic performance of the multi-stage compressor of the compressed air energy storage system under variable working conditions in the background technology is solved, and the performance prediction accuracy of the compressed air energy storage system under variable working conditions is improved through the dynamic simulation method.

[0045] The third aspect of the present application provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the multi-stage compressor dynamic simulation method of the compressed air energy storage system as described in the above embodiment.

[0046] The fourth aspect of the present application provides a computer-readable storage medium having a computer program stored thereon, which is executed by a processor to implement the multi-stage compressor dynamic simulation method of the compressed air energy storage system as described in the above embodiment.

[0047] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0049] Figure 1 A flowchart of a multi-stage compressor dynamic simulation method for a compressed air energy storage system provided according to an embodiment of the present application;

[0050] Figure 2 A schematic diagram of a compression-side simulation framework of an embedded LSTM neural network according to an embodiment of the present application;

[0051] Figure 3 Schematic diagram of a block diagram of a multi-stage compressor dynamic simulation device of a compressed air energy storage system according to an embodiment of the present application;

[0052] Figure 4 Schematic diagram of the structure of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0053] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0054] The following describes a method for dynamic simulation of a multi-stage compressor of a compressed air energy storage system according to an embodiment of the present application with reference to the accompanying drawings. In view of the problem that it is difficult to accurately simulate and optimize the dynamic performance of a multi-stage compressor of a compressed air energy storage system under variable working conditions mentioned in the above background technology, the present application provides a method for dynamic simulation of a multi-stage compressor of a compressed air energy storage system, and constructs a static thermodynamic model framework of the compression side of the compressed air energy storage system according to the static thermodynamic model of the multi-stage compressor and the static thermodynamic model of the interstage heat exchanger; embeds a pre-trained LSTM neural network into the static thermodynamic model framework of the compression side of the compressed air energy storage system, and performs a real-time variable working condition nonlinear characteristic simulation in the dynamic process of the system after the embedding is completed, wherein the pre-trained LSTM neural network is obtained by training a preset LSTM neural network with a compressor unit characteristic data set. Thus, the problem that it is difficult to accurately simulate and optimize the dynamic performance of a multi-stage compressor of a compressed air energy storage system under variable working conditions in the background technology is solved, and the performance prediction accuracy of the compressed air energy storage system under variable working conditions is improved by a dynamic simulation method.

[0055] Specifically, Figure 1 A schematic flow chart of a method for dynamic simulation of a multi-stage compressor of a compressed air energy storage system provided in an embodiment of the present application.

[0056] like Figure 1 As shown, the multi-stage compressor dynamic simulation method of the compressed air energy storage system includes the following steps:

[0057] In step S101, a static thermodynamic model of a multi-stage compressor and a static thermodynamic model of an inter-stage heat exchanger are constructed.

[0058] Furthermore, in some embodiments, a static thermodynamic model of a multi-stage compressor and a static thermodynamic model of an inter-stage heat exchanger are constructed, including: calculating the pressure ratio, outlet temperature, and isentropic efficiency of the multi-stage compressor, and calling a first preset physical property library to construct a static thermodynamic model of the multi-stage compressor; based on the compressor inter-stage heat exchanger equation, and calling a second preset physical property library, a static thermodynamic model of the inter-stage heat exchanger is constructed.

[0059] Among them, the first preset physical property library is a real physical property library, and the second preset physical property library is a heat exchange fluid physical property library.

[0060] Specifically, firstly, a static thermodynamic model of a multi-stage compressor is constructed, where the pressure ratio of the k-th stage compressor is:

[0061]

[0062] Among them, β c,k is the pressure ratio of the k-th compressor, is the inlet pressure of the kth compressor, is the outlet pressure of the kth stage compressor.

[0063] Furthermore, the ratio of the outlet temperature to the inlet temperature of the k-th compressor can be approximately calculated based on the pressure ratio:

[0064]

[0065] in, is the inlet air temperature of the kth compressor, is the outlet air temperature of the kth stage compressor, and γ is the multivariate coefficient.

[0066] Furthermore, considering air as an ideal gas, the isentropic efficiency of the k-th compressor can be obtained as:

[0067]

[0068] Among them, η ci,k is the isentropic efficiency of the k-th compressor, is the outlet temperature of the kth compressor isentropic process, is the inlet temperature of the kth compressor isentropic process, is the outlet air temperature of the kth compressor, is the inlet air temperature of the kth stage compressor.

[0069] Combining the above equations (1), (2) and (3), we can get the outlet air temperature and power of the k-th compressor as follows:

[0070]

[0071] in, is the outlet air temperature of the kth compressor, η ci,k is the isentropic efficiency of the k-th compressor, is the inlet air temperature of the kth compressor, β c,k is the compression ratio of the kth compressor, γ is the polytropic coefficient, P c,k is the power of the k-th compressor, η cm,k is the mechanical efficiency, c p,air is the specific heat capacity of air at constant pressure, is the air mass flow rate.

[0072] In the process of constructing a static thermodynamic model of a multi-stage compressor, the specific heat capacity involved can be called by REFPROP. By inputting the temperature and pressure at the corresponding time, the specific heat capacity of the air at the corresponding time can be calculated respectively.

[0073] Secondly, a static thermodynamic model of the interstage heat exchanger is constructed.

[0074] Wherein, in some embodiments, the equation for the compressor interstage heat exchanger is:

[0075]

[0076] in, is the outlet air temperature of the heat exchanger, is the inlet air temperature of the heat exchanger, ΔT is the temperature change, is the outlet heat transfer medium temperature, is the inlet temperature of the heat exchange medium, is the derivative of temperature change, τ r is the heat transfer inertia time constant, ε r is the energy efficiency coefficient of the heat exchanger.

[0077] In the process of constructing the static thermodynamic model of the interstage heat exchanger, the specific heat capacity involved can be called by REFPROP. By inputting the temperature and pressure at the corresponding time, the specific heat capacity of the air and the heat exchange medium at the corresponding time can be calculated respectively.

[0078] Therefore, the embodiment of the present application constructs the functional relationship between pressure and temperature at the inlet and outlet of the compressor based on the pressure ratio and isentropic efficiency respectively, expresses the pressure ratio and isentropic efficiency as implicit functions of the speed and mass flow rate, and introduces a real physical property library to correct the ideal gas process; constructs a thermodynamic static model of the interstage heat exchanger based on the ε-NTU method, and introduces a heat exchange fluid property library to correct the thermal parameters.

[0079] In step S102, a static thermodynamic model framework of the compression side of the compressed air energy storage system is constructed according to the static thermodynamic model of the multi-stage compressor and the static thermodynamic model of the inter-stage heat exchanger.

[0080] Specifically, the embodiment of the present application can combine multi-stage compressor and heat exchanger models based on the boundary mass flow rate and pressure consistency principle to construct a static thermodynamic model framework of the compression side of the compressed air energy storage system, that is, based on the principle that the air mass flow rate at the outlet of each compressor is the same as the mass flow rate at the inlet of the interstage heat exchanger and the pressure at the interface is equal, the compressors at each stage and the interstage heat exchangers are connected to complete the construction of the static thermodynamic model of the compression side of the compressed air energy storage system.

[0081] In step S103, the pre-trained LSTM neural network is embedded in the static thermodynamic model framework of the compression side of the compressed air energy storage system, and after the embedding is completed, the real-time variable operating condition nonlinear characteristics simulation of the system dynamic process is performed, wherein the pre-trained LSTM neural network is obtained by training the preset LSTM neural network with the compressor unit characteristic data set.

[0082] Among them, in some embodiments, before the pre-trained LSTM neural network is embedded in the static thermodynamic model framework of the compression side of the compressed air energy storage system, it also includes: collecting the pressure ratio, inlet and outlet temperature, mass flow rate and speed of the multi-stage compressor under various working conditions, and constructing a compressor unit characteristic data set according to the pressure ratio, inlet and outlet temperature, mass flow rate and speed of the multi-stage compressor under various working conditions; preprocessing the compressor unit characteristic data set to obtain a preprocessed compressor unit characteristic data set; dividing the preprocessed compressor unit characteristic data set into a training set and a verification set according to a preset division ratio; using the training set to train the preset LSTM neural network, and after the training is completed, using the verification set to verify the preset LSTM neural network to obtain a verification result, if the verification result does not meet the preset verification condition, then the network structure and / or hyperparameters of the preset LSTM neural network are optimized and adjusted until the verification result of the preset LSTM neural network meets the preset verification condition, and the pre-trained LSTM neural network is obtained.

[0083] The preset division ratio may be a division ratio preset by a person skilled in the art, such as 7:3, which is not specifically limited here. The preset verification condition may also be a verification condition preset by a person skilled in the art, such as accuracy reaching a preset threshold, which is not specifically limited here.

[0084] Specifically, the embodiments of the present application can collect the measured values ​​of the pressure ratio, inlet and outlet temperature, mass flow rate and speed of the compressor provided by the manufacturer under various working conditions to construct a compressor unit characteristic data set, and its working condition range needs to cover start-stop, rated power operation, wide-condition sliding pressure operation, surge and blockage, etc. The data in the compressor unit characteristic data set is preprocessed, including data cleaning, removal of outliers, data standardization, etc., to ensure the quality of the data and suitability for training neural networks, and the preprocessed compressor unit characteristic data set is divided into a training set and a validation set according to a preset division ratio (such as 7:3) for subsequent use.

[0085] Furthermore, in order to accurately represent the nonlinear characteristics of the multi-stage compressor, a preset LSTM neural network is constructed to nonlinearly fit the coupling relationship between the speed, mass flow rate, pressure ratio, and isentropic efficiency. The network includes an input layer, multiple hidden layers, and an output layer. The input layer is used to receive the speed and mass flow rate time series signals of the compressors at different operating conditions as input; each hidden layer contains 64 hidden units and uses the ReLU (Rectified Linear Unit) activation function; the output layer outputs the pressure ratio and isentropic efficiency of the compressors at all levels. Therefore, the LSTM with this structure can translate the nonlinear characteristics of the compressor and the coupling relationship between the various levels by virtue of its powerful nonlinear time series fitting ability.

[0086] In addition, there are multiple hyperparameters involved in the LSTM neural network. In order to determine the appropriate hyperparameters, the grid search method can be used to optimize the hyperparameters. Specifically, the number of layers, the number of hidden units, and the learning rate of the network are combined to form different network structures. Then, each network structure is trained using the training set, and its performance is evaluated using the validation set. Based on the validation results, the network structure with the best performance is selected as the final model. In addition, the learning rate decay strategy can also be used to gradually reduce the learning rate during the training process to improve the convergence and stability of the model.

[0087] Further, after the preset LSTM neural network is built, the preset LSTM neural network that reflects the nonlinear characteristics of the variable working condition of the compressor is trained. The embodiment of the present application uses the constructed training set to train the preset LSTM neural network. During the training process, the network's weights and biases are iterated and adjusted multiple times through forward propagation and back propagation. In each iteration, the gap between the predicted value and the actual value of the network is calculated, and the weights and biases of the network are updated using the back propagation algorithm. In addition, the batch gradient descent method is used, and a small batch of samples are used for calculation each time the weight is updated to speed up the training process.

[0088] Furthermore, after the training is completed, the preset LSTM neural network model is evaluated using the validation set. If the performance of the preset LSTM neural network model does not meet the preset validation conditions, the network structure and hyperparameters need to be adjusted and optimized so that the neural network can accurately represent the pressure ratio and isentropic efficiency based on the mass flow rate and the rotation speed. Specifically, the number of layers or hidden units of the network can be increased, or the learning rate can be adjusted. In addition, regularization and other techniques can be used to prevent overfitting. Through continuous iteration and optimization, until the performance of the preset LSTM neural network on the validation set meets the preset validation conditions, a pre-trained LSTM neural network is obtained, thereby achieving accurate representation of the nonlinear characteristics of the compressor under variable operating conditions.

[0089] Furthermore, after obtaining the pre-trained LSTM neural network, the trained LSTM neural network model is integrated into the thermodynamic model framework of the compressed air energy storage system to replace the implicit pressure ratio and isentropic efficiency functions, so that the nonlinear operating conditions can be matched in real time by inputting discrete signals of mass flow rate and speed; during iterative calculation, the trained neural network needs to be called in each cycle to realize real-time updating of the variable operating conditions and nonlinear real characteristics of the compressor.

[0090] Furthermore, in some embodiments, after performing the real-time variable operating condition nonlinear characteristic simulation, it also includes: obtaining multiple sets of time-series acquisition data of mass flow rate, speed and pressure, setting the initial time constant parameter estimate and the corresponding error covariance matrix, and obtaining the predicted data at the current moment based on the system model and the time constant parameter estimate at the previous moment; calculating the Kalman gain based on the time-series acquisition data and the predicted data, and updating the estimated value of the time constant parameter and the error covariance matrix based on the Kalman gain; if the updated estimated value of the time constant parameter does not meet the preset convergence condition, re-execute the step of obtaining multiple sets of time-series acquisition data of mass flow rate, speed and pressure until the updated estimated value of the time constant parameter meets the preset convergence condition, construct an inertia link based on the converged time constant parameter estimate, and embed the inertia link into the static thermodynamic model framework of the compression side of the compressed air energy storage system to perform simulation fitting of the dynamic process.

[0091] The preset convergence condition may be a convergence condition preset by a person skilled in the art, and is not specifically limited here.

[0092] Specifically, in the embodiment of the present application, after the trained LSTM neural network is embedded in the pressure ratio and isentropic efficiency functions in the thermodynamic model framework of the compressed air energy storage system to realize the real-time variable operating condition nonlinear characteristic simulation in the dynamic process of the system, the time constant parameters of the system dynamic process are identified based on the actual measured parameters through the Kalman filter algorithm, and a first-order or high-order inertia link is constructed based on the identification constant, which is embedded in the iterative process of pressure, temperature and mass flow rate to realize high-precision fitting modeling of the dynamic response performance of the system, and finally a variable operating condition nonlinear dynamic simulation system of the compression side of the compressed air energy storage system is constructed.

[0093] Exemplarily, first, the embodiment of the present application can collect multiple sets of time-series data of mass flow rate, rotation speed and pressure based on the experimental data provided by the equipment manufacturer and the measured data of the on-site unit, and initialize the Kalman filter algorithm, set the initial time constant parameter estimate and the corresponding error covariance matrix, and then perform the prediction step of the Kalman filter, and predict the system state and parameters at the current moment according to the system model and the time constant parameter estimate at the previous moment. Then perform the update step of the Kalman filter, calculate the Kalman gain according to the actual observation data and the predicted data, and update the estimated value of the time constant parameter and the error covariance matrix.

[0094] Furthermore, the prediction step of the Kalman filter and the update step of the Kalman filter are repeatedly executed until the estimated value of the time constant parameter meets the preset convergence condition. Finally, a first-order or high-order inertia link is constructed according to the estimated value of the time constant parameter that meets the preset convergence condition. The inertia link is embedded in the static thermodynamic model framework of the compressed air energy storage system compression side constructed previously. The state parameters must pass through the inertia link in each cycle update, thereby realizing the simulation fitting of the dynamic process.

[0095] Secondly, a nonlinear dynamic model of the compression side variable working conditions of the compressed air energy storage system is constructed.

[0096] Specifically, based on the constructed static thermodynamic model framework of the compression side of the compressed air energy storage system, a preset LSTM neural network is embedded to reflect the fitting relationship between the speed, mass flow rate, pressure ratio, and isentropic efficiency, and the inertia time constant of the dynamic process of pressure, speed and mass flow rate is fitted through the Kalman filter algorithm. The combination of the above links can jointly constitute a nonlinear dynamic model of the compression side of the compressed air energy storage system with variable operating conditions, and store it in the hardware medium for use.

[0097] Therefore, based on the static thermodynamic model framework of the compression side of the compressed air energy storage system, the present application maps the implicit functions of the pressure ratio and isentropic efficiency with the speed and mass flow rate through a trained LSTM neural network, and identifies the inertia time constant of the dynamic process through the Kalman filter algorithm, thereby achieving accurate simulation of the nonlinear coupled dynamic process of the compression side of the compressed air energy storage system under different working conditions.

[0098] In order to facilitate those skilled in the art to more clearly and intuitively understand the multi-stage compressor dynamic simulation method of the compressed air energy storage system proposed in this application, the following is combined with Figure 2 Provide detailed explanation.

[0099] Specifically, if Figure 2As shown in the figure, first, a static thermodynamic model framework including a compressor and an interstage heat exchanger is established, in which the pressure ratio and isentropic efficiency are modeled as implicit functions of mass flow rate and speed; then, based on the measured values ​​of pressure ratio and isentropic efficiency of the compressor under various working conditions provided by the manufacturer (i.e., test data), a compressor unit characteristic data set is constructed, and the data set is used to train an LSTM neural network representing the nonlinear characteristics of a multi-stage compressor; finally, on the basis of the thermodynamic model framework of the compressed air energy storage system, the trained LSTM neural network is embedded to reflect the fitting relationship between speed, mass flow rate, pressure ratio, and isentropic efficiency, and the time constant parameters of the system dynamic process are identified based on the test data through the Kalman filter algorithm, and a first-order or high-order inertia link is constructed based on the identification constant, and embedded in the iterative process of mass flow rate, speed, pressure, and temperature to realize the simulation modeling of the thermodynamic steady-state and nonlinear variable working condition dynamic characteristics of the compression side of the compressed air energy storage system.

[0100] According to the dynamic simulation method of the multi-stage compressor of the compressed air energy storage system of the embodiment of the present application, the static thermodynamic model framework of the compression side of the compressed air energy storage system is constructed according to the static thermodynamic model of the multi-stage compressor and the static thermodynamic model of the inter-stage heat exchanger; the pre-trained LSTM neural network is embedded in the static thermodynamic model framework of the compression side of the compressed air energy storage system, and after the embedding is completed, the real-time variable working condition nonlinear characteristic simulation of the system dynamic process is performed, wherein the pre-trained LSTM neural network is obtained by training the preset LSTM neural network with the compressor unit characteristic data set. Thus, the problem that it is difficult to accurately simulate and optimize the dynamic performance of the multi-stage compressor of the compressed air energy storage system under variable working conditions in the background technology is solved, and the performance prediction accuracy of the compressed air energy storage system under variable working conditions is improved through the dynamic simulation method.

[0101] Next, a multi-stage compressor dynamic simulation device for a compressed air energy storage system proposed in an embodiment of the present application will be described with reference to the accompanying drawings.

[0102] Figure 3 It is a block diagram of a multi-stage compressor dynamic simulation device of a compressed air energy storage system in an embodiment of the present application.

[0103] like Figure 3 As shown, the multi-stage compressor dynamic simulation device 10 of the compressed air energy storage system includes: a first building module 100, a second building module 200 and a simulation module 300.

[0104] Among them, the first construction module 100 is used to construct a static thermodynamic model of a multi-stage compressor and a static thermodynamic model of an inter-stage heat exchanger; the second construction module 200 is used to construct a static thermodynamic model framework of the compression side of a compressed air energy storage system based on the static thermodynamic model of the multi-stage compressor and the static thermodynamic model of the inter-stage heat exchanger; the simulation module 300 is used to embed a pre-trained LSTM neural network into the static thermodynamic model framework of the compression side of the compressed air energy storage system, and after the embedding is completed, perform a real-time variable operating condition nonlinear characteristic simulation in the dynamic process of the system, wherein the pre-trained LSTM neural network is obtained by training a preset LSTM neural network with a compressor unit characteristic data set.

[0105] Furthermore, in some embodiments, after performing the real-time variable operating condition nonlinear characteristic simulation, the simulation module 300 is also used to: obtain multiple sets of time-series acquisition data of mass flow rate, rotation speed and pressure, set the initial time constant parameter estimate and the corresponding error covariance matrix, and obtain the predicted data at the current moment based on the system model and the time constant parameter estimate at the previous moment; calculate the Kalman gain based on the time-series acquisition data and the predicted data, and update the estimated value of the time constant parameter and the error covariance matrix based on the Kalman gain; if the updated estimated value of the time constant parameter does not meet the preset convergence condition, re-execute the step of obtaining multiple sets of time-series acquisition data of mass flow rate, rotation speed and pressure until the updated estimated value of the time constant parameter meets the preset convergence condition, construct an inertia link based on the converged time constant parameter estimate, and embed the inertia link into the static thermodynamic model framework of the compression side of the compressed air energy storage system to perform simulation fitting of the dynamic process.

[0106] Furthermore, in some embodiments, before the pre-trained LSTM neural network is embedded in the static thermodynamic model framework of the compression side of the compressed air energy storage system, the simulation module 300 is also used to: collect the pressure ratio, inlet and outlet temperature, mass flow rate and speed of the multi-stage compressor under various working conditions, and construct a compressor unit characteristic data set according to the pressure ratio, inlet and outlet temperature, mass flow rate and speed of the multi-stage compressor under various working conditions; pre-process the compressor unit characteristic data set to obtain a pre-processed compressor unit characteristic data set; divide the pre-processed compressor unit characteristic data set into a training set and a verification set according to a preset division ratio; use the training set to train the preset LSTM neural network, and after the training is completed, use the verification set to verify the preset LSTM neural network to obtain a verification result. If the verification result does not meet the preset verification condition, the network structure and / or hyperparameters of the preset LSTM neural network are optimized and adjusted until the verification result of the preset LSTM neural network meets the preset verification condition to obtain the pre-trained LSTM neural network.

[0107] Furthermore, in some embodiments, the first construction module 200 is used to: calculate the pressure ratio, outlet temperature, and isentropic efficiency of the multi-stage compressor, and call the first preset physical property library to construct a static thermodynamic model of the multi-stage compressor; based on the compressor inter-stage heat exchanger equation, and calling the second preset physical property library, construct a static thermodynamic model of the inter-stage heat exchanger.

[0108] Further, in some embodiments, the compressor interstage heat exchanger equation is:

[0109]

[0110] in, is the outlet air temperature of the heat exchanger, is the inlet air temperature of the heat exchanger, ΔT is the temperature change, is the outlet heat transfer medium temperature, is the inlet temperature of the heat exchange medium, is the derivative of temperature change, τ r is the heat transfer inertia time constant, ε r is the energy efficiency coefficient of the heat exchanger.

[0111] It should be noted that the above explanation of the embodiment of the multi-stage compressor dynamic simulation method of the compressed air energy storage system is also applicable to the multi-stage compressor dynamic simulation device of the compressed air energy storage system of this embodiment, and will not be repeated here.

[0112] According to the multi-stage compressor dynamic simulation device of the compressed air energy storage system of the embodiment of the present application, a static thermodynamic model framework of the compression side of the compressed air energy storage system is constructed according to the static thermodynamic model of the multi-stage compressor and the static thermodynamic model of the inter-stage heat exchanger; the pre-trained LSTM neural network is embedded in the static thermodynamic model framework of the compression side of the compressed air energy storage system, and after the embedding is completed, the real-time variable working condition nonlinear characteristic simulation of the system dynamic process is performed, wherein the pre-trained LSTM neural network is obtained by training the preset LSTM neural network with the compressor unit characteristic data set. Thus, the problem that it is difficult to accurately simulate and optimize the dynamic performance of the multi-stage compressor of the compressed air energy storage system under variable working conditions in the background technology is solved, and the performance prediction accuracy of the compressed air energy storage system under variable working conditions is improved through the dynamic simulation method.

[0113] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device may include:

[0114] Memory 401 , processor 402 , and a computer program stored in the memory 401 and executable on the processor 402 .

[0115] When the processor 402 executes the program, the multi-stage compressor dynamic simulation method of the compressed air energy storage system provided in the above embodiment is implemented.

[0116] Furthermore, the electronic device further comprises:

[0117] The communication interface 403 is used for communication between the memory 401 and the processor 402 .

[0118] The memory 401 is used to store computer programs that can be executed on the processor 402 .

[0119] The memory 401 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0120] If the memory 401, the processor 402 and the communication interface 403 are implemented independently, the communication interface 403, the memory 401 and the processor 402 can be connected to each other through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0121] Optionally, in a specific implementation, if the memory 401, the processor 402 and the communication interface 403 are integrated on a chip, the memory 401, the processor 402 and the communication interface 403 can communicate with each other through an internal interface.

[0122] The processor 402 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0123] An embodiment of the present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned multi-stage compressor dynamic simulation method for a compressed air energy storage system.

[0124] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0125] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0126] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for dynamic simulation of a multi-stage compressor of a compressed air energy storage system, characterized in that: The following steps are involved: Construct static thermodynamic model of multi-stage compressor and static thermodynamic model of inter-stage heat exchanger; Constructing a compression-side static thermodynamic model framework of a compressed air energy storage system according to the static thermodynamic model of the multi-stage compressor and the static thermodynamic model of the inter-stage heat exchanger; A pre-trained LSTM neural network is embedded into the static thermodynamic model framework of the compression side of the compressed air energy storage system, and after the embedding is completed, a real-time variable operating condition nonlinear characteristic simulation is performed in the dynamic process of the system, wherein the pre-trained LSTM neural network is obtained by training a preset LSTM neural network with a compressor unit characteristic data set.

2. The method according to claim 1, characterized in that After the real-time nonlinear characteristic simulation of variable working conditions, it also includes: Acquire multiple sets of time series data of mass flow rate, speed and pressure, set the initial time constant parameter estimation value and the corresponding error covariance matrix, and obtain the prediction data at the current moment based on the system model and the time constant parameter estimation value at the previous moment; Calculating a Kalman gain according to the time series acquisition data and the prediction data, and updating an estimated value of a time constant parameter and an error covariance matrix based on the Kalman gain; If the estimated value of the updated time constant parameter does not meet the preset convergence condition, the step of obtaining multiple sets of time-series acquisition data of mass flow rate, rotation speed and pressure is re-executed until the estimated value of the updated time constant parameter meets the preset convergence condition, and an inertia link is constructed according to the converged estimated value of the time constant parameter, and the inertia link is embedded in the static thermodynamic model framework of the compression side of the compressed air energy storage system to perform simulation fitting of the dynamic process.

3. The method according to claim 1, characterized in that Before embedding the pre-trained LSTM neural network into the static thermodynamic model framework of the compressed air energy storage system compression side, it also includes: Collecting the pressure ratio, inlet and outlet temperatures, mass flow rate and speed of the multi-stage compressor under various working conditions, and constructing a compressor unit characteristic data set according to the pressure ratio, inlet and outlet temperatures, mass flow rate and speed of the multi-stage compressor under various working conditions; Preprocessing the compressor unit characteristic data set to obtain a preprocessed compressor unit characteristic data set; Dividing the preprocessed compressor unit characteristic data set into a training set and a validation set according to a preset division ratio; The preset LSTM neural network is trained using the training set, and after the training is completed, the preset LSTM neural network is verified using the verification set to obtain a verification result, and if the verification result does not meet the preset verification condition, the network structure and / or hyperparameters of the preset LSTM neural network are optimized and adjusted until the verification result of the preset LSTM neural network meets the preset verification condition, thereby obtaining the pre-trained LSTM neural network.

4. The method according to claim 1, characterized in that: The construction of the static thermodynamic model of the multi-stage compressor and the static thermodynamic model of the inter-stage heat exchanger includes: Calculating the pressure ratio, outlet temperature, and isentropic efficiency of the multi-stage compressor, and calling a first preset physical property library to construct a static thermodynamic model of the multi-stage compressor; Based on the compressor interstage heat exchanger equation and calling the second preset physical property library, a static thermodynamic model of the interstage heat exchanger is constructed.

5. The method according to claim 4, characterized in that The compressor interstage heat exchanger equation is: in, is the outlet air temperature of the heat exchanger, is the inlet air temperature of the heat exchanger, ΔT is the temperature change, is the outlet heat transfer medium temperature, is the inlet temperature of the heat exchange medium, is the derivative of temperature change, τ r is the heat transfer inertia time constant, ε r is the energy efficiency coefficient of the heat exchanger.

6. A multi-stage compressor dynamic simulation device for a compressed air energy storage system, characterized in that: include: The first building module is used to build a static thermodynamic model of a multi-stage compressor and a static thermodynamic model of an inter-stage heat exchanger; A second construction module is used to construct a static thermodynamic model framework of the compression side of the compressed air energy storage system according to the static thermodynamic model of the multi-stage compressor and the static thermodynamic model of the inter-stage heat exchanger; The simulation module is used to embed a pre-trained LSTM neural network into the static thermodynamic model framework of the compression side of the compressed air energy storage system, and after the embedding is completed, simulate the real-time variable working condition nonlinear characteristics of the system in the dynamic process, wherein the pre-trained LSTM neural network is obtained by training a preset LSTM neural network with a compressor unit characteristic data set.

7. The device according to claim 6, characterized in that After performing the real-time variable operating condition nonlinear characteristic simulation, the simulation module is further used to: Acquire multiple sets of time series data of mass flow rate, speed and pressure, set the initial time constant parameter estimation value and the corresponding error covariance matrix, and obtain the prediction data at the current moment based on the system model and the time constant parameter estimation value at the previous moment; Calculating a Kalman gain according to the time series acquisition data and the prediction data, and updating an estimated value of a time constant parameter and an error covariance matrix based on the Kalman gain; If the estimated value of the updated time constant parameter does not meet the preset convergence condition, the step of obtaining multiple sets of time-series acquisition data of mass flow rate, rotation speed and pressure is re-executed until the estimated value of the updated time constant parameter meets the preset convergence condition, and an inertia link is constructed according to the converged estimated value of the time constant parameter, and the inertia link is embedded in the static thermodynamic model framework of the compression side of the compressed air energy storage system to perform simulation fitting of the dynamic process.

8. The device according to claim 6, characterized in that Before embedding the pre-trained LSTM neural network into the static thermodynamic model framework of the compressed air energy storage system compression side, the simulation module is also used to: Collecting the pressure ratio, inlet and outlet temperatures, mass flow rate and speed of the multi-stage compressor under various working conditions, and constructing a compressor unit characteristic data set according to the pressure ratio, inlet and outlet temperatures, mass flow rate and speed of the multi-stage compressor under various working conditions; Preprocessing the compressor unit characteristic data set to obtain a preprocessed compressor unit characteristic data set; Dividing the preprocessed compressor unit characteristic data set into a training set and a validation set according to a preset division ratio; The preset LSTM neural network is trained using the training set, and after the training is completed, the preset LSTM neural network is verified using the verification set to obtain a verification result, and if the verification result does not meet the preset verification condition, the network structure and / or hyperparameters of the preset LSTM neural network are optimized and adjusted until the verification result of the preset LSTM neural network meets the preset verification condition, thereby obtaining the pre-trained LSTM neural network.

9. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the multi-stage compressor dynamic simulation method of the compressed air energy storage system according to any one of claims 1 to 5.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement a multi-stage compressor dynamic simulation method for a compressed air energy storage system as described in any one of claims 1 to 5.

Images