A method and apparatus for predicting compressor surge boundaries

Abstract

A method and device for predicting a compressor surge boundary, the method comprising: constructing a compressor model for the compressor itself, and constructing a pipeline volume model for a pipeline volume system downstream of the compressor; solving the compressor model and the pipeline volume model to obtain simulation results of the compressor at multiple operating points; and determining whether the operating points cause a surge according to a flow rate and a pressure ratio of the compressor in the simulation results; wherein the compressor model is a three-dimensional volume force model, and the pipeline volume model is a coupling of a one-dimensional model and a lumped parameter model. The embodiments of the present disclosure improve the prediction accuracy and calculation efficiency of the compressor surge.

Description

[0001] Cross-reference to related applications

[0002] This disclosure claims priority to Chinese patent application No. 202210523898.5, filed with the China National Intellectual Property Administration on May 13, 2022, entitled "A Method and Apparatus for Predicting Compressor Surge Boundary", the contents of which are incorporated herein by reference. Technical Field

[0003] This disclosure relates to, but is not limited to, compressors; more specifically, it relates to a method and apparatus for predicting compressor surge boundaries. Background Technology

[0004] Air compressors are crucial components of aircraft engines, vehicle turbochargers, and industrial compressors. Through the work done by their rotating parts, air compressors convert mechanical energy into the kinetic and potential energy of the pressurized gas. However, due to the limitations of the compressor blades' pressurization capacity, when the operating flow rate decreases to a certain value, the compressor may experience surge instability, leading to a decline in performance and even damage to its structural integrity.

[0005] To avoid surge instability during operation, it is crucial to know the compressor's surge boundary in advance—that is, under what operating conditions the compressor will experience surge. While experimental measurements can obtain relatively accurate surge boundary information, they are costly, time-consuming, and risky. In contrast, numerical simulation methods for predicting the surge boundary are low-cost, fast, and eliminate the risks associated with experiments. However, all currently proposed numerical simulation methods suffer from significant deviations in prediction accuracy compared to experimental results. Summary of the Invention

[0006] The following is an overview of the subject matter described in detail herein. This overview is not intended to limit the scope of the claims.

[0007] One embodiment of this disclosure provides a method for predicting compressor surge boundary, including:

[0008] Build a compressor model for the compressor itself, and a pipeline cavity model for the pipeline cavity system downstream of the compressor;

[0009] Solve the compressor model and the pipeline cavity model to obtain the simulation results of the compressor at multiple operating points;

[0010] Determine whether surge occurs at the operating point based on the compressor's flow rate and pressure ratio as described in the simulation results;

[0011] The compressor model is a three-dimensional volume force model, and the pipeline cavity model is a coupling of a one-dimensional model and a lumped parameter model. The one-dimensional model is constructed based on the compressor outlet pipeline, and the lumped parameter model is constructed based on the cavity and valves downstream of the compressor.

[0012] An embodiment of this disclosure also provides a compressor surge boundary prediction device, including a processor and a memory storing a computer program, wherein when the processor executes the computer program, it can implement the compressor surge boundary prediction method as described in any embodiment of this disclosure.

[0013] The embodiments disclosed herein can directly simulate surge phenomena and consider the influence of the cavity and piping system on compressor surge prediction. At the same time, the method uses a volume force model to accelerate the calculation speed of the full three-dimensional numerical simulation method, and uses a method of coupling one-dimensional and three-dimensional models to parameterize the pipeline cavity system downstream of the compressor, thereby enabling compressor surge prediction both quickly and accurately.

[0014] After reading and understanding the accompanying diagrams and detailed descriptions, other aspects can be understood. Attached Figure Description

[0015] Figure 1 This is a flowchart of a method for predicting compressor surge boundary according to an embodiment of this disclosure;

[0016] Figure 2 This is a schematic diagram of a compressor surge boundary prediction device according to an embodiment of the present disclosure;

[0017] Figure 3 This is a block diagram of a compression system model according to an embodiment of this disclosure;

[0018] Figure 4 This is a schematic diagram of a three-dimensional volumetric force model of a compressor according to an embodiment of this disclosure;

[0019] Figure 5A and Figure 5B This is a schematic diagram illustrating the bidirectional information transfer between a compressor model (three-dimensional model) and a pipeline model (one-dimensional model) according to an embodiment of this disclosure;

[0020] Figure 6 This is a schematic diagram of a stable point and surge boundary determined by compressor speed and valve opening coefficient as control parameters according to an embodiment of this disclosure;

[0021] Figure 7 This is a schematic diagram of an exemplary surge coil obtained according to an embodiment of the present disclosure. Detailed Implementation

[0022] The technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this disclosure, and not all of them. Based on the embodiments of this disclosure, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this disclosure.

[0023] In the description of this disclosure, the words "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment described as "exemplary" or "for example" in this disclosure should not be construed as being more preferred or advantageous than other embodiments. The word "and / or" in this document describes an association relationship between related objects, indicating that three relationships may exist, for example, A and / or B, which can represent: A alone, A and B simultaneously, and B alone. In the description of this disclosure, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0024] The technical solutions of the various embodiments of this disclosure can be combined with each other, but are based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection claimed by this disclosure.

[0025] Currently, the accuracy of numerical simulation methods for predicting surge boundaries deviates significantly from experimental results. The main reasons for this are: 1) Traditional full 3D numerical simulation methods cannot directly simulate surge phenomena, thus relying solely on numerical divergence as the criterion for surge determination; 2) Lumped parameter methods require the user to provide the complete characteristic curve of the compressor as input, which often requires complex experiments, naturally introducing errors into the model-based prediction, fitting, and interpolation methods; 3) The surge boundary location is significantly affected by the compressor outlet cavity and piping dimensions, while prediction methods often fail to consider the influence of the actual downstream system of the compressor. Furthermore, as the number of compressor stages increases, the computational cost and time for prediction using simulation methods increase significantly, greatly diminishing the computational economy and efficiency advantages of simulation.

[0026] In view of this, one embodiment of this disclosure provides a method for predicting compressor surge boundary, such as... Figure 1 As shown, it includes:

[0027] Step 110: Construct a compressor model for the compressor itself, and construct a pipeline cavity model for the pipeline cavity system downstream of the compressor.

[0028] Step 120: Solve the compressor model and the pipeline cavity model to obtain the simulation results of the compressor at multiple operating points;

[0029] Step 130: Determine whether surge occurs at the operating point based on the compressor's flow rate and pressure ratio as described in the simulation results;

[0030] The compressor model is a three-dimensional volume force model, and the pipeline cavity model is a coupling of a one-dimensional model and a lumped parameter model. The one-dimensional model is constructed based on the compressor outlet pipeline, and the lumped parameter model is constructed based on the cavity and valves downstream of the compressor.

[0031] In this embodiment, the compressor and its downstream piping cavity system can be collectively referred to as the compression system, such as... Figure 3 As shown, the compression system model 30 established for this compression system includes a compressor model 301 based on the compressor itself, and a pipeline cavity model based on the pipeline cavity system downstream of the compressor. The compressor model 301 can be a three-dimensional volumetric force model, solved using the finite volume method. The pipeline cavity model includes a pipeline model 303 based on the compressor outlet pipeline and a cavity valve model 305 based on the cavity and valves downstream of the compressor. The pipeline model 303 is a one-dimensional model, while the cavity valve model 305 is a lumped parameter model. The pipeline model 303 can be solved using the method of characteristics, but is not limited to this method. The cavity valve model is solved coupled to the outlet boundary conditions of the pipeline model.

[0032] Taking a centrifugal compressor as an example, when constructing a three-dimensional volumetric force model for the compressor, multiple grid points are set in both the blade region and the non-blade region of the fluid computational domain (hereinafter referred to as the computational domain or solution domain) of the compressor model. When simulating the compressor by solving the compressor model, the flow parameters of these grid points are iteratively calculated according to the set algorithm, initial field, and iteration step size. The flow parameters of the grid points can include any one or more of the following: airflow velocity, flow direction velocity, gas static density, gas static pressure, gas static temperature, total temperature, total pressure, pressure ratio, and airflow angle.

[0033] An exemplary computational domain for a compressor model is as follows: Figure 4 As shown, the computational domain consists of grid points (each with a certain area, also called grid blocks), mainly comprising three regions: impeller 1, diffuser 2, and volute 3. The compressor outlet 31 connects to the inlet of the compressor outlet pipeline (referred to as the pipeline). From the model perspective, the outlet boundary of the compressor model is coupled to the inlet boundary of the pipeline model; therefore, the outlet boundary of the compressor model can be considered the same as the inlet boundary of the pipeline model, only with different dimensions. Although this embodiment uses a centrifugal compressor as an example, it can also be applied to multi-stage axial compressors, multi-stage axial and centrifugal combined compressors, etc., demonstrating a certain degree of universality.

[0034] The pipeline model constructed in this embodiment is used to simulate the pressure wave propagation in the compressor outlet pipeline (also referred to as the outlet pipe). The computational domain of the pipeline model is one-dimensional, consisting of multiple nodes on the same straight line, including one node at the inlet boundary and one node at the outlet boundary. When simulating the pipeline by solving the pipeline model, the flow parameters of these nodes are iteratively calculated according to the set algorithm, initial field, and iteration step size. The iteration step size when solving the compressor model is an integer multiple of the iteration step size when solving the pipeline model.

[0035] The input parameters for solving the compressor model 301 include engine speed. The input parameters for solving the pipeline cavity model include pipeline parameters, cavity parameters, and valve opening coefficients. Pipeline parameters, such as pipeline length and cross-sectional area, are used to solve the pipeline model. Cavity parameters (e.g., cavity size) and valve opening coefficients are used to solve the cavity valve model. However, this is merely an example; the input parameters are not limited to these. For instance, the compressor model's input parameters also include the point cloud of each stage of the compressor blades, and the profiles of the compressor casing and hub surfaces. The output of the compression system model 30 includes the compressor inlet flow rate and pressure ratio, which can be the inlet and outlet pressure ratio but is not limited to this. The compressor inlet flow rate and pressure ratio can be used to determine whether the compressor is experiencing surge.

[0036] The embodiments disclosed herein can directly simulate surge phenomena and take into account the influence of the cavity and piping system on compressor surge prediction. At the same time, the use of a volume force model accelerates the calculation speed of the full three-dimensional numerical simulation method, enabling compressor surge prediction to be performed quickly and accurately.

[0037] In an exemplary embodiment of this disclosure, when solving the compression system model, the finite volume method is used to solve the compressor model, and the characteristic line method is used to solve the pipeline model. The compressor model and the pipeline cavity model are bidirectionally coupled, and bidirectional information is transferred between the two models, and calculations are performed alternately.

[0038] In this embodiment, solving the compressor model and the pipeline cavity model includes:

[0039] Based on the initial field or the result of the previous iteration, the computational domain of the compressor model is calculated in one iteration using the finite volume method, and the first information of the compressor model outlet boundary is transmitted to the pipeline model inlet boundary.

[0040] The flow parameters at the inlet boundary of the pipeline model are updated based on the first information. The computational domain of the pipeline model is iterated k times using the method of characteristics. The second information of the inlet boundary of the pipeline model is then passed to the outlet boundary of the compressor model, where k is a positive integer.

[0041] The flow parameters at the outlet boundary of the compressor model are updated based on the second information for the next iteration calculation of the computational domain of the compressor model.

[0042] In this model, the lumped parameter model is coupled to solve the outlet boundary conditions of the pipeline model, and the iteration step size of the compressor model is k times the iteration step size of the pipeline model.

[0043] In one example of this embodiment, the first information includes information on the total temperature, total pressure, static pressure, and flow rate at the outlet boundary of the compressor model;

[0044] Updating the flow parameters of the pipeline model inlet boundary based on the first information includes: when the airflow is in a co-current state, updating the flow parameters of the nodes at the pipeline model inlet boundary based on the total temperature, total pressure, and flow rate information in the first information; when the airflow is in a recirculation state, updating the flow parameters of the nodes at the pipeline model inlet boundary based on the static pressure information in the first information.

[0045] The second information includes the total temperature, total pressure, and static pressure information at the inlet boundary of the pipeline model; updating the flow parameters at the outlet boundary of the compressor model based on the second information includes: when the airflow is in a co-current state, updating the flow parameters of the grid points at the outlet boundary of the compressor model based on the static pressure information in the second information; when the airflow is in a recirculation state, updating the flow parameters of the grid points at the outlet boundary of the compressor model based on the total temperature and total pressure information in the second information.

[0046] like Figure 5A and Figure 5B As shown in the figure, the line marked 0 represents the current time step. Δt3 is the iteration step size when iterating the computational domain of the compressor model, and Δt1 is the iteration step size when iterating the computational domain of the pipeline model. Δt3 = kΔt1. As shown, at the current time step, the first information of the compressor model's outlet boundary needs to be passed to the pipeline model's inlet boundary. The black dot marked 1 represents a node at the pipeline model's inlet boundary, and the black dot marked 2 represents a node adjacent to that node. In the example shown, k ≥ 2; the specific value can be set as needed to meet the requirement of synchronous calculation of the two models.

[0047] Figure 5A This represents a pipeline model where the airflow is in a downstream state (flowing from inside the compressor to the compressor outlet). The model utilizes the transferred total temperature T. t Total pressure p t and traffic Update the flow parameters at the nodes at the inlet boundary of the pipeline model, such as based on T. t p t and After conversion, the updated static pressure p, static temperature T, and pipeline flow velocity v of the node are obtained. At this time, the pipeline model can be used to simulate the right-hand traveling wave and entropy wave in the pipeline at the compressor outlet. Figure 5B This indicates that when the airflow is in a recirculation state (flowing from the compressor outlet pipe to the compressor interior), the static pressure of the nodes at the inlet boundary of the pipe model is updated using the transmitted static pressure p. In this case, the pipe model can be used to simulate the right-hand traveling wave in the compressor outlet pipe.

[0048] In this embodiment, when transmitting the first information, information such as total temperature, total pressure, static pressure, and flow rate at the compressor model outlet boundary can be transmitted to the pipeline model inlet boundary. The pipeline model can then determine whether the airflow is in a co-current or re-current state based on the transmitted flow rate (including flow direction information), thereby extracting the corresponding parameters to update the flow parameters of the nodes at the pipeline model inlet boundary.

[0049] like Figure 5A and Figure 5B As shown, after updating the flow parameters of the nodes at the inlet boundary of the pipeline model based on the first information, the computational domain of the pipeline model is iterated for k time steps, with a total iteration time of kΔt1. Then, the second information obtained from the inlet boundary of the pipeline model is passed to the outlet boundary of the compressor model. This second information includes the total temperature, total pressure, and static pressure of the inlet boundary of the pipeline model (i.e., the total temperature, total pressure, and static pressure of the nodes at the inlet boundary of the pipeline model obtained after k iterations). After receiving the second information, if the airflow is in a downstream state (still based on the airflow direction calculated in this iteration), the static pressure of the grid points at the outlet boundary of the compressor model is updated according to the static pressure p. If the airflow is in a reversible state, the static temperature and static pressure are calculated based on the total temperature and total pressure information in the second information, and the corresponding parameters of the grid points at the outlet boundary of the compressor model are updated.

[0050] By transmitting information bidirectionally between the compressor model and the pipeline model, the flow parameters of the grid points at the outlet boundary of the compressor model are updated, and the computational domain of the compressor model is then used for the next iteration calculation based on the updated flow parameters.

[0051] The compressor model is a three-dimensional model, and its outlet boundary, defined within its three-dimensional computational domain, can include multiple grid points. The pipeline model, on the other hand, is a one-dimensional model, with one node at the inlet boundary of its one-dimensional computational domain. In this embodiment, when transferring the first information of the compressor model's outlet boundary to the pipeline model's inlet boundary, the corresponding parameter values ​​of all grid points on the compressor model's outlet boundary can be averaged before being used to update the parameters of the nodes on the pipeline model's inlet boundary. Conversely, when transferring the second information of the pipeline model's inlet boundary to the compressor model's outlet boundary, the corresponding parameter values ​​of the nodes on the pipeline model's inlet boundary can be assigned to each grid point on the compressor model's outlet boundary, thus achieving data conversion between the one-dimensional and three-dimensional computational domains.

[0052] This embodiment uses a method of coupling one-dimensional and three-dimensional models to parameterize the pipeline cavity system downstream of the compressor, which can accurately predict compressor surge.

[0053] In an exemplary embodiment of this disclosure, when performing k iterations of the computational domain of the pipeline model using the method of characteristics, the equations requiring a solution include the following set of flow control equations:

[0054]

[0055]

[0056]

[0057] Where: p is static pressure; ρ is static density; A is the cross-sectional area of ​​the pipe; v is the flow velocity in the pipe direction; x is the pipe length coordinate; t is time; G is friction force; a is the speed of sound.

[0058] When the lumped parameter model is coupled to the outlet boundary conditions of the pipeline model for solution, the equations to be solved include:

[0059]

[0060]

[0061]

[0062] Where, p p The cavity pressure is ρ. p T is the cavity density; p The temperature of the cavity; and These are the pipeline outlet flow rate and the valve flow rate, respectively, T c V represents the temperature of the gas exiting the pipeline. p K represents the size of the cavity. t c is the valve opening coefficient. v and c p ρ0 represents the specific heat capacity of the gas at constant volume and constant pressure, respectively, and p0 represents the ambient pressure.

[0063] When solving a one-dimensional model based on a set time step, the flow control equations can be solved using the method of characteristics for each node in the computational domain. For nodes at the outlet boundary in the computational domain, the equations required by the lumped parameter model and the flow control equations required by the pipeline model are combined to obtain the results of iterative calculation.

[0064] In this embodiment, the above equation serves as the outlet boundary condition for the pipeline model, used for iterative calculation of the flow parameters at the nodes of the pipeline model's outlet boundary. The cavity pressure P... p The ideal gas law P p =ρ p RT p The calculation shows that R is a constant.

[0065] In an exemplary embodiment of this disclosure, the control parameters of the compression system model include compressor speed and valve opening coefficient. Determining whether surge occurs at the operating point based on the compressor flow rate and pressure ratio in the simulation results includes:

[0066] For a given operating point, if the curves showing the changes in the compressor's inlet flow rate and inlet / outlet pressure ratio over time in the simulation results form a surge loop, then surge is determined to have occurred, and this operating point is called the surge point; wherein, a given operating point corresponds to a combination of the input valve opening coefficient and the compressor speed.

[0067] In one example of this embodiment, solving the compressor model and the pipeline cavity model to obtain simulation results of the compressor at multiple operating points; determining whether surge occurs at the operating point based on the compressor's flow rate and pressure ratio in the simulation results includes:

[0068] For any given compressor speed, the corresponding surge boundary point is determined as follows:

[0069] Using the compressor speed and an initial valve opening coefficient as input, the compressor model and the pipeline cavity model are solved to obtain the simulation results under the steady operating point.

[0070] Using the simulation results under the stable operating point as the initial field, the valve opening coefficient is increased, and the compression system model and the pipeline cavity model are solved to obtain the simulation results.

[0071] Based on the compressor flow rate and pressure ratio in the simulation results, if surge is determined, the surge boundary point is determined based on the current operating point; if surge does not occur, the valve opening coefficient is increased and the compression system model and the pipeline cavity model are solved until surge occurs.

[0072] In one example of this embodiment, multiple compressor speeds can be set, and for each compressor speed, a corresponding surge boundary point can be determined. Then, in a graph with inlet flow rate and inlet / outlet pressure ratio as the horizontal and vertical axes, the surge boundary points corresponding to the multiple speeds can be connected to obtain the surge boundary of the compressor.

[0073] In this embodiment, the control parameters for solving the compressor model include compressor speed and valve opening coefficient. Any combination of compressor speed and valve opening coefficient constitutes an operating point of the compressor. By solving the model, the changes in the compressor's inlet flow rate and inlet / outlet pressure ratio over time can be obtained when the compressor is operating at that point. If the compressor's inlet flow rate and inlet / outlet pressure ratio do not change over time, then this point is a stable point, and no surge instability has occurred. If the curves of the compressor's inlet flow rate and inlet / outlet pressure ratio form a surge loop, then this operating point is a surge point. For a set compressor speed, the operating point where surge first occurs during the transition from stable operating condition to surge operating condition is the surge boundary point corresponding to that compressor speed.

[0074] In an exemplary embodiment of this disclosure, the compressor itself is modeled using a three-dimensional volumetric force model. This three-dimensional volumetric force model adopts the idea of ​​real-time partitioning based on local flow parameters to address compressor surge phenomena, which can simulate the evolution process of the surge recirculation region to the greatest extent, thereby completely simulating the entire process of the compressor from steady state to surge instability. The three-dimensional volumetric force model and solution method of this embodiment can use the volumetric force model and solution method disclosed in CN114662218A (titled: Turbomachinery Simulation, Instability Prediction Method and Device Based on Volumetric Force Model). The compressor is a type of turbomachinery, and solving the three-dimensional volumetric force model means simulating the compressor based on the three-dimensional volumetric force model.

[0075] In this embodiment, solving the compressor model includes:

[0076] Based on multiple grid points set in both the blade region and non-blade region of the compressor computational domain, the flow parameters in the compressor solution domain are iteratively calculated according to a set time step; wherein, the iterative calculation based on the grid points in the blade region includes:

[0077] Step 1: Determine the flow condition at the grid point location based on the flow parameters obtained in the initial field or the previous iteration.

[0078] Step 2: Determine the ideal velocity field at the grid point location based on the flow conditions, and construct the volume force vector field at the grid point location based on the ideal velocity field;

[0079] Step 3: Solve the flow control equations based on the volume force vector field to obtain the flow parameters at the grid point locations in this iteration.

[0080] In this embodiment, when simulating based on a three-dimensional volumetric force model of the compressor, the point cloud data of the compressor impeller and diffuser blades, the profile data of the compressor hub and casing surfaces, and the compressor speed (also known as the compressor operating speed) are used as inputs for preprocessing, including: mesh generation, calculation of mesh geometry, calculation of blade geometry, and calculation of the initial field (such as p, T, U). Then, calculation parameters, such as the Corange number, speed, and boundary conditions, are set. Before iterative calculation of the mesh points in the fluid computational domain, the boundary conditions of the fluid computational domain need to be set. After completing one iteration, the set boundary conditions can be updated. The boundaries of the fluid computational domain include the inlet boundary, outlet boundary, hub wall, and casing wall. When simulating surge conditions, the inlet boundary conditions are selected as total temperature and total pressure conditions in the downstream flow and static pressure conditions in the reflow flow, while the outlet boundary conditions are all selected as pipe cavity conditions, i.e., calculations are based on a coupled pipe cavity model.

[0081] After setting the calculation parameters, the finite volume method can be used to solve the three-dimensional volume force model. Specifically, based on the input parameters and boundary conditions, the flow parameters of the grid points in the fluid computational domain can be iteratively calculated step by step according to the set iteration step size (i.e., the time interval between the current iteration and the next iteration). The blade region is the domain of volume force action; during step-by-step iterative calculation, it is necessary to construct the volume force vector field at the grid point locations in the blade region as the volume force source term in the flow control equations. The non-blade region uses the same control equations, but the volume force source term is 0, and no volume force solution is required. The iterative calculation based on the grid points in the blade region can be performed as described in steps one through three above.

[0082] In this embodiment, when calculating the ideal velocity direction at a grid point location, the relationship model between the airflow angle of attack and the lag angle is established based on the flow conditions at that grid point location. The lag angle is derived from the calculated airflow angle of attack. Therefore, the relationship model used will differ depending on the flow conditions at the grid point location, and the calculation of the ideal velocity field will also differ. The relationship model under different flow conditions can use empirical formulas for the corresponding scenarios. Therefore, this embodiment can distinguish local flow parameters to achieve fine simulation, making the simulation results closer to reality.

[0083] In one example of this embodiment, the flow control equations are solved based on the volume force vector field, and the flow parameters at the grid point locations in this iteration are obtained based on the solution results, including:

[0084] The flow control equations are discretized based on grid points in the fluid computational domain, and then numerically solved. The flow control equations include the following equations:

[0085]

[0086]

[0087]

[0088] in, Here, ρ is the gas velocity, ρ is the gas static density, p is the gas static pressure, and h is the gas static pressure. t Let t be the total enthalpy of the gas, and t be time. The entrainment speed generated by the rotation of the compressor. Let be the volume force per unit mass of fluid, and b be the blockage factor.

[0089] The aforementioned flow control equations are unsteady Euler equations with source terms in an absolute coordinate system, including the mass conservation equation, momentum conservation equation, and energy conservation equation. The physical quantities (i.e., flow parameters) in the flow control equations for a grid point are represented by the physical quantities of that grid point and its surrounding grid points at the previous time step (previous iteration) and at the current time step (this iteration). Therefore, the original control equations can be expressed as the relationships between the physical quantities at each discrete point.

[0090] The above governing equations can be solved by combining the ideal gas law p = ρRT and the following formula for calculating the total enthalpy of a gas, to obtain the static temperature T of the gas:

[0091]

[0092] Where R is a constant, c p This is the specific heat capacity of the gas at constant pressure.

[0093] In this paper, the parameters and variables in the formulas used to calculate the flow parameters at the grid point location all refer to the parameters and variables at that grid point.

[0094] In another example of this embodiment, the third equation of the above flow control equation set can be replaced by the following equation:

[0095]

[0096] Where E represents the total energy, and the other parameters have the same meanings as those in the flow control equations.

[0097] In this embodiment, the volume force experienced by a unit mass of fluid is... It consists of two parts: deflection force f ζ and loss force f ηThe volume force vector field at the grid point location is constructed based on the ideal velocity field, including: calculating the deflection force component and loss force component of the volume force at the grid point location based on the ideal velocity direction at the grid point location, and obtaining the volume force at the grid point location based on the deflection force component and loss force component; wherein, when the grid point is located in the downstream region, the direction of the volume force at the grid point location is determined by the directions of the deflection force component and the loss force component; when the grid point is located in the recirculation region, the direction of the volume force at the grid point location is always the blade normal.

[0098] For example, the value of the deflection force component at this grid point location can be calculated using the following formula:

[0099]

[0100] Among them, f ζ K represents the value of the deflection force component. n Where ρ is the proportional gain coefficient and ρ is the static density of the airflow. W represents the relative velocity of the airflow. m for The component along the airflow direction, W n for The components of the intersection line between the plane perpendicular to the ideal velocity direction and the S1 flow surface, R, s and β are the radius of curvature, pitch and blade angle of the grid points on the blade, respectively;

[0101] When the grid point is located in the downstream region, the direction of the deflection force component is perpendicular to...

[0102] For example, the value of the loss force component at this grid point location can be calculated using the following formula:

[0103]

[0104] Among them, f η K represents the value of the loss force component. t1 and K t2 It is a coefficient (loss coefficient) calculated based on the angle of attack of the airflow, where ρ is the static density of the airflow. The velocity is the entrainment velocity generated by the compressor rotation, and p is the static pressure of the airflow. For airflow velocity, The relative velocity of the airflow is given by m, where m is the direction of flow.

[0105] When the grid point is located in the downstream region, the direction of the loss force component is... parallel.

[0106] One embodiment of this disclosure also provides a device for predicting compressor surge boundaries, such as... Figure 2As shown, it includes a processor 60 and a memory 50 storing a computer program, wherein the processor 60, when executing the computer program, is capable of implementing a compressor surge boundary prediction method as described in any embodiment of this disclosure.

[0107] The processor in this disclosure can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), a microprocessor, etc., or other conventional processors. The processor can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), discrete logic or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or combinations of the above devices. That is, the processor in the above embodiments can be any processing device or combination of devices that implements the methods, steps, and logic block diagrams disclosed in the embodiments of this invention. If this disclosure is implemented in part in software, instructions for software can be stored in a suitable non-volatile computer-readable storage medium, and one or more processors can be used to execute the instructions in hardware to implement the methods of this disclosure.

[0108] The prediction method of the above embodiments of this disclosure is based on the compressor geometry and the geometry of the compressor outlet pipeline cavity system to construct a compression system model. A numerical method combining the finite volume method and the method of characteristics is used to solve the compression system model, obtaining the compressor characteristics at different operating points. By observing whether a surge ring composed of flow rate and pressure ratio appears in the simulation results, it is determined whether surge occurs at that operating point. Surge points can be obtained by performing the same simulation calculations at various operating speeds. Connecting the surge points at each speed yields the compressor surge boundary.

[0109] The method and apparatus of this disclosure address the shortcomings of existing numerical simulation methods in predicting compressor surge boundaries. They can directly simulate surge phenomena and consider the influence of the cavity and piping system on compressor surge prediction. At the same time, the use of a volume force model accelerates the calculation speed of the full three-dimensional numerical simulation method, and the use of a coupling method of one-dimensional and three-dimensional models parameterizes the pipeline cavity system downstream of the compressor. Thus, it is possible to predict compressor surge quickly and accurately and obtain surge boundaries.

[0110] The following uses a centrifugal compressor with a volute as an example to illustrate the process of surge boundary prediction in this embodiment of the disclosure. The surge boundary prediction method for this centrifugal compressor includes the following steps:

[0111] The first step is to model the centrifugal compressor based on a three-dimensional volumetric force model, and calculate the mesh as follows: Figure 4As shown, the input parameters of the compressor model include the point cloud data of the compressor impeller and diffuser blades, the profile data of the compressor hub surface and casing surface, and the compressor speed.

[0112] The second step involves modeling the compressor outlet pipeline using a one-dimensional model, while the cavity and valves are modeled using a lumped parameter model.

[0113] This embodiment uses a coupled three-dimensional model (compressor model) and a one-dimensional model (pipeline model) to model the compressor and its outlet pipeline cavity system. Three-dimensional unsteady simulation is performed on the compressor, and one-dimensional unsteady simulation is performed on the compressor outlet pipeline. The one-dimensional model constructed based on the compressor outlet pipeline has input parameters mainly including pipeline length and cross-sectional area. The lumped parameter model constructed for the downstream cavities and valves of the compressor is coupled and solved at the boundary nodes of the pipeline model solved using the method of characteristics. The input parameters of this lumped parameter model include the size of the cavity, such as its volume.

[0114] The third step is to set the compressor operating speed according to the actual application scenario. The compressor model is solved using the finite volume method, and the one-dimensional model constructed by the compressor outlet pipeline is solved using the method of characteristics. The lumped parameter model constructed by the cavity and valve is coupled and solved in the boundary conditions of the one-dimensional model.

[0115] The method of characteristics is an approximate computational method based on characteristic theory for solving hyperbolic partial differential equations. It has been widely used in problems involving one-dimensional unsteady flow and two-dimensional steady flow. The method of characteristics is used to solve for Riemann invariants.

[0116] In this step, for a given compressor speed, a relatively small valve opening coefficient can be set first (i.e., the valve throat area is relatively large). The steady-state operating point is then calculated. Figure 6 Point B in the diagram is referred to as the steady-state point. The criterion for determining a steady-state operating point is that the compressor inlet and outlet flow rates and pressures do not change over time, i.e., they are essentially constant values. For example, if the residual threshold is set to 0.0001, when the residuals of the flow parameters at all monitoring points are less than 0.0001, the simulation of the steady-state operating condition is considered to have converged, and the current operating point is considered the steady-state operating point. The monitoring points can be any one or more grid points in the compressor fluid computational domain.

[0117] Then, using the calculation results of the stable operating point (such as the flow parameters of each grid point) as the initial field, the input valve opening coefficient is adjusted to a larger value (the value of each increase can be set as needed), and the compression system model is solved to obtain the simulation results of the compressor at the current operating point. Based on the calculated changes in the compressor inlet and outlet parameters (such as flow rate, pressure ratio, etc.), it is determined whether surge has occurred at the current operating point. If surge has occurred, the surge boundary point is determined based on the current operating point (defined by the combination of the currently set compressor speed and valve opening coefficient). If a surge cycle is not formed, the valve opening coefficient is adjusted to a larger value again, and this process is repeated until the compressor surges.

[0118] Figure 6 Point A in the diagram represents the surge boundary point. The criterion for determining whether surge has occurred at the current operating point is whether the curves showing the compressor inlet flow rate and inlet / outlet pressure ratio changing over time constitute a surge loop. A surge loop is shown in the diagram. Figure 7 As shown, when the curves of the compressor's inlet flow rate and the inlet / outlet pressure ratio change over time form a surge loop, it can be determined that surge has occurred at the current operating point.

[0119] In actual calculations, when determining the surge boundary point based on the current operating point, depending on the required surge boundary accuracy, the current operating point can be directly determined as the surge boundary point; alternatively, the current operating point can be used as the surge boundary point obtained from the initial calculation, and multiple simulations can be performed near this surge boundary point based on the set compressor speed and multiple different valve opening coefficients. The operating point where surge first occurs can be used to determine the final surge boundary point corresponding to the compressor speed, in order to obtain a more accurate location of the surge boundary point.

[0120] This disclosure is not limited to the centrifugal compressor type mentioned above, but can also be applied to multi-stage axial compressors and multi-stage axial / centrifugal combined compressors, and has a certain degree of universality.

[0121] As described above, to address the issue of low accuracy in existing surge boundary simulation prediction methods, this disclosure proposes a method based on the coupling of a three-dimensional volume force model and a one-dimensional model for predicting compressor surge. This method can directly simulate the three-dimensional unsteady surge instability process and obtain the compressor surge boundary. This embodiment considers the influence of the compressor outlet cavity and piping system. The computational accuracy and versatility are comparable to traditional full three-dimensional numerical simulation methods, while the computational efficiency is significantly improved.

[0122] In one or more exemplary embodiments of this disclosure, the described functionality may be implemented in hardware, software, firmware, or any combination thereof. If implemented in software, the functionality may be stored as one or more instructions or code on or transmitted via a computer-readable medium and executed by a hardware-based processing unit. The computer-readable medium may comprise a computer-readable storage medium corresponding to a tangible medium, such as a data storage medium, or a communication medium comprising any medium facilitating the transfer of a computer program from one place to another, such as according to a communication protocol. In this manner, a computer-readable medium may generally correspond to a non-transitory tangible computer-readable storage medium or a communication medium, such as a signal or carrier wave. A data storage medium may be any available medium accessible by one or more computers or one or more processors to retrieve instructions, code, and / or data structures for implementing the techniques described in this disclosure. A computer program product may comprise a computer-readable medium.

[0123] For example, and not as a limitation, such computer-readable storage media may include RAM, ROM, EEPROM, CD-ROM or other optical disc storage devices, magnetic disk storage devices or other magnetic storage devices, flash memory, or any other medium that can be used to store desired program code in the form of instructions or data structures and is accessible by a computer. Furthermore, any connection may also be referred to as a computer-readable medium. For example, if instructions are transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, then coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of medium. However, it should be understood that computer-readable storage media and data storage media do not include connections, carrier waves, signals, or other transient media, but rather refer to non-transient tangible storage media. As used herein, disks and optical discs include compact optical discs (CDs), laser discs, optical discs, digital versatile discs (DVDs), floppy disks, or Blu-ray discs, where disks typically reproduce data magnetically, while optical discs use lasers to reproduce data optically. The above combinations should also be included within the scope of computer-readable media.

[0124] Instructions can be executed by one or more processors, such as one or more digital signal processors (DSPs), general-purpose microprocessors, application-specific integrated circuits (ASICs), field-programmable arrays (FPGAs), or other equivalent integrated or discrete logic circuits. Therefore, the term "processor" as used herein can refer to any of the above-described structures or any other structures suitable for implementing the techniques described herein. Additionally, in some aspects, the functionality described herein can be provided within dedicated hardware and / or software modules configured for encoding and decoding, or incorporated into combined codecs. Furthermore, the techniques can be fully implemented in one or more circuit or logic elements.

[0125] The technical solutions of the embodiments of this disclosure can be implemented in a wide variety of devices or equipment, including wireless mobile phones, integrated circuits (ICs), or a set of ICs (e.g., chipsets). Various components, modules, or units are described in the embodiments of this disclosure to emphasize functional aspects of a device configured to perform the described techniques, but they do not necessarily need to be implemented through different hardware units. Rather, as described above, the various units can be combined in codec hardware units or provided by a collection of interoperable hardware units (including one or more processors as described above) combined with suitable software and / or firmware.

Claims

1. A method for predicting compressor surge boundary, comprising: A compression system model is established, which includes a compressor model based on the compressor itself and a pipeline cavity model based on the pipeline cavity system downstream of the compressor. Solve the compressor model and the pipeline cavity model to obtain the simulation results of the compressor at multiple operating points; Determine whether surge occurs at the operating point based on the compressor's flow rate and pressure ratio as described in the simulation results; The compressor model is a three-dimensional volume force model, and the pipeline cavity model is a coupling of a one-dimensional model and a lumped parameter model. The one-dimensional model is a pipeline model constructed based on the pipeline at the compressor outlet, and the lumped parameter model is constructed based on the cavity and valves downstream of the compressor. The process of solving the compressor model and the pipeline cavity model includes: Based on the initial field or the result of the previous iteration, the computational domain of the compressor model is calculated in one iteration using the finite volume method, and the first information of the compressor model outlet boundary is transmitted to the pipeline model inlet boundary. The flow parameters at the inlet boundary of the pipeline model are updated based on the first information. The computational domain of the pipeline model is iterated k times using the method of characteristics. The second information of the inlet boundary of the pipeline model is then passed to the outlet boundary of the compressor model, where k is a positive integer. The flow parameters at the outlet boundary of the compressor model are updated based on the second information for the next iteration calculation of the computational domain of the compressor model. The lumped parameter model is solved in the outlet boundary conditions of the pipeline model, and the iteration step size of the compressor model is k times the iteration step size of the pipeline model. Solving the compressor model includes: Based on multiple grid points set in both the blade region and non-blade region of the compressor computational domain, the flow parameters in the compressor solution domain are iteratively calculated according to a set time step; wherein, the iterative calculation based on the grid points in the blade region includes: Based on the flow parameters at the grid point locations obtained in the initial field or the previous iteration, determine the flow conditions at the grid point locations; Based on the flow conditions, determine the ideal velocity field at the grid point location, and construct the volume force vector field at the grid point location based on the ideal velocity field; The flow control equations are solved based on the volume force vector field to obtain the flow parameters at the grid point locations in this iteration.

2. The method as described in claim 1, characterized in that: The first information includes information on the total temperature, total pressure, static pressure, and flow rate at the outlet boundary of the compressor model; updating the flow parameters at the inlet boundary of the pipeline model based on the first information includes: when the airflow is in a co-current state, updating the flow parameters of the nodes at the inlet boundary of the pipeline model based on the total temperature, total pressure, and flow rate information in the first information; when the airflow is in a recirculation state, updating the flow parameters of the nodes at the inlet boundary of the pipeline model based on the static pressure information in the first information; The second information includes the total temperature, total pressure, and static pressure information at the inlet boundary of the pipeline model; updating the flow parameters at the outlet boundary of the compressor model based on the second information includes: when the airflow is in a co-current state, updating the flow parameters of the grid points at the outlet boundary of the compressor model based on the static pressure information in the second information; when the airflow is in a recirculation state, updating the flow parameters of the grid points at the outlet boundary of the compressor model based on the total temperature and total pressure information in the second information.

3. The method as described in claim 1, characterized in that: When performing k iterations of the computational domain calculation of the pipeline model using the method of characteristics, the equations to be solved include: in: It is static pressure; It is the static density; A This refers to the cross-sectional area of ​​the pipeline. v The velocity is the flow direction in the pipeline. x The coordinates represent the length of the pipeline; For time; G Friction; a The speed of sound.

4. The method as described in claim 1, characterized in that: When the lumped parameter model is coupled to the outlet boundary conditions of the pipeline model for solution, the equations to be solved include: in, p p The pressure in the cavity; ρ p The density of the cavity; T p The temperature of the cavity; and These are the pipeline outlet flow rate and the valve flow rate, respectively. The temperature of the gas at the pipeline outlet; V p The size of the cavity; K t This is the valve opening coefficient. and These are the isochoric and isobaric specific heat capacities of gases, respectively. p 0 Due to environmental pressures.

5. The method as described in claim 1, characterized in that: The control parameters of the compression system model include compressor speed and valve opening coefficient. Based on the compressor flow rate and pressure ratio in the simulation results, it is determined whether surge occurs at the operating point, including: For a given operating point, if the curves showing the changes in the compressor's inlet flow rate and inlet / outlet pressure ratio over time in the simulation results form a surge loop, then surge is determined to have occurred, and this operating point is called the surge point; wherein, a given operating point corresponds to a combination of the input valve opening coefficient and the compressor speed.

6. The method as described in claim 5, characterized in that: The process involves solving the compressor model and the pipeline cavity model to obtain simulation results of the compressor at multiple operating points; determining whether surge occurs at each operating point based on the compressor's flow rate and pressure ratio from the simulation results includes: For any given compressor speed, the corresponding surge boundary point is determined as follows: Using the compressor speed and an initial valve opening coefficient as input, the compressor model and the pipeline cavity model are solved to obtain the simulation results under the steady operating point. Using the simulation results under the stable operating point as the initial field, the valve opening coefficient is increased, and the compression system model and the pipeline cavity model are solved to obtain the simulation results. Based on the compressor flow rate and pressure ratio in the simulation results, if surge is determined, the surge boundary point is determined based on the current operating point; if surge does not occur, the valve opening coefficient is increased and the compression system model and the pipeline cavity model are solved until surge occurs.

7. The method as described in claim 6, characterized in that, Also includes: Set multiple compressor speeds, and for each compressor speed, determine the corresponding surge boundary point; In a graph with inlet flow rate and inlet / outlet pressure ratio as the horizontal and vertical axes, the surge boundary points corresponding to the multiple rotational speeds are connected to obtain the surge boundary of the compressor.

8. A device for predicting compressor surge boundary, comprising a processor and a memory storing a computer program, wherein, When the processor executes the computer program, it can implement the compressor surge boundary prediction method as described in any one of claims 1 to 7.

Images