A method and device for generating flow channel planning of a hydraulic housing

Abstract

This application provides a method and apparatus for generating flow channels in a hydraulic housing, belonging to the field of hydraulic fluid control. The method includes: acquiring and extracting global design information of the hydraulic housing; initializing a reinforcement learning network based on the extracted global design information, and sequentially planning dual-interface flow path, single-interface flow path, and interfaceless flow path in conjunction with the constructed hydraulic housing environment to complete the overall flow path planning; determining whether the overall flow path planning is complete; if not, returning to the flow path planning process; if complete, establishing a hydraulic housing model based on the overall flow path planning. The hydraulic housing flow channel planning generation method and apparatus provided in this application can plan dual-interface, single-interface, and interfaceless flow channels in hydraulic housings and improve planning efficiency.

Description

Technical Field

[0001] The present application relates to the field of hydraulic fluid control technology, and in particular to a method and device for generating hydraulic housing flow channel planning. Background Art

[0002] The hydraulic housing is one of the core components of the hydraulic system. It distributes hydraulic oil by connecting the cylinder and other control valves to control the actuator to perform different actions. It plays an important role in engineering machinery, transportation and even modern industry.

[0003] At present, the traditional design method still mainly relies on manual design and modeling, which has a large amount of repetitive work, a high degree of reliance on experience, low design efficiency, and unstable design quality. Modern industrial systems have higher and higher performance requirements for hydraulic control systems, resulting in more complex shapes of hydraulic housing parts and an increasing application of new materials.

[0004] With the development of additive manufacturing technology, additive manufacturing of hydraulic housings is also being explored gradually. Limited by the characteristics of additive manufacturing, additive manufacturing products often need to be fine-machined to meet higher surface roughness requirements in order to meet design requirements. Therefore, additive manufacturing products should also take into account the machinability requirements of subtractive manufacturing in the design engineering.

[0005] Among them, the hydraulic housing contains a large number of flow channels. According to known design factors, the current hydraulic housing oil flow channel characteristics can be divided into double-interface flow channels, single-interface flow channels and non-interface flow channels.

[0006] At present, the commonly used hydraulic shell flow channel planning methods are mostly manual design, or the A* algorithm is used to plan the dual-interface flow channel. Among them, the manual design efficiency is low, and the currently used A* algorithm can only plan for dual-interface flow channel characteristics, and lacks planning research on complex flow channel shapes such as single-interface flow channel characteristics and non-interface flow channel characteristics. Summary of the invention

[0007] In view of this, the present application provides a method and device for planning and generating a hydraulic housing flow channel, which can plan dual-interface, single-interface and non-interface flow channels of a hydraulic housing and improve planning efficiency.

[0008] Specifically, the present application is implemented through the following technical solutions:

[0009] In a first aspect, the present application provides a method for generating a hydraulic housing flow channel plan, the method comprising:

[0010] Acquiring global design information of the hydraulic housing and extracting the global design information;

[0011] Initialize the reinforcement learning network based on the extracted global design information, and combine with the constructed hydraulic shell environment to sequentially plan the dual-interface flow path, the single-interface flow path and the non-interface flow path to complete the overall flow path planning; wherein, determine the interface distribution in the hydraulic shell environment according to the global design information, generate interface data based on the interface distribution, the interface data is used to characterize the flow path position, plan the flow path based on the interface data, select any point in the adjacent flow path of the non-interface flow path as the auxiliary interface of the non-interface flow path, convert the non-interface flow path planning into the single-interface flow path planning, and after each flow path planning is completed, write the currently planned flow path into the hydraulic shell environment as a constraint, and update the environment of the subsequent flow path planning so that the newly planned flow path does not conflict with the previously planned flow path;

[0012] It is determined whether the overall flow path planning is completed. If not, the process of planning the flow path is returned to again. If completed, a hydraulic housing model is established based on the overall flow path planning.

[0013] A second aspect of the present application provides a hydraulic housing flow channel planning and generation device, the device comprising an extraction module, a planning module and a processing module; wherein:

[0014] The extraction module is used to obtain the global design information of the hydraulic housing and extract the global design information;

[0015] The planning module is used to initialize the reinforcement learning network based on the extracted global design information, and plan the dual-interface flow path, the single-interface flow path and the non-interface flow path in sequence in combination with the constructed hydraulic housing environment to complete the overall flow path planning; wherein, the interface distribution in the hydraulic housing environment is determined according to the global design information, interface data is generated based on the interface distribution, the interface data is used to characterize the flow channel position, the flow channel is planned based on the interface data, any point in the adjacent flow channel of the non-interface flow channel path is selected as the auxiliary interface of the non-interface flow channel path, the non-interface flow channel path planning is converted into the single-interface flow channel path planning, and after each flow path planning is completed, the currently planned flow channel path is used as a constraint and written into the hydraulic housing environment, and the environment of the subsequent flow channel planning is updated so that the newly planned flow channel does not conflict with the previously planned flow channel;

[0016] The processing module is used to determine whether the overall flow path planning is completed. If not, it returns to the process of planning the flow path; if it is completed, it establishes a hydraulic housing model based on the overall flow path planning.

[0017] The hydraulic housing flow channel planning generation method and device provided by the present invention can use reinforcement learning to automatically complete the planning of dual-interface, single-interface and non-interface flow channels, can adapt to the planning of various complex hydraulic housing flow channels, improve the adaptability of the flow channel planning method, and different types of flow channels are constrained to each other, that is, the flow channels obtained by historical planning will continuously update the environment space, and the real-time updated environment space can avoid conflicts caused by the previous and next flow channel planning, and can also continuously reduce the environment space of the next flow channel planning, thereby improving the scientificity and efficiency of flow channel planning. Compared with traditional manual planning and modeling, the automatic planning algorithm using reinforcement learning greatly improves the design efficiency and shortens the design cycle in terms of efficiency; in terms of quality, the algorithm is used to perform global iteration of the design, which improves the design quality and avoids the limitations of manual experience. In addition, the method can also design dual-interface flow channel features, single-interface flow channel features and non-interface flow channel features at the same time, filling the gap in current related technologies. Furthermore, the method provided by the present invention marks each flow channel with an interface, plans the storage format of the flow channel data, and simultaneously uses preset points to transform the design of flow channels without interfaces into that of flow channels with interfaces, thereby simplifying the design process of flow channels without interfaces and improving the compatibility and efficiency of the design. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A schematic diagram of an internal flow channel of a hydraulic housing according to an exemplary embodiment of the present application;

[0019] Figure 2 A flowchart of a first embodiment of a method for planning and generating a flow channel for a hydraulic housing provided in the present application;

[0020] Figure 3 This is an interface data information diagram shown in an exemplary embodiment of the present application;

[0021] Figure 4 A schematic diagram of an annular groove flow channel shown in an exemplary embodiment of the present application;

[0022] Figure 5 A schematic diagram of flow channel information shown in an exemplary embodiment of the present application;

[0023] Figure 6 A demonstration diagram of dual-interface flow channel path planning shown in an exemplary embodiment of the present application;

[0024] Fig. 7A A schematic diagram showing the intersection of an extension line and a flow path according to an exemplary embodiment of the present application;

[0025] Figure 7B This is a schematic diagram showing an exemplary embodiment of the present application in which a line connecting an interface and a turning point of a flow path is an acute angle;

[0026] Figure 7C This is a schematic diagram showing an exemplary embodiment of the present application in which a line connecting an interface and a turning point of a flow path is an obtuse angle;

[0027] Figure 8 A schematic diagram of the structure of a first embodiment of a hydraulic housing flow channel planning and generating device provided in the present application;

[0028] Reference numerals:

[0029] 0: interface number 0;

[0030] 1: interface number 1;

[0031] 2: interface number 2;

[0032] L001-01: flow channel number L001-01;

[0033] L001-03: Flow channel serial number L001-03. DETAILED DESCRIPTION

[0034] Here, exemplary embodiments are described in detail, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application.

[0035] The terms used in this application are only for the purpose of describing specific embodiments and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in this article refers to and includes any or all possible combinations of one or more associated listed items.

[0036] It should be understood that although the terms first, second, third, etc. may be used in the present application to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0037] Specific embodiments are given below to introduce the technical solution of the present application in detail.

[0038] In order to facilitate a better understanding of the present solution, the characteristics of the double-interface flow channel, the single-interface flow channel and the non-interface flow channel inside the hydraulic housing are first described here. Figure 1This is a schematic diagram of the internal flow channel of the hydraulic housing according to an exemplary embodiment of the present application. Figure 1 A double-interface flow channel refers to a flow channel with interfaces at both ends, that is, the two interface coordinates of the flow channel are known, allowing the fluid to flow in from one interface, flow through the flow channel to the other interface and then flow out, as shown in a in the figure. The figure shown includes two double-interface flow channels, among which the leftmost and rightmost blocks of the double-interface flow channel a in the upper half of the figure and the double-interface flow channel a in the lower half are the positions of the two interfaces; a single-interface flow channel refers to a known flow channel with one interface and the other interface on a certain flow channel (the fluid can only flow in or out), as shown in b in the figure, and the other interface of the single-interface flow channel is on the double-interface flow channel a; a non-interface flow channel refers to a flow channel without fixed interface points at both ends, that is, both ends of the flow channel are on other flow channels, as shown in c in the figure.

[0039] Figure 2 This is a flow chart of the first embodiment of the method for planning and generating flow channels of a hydraulic housing provided by this application. Figure 2 , the method provided in this embodiment may include:

[0040] S201, obtaining global design information of a hydraulic housing, and extracting the global design information.

[0041] It should be noted that the global design information of the hydraulic housing is integrated in a UG plug-in. By picking up from the relevant model, all relevant design documents (including CAD drawings, design instructions, technical specifications, etc.) can be obtained to obtain the global design information of the hydraulic housing. Then, through analysis and organization, the required global design information can be extracted from it.

[0042] Specifically, in the present application, the extracted global design information at least includes design space information, interface data information, flow channel information and global parameters. Among them, the design space is used to determine the overall geometric shape and size range of the hydraulic housing model, which can ensure that all components can be placed and operated within a limited space; the design process of the design space includes: abstracting the design space into a combination of rectangular cubes, wherein the information of each rectangular cube includes the coordinate information of each vertex; based on the coordinate information, the diagonal points are stored to obtain a design space data set. Table 1 is a design space data set shown in an exemplary embodiment of the present application (a part of it is shown as an example), and specifically, Table 1 is as follows:

[0043] Table 1 Design space dataset

[0044] Design Space Diagonal point 1 Diagonal point 2 1 （0,0,0） （10,10,10） 2 （10,0,0） （20,20,20）

[0045] According to Table 1, the storage method of the design space can be clearly obtained.

[0046] Among them, the interface data information describes the connection mode and position between the hydraulic housing and the external system or other components, and is the key point of fluid input and output. Specifically, the design process of the interface data includes: determining the flow channel interface information in the design target, marking the interface serial number for each flow channel interface information; based on the interface serial number, determining the interface coordinates and vectors corresponding to each interface serial number; determining the flow channel parameters corresponding to each interface serial number based on the design requirements; the flow channel parameters at least include the interface diameter, design pressure and design flow; marking the special attributes corresponding to each interface serial number; the special attributes are the geometric attributes of the annular groove, and the special attributes at least include whether it is an annular groove, the main valve hole where the annular groove is located, and the interface type. For example, in one embodiment, the interface serial number is marked for each flow channel interface information, interface serial number 0 represents a normal interface, interface serial number 1 represents an annular groove interface, interface serial number 2 represents a flow channel middle section interface, interface serial number 3 represents a flow channel inflection point interface, and interface serial number 4 represents a specified interface. Figure 3 For an interface data information diagram shown in an exemplary embodiment of the present application, please refer to Figure 3 , the interface coordinates are the coordinates of the interface on the surface of the hydraulic housing. Since the interface usually has a certain depth, the interface vector is the vector from the interface coordinates to the start point of the flow channel. In particular, Figure 4 As shown in the figure, if the interface is a ring groove of the valve sleeve hole, the ring groove mark is 1 (i.e., interface serial number 1). Due to the special structure of the ring groove, the flow channel L2 must be perpendicular to the valve sleeve hole L1, so the interface coordinate is the center coordinate C of the ring groove, the path finding coordinate is the radius R of the ring groove, and the interface diameter is the depth L of the ring groove.

[0047] It should also be noted that the flow channel information describes the fluid channel in the hydraulic housing to ensure that the hydraulic oil can flow inside the housing according to the design requirements. The details will be described in detail below and will not be repeated here. The global parameters cover the overall design requirements and performance indicators of the hydraulic housing, including the minimum spacing of the hydraulic flow channel, the maximum aspect ratio of the flow channel, and the maximum spark hole length.

[0048] S202, initializing a reinforcement learning network based on the extracted global design information, and planning a dual-interface flow path, a single-interface flow path, and a non-interface flow path in sequence in combination with the constructed hydraulic housing environment to complete the overall flow path planning.

[0049] Among them, the interface distribution in the hydraulic housing environment is determined according to the global design information, and interface data is generated based on the interface distribution, the interface data is used to characterize the flow channel position and force conditions, and the flow channel is planned based on the interface data, and any point in the adjacent flow channels of the interface-free flow channel path is selected as the auxiliary interface of the interface-free flow channel path, and the interface-free flow channel path planning is converted into single-interface flow channel path planning, and after each flow channel path planning is completed, the currently planned flow channel path is used as a constraint and written into the hydraulic housing environment, and the environment of the subsequent flow channel planning is updated, so that the newly planned flow channel does not conflict with the previously planned flow channel.

[0050] It should be noted that the reinforcement learning network is initialized based on the extracted global design information mainly using the attribute information of the flow channel information in the global design information, wherein the attribute information of the flow channel information at least includes the flow channel number, interface number, flow channel diameter and flow rate.

[0051] It should also be noted that the reinforcement learning network can be an A3C reinforcement learning network, where A3C (Asynchronous Advantage Actor-Critic) is an advanced algorithm for reinforcement learning. It plays a significant role in solving decision-making problems in large-scale and complex environments. Its core idea is to accelerate the learning process and improve the stability and robustness of the strategy by training multiple agents in parallel. By using the A3C reinforcement learning network in the modeling process, learning can be performed simultaneously (in parallel) in different environments to improve design efficiency. It can also handle multiple design goals and provide real-time feedback to adjust parameters to improve design quality. In addition, there are many complex constraints in the design of hydraulic housings, such as the minimum flow channel spacing, the maximum aspect ratio, etc. The A3C reinforcement learning network can handle complex design constraints.

[0052] It should also be noted that the construction of the hydraulic housing environment is completed based on the global design information. For the specific implementation process, please refer to the introduction of related technologies, which will not be repeated here.

[0053] The following describes the planning of dual-interface flow path, single-interface flow path, and non-interface flow path:

[0054] Specifically, planning the dual-interface flow path includes:

[0055] (1) Obtain the property information of the flow channel based on the extracted global design information.

[0056] In combination with the above description, the attribute information of the flow channel information at least includes the flow channel number, the interface serial number, the flow channel diameter and the flow rate.

[0057] (2) Enter the serial numbers of the flow channel interfaces at both ends. The program obtains detailed information of the selected interface from the interface data table based on the serial numbers of the flow channel interfaces entered.

[0058] According to the interface serial number in the obtained flow channel information, information corresponding to the interface serial number can be obtained from a predefined interface data table. Figure 5 This is a schematic diagram of flow channel information shown in an exemplary embodiment of the present application, please refer to Figure 5 , where interface number 1 and interface number 2 represent the two interfaces at both ends of the flow channel. From the figure, we can see that the flow channels with flow channel numbers 1101 and 1102 are double-interface flow channels, the flow channel with flow channel number 1103 is a single-interface flow channel, and the flow channel with flow channel number 1104 is a non-interface flow channel. Combined with the above example of marking the interface number for each flow channel interface information, Figure 5 The flow channel with the flow channel number 1104 is a flow channel without an interface. The number 5 in the second column corresponding to this row represents a manually defined interface. The type of this interface is 4, that is, the interface number 4, which represents a specified interface (in this application, all flow channel information related content can be found in Figure 5 ).

[0059] (3) Based on the obtained interface detailed information, the flow channel information is initialized, including at least determining the flow channel diameter and flow rate.

[0060] Combination Figure 5 In the example shown, if the flow channel interface numbers at both ends of the input are 0 and 1, the obtained flow channel diameter is 10 (unit: mm), the flow rate is 5 (unit: L / min), and it can be concluded that the flow channel belongs to a dual-interface flow channel, and the flow channel information is initialized accordingly.

[0061] (4) Reinforcement learning is performed based on the initialized flow channel information to complete dual-interface flow channel path planning.

[0062] Specifically, this process can effectively perform dual-interface flow channel path planning and improve design efficiency and quality.

[0063] In addition, after reinforcement learning is performed based on the initialized flow channel information to complete the dual-interface flow channel path planning, it includes: obtaining the planned dual-interface flow channel path information; when it is checked that there are two dual-interface flow channel paths and there is a redundant relationship, relaxing the minimum flow channel spacing requirement; when the two dual-interface flow channel paths intersect and the flow area is greater than the preset flow area, removing one of the redundant flow channels.

[0064] Figure 6 This is a dual-interface flow path planning demonstration diagram shown in an exemplary embodiment of the present application, please refer to Figure 6, Figure A in the figure is a schematic diagram in which no redundant relationship is detected between the two dual-interface flow paths, and Figure B in the figure is a schematic diagram in which a redundant relationship is detected between the two dual-interface flow paths. Specifically, when there are two dual-interface flow paths and there is a redundant relationship (communication relationship), the minimum flow path spacing requirement can be appropriately relaxed or even allowed to intersect, and when the two dual-interface flow paths intersect and the flow area is greater than the preset flow area, the redundant flow path between the two dual-interface flow paths can be removed.

[0065] Specifically, planning a single-interface flow path includes:

[0066] (1) According to the attribute information of the flow channel, a point is selected on the planned dual-interface flow channel path as the flow channel interface.

[0067] For example, please refer to Figure 5 According to the information of the single-interface flow channel (flow channel number 1103), a point is specified on its connecting flow channel (flow channel number 1101) as another interface of the flow channel with flow channel number 1103. The interface number is represented by 2, that is, interface number 2.

[0068] (2) An extension line is generated based on the initial direction of the interface, and a single-interface flow path is planned according to the intersection relationship between the extension line and the dual-interface flow path.

[0069] It should be noted that the initial direction refers to the predetermined extension direction of the flow channel or the direction of fluid flow from the path finding starting point, which is usually represented by a vector. The path finding starting point refers to the starting point for searching and planning the path during the flow channel path planning process, which is the location of interface number 2 in this case.

[0070] Specifically, according to the intersection relationship between the extension line and the dual-interface flow path, a single-interface flow path is planned, including: determining whether the extension line intersects with the dual-interface flow path; when the extension line in the initial direction of the interface intersects with the dual-interface flow path, planning the single-interface flow path based on the intersection point; when the extension line in the initial direction of the interface does not intersect with the dual-interface flow path, determining the angular relationship between the line connecting the inflection points of the interface and the dual-interface flow path, and completing the planning of the single-interface path based on the angular relationship; the angular relationship includes an acute angle relationship and an obtuse angle relationship.

[0071] In specific implementation, the planning of the single flow channel path is completed based on the angle relationship, including: if the angle formed is an acute angle, the planning of the single flow channel path is completed based on the inflection point; if the angle formed is an obtuse angle, based on the location of the interface, and with the initial direction of the interface and the dual-interface flow channel path as a common perpendicular; the planning of the single flow channel path is completed with the intersection of the common perpendicular and the dual-interface flow channel path.

[0072] Fig. 7A This is a schematic diagram showing the intersection of an extension line and a flow path according to an exemplary embodiment of the present application. Figure 7B This is a schematic diagram showing an exemplary embodiment of the present application in which the line connecting the interface and the inflection point of the flow path is an acute angle. Figure 7C This is a schematic diagram showing an example embodiment of the present application in which the line connecting the interface and the inflection point of the flow path is an obtuse angle. Fig. 7A , 0, 1, 2 in the figure represent interface number 0, interface number 1, and interface number 2 respectively, L001-01 and L001-03 represent flow channel number L001-01 and flow channel number L001-03 respectively. If the extension line of the initial direction of interface number 2 intersects with the flow channel path, the interface is taken as the intersection to form a single interface flow channel with flow channel number L001-03; please also refer to Figure 7B and Figure 7C , the extension line of the initial direction of interface number 2 does not intersect the flow path, at this time through Figure 7B It can be seen that the angle between the interface number 2 and the inflection point of the dual-interface flow channel path is an acute angle, and the inflection point of the dual-interface flow channel of flow channel number L001-01 is selected as the other interface of the single-interface flow channel of flow channel number L001-03. Figure 7C It can be seen that the angle of the line connecting interface number 2 and the turning point of the double-interface flow path is an obtuse angle. Based on the location of interface number 2, a common perpendicular line is made between the initial direction of interface number 2 and flow channel number L001-01, and the planning of the single flow path is completed at the intersection of the common perpendicular line and the double-interface flow path.

[0073] Specifically, planning a non-interface flow channel path includes: selecting a point on a single-interface flow channel path that is adjacent to the non-interface flow channel path and has been planned, and inputting the information of the point into a reinforcement learning network, converting the non-interface flow channel path problem into a single-interface flow channel path planning problem, and the planned single-interface flow channel path uses the interface as a point inside the flow channel, thereby converting the single-interface flow channel path into a non-interface flow channel path. It should be noted that a non-interface flow channel usually refers to a flow channel that does not have a fixed starting point or end point in the flow channel network. Such a flow channel may have multiple reasonable connection schemes and path selections. Due to the existence of multiple solutions, directly processing the non-interface flow channel will increase the computational complexity and uncertainty. Therefore, through direct manual selection, the specified point is regarded as the starting point or end point of the single-interface flow channel, and the other parts are regarded as the path of the flow channel, and this new single-interface flow channel information is input into the reinforcement learning network for optimization.

[0074] It should also be noted that after each flow path planning is completed, the currently planned flow path is used as a constraint and written into the hydraulic housing environment. Specifically, the currently planned flow path is used as an obstacle when planning other flow paths.

[0075] S203, judging whether the overall flow path planning is completed, if not, returning to the process of planning the flow path; if completed, establishing a hydraulic housing model based on the overall flow path planning.

[0076] It should be noted that whether the overall flow path planning is completed is determined based on preset requirements (in specific implementation, the program will traverse the design information of each flow channel). If not completed, the process of planning the flow path is returned to until all flow channels are planned. After the overall flow path planning is completed, the feature library template is called according to the flow channel characteristics. For example, the NX Open script can be used to achieve modeling of the hydraulic housing through parameter drive.

[0077] The hydraulic housing flow path planning generation method provided in the present application obtains the global design information of the hydraulic housing and extracts the global design information, and then initializes the reinforcement learning network based on the extracted global design information, and combines with the constructed hydraulic housing environment to sequentially plan the dual-interface flow path, the single-interface flow path and the non-interface flow path to complete the overall flow path planning; wherein, after each flow path planning is completed, the currently planned flow path is used as a constraint and written into the hydraulic housing environment, and finally it is determined whether the overall flow path planning is completed. If not, the process of planning the flow path is returned to again; if completed, a hydraulic housing model is established based on the overall flow path planning. The hydraulic housing flow channel planning generation method and device provided by the present invention can use reinforcement learning to automatically complete the planning of dual-interface, single-interface and non-interface flow channels, can adapt to the planning of various complex hydraulic housing flow channels, improve the adaptability of the flow channel planning method, and different types of flow channels are constrained to each other, that is, the flow channels obtained by historical planning will continuously update the environment space, and the real-time updated environment space can avoid conflicts caused by the previous and next flow channel planning, and can also continuously reduce the environment space of the next flow channel planning, thereby improving the scientificity and efficiency of flow channel planning. Compared with traditional manual planning and modeling, the automatic planning algorithm using reinforcement learning greatly improves the design efficiency and shortens the design cycle in terms of efficiency; in terms of quality, the algorithm is used to perform global iteration of the design, which improves the design quality and avoids the limitations of manual experience. In addition, the method can also design dual-interface flow channel features, single-interface flow channel features and non-interface flow channel features at the same time, filling the gap in current related technologies. Furthermore, the method provided by the present invention marks each flow channel with an interface, plans the storage format of the flow channel data, and simultaneously uses preset points to transform the design of flow channels without interfaces into that of flow channels with interfaces, thereby simplifying the design process of flow channels without interfaces and improving the compatibility and efficiency of the design.

[0078] Corresponding to the aforementioned embodiment of a method for generating a hydraulic housing flow channel plan, the present application also provides an embodiment of a device for generating a hydraulic housing flow channel plan.

[0079] Figure 8 This is a structural diagram of the first embodiment of the hydraulic housing flow channel planning and generation device provided by this application. Figure 8 The device provided in this embodiment includes an extraction module 810, a planning module 820 and a processing module 830; wherein,

[0080] The extraction module 810 is used to obtain the global design information of the hydraulic housing and extract the global design information;

[0081] The planning module 820 is used to initialize the reinforcement learning network based on the extracted global design information, and plan the dual-interface flow path, the single-interface flow path and the non-interface flow path in sequence in combination with the constructed hydraulic housing environment to complete the overall flow path planning; wherein, the interface distribution in the hydraulic housing environment is determined according to the global design information, interface data is generated based on the interface distribution, the interface data is used to characterize the flow channel position, the flow channel is planned based on the interface data, any point in the adjacent flow channel of the non-interface flow channel path is selected as the auxiliary interface of the non-interface flow channel path, the non-interface flow channel path planning is converted into the single-interface flow channel path planning, and after each flow path planning is completed, the currently planned flow channel path is used as a constraint and written into the hydraulic housing environment, and the environment of the subsequent flow channel planning is updated so that the newly planned flow channel does not conflict with the previously planned flow channel;

[0082] The processing module 830 is used to determine whether the overall flow path planning is completed. If not, the process of planning the flow path is returned to again; if completed, a hydraulic housing model is established based on the overall flow path planning.

[0083] The device of this embodiment can be used to perform Figure 2 The steps, specific implementation principles and implementation processes of the method embodiment shown are similar and will not be repeated here.

[0084] The implementation process of the functions and effects of each unit in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, and will not be repeated here.

[0085] For the device embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The device embodiment described above is only schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present application scheme. A person of ordinary skill in the art can understand and implement it without paying any creative work.

[0086] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A method for generating flow channel planning for a hydraulic housing, characterized in that: The method comprises: Acquiring global design information of the hydraulic housing and extracting the global design information; Initializing the reinforcement learning network based on the extracted global design information, and combining with the constructed hydraulic housing environment, planning the dual-interface flow path, the single-interface flow path and the non-interface flow path in sequence to complete the overall flow path planning; wherein, determining the interface distribution in the hydraulic housing environment according to the global design information, generating interface data based on the interface distribution, the interface data being used to characterize the flow channel position, planning the flow channel based on the interface data, and writing the currently planned flow channel path into the hydraulic housing environment as a constraint after each flow channel path planning is completed, and updating the environment of the subsequent flow channel planning so that the newly planned flow channel does not conflict with the previously planned flow channel; Determine whether the overall flow path planning is completed, and if not, return to the process of planning the flow path; if completed, establish a hydraulic housing model based on the overall flow path planning; Plan the dual-interface flow path, including: Acquire property information of the flow channel based on the extracted global design information; Input the serial numbers of the flow channel interfaces at both ends, and the program will obtain detailed information of the selected interface from the interface data table according to the serial numbers of the flow channel interfaces input; Initialize the flow channel information based on the obtained interface detailed information, including at least determining the flow channel diameter and flow rate; Reinforcement learning is performed based on the initialized flow channel information to complete the dual-interface flow channel path planning; Planning the single-interface flow path includes: According to the attribute information of the flow channel, a point is selected on the planned dual-interface flow channel path as the flow channel interface; Generate an extension line based on the initial direction of the interface, and plan a single-interface flow path according to the intersection relationship between the extension line and the dual-interface flow path; Planning of non-interface flow path, including: A point is selected on the single-interface flow path that is adjacent to the no-interface flow path and has been planned, and the information of the point is input into the reinforcement learning network, so as to transform the no-interface flow path problem into the single-interface flow path planning problem. The planned single-interface flow path uses the interface as a point inside the flow channel, thereby transforming the single-interface flow path into a no-interface flow path.

2. The method according to claim 1, characterized in that: The process of performing reinforcement learning based on the initialized flow channel information to complete the dual-interface flow channel path planning includes: Obtain the planned dual-interface flow channel path information; When it is detected that there are two dual-interface flow path paths and there is a redundant relationship, the minimum flow path spacing requirement is relaxed; When two dual-interface flow channel paths intersect and the flow area is greater than the preset flow area, one of the redundant flow channels is removed.

3. The method according to claim 1, characterized in that: The planning of the single-interface flow channel path according to the intersection relationship between the extension line and the dual-interface flow channel path includes: Determining whether the extension line intersects with the dual-interface flow channel path; When the extension line of the initial direction of the interface intersects with the dual-interface flow channel path, planning a single-interface flow channel path based on the intersection point; When the extension line of the initial direction of the interface does not intersect with the dual-interface flow path, the angular relationship between the interface and the line connecting the inflection points of the dual-interface flow path is determined, and the planning of the single-interface flow path is completed based on the angular relationship; the angular relationship includes an acute angle relationship and an obtuse angle relationship.

4. The method according to claim 3, characterized in that The planning of a single flow channel path based on the angle relationship includes: If the angle formed is an acute angle, the single-interface flow channel path is planned based on the inflection point; If the angle formed is an obtuse angle, based on the location of the interface, the initial direction of the interface and the dual-interface flow path are used as a common perpendicular line; The planning of the single-interface flow channel path is completed by the intersection of the common vertical line and the dual-interface flow channel path.

5. The method according to claim 1, characterized in that The global design information at least includes a design space, and the design space is used to determine the overall geometric shape and size range of the hydraulic housing model; The design process of the design space includes: Abstracting the design space into a combination of rectangular cubes, wherein information of each rectangular cube includes coordinate information of each vertex; Based on the coordinate information, the corner points are stored to obtain a design space data set.

6. The method according to claim 1, characterized in that The global design information at least includes interface data, and the design process of the interface data includes: Determine the flow channel interface information in the design target, and mark the interface serial number for each flow channel interface information; Based on the interface serial numbers, determining the interface coordinates and vectors corresponding to the respective interface serial numbers; Determine the flow channel parameters corresponding to each interface number based on the design requirements; the flow channel parameters at least include the interface diameter, the design pressure and the design flow rate; Mark the special attributes corresponding to each interface serial number; the special attributes are the geometric attributes of the ring groove, and the special attributes at least include whether it is a ring groove, the main valve hole where the ring groove is located, and the interface type.

7. A hydraulic housing flow channel planning and generating device, characterized in that: The device comprises an extraction module, a planning module and a processing module; wherein, The extraction module is used to obtain the global design information of the hydraulic housing and extract the global design information; The planning module is used to initialize the reinforcement learning network based on the extracted global design information, and plan the dual-interface flow path, the single-interface flow path and the non-interface flow path in sequence in combination with the constructed hydraulic housing environment to complete the overall flow path planning; wherein, the interface distribution in the hydraulic housing environment is determined according to the global design information, interface data is generated based on the interface distribution, the interface data is used to characterize the flow channel position, the flow channel is planned based on the interface data, and after each flow channel path planning is completed, the currently planned flow channel path is used as a constraint and written into the hydraulic housing environment, and the environment of the subsequent flow channel planning is updated so that the newly planned flow channel does not conflict with the previously planned flow channel; The processing module is used to determine whether the overall flow path planning is completed, and if not, return to the process of planning the flow path; if completed, establish a hydraulic housing model based on the overall flow path planning; Plan the dual-interface flow path, including: Acquire property information of the flow channel based on the extracted global design information; Input the serial numbers of the flow channel interfaces at both ends, and the program will obtain detailed information of the selected interface from the interface data table according to the serial numbers of the flow channel interfaces input; Initialize the flow channel information based on the obtained interface detailed information, including at least determining the flow channel diameter and flow rate; Reinforcement learning is performed based on the initialized flow channel information to complete the dual-interface flow channel path planning; Planning the single-interface flow path includes: According to the attribute information of the flow channel, a point is selected on the planned dual-interface flow channel path as the flow channel interface; Generate an extension line based on the initial direction of the interface, and plan a single-interface flow path according to the intersection relationship between the extension line and the dual-interface flow path; Planning of non-interface flow path, including: A point is selected on the single-interface flow path that is adjacent to the no-interface flow path and has been planned, and the information of the point is input into the reinforcement learning network, so as to transform the no-interface flow path problem into the single-interface flow path planning problem. The planned single-interface flow path uses the interface as a point inside the flow channel, thereby transforming the single-interface flow path into a no-interface flow path.

Images