Mechanism model optimization method

Abstract

The invention provides a mechanism model optimization method, and relates to the technical field of mechanism models. According to the mechanism model optimization method provided by the embodiment of the invention, the process of iterative calculation which is relatively long in time consumption and complicated in calculation is adopted in the initial mechanism model, and parameters with relatively high correlation are utilized for fitting, so that the empirical formula is obtained and substituted into the initial mechanism model, and the optimization mechanism model is obtained; the optimized mechanism model can obtain the result of the preset parameter through the empirical formula, and the calculation speed is higher, so that the model practicability is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of mechanism models, and in particular to a mechanism model optimization method. Background Art

[0002] The mechanism model is an accurate mathematical model established based on the internal mechanism of the object or production process or the transfer mechanism of the substance. It can be a mathematical model of the object or process based on the mass conservation equation, energy conservation equation, momentum balance equation, phase balance equation, and certain physical property equations, chemical reaction laws, or basic circuit laws. It has been widely used in simulations of various industries. For example, the turbine model is a commonly used mechanism model in the supercharging model.

[0003] In the existing turbine performance models, the zero-dimensional model generally equates the turbine to an equivalent nozzle model to calculate the flow and torque generated by the turbine, or based on the existing turbine performance MAP, interpolates the other two known performance parameters according to the turbine flow, efficiency, expansion ratio, and supercharger speed. The zero-dimensional model has a fast calculation speed and low computing power requirements, but the calculation accuracy is not high and the generalization is not good. The three-dimensional CFD model considers the changes in the flow field at each grid node in the internal space of the turbine in detail, but has high requirements for computing resources, long calculation time, high cost, and cannot meet the real-time requirements. In the process of combining with the consistency correction of the digital twin optimization algorithm, the traditional one-dimensional model takes too long to iterate multiple times, and cannot meet the function of the self-correction model to correct the digital model regularly, and the model is of low practicality. Summary of the invention

[0004] A brief summary of one or more aspects is given below to provide a basic understanding of these aspects. This summary is not an exhaustive overview of all conceived aspects, and is neither intended to identify the key or critical elements of all aspects nor to define the scope of any or all aspects. Its only purpose is to give some concepts of one or more aspects in a simplified form as a prelude to a more detailed description that will be given later.

[0005] The purpose of the present invention is to provide a mechanism model optimization method, which can reduce the calculation time of the mechanism model, improve the operation efficiency of the mechanism model, and help meet practical requirements.

[0006] The embodiments of the present invention can be implemented in the following ways:

[0007] A mechanism model optimization method, the mechanism model optimization method comprising:

[0008] Acquire an initial mechanism model; wherein the initial mechanism model has a first control surface and a second control surface;

[0009] The data of the first control surface and the data of the second control surface under the preset parameters under multiple working conditions are calculated by using the initial mechanism model to form a corresponding data set; wherein the corresponding data set includes the corresponding data of the data of the first control surface and the working conditions and the corresponding data of the data of the second control surface and the working conditions; the data of the first control surface is obtained by measured data or calculated by using the measured data; the data of the second control surface is obtained by iterative calculation of the initial mechanism model;

[0010] determining a correlation between data of the first control surface and data of the second control surface under the same operating condition;

[0011] If the correlation between the data of the first control plane and the data of the second control plane meets a preset requirement, fitting the data of the first control plane and the data of the second control plane to obtain an empirical formula;

[0012] The empirical formula is substituted into the initial mechanism model to replace the iterative calculation process of the preset parameters of the second control surface, so as to obtain the optimized mechanism model.

[0013] Optionally, the step of determining the correlation between the data of the first control plane and the data of the second control plane includes:

[0014] The correlation value between the data of the first control plane and the data of the second control plane is calculated by the following formula:

[0015]

[0016] Among them, x i data of the first control plane; is the mean value of the first control surface data; i data of the second control plane; is the mean value of the second control plane data.

[0017] Optionally, the preset requirement is: the correlation value is greater than 95%.

[0018] Optionally, after the step of fitting the data of the first control surface and the data of the second control surface to obtain an empirical formula, the mechanism model optimization method further includes:

[0019] Calculate the goodness of fit of the empirical formula obtained by fitting:

[0020]

[0021] Among them, R 2 is the goodness of fit; i is the actual value data of the second control surface, fitting the predicted data for the second control surface, is the mean value of the second control plane data;

[0022] The fitting effect is judged according to the goodness of fit. If the goodness of fit reaches a preset value, the empirical formula is substituted into the initial mechanism model to replace the iterative calculation process of the preset parameters of the second control surface.

[0023] Optionally, the initial mechanism model further has a third control surface, and the third control surface is located on a side of the second control surface away from the first control surface; after the step of substituting the empirical formula into the initial mechanism model to replace the iterative calculation process of the preset parameters of the second control surface to obtain the optimized mechanism model, the mechanism model optimization method further includes:

[0024] The optimization mechanism model and the initial mechanism model are respectively used to calculate the data value of the third control surface under the same working condition to obtain an optimized data value and an initial data value; wherein the optimized data value is the data value of the third control surface calculated by using the optimization mechanism model; and the initial data value is the data value of the third control surface calculated by using the initial mechanism model;

[0025] Calculating a data error between the optimized data value and the initial data value;

[0026] If the data error is less than 1%, it is determined that the optimization mechanism model meets the requirements.

[0027] Optionally, after the step of determining the correlation between the data of the first control plane and the data of the second control plane, the mechanism model optimization method further includes:

[0028] If the correlation between the data of the first control plane and the data of the second control plane does not meet the preset requirement, other parameters are replaced as preset parameters for optimization calculation.

[0029] Optionally, the preset parameter is a fluid density parameter.

[0030] Optionally, the mechanism model optimization method further includes model empirical parameter optimization, and the step of model empirical parameter optimization includes:

[0031] Selecting a parameter to be optimized from the empirical parameters of the initial mechanism model; wherein the parameter to be optimized is a parameter in the empirical parameters that affects the model accuracy;

[0032] An initial value is set for each parameter to be optimized; the initial value is an average value of a value range of the parameter to be optimized;

[0033] The initial value is used to perform prediction calculation, and the initial value is iterated according to the calculation result to obtain the optimized parameter value.

[0034] Optionally, the step of selecting parameters to be optimized from the empirical parameters of the initial mechanism model includes:

[0035] Performing sensitivity analysis on the empirical parameters and obtaining corresponding sensitivity values;

[0036] If the sensitivity value is greater than a preset value, the empirical parameter corresponding to the sensitivity is selected as the parameter to be optimized.

[0037] Optionally, the step of using the initial value to perform prediction calculation, and iterating the initial value according to the calculation result to obtain the optimized parameter value includes:

[0038] The initial value is iterated using a particle swarm optimization algorithm to obtain an optimized parameter value.

[0039] The beneficial effects of the mechanism model optimization method provided by the embodiment of the present invention include:

[0040] An embodiment of the present invention provides a mechanism model optimization method, which includes obtaining an initial mechanism model; wherein the initial mechanism model has a first control surface and a second control surface; using the initial mechanism model to obtain data of the first control surface and the second control surface under preset parameters under multiple working conditions to form a corresponding data set; wherein the corresponding data set includes data of the first control surface and corresponding data of the working conditions and data of the second control surface and corresponding data of the working conditions; the data of the first control surface is obtained by measured data or calculated using measured data; the data of the second control surface is obtained by iterative calculation using the initial mechanism model; judging the correlation between the data of the first control surface and the data of the second control surface; if the correlation between the data of the first control surface and the data of the second control surface meets the preset requirements, fitting the data of the first control surface and the data of the second control surface to obtain an empirical formula; substituting the empirical formula into the initial mechanism model to replace the iterative calculation process of the preset parameters of the second control surface to obtain an optimized mechanism model. This method obtains an empirical formula by fitting the time-consuming and computationally complex iterative calculation process in the initial mechanism model with parameters with high correlation, and then substitutes the empirical formula into the initial mechanism model to obtain an optimized mechanism model. The optimized mechanism model can obtain the results of preset parameters through the empirical formula, with faster calculation speed, thereby improving the practicality of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The above features and advantages of the present invention can be better understood after reading the detailed description of the embodiments of the present disclosure in conjunction with the following drawings. In the drawings, the components are not necessarily drawn to scale, and components with similar related properties or features may have the same or similar reference numerals.

[0042] Figure 1 A step diagram of a mechanism model optimization method provided according to one aspect of the present invention is shown;

[0043] Figure 2 A schematic structural diagram of an initial mechanism model provided according to one aspect of the present invention is shown;

[0044] Figure 3 The optimization process for calculation in the mechanism model optimization method provided according to one aspect of the present invention is shown;

[0045] Figure 4 A flow chart of a particle swarm optimization algorithm provided according to one aspect of the present invention is shown.

[0046] 100-initial mechanism model; 111-volute inlet surface; 112-impeller inlet surface; 113-impeller outlet surface; 114-volute section; 115-impeller section; 116-diffuser section. DETAILED DESCRIPTION

[0047] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. Note that the aspects described below in conjunction with the accompanying drawings and specific embodiments are only exemplary and should not be construed as limiting the scope of protection of the present invention in any way.

[0048] In the description of the present invention, it should be noted that if the terms "upper", "lower", "inside", "outside", "vertical" and the like appear, the orientation or position relationship indicated is based on the orientation or position relationship shown in the drawings, or is the orientation or position relationship in which the product of the invention is usually placed when used, rather than indicating or implying that the device or element referred to must have a specific orientation or be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0049] At the same time, it should be noted that the terms "first", "second", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0050] In the description of the present invention, it is also necessary to explain that, unless otherwise clearly specified or limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, an integral connection, or a detachable connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, or the internal communication of two components, etc. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0051] Figure 1 The step diagram of the mechanism model optimization method provided in this embodiment, Figure 2 This is a schematic diagram of the structure of the initial mechanism model provided in this embodiment. Figure 1 and Figure 2 This embodiment provides a mechanism model optimization method, and in this embodiment, the mechanism model optimization method uses a turbine model as an example for explanation. It can be understood that in some other embodiments, the mechanism model optimization method can also be used to optimize other mechanism models. In the description of this embodiment, the initial mechanism model is an average streamline model of an existing turbine model, which has a volute section 114, an impeller section 115 and a diffuser section 116.

[0052] The following describes the mechanism model optimization method provided by this embodiment by taking a turbine model as an example. The mechanism model optimization method includes optimizing the calculation in the mechanism model. Figure 3 The optimization process for calculation in the mechanism model optimization method provided in this embodiment is shown, please refer to Figure 1-Figure 3 , the optimization of this calculation includes the following steps:

[0053] S01: Obtain the initial mechanism model.

[0054] The initial mechanism model 100 is the model to be optimized, which can be constructed according to the construction method in the prior art. Taking the turbine model provided as the initial mechanism model 100 in this embodiment as an example, the one-dimensional average streamline method is used for its construction. The one-dimensional average streamline method simplifies the flow process in the three-dimensional space inside the turbine into a one-dimensional flow on the streamline according to the flow characteristics inside the turbine. The flow segments are divided according to the internal structure of the turbine and the changes in the exhaust gas flow characteristics, and the cross-section with the change in flow characteristics is selected as the control surface. The average streamline model divides the flow in the turbine according to the flow mode, and realizes the thermodynamic calculation of the turbine flow by exchanging the control surface parameters. The parameters of the control surface adopt the average value in the flow channel cross section, assuming that the flow in the turbine is a one-dimensional quasi-steady flow; the fluid is an ideal working fluid, and the specific heat ratio of the fluid remains unchanged in the turbine flow; the fluid is an adiabatic fluid, and no heat exchange occurs with the outside during the flow process. The inlet temperature and pressure, mass flow rate, geometric parameters and supercharger speed of the turbocharger turbine part are known. The control equations of the one-dimensional streamline model include: mass conservation equation, momentum conservation equation, energy conservation equation and ideal gas state equation, etc. After multiple iterative solution processes, various parameters of the volute inlet surface 111, the impeller inlet surface 112 and the impeller outlet surface 113 and the velocity triangle of the impeller inlet and outlet can be obtained, and the performance parameters of the turbine can be calculated. It should be noted that the impeller inlet surface 112 can also be called the volute outlet surface.

[0055] The initial mechanism model 100 has a first control surface and a second control surface. Specifically, the specific positions of the first control surface and the second control surface in the initial mechanism model 100 can be set according to the requirements, and the parameters satisfying the first control surface can be directly obtained by actual measurement or actual measurement combined with geometric formula calculation, and the data of the second control surface are obtained by iterative calculation in the initial mechanism model. For example, in the turbine model, the first control surface can be set as the volute inlet surface 111, and the second control surface can be set as the impeller inlet surface 112.

[0056] Specifically, the temperature, pressure, flow rate and other data at the volute inlet surface 111 can be obtained by direct detection and combined with the turbine geometry formula; according to the first theorem of thermodynamics, the total temperature at the impeller inlet surface 112 and the total temperature at the turbine inlet surface can be regarded as equal. According to the known parameter values ​​at the volute inlet surface 111, combined with iteration, geometric structure and conservation equations, the various parameter values ​​at the impeller inlet surface 112 can be further calculated. Specifically, the fluid density ρ2 at the impeller inlet surface 112 is calculated first. First, ρ2 is assumed to represent the ideal outlet fluid density of the impeller inlet surface 112. Combined with multiple loss analyses and mass conservation equations, ρ2 is corrected through iterative calculation to calculate the final impeller outlet fluid density value. After calculating the fluid density ρ2 at the impeller inlet, the velocity triangle at the impeller inlet combined with the momentum conservation equation can be used to obtain the flow velocity data at the impeller inlet, and then other parameter values ​​at the control surface 112 can be calculated; for the impeller outlet surface 113, with the impeller as the reference coordinate system, three relative quantities are defined, namely relative Mach number, relative stagnation temperature, and relative stagnation pressure; four types of losses are mainly considered, namely angle of attack loss, impeller channel loss, friction loss, and tip leakage loss. Combining relative quantity and loss analysis, the impeller outlet pressure P3 is iteratively solved according to the law of conservation of mass.

[0057] In this embodiment, the first control surface is configured as the volute inlet surface 111 .

[0058] S02: using the initial mechanism model to obtain data of the first control surface and the second control surface under multiple working conditions with preset parameters to form a corresponding data set.

[0059] Specifically, since in this embodiment, the initial mechanism model is a turbine model, the data of the first control surface can be obtained by actual measurement data or by using the actual measurement data and calculating according to the turbine geometry formula, so as to obtain data of multiple first control surfaces, and each data of the first control surface corresponds to one working condition; at the same time, the data of the second control surface under multiple working conditions with preset parameters is calculated by using the initial mechanism model, so as to obtain data of multiple second control surfaces, and each data of the second control surface corresponds to one working condition. The data of multiple first control surfaces and the corresponding data of multiple working conditions and the data of multiple second control surfaces and the corresponding data of multiple working conditions constitute corresponding data sets.

[0060] In this embodiment, the preset parameter is a fluid density parameter. In other words, the data of the first control surface obtained by executing step S02 in this embodiment is the fluid density of the first control surface, and the data of the second control surface is the fluid density of the second control surface.

[0061] Furthermore, in order to ensure that the obtained initial mechanism model can be optimized using the mechanism optimization method, before executing step S03, the initial mechanism model may also be verified, and the verification step includes:

[0062] The calculation data of the third control surface is obtained according to the initial mechanism model. Specifically, the initial mechanism model also has a third control surface, that is, the third control surface is located in the initial mechanism model, and the third control surface is located on the side of the second control surface away from the first control surface. It should be noted that in the description here, "away" does not refer to the distance in spatial position, but is based on the direction of action during the operation of the initial mechanism model. For example, in the turbine model provided in this embodiment, the third control surface is located at a position downstream of the airflow of the second control surface. Specifically, in this embodiment, the third control surface is the impeller outlet surface 113.

[0063] The third control surface parameter data calculated by the initial mechanism model is compared with the measured third control surface data. If the error between the two is within the preset range, the initial mechanism model is judged to be basically correct and can be optimized; if the error between the two is not within the preset range, the initial mechanism model is judged to have a large error and the mechanism formula therein needs to be adjusted. Optionally, the preset range can be set between 0-5%, and its specific value can be set according to demand, and no specific requirements are made here.

[0064] S03: Determine the correlation between the data of the first control plane and the data of the second control plane.

[0065] The correlation between the data of the first control plane and the data of the second control plane can be judged by calculating the correlation value in the following manner, and then judging the strength of the correlation by the size of the correlation value:

[0066]

[0067] Among them, x i is data of the first control plane; is the mean value of the first control surface data; i Data of the second control plane; is the mean value of the second control surface data; r is the correlation value. n is the number of operating conditions, and accordingly, also represents the number of data on the first control surface and the number of data on the second control surface.

[0068] Whether the correlation between the data of the first control plane and the data of the second control plane meets the preset requirement is determined according to the size of the correlation value. Optionally, in this embodiment, the preset requirement is that the correlation value is greater than 95%, that is, if the calculated correlation value is greater than 95%, it can be considered that the correlation between the data of the first control plane and the data of the second control plane meets the preset requirement, otherwise, if the calculated correlation value is less than or equal to 95%, it can be considered that the correlation between the data of the first control plane and the data of the second control plane does not meet the preset requirement.

[0069] If the correlation between the data of the first control surface and the data of the second control surface meets a preset requirement, the data of the first control surface and the data of the second control surface are fitted to obtain an empirical formula.

[0070] Specifically, when the data of the first control surface and the data of the second control surface are obtained, curve fitting can be performed by least square method, support vector machine and other methods to obtain an empirical formula with the data of the first control surface and the data of the second control surface as variables.

[0071] Furthermore, after the step of fitting the data of the first control surface and the data of the second control surface to obtain an empirical formula, the mechanism model optimization method further includes:

[0072] Calculate the goodness of fit of the empirical formula obtained by fitting:

[0073]

[0074] Among them, R 2 is the goodness of fit; i is the actual value data of the second control surface, Fit the predicted data to the second control surface, is the mean value of the second control surface data. Since in this embodiment, the preset parameter is the fluid density parameter, y i is the actual value data of the fluid density of the second control surface, Fit the predicted data for the second control surface fluid density, is the overall mean data of the fluid density of the second control surface.

[0075] The fitting effect is judged according to the goodness of fit. If the goodness of fit reaches a preset value, step S04 can be executed; if the goodness of fit does not reach the preset value, the empirical formula is refitted until the goodness of fit reaches the preset value. Optionally, the preset value is 95%, that is, if the goodness of fit reaches more than 95%, it is considered that the empirical formula can express the relationship between the data of the first control surface and the data of the second control surface.

[0076] If the correlation between the data of the first control surface and the data of the second control surface does not meet the preset requirements, other parameters can be replaced as preset parameters for optimization calculation. Specifically, the optimization calculation is the calculation process of steps S02 to S04. For example, flow rate, temperature, Mach number and other parameters are selected as preset parameters.

[0077] Furthermore, before executing the step of calculating the correlation value through the formula to determine the correlation, the correlation can be preliminarily determined by comparing the data of the first control plane with the data of the second control plane and observing the changing patterns of the data of the first control plane and the data of the second control plane. If the data of the first control plane and the data of the second control plane have certain similar changing patterns, the correlation value can be further calculated through the above formula to finally determine whether the correlation meets the preset requirements; accordingly, if it can be seen through comparison that the data of the first control plane and the data of the second control plane do not have certain similar changing patterns, other parameters can be replaced as preset parameters for optimization calculation.

[0078] It should be noted that, in some embodiments, multiple parameters can be selected as preset parameters when executing step S02, and then the parts with similar change patterns are selected by observing the change patterns of the data of the first control plane and the data of the second control plane to execute the step of calculating the correlation value through a formula to determine the correlation.

[0079] S04: Substituting the empirical formula into the initial mechanism model to replace the iterative calculation process of the preset parameters of the second control surface, so as to obtain the optimized mechanism model.

[0080] Substitute the empirical formula into the initial mechanism model to replace the iterative calculation process of the preset parameters of the second control surface in the initial mechanism model (such as the fluid density parameter in this embodiment) to obtain the optimized mechanism model. In other words, in this embodiment, the calculation of the fluid density parameter of the second control surface in the optimized mechanism model is obtained by the empirical formula, which eliminates the complex iterative process, thereby helping to reduce the time consumed by the calculation and improve the calculation efficiency, so as to meet the large self-correction requirements of the digital twin optimization calculation, effectively reduce the online dynamic optimization time of the digital model, and improve the practicality of the model.

[0081] Furthermore, after obtaining the optimization mechanism model, a step of verifying the optimization mechanism model may be performed, which step includes:

[0082] The data value of the third control surface under the same working condition is calculated using the optimization mechanism model and the initial mechanism model, respectively, so as to obtain the optimized data value and the initial value. Among them, the optimized data value is the data value of the third control surface calculated using the optimization mechanism model; the initial data value is the data value of the third control surface calculated using the initial mechanism model. In this embodiment, the third control surface is the impeller outlet surface 113, so correspondingly, the data value of the third control surface is the impeller outlet parameter value.

[0083] It should be noted that the parameters used for verifying the optimization mechanism model are not limited to the preset parameters, and other parameters may also be used. Optionally, the impeller outlet pressure value may be used for the verification step.

[0084] The data error of the optimized data value compared to the initial data value is calculated. For example, the data error can be the ratio of the difference between the optimized data value and the initial data value to the initial data value.

[0085] If the data error is less than 1%, it is judged that the optimization mechanism model meets the requirements and can be applied, for example, as a self-correcting model for digital twin optimization.

[0086] Furthermore, since a variety of loss models are applied in the turbine model provided as the initial mechanism model in this embodiment, it is necessary to select appropriate empirical parameter values ​​in the loss model and calibrate it in combination with the data of actual turbine operation. The model has a large number of empirical parameters, and each parameter affects each other. The practicality of the traditional parameter calibration method is low. Therefore, the mechanism model optimization method provided in the embodiment of the present invention also includes model empirical parameter optimization. The step of model empirical parameter optimization is performed before the step of calculation optimization, that is, the initial mechanism model used when executing step S02 is the model obtained after the model empirical parameter optimization. Specifically, the step of model empirical parameter optimization includes:

[0087] S05: Select parameters to be optimized from the empirical parameters of the initial mechanism model.

[0088] There are many empirical parameters in the initial mechanism model, and some of them are selected as parameters to be optimized, so as to perform subsequent steps to optimize the selected parameters to be optimized. The selection standard of the parameters to be optimized is to select parameters that have a great impact on the model accuracy from the empirical parameters adopted by the initial mechanism model. Specifically, step S05 includes:

[0089] A sensitivity analysis is performed on the empirical parameters to obtain corresponding sensitivity values. Specifically, the empirical parameters can be substituted into the software for performing sensitivity analysis, so that the corresponding sensitivity values ​​can be obtained through the software.

[0090] If the sensitivity value is greater than the preset value, the empirical parameter corresponding to the sensitivity is selected as the parameter to be optimized.

[0091] S06: Setting an initial value for each parameter to be optimized.

[0092] An initial value is set for each parameter to be optimized. The initial value is the initial value used in subsequent iterative optimization. Specifically, the initial value can be set using the average value of the value range of the parameter to be optimized. The value range of the parameter to be optimized can be obtained based on experience or known materials. For example, the impeller channel loss parameter Kp has a value range of 0.2 to 0.5, and the initial value of Kp is set to 0.35.

[0093] S07: Iteratively optimize the selected empirical parameters through an optimization algorithm.

[0094] The initial value is used in combination with the law of conservation of mass for prediction calculation. The initial value of the empirical parameter is used for calculation. After the calculation process, it is determined whether the corresponding mass conservation equation is satisfied. If it is satisfied, the next step of calculation can be performed. If not, the initial value must be changed, and then the iterative process is repeated until the mass conservation equation is satisfied. It is determined that the empirical parameter is optimized and the optimized empirical parameter value is obtained.

[0095] The specific process of iteratively calculating the initial value of the empirical parameter to obtain the optimized empirical parameter value can be performed using a particle swarm optimization algorithm (PSO). Specifically, the process of optimizing the parameter value using the particle swarm optimization algorithm can be referred to Figure 4 The process shown.

[0096] After obtaining the optimized empirical parameter value, the optimized empirical parameter value can be used to replace the empirical parameter mean value assumed in S06 before the iterative calculation to obtain the initial mechanism model after the empirical parameter optimization.

[0097] After obtaining the initial mechanism model after empirical parameter optimization, its optimization effect can be verified through the predicted performance of the optimized turbine.

[0098] This mechanism model optimization method effectively improves the calculation speed and accuracy of the mechanism model by optimizing empirical parameters and model calculation, reduces the time loss of model calculation, helps to perform online dynamic correction analysis of the digital model, reduces model calculation time, and improves the practicality of the model.

[0099] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technology in the field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A mechanism model optimization method, characterized in that: The mechanism model optimization method comprises: Acquire an initial mechanism model; wherein the initial mechanism model has a first control surface and a second control surface; The initial mechanism model is used to obtain data of a first control surface and data of a second control surface under a plurality of working conditions with preset parameters, so as to form a corresponding data set; wherein the corresponding data set includes corresponding data of the data of the first control surface and the working conditions and corresponding data of the data of the second control surface and the working conditions; the data of the first control surface is obtained by measured data or calculated by using the measured data; the data of the second control surface is obtained by iterative calculation of the initial mechanism model; determining a correlation between data of the first control surface and data of the second control surface under the same operating condition; If the correlation between the data of the first control plane and the data of the second control plane meets a preset requirement, fitting the data of the first control plane and the data of the second control plane to obtain an empirical formula; The empirical formula is substituted into the initial mechanism model to replace the iterative calculation process of the preset parameters of the second control surface, so as to obtain the optimized mechanism model.

2. The mechanism model optimization method according to claim 1, characterized in that: The step of determining the correlation between the data of the first control plane and the data of the second control plane includes: The correlation value between the data of the first control plane and the data of the second control plane is calculated by the following formula: Among them, x i data of the first control plane; is the mean value of the first control surface data; i data of the second control plane; is the mean value of the second control plane data.

3. The mechanism model optimization method according to claim 2, characterized in that: The preset requirement is that the correlation value is greater than 95%.

4. The mechanism model optimization method according to claim 1, characterized in that: After the step of fitting the data of the first control surface and the data of the second control surface to obtain an empirical formula, the mechanism model optimization method further includes: Calculate the goodness of fit of the empirical formula obtained by fitting: Among them, R 2 is the goodness of fit, y i is the actual value data of the second control surface, fitting the predicted data for the second control surface, is the mean value of the second control plane data; The fitting effect is judged according to the goodness of fit. If the goodness of fit reaches a preset value, the empirical formula is substituted into the initial mechanism model to replace the iterative calculation process of the preset parameters of the second control surface.

5. The mechanism model optimization method according to claim 1, characterized in that: The initial mechanism model further has a third control surface, and the third control surface is located on a side of the second control surface away from the first control surface; after the step of substituting the empirical formula into the initial mechanism model to replace the iterative calculation process of the preset parameters of the second control surface to obtain the optimized mechanism model, the mechanism model optimization method further includes: The optimization mechanism model and the initial mechanism model are respectively used to calculate the data value of the third control surface under the same working condition to obtain an optimized data value and an initial data value; wherein the optimized data value is the data value of the third control surface calculated by using the optimization mechanism model; and the initial data value is the data value of the third control surface calculated by using the initial mechanism model; Calculating a data error between the optimized data value and the initial data value; If the data error is less than 1%, it is determined that the optimization mechanism model meets the requirements.

6. The mechanism model optimization method according to claim 1, characterized in that: After the step of determining the correlation between the data of the first control surface and the data of the second control surface, the mechanism model optimization method further includes: If the correlation between the data of the first control plane and the data of the second control plane does not meet the preset requirement, other parameters are replaced as preset parameters for optimization calculation.

7. The mechanism model optimization method according to claim 1, characterized in that: The preset parameter is a fluid density parameter.

8. The mechanism model optimization method according to claim 1, characterized in that: The mechanism model optimization method also includes model empirical parameter optimization, and the steps of model empirical parameter optimization include: Selecting a parameter to be optimized from the empirical parameters of the initial mechanism model; wherein the parameter to be optimized is a parameter in the empirical parameters that affects the model accuracy; An initial value is set for each parameter to be optimized; the initial value is an average value of a value range of the parameter to be optimized; The initial value is used to perform prediction calculation, and the initial value is iterated according to the calculation result to obtain the optimized parameter value.

9. The mechanism model optimization method according to claim 8, characterized in that: The step of selecting parameters to be optimized from the empirical parameters of the initial mechanism model comprises: Performing sensitivity analysis on the empirical parameters and obtaining corresponding sensitivity values; If the sensitivity value is greater than a preset value, the empirical parameter corresponding to the sensitivity is selected as the parameter to be optimized.

10. The mechanism model optimization method according to claim 8, characterized in that: The steps of using the initial value to perform prediction calculation and iterating the initial value according to the calculation result to obtain the optimized parameter value include: The initial value is iterated using a particle swarm optimization algorithm to obtain an optimized parameter value.

Images