Automatic production method and production system for pump

Abstract

The invention belongs to the technical field of automatic production of pumps, and particularly relates to an automatic production method and system of a pump. An automatic production method of a pump comprises the following steps that S1, according to target performance, final optimization parameters of a basic impeller in a basic pump are calculated; s2, casting the special impeller according to the final optimization parameters; and S3, the specially-made impeller and other parts in the basic pump are assembled, and then the specially-made pump is obtained. Intelligent design is integrated into the production process, the development and production period of the special pump is shortened, and the overall cost of the special pump is reduced.

Description

An automated production method and system for pumps Technical Field

[0001] This application belongs to the field of automated pump production technology, and particularly relates to an automated pump production method and production system. Background Technology

[0002] Pumps, as key equipment in the equipment manufacturing field, are widely used in important sectors of the national economy such as energy, metallurgy, and water supply and drainage. They are fundamental supporting equipment for industrial production and people's livelihood.

[0003] Current production lines typically mass-produce general-purpose pumps. To enable pumps to better fulfill their role as fundamental supporting equipment, it's necessary to optimize the impeller parameters based on the specific performance requirements (head, efficiency) for each application scenario. For ease of description, these optimized pumps will be referred to as custom pumps. Custom pumps are often produced in small quantities as samples only after continuous parameter modifications and testing in the laboratory.

[0004] Therefore, the following problems exist in the process of acquiring custom pumps: ① The current production line is not intelligent enough and is not suitable for the characteristics of custom pumps with multiple varieties and variable batches; ② Custom pumps are mostly distributed from design to production, resulting in excessively long development and production cycles and excessively high overall costs.

[0005] This application content

[0006] The purpose of this application is to overcome the shortcomings of the prior art and provide an automated production method for pumps that integrates intelligent design into the production process, shortens the development and production cycle of specialized pumps, and reduces the overall cost of specialized pumps.

[0007] To achieve the above objectives, this application adopts the following technical solution: an automated production method for a pump, comprising the following steps: S1, calculating the final optimized parameters of the base impeller in the base pump according to the target performance; S2, casting a special impeller according to the final optimized parameters; S3, assembling the special impeller with other components in the base pump to obtain the special pump.

[0008] Preferably, step S1 further includes the following steps: S11, determining the target head range and / or target efficiency range based on the usage scenario; S12, recording the impeller inlet angle, outlet angle, number of blades, and wrap angle as parameters to be optimized, and recording other geometric parameters in the impeller besides the parameters to be optimized as invariant parameters, with the invariant parameters taking the values ​​of the corresponding parameters in the current basic impeller; constructing the pump's performance objective function based on the parameters to be optimized and the invariant parameters, and using the target head range and / or target efficiency range as constraints; S13, using a surrogate model to solve the pump's performance objective function based on the PSO algorithm, predicting several sets of potential solutions for the parameters to be optimized; each set of potential solutions for the parameters to be optimized... The pump performance objective function includes impeller inlet angle, outlet angle, number of blades, and wrap angle. The surrogate model is used to select several sets of solutions for the parameters to be optimized based on the PSO algorithm. These solutions, along with the invariant parameters of the base impeller, are fed into the surrogate model for head prediction and / or efficiency prediction. S14: Based on CFD flow field simulation analysis, the actual head range and actual efficiency range corresponding to the potential solutions of each set of optimized parameters are calculated. Potential solutions whose actual head range and / or actual efficiency range fall within the target head range and / or target efficiency range are selected as candidate solutions. S15: A set of candidate solutions is randomly selected as the final optimized parameters for the base impeller.

[0009] Preferably, replace S15 with S15´: S15´, calculate the score G of each group of alternative solutions, and take the group of alternative solutions with the highest score as the final optimization parameters of the basic impeller. ;in, and These represent the first weighting coefficient and the second weighting coefficient, respectively. and These represent the rated head and rated efficiency of the corresponding alternative solutions, respectively.

[0010] Preferably, in S14, the following is also included: using the potential solutions of the optimization parameters and their corresponding actual head range and actual efficiency range as samples in the training set and validation set, and using the training set and validation set to train and optimize the surrogate model.

[0011] Preferably, in S13, the surrogate model solves the pump's performance objective function based on the PSO algorithm, predicting several sets of potential solutions for the parameters to be optimized. This also includes the following: S131, the solutions for different sets of parameters to be optimized are clustered according to the numerical differences in the impeller inlet angle, outlet angle, number of blades, and wrap angle; S132, the PSO algorithm is used to select several sets of solutions for the parameters to be optimized from different regions of the clustered distribution, and these solutions, along with the invariant parameters of the basic impeller, are fed into the surrogate model for head prediction and / or efficiency prediction of each set of solutions; if the head prediction and / or efficiency prediction obtained from all the solutions for the parameters to be optimized selected by the PSO algorithm in a certain region are not within the target head range and / or target efficiency range, then the current region is filtered out; otherwise, the current region is retained; S133, after further region segmentation within the retained clustered distribution region using the PSO algorithm, S132 is repeated for further filtering, and after n iterations, the solutions for the parameters to be optimized within the retained region are output as potential solutions for the parameters to be optimized.

[0012] Preferably, after S15 or S15', S16 is also included: S16, which determines the rationality of the efficiency gain η of the special pump compared to the basic pump. ; ; ;in, and These are the first flow field stability coefficient and the second flow field stability coefficient, respectively. and These are the impeller inlet angle and outlet angle of the basic pump, respectively; k is the entropy production correction factor. Indicates the characteristic ratio; Indicates total entropy production; This represents viscous dissipative entropy production; if If the efficiency gain η is reasonable, then it is determined to be reasonable; otherwise, it is unreasonable. If the efficiency gain η is unreasonable, then S16 is re-executed after selecting a new set of alternative solutions as the final optimization parameters of the basic impeller.

[0013] Preferably, S3 is followed by S4 and S5: S4, assembling the special pump with the motor, transmission mechanism, base, control cabinet, valves and pipelines into a special centrifugal pump system; S5, performing finished product testing on the special centrifugal pump system, classifying qualified products into qualified products and unqualified products, sending qualified products to the warehouse, and reporting unqualified products for repair.

[0014] This application also provides an automated pump production system, comprising: an optimization design module, a special impeller manufacturing module, a special pump assembly module, a special centrifugal pump system assembly module, and a testing module; the optimization design module calculates the final optimized parameters of the base impeller in the base pump based on the target performance and then sends them to the special impeller manufacturing module; the special impeller manufacturing module casts the special impeller according to the final optimized parameters and then sends it to the special pump assembly module; the special pump assembly module assembles the special impeller with other components in the base pump to obtain the special pump, and then sends the special pump to the special centrifugal pump system assembly module; the special centrifugal pump system assembly module assembles the special pump with a motor, transmission mechanism, base, control cabinet, valves, and pipelines to form a special centrifugal pump system, and then sends the special centrifugal pump system to the testing module; the testing module performs finished product testing on the special centrifugal pump system; each module is programmed or configured to execute the steps of the above-described automated pump production method.

[0015] This application also provides a computer-readable storage medium, characterized in that: the computer-readable storage medium stores a computer program that is programmed or configured to perform an automated production method for a pump as described above.

[0016] This application also provides a computer program product, including a computer program / instructions, characterized in that: when the computer program / instructions are executed by a processor, they implement the steps of an automated pump production method as described above.

[0017] The beneficial effects of this application are: (1) This application provides an automated production method for pumps, which takes intelligent design as the first step in the production process, and through efficient and rapid intelligent design, shortens the development and production cycle of special pumps and reduces the overall cost of special pumps.

[0018] (2) The automated production method of this application can efficiently calculate the final optimized parameters of the basic impeller without human intervention based on performance requirements. The casting of a special impeller based on the final optimized parameters can greatly improve the performance of the special pump to which it belongs, so that the special pump produced can better meet the performance requirements of specific application scenarios.

[0019] (3) The production method of this application is highly flexible and suitable for the characteristics of multi-variety and variable batch production of special pumps.

[0020] (4) In the production method of this application, the proxy model is based on the PSO algorithm to solve the performance objective function of the pump. The resulting potential solutions of the optimization parameters are already a small number of solutions selected from tens of thousands of combinations of optimization parameters under the premise of satisfying the constraints. Even if a set is randomly selected from these potential solutions as the final optimization parameters of the basic impeller, the performance of the resulting custom pump is improved compared to the unoptimized basic pump. Furthermore, the time and computational cost of obtaining these potential solutions are significantly reduced compared to directly using CFD flow field simulation analysis. Therefore, the overall time and computational cost of further verifying the corresponding performance data and selecting alternative solutions using CFD flow field simulation analysis based on these potential solutions are also much less than those of directly using CFD flow field simulation analysis. Therefore, this application can significantly reduce the computational cost and time cost of optimizing the design of the pump.

[0021] (5) Compared with the method of directly using the surrogate model to predict the head and / or efficiency of the solutions of all different groups of optimization parameters to obtain potential solutions, this application significantly reduces the computational overhead and computation time of the surrogate model, and further improves the overall optimization design efficiency of this application.

[0022] (6) Since there may be more than one set of alternative solutions, in order to meet the difficulty of selection under the constraint that different sets of alternative solutions show different trends in efficiency and head, this application eliminates the problem of inconsistent standards when manually selecting and selects a set of alternative solutions as the final optimization parameters of the basic impeller by using a score G that comprehensively considers efficiency and head.

[0023] (7) This application further ensures the rationality of the final optimized parameters of the base impeller by judging the rationality of the efficiency gain η of the special pump compared with the base pump.

[0024] (8) This application can continuously utilize the accurate results of the potential solutions of the optimization parameters calculated by CFD flow field simulation analysis (i.e., the actual head range and the actual efficiency range) during use to further optimize the prediction effect of the surrogate model. Therefore, the production method of this application does not require a large amount of time to train the surrogate model in the early stage, and can achieve rapid production and continuously increase the prediction accuracy of the surrogate model as it is used. Attached Figure Description

[0025] Figure 1 is a flowchart of an automated production method for a pump according to this application; Figure 2 is a performance curve of a specially designed pump and a base pump before optimization; Figure 3 is a pressure distribution diagram of the impeller at 50% span and the middle of the volute under multiple operating conditions of the specially designed pump and the base pump before optimization; Figure 4 is a velocity streamline distribution diagram of the impeller midsection under multiple operating conditions of the specially designed pump and the base pump before optimization; Figure 5 is a vortex core distribution diagram of the impeller flow channel under multiple operating conditions of the specially designed pump and the base pump before optimization. Detailed Implementation

[0026] To make the technical solution of this application clearer and more explicit, the application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Solutions derived by those skilled in the art through equivalent substitution and conventional reasoning of the technical features of the technical solution of this application without creative effort all fall within the protection scope of this application.

[0027] An automated production method for a pump according to this application, as shown in Figure 1, includes the following steps: S1, calculating the final optimized parameters of the base impeller in the base pump according to the target performance; S2, casting a special impeller according to the final optimized parameters; S3, assembling the special impeller with other components in the base pump to obtain the special pump.

[0028] Optionally, S4 is included after S3: S4 assembles a special pump with a motor, transmission mechanism, base, control cabinet, valves and pipelines into a special centrifugal pump system.

[0029] Optionally, S5 can be included after S4: S5 is used to perform finished product testing on the special centrifugal pump system, classify the products as qualified and unqualified, send the qualified products to the warehouse, and report the unqualified products for repair.

[0030] S1 also includes the following steps: S11, determining the target head range and / or target efficiency range based on the usage scenario.

[0031] S12, the impeller inlet angle, outlet angle, number of blades and wrap angle are recorded as parameters to be optimized, and the other geometric parameters in the impeller other than the parameters to be optimized are recorded as constant parameters. The constant parameters are taken as the values ​​of the corresponding parameters in the current basic impeller. The pump performance objective function is constructed based on the parameters to be optimized and the constant parameters, and the target head range and / or target efficiency range are used as constraints.

[0032] S13, the surrogate model solves the pump's performance objective function based on the PSO algorithm and predicts several sets of potential solutions for the parameters to be optimized; each set of potential solutions for the parameters to be optimized includes the impeller inlet angle, outlet angle, number of blades, and wrap angle; the pump's performance objective function is an entropy production calculation formula.

[0033] Using known impeller geometry parameters and corresponding actual head and efficiency ranges to construct sample and validation sets, the LSSVR model is trained and optimized to obtain a surrogate model. The method for obtaining the surrogate model is existing technology and will not be described in detail here.

[0034] S14. After calculating the actual head range and actual efficiency range corresponding to the potential solutions of each set of optimization parameters based on CFD flow field simulation analysis, potential solutions whose actual head range and / or actual efficiency range fall within the target head range and / or target efficiency range are selected as candidate solutions.

[0035] S15, randomly select a set of alternative solutions as the final optimization parameters for the basic impeller.

[0036] It is important to emphasize that this application focuses on optimizing and improving the base impeller of a selected base pump, and the optimization and improvement are limited to the parameters to be optimized. In other words, the invariant parameters in the base impeller remain unchanged after the optimization and improvement. Invariant parameters include the impeller outer diameter and the impeller inlet diameter.

[0037] The base pump can be pre-selected by technicians (a manufacturer of a specific type of pump uses this method to better meet customer needs, so the base pump must be a specific type of pump from that manufacturer), or it can be selected based on the usage scenario.

[0038] Optionally, replace S15 with S15´: S15´, calculate the score G of each set of alternative solutions, and take the set of alternative solutions with the highest score as the final optimization parameters of the base impeller. ;in, and These represent the first weighting coefficient and the second weighting coefficient, respectively. and These represent the rated head and rated efficiency of the corresponding alternative solutions, respectively. and It is obtained in S14 by calculating the actual head range and actual efficiency range corresponding to each set of potential solutions through CFD flow field simulation analysis.

[0039] Optionally, S14 also includes the following: using the actual head range and actual efficiency range corresponding to the potential solution of the optimization parameters as samples in the training set and validation set, and using the training set and validation set to train and optimize the surrogate model in order to continuously improve the prediction accuracy of the surrogate model for the potential solution.

[0040] It's important to note that knowing a set of specific optimization parameters is equivalent to determining the parameters of a pump. CFD flow field simulation analysis can then be used to obtain the accurate actual head and efficiency ranges of the pump based on these parameters. However, calculating the actual head and efficiency ranges using CFD flow field simulation analysis is not only computationally intensive but also time-consuming. Furthermore, due to the varying values ​​of the parameters to be optimized, there can be thousands of combinations. Therefore, directly using CFD flow field simulation analysis to sequentially calculate the actual head and efficiency ranges for all parameters under different values, and then using the relationship between the actual head and efficiency ranges and the target head and / or efficiency ranges to determine the final impeller optimization parameters, is impractical.

[0041] In this application, the proxy model is based on the PSO algorithm to solve the performance objective function of the pump. The resulting potential solution of the optimization parameters is already a small number of solutions selected from tens of thousands of combinations of optimization parameters under the premise of satisfying the constraints. Even if a set is randomly selected from these potential solutions as the final optimization parameters of the basic impeller, the resulting custom pump will still have improved performance compared to the basic pump before optimization.

[0042] Furthermore, the time and computational cost of finding these potential solutions are significantly reduced compared to directly using CFD flow field simulation analysis. Therefore, the overall time and computational cost of further verifying the corresponding performance data and selecting alternative solutions using CFD flow field simulation analysis based on these potential solutions are also much less than those of directly using CFD flow field simulation analysis.

[0043] Therefore, this application can significantly reduce the computational overhead and time cost of optimizing pump design.

[0044] In S13, the surrogate model solves the pump's performance objective function based on the PSO algorithm; it predicts several sets of potential solutions for the parameters to be optimized, and also includes the following: S131, the solutions for different sets of parameters to be optimized are clustered according to the numerical differences in the impeller inlet angle, outlet angle, number of blades, and wrap angle; S132, the PSO algorithm is used to select several sets of solutions for the parameters to be optimized from different regions of the clustered distribution, and these solutions, along with the invariant parameters of the basic impeller, are fed into the surrogate model to predict the head and / or efficiency of each set of solutions; if the head and / or efficiency predictions obtained by all the solutions for the parameters to be optimized selected by the PSO algorithm in a certain region are not within the target head and / or target efficiency range, then the current region is filtered out; otherwise, the region is retained; S133, the PSO algorithm is used to further segment the retained clustered distribution regions, and S132 is repeated for filtering again. After n iterations, the solutions for the parameters to be optimized in the retained regions are output as potential solutions for the parameters to be optimized.

[0045] In this embodiment, a surrogate model is used to solve the pump's performance objective function based on the PSO algorithm. In practical engineering, approximately 50 iterations are sufficient to obtain several sets of potential solutions for the parameters to be optimized. Compared to directly using a surrogate model to predict the head and / or efficiency of all different sets of solutions for the parameters to be optimized to obtain potential solutions, this application significantly reduces the computational overhead and time of the surrogate model, further improving the overall optimization design efficiency of this application.

[0046] Because there may be more than one set of alternative solutions, assuming only the target efficiency range is a constraint, the following situation may exist: the rated efficiency of the first set of alternative solutions is slightly lower than that of the second set, but the rated head is much larger than that of the second set. To address the difficulty of selecting alternative solutions when their efficiencies and heads exhibit different trends under the constraint, this application employs S15', which eliminates the problem of inconsistent standards during manual selection. It selects a set of alternative solutions as the final optimization parameters for the base impeller based on a score G that comprehensively considers both efficiency and head.

[0047] Optionally, after S15 or S15', S16 is also included: S16 determines the reasonableness of the efficiency gain η of the special pump compared to the basic pump. ; ; ;in, and These are the first flow field stability coefficient and the second flow field stability coefficient, respectively. and These are the impeller inlet angle and outlet angle of the basic pump, respectively; k is the entropy production correction factor. Indicates the characteristic ratio; Indicates total entropy production; This represents viscous dissipative entropy production.

[0048] like If the efficiency gain η is reasonable, then it is determined to be reasonable; otherwise, it is unreasonable. If the efficiency gain η is unreasonable, then S16 is re-executed after selecting a new set of alternative solutions as the final optimization parameters of the basic impeller.

[0049] By using CFD flow field simulation analysis to calculate the actual head range and actual efficiency range corresponding to the final optimized parameters of the basic impeller, the total entropy production of the customized pump can be obtained. and viscous dissipative entropy production .

[0050] This application further ensures the rationality of the final optimized parameters of the base impeller by judging the rationality of the efficiency gain η of the special pump compared with the base pump.

[0051] Optionally, if an unreasonable efficiency gain is detected z times consecutively, an error will be reported to the technical staff.

[0052] In this embodiment, The value range is 0.82 to 0.95; The value range is 0.88 to 0.96; The value range is 26° to 45°. The value of is in the range of 15° to 40°; the value of k is in the range of 1.05 to 1.25.

[0053] In S13, the role of the surrogate model is to predict the head range and efficiency range of the corresponding pump based on the solution of the parameters to be optimized selected by the PSO algorithm.

[0054] This application can continuously utilize the accurate results of the potential solutions to the optimization parameters calculated by CFD flow field simulation analysis (i.e., the actual head range and actual efficiency range) during use to further optimize the prediction performance of the surrogate model. Therefore, the production method of this application can achieve rapid production without investing a lot of time in training the surrogate model in the early stage, and the prediction accuracy of the surrogate model can be continuously increased with use.

[0055] In S2, the process of casting a special impeller according to the final optimized parameters uses devices such as a laser sintering molding cavity, a sand mold additive manufacturing system, a mold shell baking furnace, and a centrifugal casting machine.

[0056] In S3, other components in the basic pump include the volute, impeller, and motor. First, the motor is assembled: the stator and rotor are wound and wound using a winding and embedding machine, with the winding evenly distributed along the circumference of the stator slots. Next, the volute is assembled: this includes impregnation and curing using a vacuum impregnation tank and curing oven.

[0057] In S5, the finished product inspection of the specialized centrifugal pump system includes the following: first, a visual inspection of the specialized centrifugal pump system; second, an airtightness test of the centrifugal pump system using a three-station parallel test structure; and finally, performance tests including no-load testing and turbulent kinetic energy analysis. Specialized centrifugal pump systems that pass all of the above tests are classified as qualified products; otherwise, they are classified as unqualified products.

[0058] To more intuitively demonstrate that the specially designed pump produced by the method of this application does indeed have performance improvements compared to the corresponding basic pump, as shown in Figures 2 to 5: Figure 2 shows the performance curves of the specially designed pump and the basic pump before optimization. Under the premise of falling within the target efficiency range and the target head range, the efficiency of the specially designed pump is higher than that of the basic pump, and the head before and after the rated flow is basically the same as or even higher than that of the basic pump; Figure 3 shows the pressure distribution of the impeller at 50% span and the middle of the volute under multiple operating conditions of the specially designed pump and the basic pump before optimization. It can be seen that the pressure distribution in the middle of the volute and the specially designed impeller is more uniform in the specially designed pump; Figure 4 shows the velocity streamline distribution of the impeller cross section under multiple operating conditions of the specially designed pump and the basic pump before optimization. It can be seen that the velocity streamline distribution of the specially designed impeller cross section is more uniform in the specially designed pump; Figure 5 shows the vortex core distribution in the impeller flow channel under multiple operating conditions of the specially designed pump and the basic pump before optimization. It can be seen that the number of vortex cores in the flow channel of the specially designed impeller in the specially designed pump is less and more stable.

[0059] In the production method of this application, the final optimized parameters of the basic impeller can be efficiently calculated based on performance requirements without manual intervention. Casting a special impeller according to the final optimized parameters can significantly improve the performance of the special pump to which it belongs, so that the produced special pump can better meet the performance requirements of specific application scenarios.

[0060] The production method described in this application takes intelligent design as the first step in the production process, and through efficient and rapid intelligent design, shortens the development and production cycle of special pumps and reduces the overall cost of special pumps.

[0061] The production method described in this application is highly flexible and suitable for the production of various types and batches of specialized pumps.

[0062] This application also provides an automated production system for pumps, including: an optimized design module, a special impeller manufacturing module, a special pump assembly module, a special centrifugal pump system assembly module, and a testing module.

[0063] The optimization design module calculates the final optimized parameters of the base impeller in the base pump based on the target performance and then sends them to the special impeller manufacturing module. The special impeller manufacturing module casts the special impeller according to the final optimized parameters and then sends it to the special pump assembly module. The special pump assembly module assembles the special impeller with other components in the base pump to obtain the special pump, and then sends the special pump to the special centrifugal pump system assembly module. The special centrifugal pump system assembly module assembles the special pump with the motor, transmission mechanism, base, control cabinet, valves, and pipelines to form a special centrifugal pump system, and then sends the special centrifugal pump system to the testing module. The testing module is used to perform finished product testing on the special centrifugal pump system.

[0064] Each module is programmed or configured to perform the steps of the above-described automated pump production method.

[0065] This application also provides a computer-readable storage medium storing a computer program that is programmed or configured to perform the above-described automated production method of a pump.

[0066] This application also provides a computer program product, including a computer program / instructions, which are executed by a processor to implement the steps of the above-described automated production method for a pump.

[0067] The technologies, shapes, and structures not described in detail in this application are all well-known technologies. It should also be noted that the above are merely preferred embodiments of this application and are not intended to limit the scope of this application. The components or steps in the embodiments of this application can be decomposed and / or recombined, and these decompositions and / or recombinations should be considered as equivalent solutions of this application and should all fall within the protection scope of this application.

Claims

1. An automated production method for pumps, characterized in that, Includes the following steps: S1, Calculate the final optimized parameters of the base impeller in the base pump according to the target performance; S2, Cast a special impeller according to the final optimized parameters; S3, Assemble the special impeller with other components in the base pump to obtain the special pump.

2. The automated production method for a pump according to claim 1, characterized in that, S1 also includes the following steps: S11, determining the target head range and / or target efficiency range based on the usage scenario; S12, designating the impeller inlet angle, outlet angle, number of blades, and wrap angle as parameters to be optimized, and designating other geometric parameters in the impeller besides the parameters to be optimized as invariant parameters, with the invariant parameters taking the values ​​of the corresponding parameters in the current basic impeller; constructing the pump's performance objective function based on the parameters to be optimized and the invariant parameters, and using the target head range and / or target efficiency range as constraints; S13, using a surrogate model to solve the pump's performance objective function based on the PSO algorithm, predicting several sets of potential solutions for the parameters to be optimized; each set of potential solutions for the parameters to be optimized contains... Impeller inlet angle, outlet angle, number of blades, and wrap angle; the pump's performance objective function is an entropy production calculation formula; a surrogate model is used to select several sets of solutions for the parameters to be optimized based on the PSO algorithm, and these solutions, along with the invariant parameters of the basic impeller, are fed into the surrogate model for head prediction and / or efficiency prediction of each set of solutions; S14, after calculating the actual head range and actual efficiency range corresponding to the potential solutions of each set of optimization parameters based on CFD flow field simulation analysis, potential solutions whose actual head range and / or actual efficiency range fall within the target head range and / or target efficiency range are selected as candidate solutions; S15, a set is randomly selected from the candidate solutions as the final optimization parameters of the basic impeller.

3. The automated production method for a pump according to claim 2, characterized in that, Replace S15 with S15´: S15´, calculate the score G of each group of alternative solutions, and take the group of alternative solutions with the highest score as the final optimization parameters of the basic impeller. ;in, and These represent the first weighting coefficient and the second weighting coefficient, respectively. and These represent the rated head and rated efficiency of the corresponding alternative solutions, respectively.

4. The automated production method for a pump according to claim 2, characterized in that, In S14, also This includes the following: using the potential solutions of the optimization parameters and their corresponding actual head range and actual efficiency range as samples in the training set and validation set, and using the training set and validation set to train and optimize the surrogate model.

5. The automated production method for a pump according to claim 2, characterized in that, In S13, the surrogate model solves the pump's performance objective function based on the PSO algorithm, predicting several sets of potential solutions for the parameters to be optimized. This also includes the following: S131, the solutions for different sets of parameters to be optimized are clustered according to the numerical differences in the impeller inlet angle, outlet angle, number of blades, and wrap angle; S132, the PSO algorithm is used to select several sets of solutions for the parameters to be optimized from different regions of the clustered distribution, and these solutions, along with the invariant parameters of the basic impeller, are fed into the surrogate model for head prediction and / or efficiency prediction of each set of solutions; if the head prediction and / or efficiency prediction obtained from all the solutions for the parameters to be optimized selected by the PSO algorithm in a certain region are not within the target head range and / or target efficiency range, then the current region is filtered out; otherwise, the current region is retained; S133, the PSO algorithm is used to further segment the retained clustered distribution regions, and S132 is repeated for filtering again. After n iterations, the solutions for the parameters to be optimized in the retained regions are output as potential solutions for the parameters to be optimized.

6. The automated production method for a pump according to claim 2, characterized in that, Following S15 or S15', there is also S16: S16 determines the reasonableness of the efficiency gain η of the specialized pump compared to the basic pump. ； ； ;in, and These are the first flow field stability coefficient and the second flow field stability coefficient, respectively. and These are the impeller inlet angle and outlet angle of the basic pump, respectively; k is the entropy production correction factor. Indicates the characteristic ratio; Indicates total entropy production; This represents viscous dissipative entropy production; if If the efficiency gain η is reasonable, then it is determined to be reasonable; otherwise, it is unreasonable. If the efficiency gain η is unreasonable, then S16 is re-executed after selecting a new set of alternative solutions as the final optimization parameters of the basic impeller.

7. An automated production method for a pump according to any one of claims 1-6, characterized in that, S3 is followed by S4 and S5: S4 assembles the special pump with the motor, transmission mechanism, base, control cabinet, valves and pipelines into a special centrifugal pump system; S5 performs finished product testing on the special centrifugal pump system, classifies qualified products into qualified products and unqualified products, sends qualified products to the warehouse, and reports unqualified products for repair.

8. An automated production system for pumps, characterized in that, include: Optimized design module, special impeller manufacturing module, special pump assembly module, special centrifugal pump system assembly module, and testing module; The optimization design module calculates the final optimized parameters of the base impeller in the base pump based on the target performance and then sends them to the special impeller manufacturing module. The special impeller manufacturing module casts a special impeller based on the final optimized parameters and then sends it to the special pump assembly module. The special pump assembly module assembles the special impeller with other components in the base pump to obtain a special pump, and then sends the special pump to the special centrifugal pump system assembly module. The special centrifugal pump system assembly module assembles the special pump with a motor, transmission mechanism, base, control cabinet, valves, and pipelines to form a special centrifugal pump system, and then sends the special centrifugal pump system to the testing module. The testing module performs finished product testing on the special centrifugal pump system. Each module is programmed or configured to execute the steps of an automated pump production method as described in any one of claims 1-7.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program that is programmed or configured to perform an automated production method of a pump as described in any one of claims 1-7.

10. A computer program product comprising a computer program / instructions, characterized in that: When executed by a processor, the computer program / instructions implement the steps of an automated pump production method as described in any one of claims 1-7.

Images