Cooperative optimization method for micro-interventional artificial heart

Abstract

The application relates to the technical field of medical equipment design, in particular to a collaborative optimization method of a micro interventional artificial heart, which comprises the following steps: obtaining optimization parameters and optimization targets of a subsystem and a parent system of the micro interventional artificial heart, and determining a child optimization problem of inner layer iteration and a parent optimization problem of outer layer iteration; selecting optimization algorithms which can avoid non-convex characteristics possibly occurring in a calculation process for the inner layer iteration and the outer layer iteration respectively, performing calculation of the inner layer iteration and the outer layer iteration based on the optimization algorithms to obtain a plurality of groups of front edge solution sets, demarcating a feasible region based on the plurality of groups of front edge solution sets and a design target, and selecting a design configuration of the micro interventional artificial heart and final assembly in the feasible region. Therefore, the problems of high clinical requirement limitation, difficult matching of components, more design modification times and waste of calculation and experiment resources in the related art are solved.

Description

Technical Field

[0001] This application relates to the field of medical device design technology, and in particular to a collaborative optimization method for a miniature interventional artificial heart. Background Technology

[0002] Miniature interventional artificial hearts require interventional treatment, so the size of the device is subject to very strict requirements; the size will limit the output capacity of the motor and the hydraulic efficiency of the fluid mechanical structure.

[0003] In related technologies, the structure of the fluid machinery can be optimized first to achieve certain hydraulic performance as much as possible, and then the power drive can be designed according to the power requirements of the fluid machinery structure.

[0004] However, due to the strict size limitations of miniature interventional artificial hearts, the fluid mechanical structure needs to reach very high rotational speeds to achieve the clinically required cardiac output, resulting in high loads on the blades. Under certain size constraints, it may be impossible to find an effective design point, thus requiring modification of the blade design and wasting computational and experimental resources. In addition, due to the power consumption limitations of miniature interventional artificial hearts, when the supplied power exceeds the limit, it may be impossible to guarantee the operation of the artificial heart during movement or when power is limited. Summary of the Invention

[0005] This application provides a collaborative optimization method for miniature interventional artificial hearts to address the problems in related technologies, such as high clinical requirements and difficulty in matching components when designing miniature interventional artificial hearts, which leads to numerous design modifications and waste of computational and experimental resources.

[0006] This application provides a collaborative optimization method for a miniature interventional artificial heart, comprising the following steps: obtaining optimization parameters and optimization objectives of the subsystems and parent system of the miniature interventional artificial heart; determining the sub-optimization problem of inner-layer iteration based on the optimization parameters and optimization objectives of the subsystems, and determining the parent optimization problem of outer-layer iteration based on the optimization parameters and optimization objectives of the parent system, wherein the inner-layer iteration realizes the optimization calculation of a single design component of the miniature interventional artificial heart, and the parent optimization problem of outer-layer iteration is determined based on the optimization parameters and optimization objectives of the parent system, wherein the outer-layer iteration realizes the coupling amount matching between various design components; selecting optimization algorithms that can avoid non-convex characteristics that may occur during the calculation process for both the inner-layer iteration and the outer-layer iteration; calculating multiple sets of leading edge solutions based on the optimization algorithms for the inner-layer iteration and the outer-layer iteration; delineating a feasible region based on the multiple sets of leading edge solutions and the design objectives; and selecting the design configuration and assembly of the miniature interventional artificial heart within the feasible region.

[0007] Optionally, the subsystem is a nonlinear and continuous system, and the parent system is a system oriented towards decoupling between subsystems. In this system, a set of coupling variables input into the parent system, when a single individual is passed to the subsystem, yields a set of leading edge solutions of the subsystem under the current coupling variable. The leading edge solution set contains the same-named response of another single individual in the set of coupling variables.

[0008] Optionally, the subsystem includes a fluid subsystem and an electromagnetic subsystem, wherein the first sub-optimization problem of the fluid subsystem is:

[0009]

[0010]

[0011] wrtα i (i = 1…6), N ∈ Ω1

[0012] Where P is the pump head or pressure difference between the inlet and outlet of the blood pump, HI is the hemolysis index, and f BP For the evaluation function, α i Let N be the parameter to be optimized, N be the rotational speed, and Ω1 be the union of its value range with the value range of the parent problem for rotational speed N; the second sub-optimization problem of the electromagnetic subsystem is:

[0013]

[0014]

[0015] wrtgap,r,DW,I,L,T∈Ω2

[0016] Among them, P OUT η is the motor output power, η is the motor efficiency, and g is the motor operating efficiency. BP Let be the evaluation function, gap, etc. be the parameters to be optimized, T be the load, and Ω2 be the union of the range of values ​​of Ω2 with the range of values ​​of load T in the parent problem.

[0017] Optionally, the parent optimization problem is:

[0018] min: E1 = (T2 - T1) 2

[0019] min: E2 = (N2 - N1) 2

[0020] wrtT,N∈Ω all

[0021] Where E1 and E2 are the difference functions of the coupling quantities, and the subscripts of T and N correspond to the input or output of the coupling quantities in the corresponding sub-optimization problems.

[0022] Optionally, multiple sets of leading edge solution sets are obtained by performing inner and outer iterations based on the optimization algorithm, including: setting the number of sample individuals and the number of iteration steps, cross-fertilizing parent samples to form new parent samples; defining the meaning of dominance, substituting the new parent samples into the evaluation function, sorting the sample results for non-dominated samples, determining the non-dominated level based on the number of dominant individuals in the sample results, and storing the individuals of the first non-dominated level into the non-dominated set, wherein dominance means: arbitrarily given two samples within the sampling range. For any f i All have but Dominate For each individual in the non-dominated set, each individual n in the set of individuals dominated by that individual is... j Subtract 1, if n j If -1 = 0, then individual j is stored in the target set, and individuals in the target set are assigned a second non-dominated level. The same operation is performed on individuals in the target set as on individuals in the non-dominated set, until all individuals are assigned a non-dominated level. The crowding degree of the sample results at each level is calculated, and individuals at the first and second non-dominated levels are selected and placed into a new parent population based on the crowding degree, until the new parent population is filled, resulting in multiple sets of leading edge solutions. Therefore, this application has at least the following beneficial effects:

[0023] The embodiments of this application can introduce optimization algorithms into the design process of miniature interventional artificial hearts to perform multi-objective optimization calculations, thereby avoiding non-convex characteristics that may occur during the calculation process, obtaining rich solution sets, and eliminating coupling problems between components, reducing the number of iterations. Thus, the embodiments of this application can achieve matching between system components while meeting design objectives, and quickly and economically find a suitable miniature interventional artificial heart design scheme for fabrication in the early stages of design.

[0024] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0025] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0026] Figure 1 This is a flowchart illustrating the collaborative optimization method for a miniature interventional artificial heart according to an embodiment of this application;

[0027] Figure 2 A schematic diagram of the design configuration and its assembly selected based on the calculation results of the embodiments of this application;

[0028] Figure 3 This is a schematic diagram of the leading edge solution set of the calculation results in an embodiment of this application;

[0029] Figure 4 This is a schematic diagram of the algorithm flow of an embodiment of this application. Detailed Implementation

[0030] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0031] Miniature interventional artificial hearts differ from other types of artificial hearts because they require interventional treatment, thus imposing very strict requirements on device size; however, size will limit the output capacity of the motor and the hydraulic efficiency of the fluid mechanical structure.

[0032] In related technologies, the structure of the fluid machinery can be optimized first to achieve certain hydraulic performance as much as possible, and then the power drive can be designed according to the power requirements of the fluid machinery structure.

[0033] However, when it comes to miniature interventional artificial hearts, due to strict size limitations, the fluid mechanical structure needs to reach very high rotational speeds to achieve the clinically required cardiac output, resulting in high loads on the blades. The output requirements of the motor drive may make it impossible to find an effective design point under the corresponding size constraints, thus requiring modification of the blade design and wasting computational and experimental resources.

[0034] In addition, considering the power consumption limitations, although the motor can reach the corresponding speed under the corresponding load, the power limit will be exceeded when there is a "supply exceeding demand", thus failing to guarantee the operation of the artificial heart in the case of movement or limited power.

[0035] To address the aforementioned problems, this application provides a collaborative optimization method for a miniature interventional artificial heart; specifically, Figure 1 This is a flowchart illustrating a collaborative optimization method for a miniature interventional artificial heart provided in an embodiment of this application.

[0036] like Figure 1 As shown, the collaborative optimization method for this miniature interventional artificial heart includes the following steps:

[0037] In step S101, the optimization parameters and optimization objectives of the subsystem and parent system of the miniature interventional artificial heart are obtained.

[0038] The subsystem is a nonlinear and continuous system, and the parent system is a system for decoupling between subsystems. A set of coupling variables input into the parent system, when a single individual is passed to the subsystem, results in a leading edge solution set of the subsystem under the current coupling variable. The leading edge solution set contains the same-named response of another single individual in the set of coupling variables.

[0039] It is understood that the embodiments of this application can obtain the optimization parameters and optimization objectives of the subsystems and parent systems of the miniature interventional artificial heart; wherein, the embodiments of this application can use at least one method to obtain the optimization parameters and optimization objectives, as follows:

[0040] I. Obtaining Optimization Parameters:

[0041] The core component of the miniature interventional artificial heart in this application embodiment can be a small-sized axial flow blood pump, which includes two main parts corresponding to two subsystems: the pump blade structure and the brushless hollow cup motor drive structure; wherein, the blade structure includes several blade structures and a hub structure.

[0042] (1) Inner iteration:

[0043] This application embodiment can acquire the blade curve design parameters and the motor size design parameters for use in the inner layer iteration process; wherein, to avoid conflicts during data exchange, the blade curve design parameters and the motor size design parameters only change in the inner layer iteration:

[0044] For the blade structure, since it comprises several blade structures and a hub structure, embodiments of this application can parameterize the generation curves of the corresponding structures, such as... Figure 2 As shown, the optimized parameters of the blade structure are obtained. For the brushless hollow cup motor structure, since the components have relatively fixed shapes, the embodiments of this application can directly optimize its design dimensions.

[0045] (2) Outer iteration:

[0046] In this embodiment, the motor speed under fixed size and input power is limited by the blade load, and the resistance of the fixed-shape blade, i.e. the load generated, is affected by the speed. Thus, there are two bidirectional coupling quantities: load and speed. In the design process of this embodiment, decoupling is achieved through outer layer iteration, i.e., matching between the two. Therefore, in the outer layer iteration, the two coupling quantities of load and speed are the optimization parameters.

[0047] It should be noted that in some embodiments, more design components (such as the size design of the conduit portion) may be involved, resulting in multiple subsystems. Regardless of whether there is a coupling relationship between these subsystems, they can be substituted into the algorithm of the application embodiment for optimization. For those with coupling relationships, it is necessary to clarify the coupling direction of the coupling amount to ensure that it does not appear as the overall optimized variable.

[0048] II. Optimization of target acquisition:

[0049] In this embodiment, for the outer layer iteration, the optimization objective is to make each optimization parameter approach 0; for the inner layer iteration, the optimization objective can be obtained by converting it into corresponding output parameters according to clinical needs.

[0050] In step S102, the sub-optimization problem of the inner iteration is determined according to the optimization parameters and optimization objectives of the subsystem, and the parent optimization problem of the outer iteration is determined according to the optimization parameters and optimization objectives of the parent system. The inner iteration realizes the optimization calculation of a single design component of the micro-interventional artificial heart, and the parent optimization problem of the outer iteration is determined according to the optimization parameters and optimization objectives of the parent system. The outer iteration realizes the matching of coupling between various design components.

[0051] It is understood that, in the embodiments of this application, the sub- and parent optimization problems of the inner and outer iterations can be confirmed separately based on the obtained optimization parameters and optimization objectives of the subsystem and parent system. The inner and outer iterations respectively realize the optimization calculation of individual design components of the miniature interventional artificial heart and the matching of coupling quantities between various design components. The confirmation method of the sub- and parent optimization problems in the embodiments of this application can be specifically as follows:

[0052] I. Sub-optimization problem:

[0053] In the embodiments of this application, such as Figure 3 As shown, the subsystem includes a fluid subsystem and an electromagnetic subsystem. The first sub-optimization problem of the fluid subsystem is:

[0054]

[0055]

[0056] wrtα i (i = 1…6), N ∈ Ω1

[0057] Where P is the pump head or pressure difference between the inlet and outlet of the blood pump, HI is the hemolysis index, and f BP For the evaluation function, α i Let N be the parameter to be optimized, N be the rotational speed, and Ω1 be the union of its range with the range of values ​​for rotational speed N in the parent problem; the second sub-optimization problem of the electromagnetic subsystem is:

[0058]

[0059]

[0060] wrtgap,r,DW,I,L,T∈Ω2

[0061] Among them, P OUT η is the motor output power, η is the motor efficiency, and g is the motor operating efficiency. BP Let be the evaluation function, gap, etc. be the parameters to be optimized, T be the load, and Ω2 be the union of the range of values ​​of Ω2 with the range of values ​​of load T in the parent problem.

[0062] II. Parent Optimization Problem:

[0063] In this embodiment of the application, the parent optimization problem is:

[0064] min: E1 = (T2 - T1) 2

[0065] min: E2 = (N2 - N1) 2

[0066] wrtT,N∈Ω all

[0067] Where E1 and E2 are the difference functions of the coupling quantities, and the subscripts of T and N correspond to the input or output of the coupling quantities in the corresponding sub-optimization problems, respectively. The interpolation function in this application embodiment can be measured by the square of the difference or by the norm, such as the Euclidean norm, etc., without making specific limitations.

[0068] It should be noted that, in order to conduct targeted design, the embodiments of this application may also use an experimental evaluation method to limit the design scope; wherein, the experimental evaluation may be completed jointly based on finite element computer simulation experiments and actual prototype testing experiments, or it may be an actual experimental test of the prototype after the design scheme is implemented, without specific limitations.

[0069] Specifically, the embodiments of this application can perform experimental evaluations to obtain a dataset on the optimized parameters and the optimization target. Then, the relational function between the optimized parameters and the optimization target can be found in the database as the evaluation function, and the range of optimized parameters can be determined from the same database.

[0070] It is understood that the embodiments of this application can use at least one method to solve the relationship function between the optimization parameters and the optimization objective, such as interpolation, regression, or the construction of an artificial neural network. When constructing an artificial neural network, the embodiments of this application can introduce a sampling step according to the design principles of computer experiments. The sampling method can be Sobol sampling sequence or Latin hypercube sampling, etc., without specific limitations.

[0071] In step S103, optimization algorithms that can avoid non-convex characteristics that may occur during the calculation process are selected for the inner and outer iterations respectively. Multiple sets of leading edge solutions are obtained by calculating the inner and outer iterations based on the optimization algorithms. The feasible region is delineated based on the multiple sets of leading edge solutions and the design target. The design configuration and assembly of the miniature interventional artificial heart are selected according to the feasible region.

[0072] It is understandable that, in order to avoid non-convex characteristics that may occur during the calculation process, the embodiments of this application may select appropriate optimization algorithms for the inner and outer layer iterations respectively; such as Figure 4 As shown, inner and outer layer iterative calculations are performed based on the selected optimization algorithm to obtain multiple sets of leading edge solutions; the feasible region is planned based on the leading edge solutions and design objectives, and the design configuration and assembly of the miniature interventional artificial heart are selected from the feasible region.

[0073] Specifically, firstly, select the appropriate optimization algorithm:

[0074] In this embodiment, suitable optimization algorithms can be selected for the inner and outer iterations to select the leading solution set composed of all valid solutions. The inner loop can use particle swarm optimization (such as ant colony optimization), while the outer loop can use augmented Lagrange algorithm or simulated annealing algorithm, which can further save computational resources. In the following embodiments, the NSGA-II algorithm will be used as the basis for the description.

[0075] II. Multiple sets of leading edge solution sets were obtained through calculation:

[0076] In this embodiment, multiple sets of leading edge solution sets are obtained by performing inner and outer iterations based on the optimization algorithm, including: setting the number of sample individuals and the number of iteration steps; cross-pollinating parent samples to form new parent samples; defining the meaning of dominance; substituting the new parent samples into the evaluation function; sorting the sample results for non-dominated samples; determining the non-dominated level based on the number of dominant individuals in the sample results; and storing the individuals of the first non-dominated level into the non-dominated set. The meaning of dominance is: given any two samples within the sampling range... For any f i All have but Dominate For each individual in the non-dominated set, assign each individual n in the set of individuals dominated by that individual. j Subtract 1, if n j If -1 = 0, then individual j is stored in the target set, and individuals in the target set are assigned the second non-dominated level. The same operation is performed on individuals in the target set as on individuals in the non-dominated set, until all individuals are assigned the non-dominated level. The crowding degree of the sample results at each level is calculated, and individuals at the first and second non-dominated levels are selected according to the crowding degree and placed into the new parent population, until the new parent population is filled, resulting in multiple sets of leading edge solutions.

[0077] Specifically, based on the definition of the Pareto leading edge, embodiments of this application can set x * For any x ∈ D, if there exists f(x) ≤ f(x) * That is, the following conditions are not met:

[0078]

[0079] Then x * It is considered an efficient solution to multi-objective optimization problems.

[0080] The specific iterative process of the optimization algorithm (NSGA-II algorithm) in this application embodiment is as follows:

[0081] (1) Generate the parent generation, set the number of sample individuals to N0, set the number of iteration steps to K0, and perform crossover on the parent generation samples:

[0082]

[0083]

[0084] Where, p 1，k+1 and p 2，k+1 It is the (k+1)th generation individual generated after crossover; p 1，k and p 2，k It is the selected kth generation individual; β qi It is the uniform distribution factor, and its calculation method is as follows:

[0085]

[0086]

[0087] Among them, u i It is a random number belonging to [0, 1), and η is the crossover exponent; mutation operations will occur in some computation steps:

[0088]

[0089] Where, p k It is the selected kth generation individual; p k+1 It is the (k+1)th generation individual obtained by mutating pk; p max k and p min k These are the upper and lower bounds of the decision variable, respectively; δ k The calculation formula is as follows:

[0090]

[0091] Where, r k η is a uniformly distributed random number in [0, 1]. m This is the variation distribution index.

[0092] (2) Definition of dominance: A vector consisting of multiple evaluation functions Given any two samples within the sampling range For any f i All have but Dominate

[0093] In this embodiment, the sample results can be non-dominated and ordered in the evaluation function after sample operations are performed: each individual i in the sample results has two parameters n. i and p i n i p represents the number of individuals that dominate individual i in the sample results. i It is the set of individuals dominated by individual i; find all n in the population through iterative comparison. i Individuals with a dominance level of 0 are assigned a non-dominance rank of 1 and stored in the non-dominance set rank1. For each individual in set rank1, the non-dominance rank of each individual in the set it dominates is set to rank1. j Subtract 1 from all n, if n j If -1 = 0, store individual j in set rank2 and assign it a non-dominated rank of 2. Repeat the above operation for individuals in rank2 until all individuals are assigned a non-dominated rank, which is called the Pareto rank.

[0094] (3) Calculate the crowding degree for the sample results at each level, which is usually defined as: in and For each of the two points closest to the observation point i, the points are calculated and then sorted from highest to lowest within the same Pareto level.

[0095] (4) Individuals with Pareto level 1 and level 2 are placed into the new parent population, and so on, until N0 is filled. If level k cannot be filled during the filling process, but level k+1 is filled beyond N0, then fill them one by one according to the crowding degree from high to low; then start the next iteration.

[0096] In this embodiment, the above operations (1)-(4) are performed in both the inner and outer loops. The corresponding evaluation functions, optimized parameters, and optimization objectives are described in the optimization problem description. After iterative calculation, this embodiment obtains multiple sets of leading edge solution sets. The leading edge solution set obtained by the outer iteration corresponds to the coupling situation and is a set of solutions close to the coordinate axis. The other leading edge solution sets correspond to the leading edge solutions obtained by solving each subproblem. Each set of other leading edge solution sets corresponds to each point on the leading edge solution set obtained by the outer iteration, such as... Figure 3 As shown.

[0097] III. Delineate the feasible region, and select the design configuration and assembly of the miniature interventional artificial heart based on the feasible region:

[0098] In this embodiment, a feasible region can be defined based on the design objectives, and design points within the feasible region can be selected for further comparison to identify feasible design points. The obtained points and their corresponding final design configurations and assembly can be as follows: Figure 2 As shown.

[0099] In summary, the collaborative optimization method for miniature interventional artificial hearts proposed in the embodiments of this application can introduce optimization algorithms into the design process of miniature interventional artificial hearts, perform multi-objective optimization calculations, thereby avoiding non-convex characteristics that may occur during the calculation process, obtaining rich solution sets, and eliminating coupling problems between components, reducing the number of iterations. Thus, under the premise of meeting the design objectives, it is possible to achieve matching between the components of the system, and quickly and economically find a suitable design scheme for miniature interventional artificial hearts to be manufactured in the early stage of design.

[0100] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0101] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0102] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0103] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (FPGAs), field-programmable gate arrays (FPGAs), etc.

[0104] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0105] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A collaborative optimization method for a miniature interventional artificial heart, characterized in that, Includes the following steps: Obtain the optimization parameters and optimization objectives of the subsystems and parent system of the miniature interventional artificial heart; The inner iteration sub-optimization problem is determined based on the optimization parameters and optimization objectives of the subsystem, and the outer iteration parent optimization problem is determined based on the optimization parameters and optimization objectives of the parent system. The inner iteration realizes the optimization calculation of a single design component of the micro-interventional artificial heart, and the outer iteration realizes the matching of coupling between various design components. Optimization algorithms that can avoid non-convex characteristics that may occur during the calculation process are selected for inner and outer iterations respectively. Multiple sets of leading edge solutions are obtained by calculating inner and outer iterations based on the optimization algorithms. Feasible regions are delineated based on the multiple sets of leading edge solutions and the design target. The design configuration and assembly of the miniature interventional artificial heart are selected according to the feasible regions.

2. The collaborative optimization method for a miniature interventional artificial heart according to claim 1, characterized in that, The subsystem is a nonlinear and continuous system, and the parent system is a system oriented towards decoupling between subsystems. In the parent system, a set of coupling variables is input, and after a single individual is passed to the subsystem, a set of leading edge solutions of the subsystem under the current coupling variable is obtained. The leading edge solution set contains the same-named response of another single individual in the set of coupling variables.

3. The collaborative optimization method for a miniature interventional artificial heart according to claim 1, characterized in that, The subsystem includes a fluid subsystem and an electromagnetic subsystem, wherein... The first sub-optimization problem of the fluid subsystem is: Where P is the pump head or pressure difference between the inlet and outlet of the blood pump, HI is the hemolysis index, and f BP For the evaluation function, α i Here, N is the rotational speed, and Ω1 is the union of its value range with the value range of rotational speed N in the parent problem. The second sub-optimization problem of the electromagnetic subsystem is: Among them, P OUT η is the motor output power, η is the motor efficiency, and g is the motor operating efficiency. BP The evaluation function is defined as follows: gap is the air gap width, r is the wire diameter, DW is the width of the silicon steel sheet, I is the current amplitude, L is the total winding length, T is the load, and Ω2 is the union of the range of values ​​of Ω2 with the range of values ​​of load T in the parent problem.

4. The collaborative optimization method for a miniature interventional artificial heart according to claim 3, characterized in that, The parent optimization problem is: Where E1 and E2 are the difference functions of the coupling quantities, and the subscripts of T and N correspond to the input or output of the coupling quantities in the respective sub-optimization problems. The range of values ​​for T and N in the parent problem.

5. The collaborative optimization method for a miniature interventional artificial heart according to claim 1, characterized in that, The calculation based on the optimization algorithm for inner and outer iterations yields multiple sets of leading edge solutions, including: Set the number of individual samples and the number of iterations, and cross the parent samples to form new parent samples; After defining the meaning of dominance, the new parent sample is substituted into the evaluation function to sort the sample results into non-dominated categories. The non-dominated level is determined based on the number of dominant individuals in the sample results, and individuals of the first non-dominated level are stored in the non-dominated set. Here, dominance means: given any two samples within the sampling range... For any objective evaluation function All have ,but Dominate ; For each individual in the non-dominated set, each individual n in the set of individuals dominated by that individual is... j Subtract 1, if n j If -1=0, then store individual j in the target set, assign the second non-dominated level to the individuals in the target set, and perform the same operation on the individuals in the target set as on the individuals in the non-dominated set, until all individuals are assigned the non-dominated level; Calculate the crowding degree of the sample results at each level, and select individuals from the first non-dominated level and the second non-dominated level according to the crowding degree to put into the new parent population until the new parent population is filled, thus obtaining multiple sets of front edge solutions.

Images