Numerical control machining optimization algorithm based on knowledge graph and depth-first search
Through the CNC machining optimization algorithm based on knowledge graph and depth-first search, the problem of traditional process optimization methods relying on experience and lack of systematicity and scientificity is solved, and the scientificity, systematicity, efficiency and accuracy of process optimization is improved, and the process parameters and processing paths suitable for specific parts can be quickly and accurately determined.
Patent Information
- Application Number
- CN202411884045.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-20
- Publication Date
- 2025-05-06
AI Technical Summary
Traditional CNC machining process optimization methods rely on experience, lack systematicity and scientificity, it is difficult to find the optimal process solution, it is impossible to effectively integrate multi-source data, and it is difficult to quickly and accurately determine the process parameters and processing paths suitable for specific parts.
Using CNC machining optimization algorithm based on knowledge graph and depth-first search, through data collection, sorting and knowledge graph construction, the depth-first search algorithm is used to search for possible processing paths in the knowledge graph, and quickly find process solutions suitable for specific parts.
It improves the scientificity and systematicity of process optimization, improves the efficiency and accuracy of process optimization, and can quickly and accurately determine the process parameters and processing paths suitable for specific parts, shortens the time for process optimization, and improves production efficiency.
Abstract
Description
Technical Field
[0001] The present invention relates to the field of CNC machining technology, and in particular to a CNC machining optimization algorithm based on knowledge graph and depth-first search. Background Art
[0002] In CNC machining, determining the appropriate process parameters and machining paths is crucial to improving machining quality, efficiency and reducing costs. Traditional process optimization methods often rely on experience and trial and error, which are inefficient and difficult to guarantee the optimal process solution. With the continuous accumulation of CNC machining data and the development of information technology, a more scientific and efficient CNC machining process optimization method is needed.
[0003] At present, there is no relevant solution on the market to solve the following technical problems existing in the existing CNC machining process optimization: Traditional process optimization methods rely on experience, lack of systematicity and scientificity, and it is difficult to find the optimal process solution. It is impossible to effectively integrate multi-source CNC machining data and it is difficult to fully explore the potential knowledge in the data. It is difficult to quickly and accurately determine the process parameters and processing paths suitable for the processing of specific parts. Summary of the invention
[0004] In response to the above technical problems in the related technology, the present invention proposes a CNC machining optimization algorithm based on knowledge graph and depth-first search, which can overcome the above shortcomings of the prior art.
[0005] To achieve the above technical objectives, the technical solution of the present invention is implemented as follows: A CNC machining optimization algorithm based on knowledge graph and depth-first search includes the following steps: S1 Data Collection: Collect CNC machining process data, including machining equipment parameters, machining quality inspection data, process route, tool information, etc.; S2 data collation: S2.1: Cleaning and preprocessing the CNC machining process data, removing noise and erroneous data, and unifying the data format and unit; S2.2: Establish corresponding data standards for the CNC machining process data, and associate the CNC machining process data with evaluation indicators, wherein the evaluation indicators include machining quality, efficiency, cost, process stability, etc.; S3 constructs a knowledge graph: extracts a number of entities from the CNC machining process data, and determines the relationship between different entities; merges the newly extracted entities with the existing knowledge graph, removes duplicate and conflicting information, and thus completes the construction of the knowledge graph; S4 prepares optimization conditions: presets the search strategy and target conditions required for the search, wherein the search strategy includes specifying an entity corresponding to the part as a starting point; S5 depth-first search for process optimization: S5.1 Search: In the knowledge graph, possible processing paths are searched based on the search strategy; during the search, the processing paths are judged and screened according to the target conditions; S5.2 Optimization: If a processing path that meets the target condition is found, the process plan corresponding to the processing path is recorded, including the processing method, tool selection, cutting parameters, etc., and S5.3 is continued; otherwise, the search strategy is adjusted or the target condition is relaxed, so the target condition after relaxation is reset, and S5.1 is returned; S5.3 Summary: If the knowledge graph has been completely searched and all the recorded process plans are summarized, then continue to S5; otherwise, it means that there may be other processing paths that meet the current target conditions, which have not been searched yet, so return to S5.3; S6: Output all the process plans recorded in summary, that is, the optimized process plan, including specific processing methods, processing sequence, tool selection, cutting parameter settings, etc.; S7: Analyze the optimized process plan to evaluate whether it meets the actual production requirements. If so, end the operation; otherwise, further adjust the search strategy or relax the target conditions, and return to S5.
[0006] Preferably, in S1, the CNC machining process data is collected from multiple channels such as machining equipment control systems, quality inspection departments, process documents, tool suppliers and operators.
[0007] Preferably, the S2 comprises the following steps: S2.1 Entity recognition and extraction: extracting a number of entities from the CNC machining process data, including parts, equipment, tools, process parameters, etc.; S2.2 Determine entity relationships: Determine the relationships between different entities, such as "parts - used tools", "equipment - processed parts", etc. S2.3 Knowledge fusion: The newly extracted entities are integrated with the existing knowledge graph to construct a knowledge graph, thereby removing duplicate and conflicting information; S2.4 Knowledge storage: storing the knowledge graph to facilitate quick query and reasoning.
[0008] Preferably, in S2.4, a graph database is used to store the knowledge graph.
[0009] Preferably, in S5, a depth-first search algorithm is used to search for possible processing paths in the knowledge graph.
[0010] Preferably, in S7, the optimized process scheme is analyzed in terms of quality, efficiency, cost, etc.
[0011] Preferably, the search strategy is path query.
[0012] Preferably, the target conditions include processing quality requirements, efficiency requirements, cost requirements, etc.
[0013] Compared with the prior art, the present disclosure has the following beneficial effects: (A) It improves the scientificity and systematicity of process optimization, integrates knowledge and experience in the field of CNC machining by using knowledge graphs, avoids the limitation of traditional process optimization methods that rely on experience, and makes the process optimization process more scientific and systematic. The process scheme suitable for specific parts processing can be quickly found in the knowledge graph through the depth-first search algorithm, which improves the efficiency and accuracy of process optimization.
[0014] (B) Effectively integrates multi-source CNC machining data, can collect CNC machining process data from multiple channels, and clean, pre-process and integrate them, fully mining the potential knowledge in multi-source data. This helps to improve the quality and reliability of process solutions, and also provides more comprehensive data support for corporate decision-making.
[0015] (C) Rapidly and accurately determine the process parameters and processing paths. Through the depth-first search algorithm, possible processing paths can be quickly searched in the knowledge graph and screened according to the preset target conditions, so as to quickly and accurately determine the process parameters and processing paths suitable for specific parts processing. This greatly shortens the process optimization time and improves production efficiency.
[0016] (D) Improved flexibility and adaptability. The search strategy and target conditions can be adjusted according to actual production needs. It has strong flexibility and adaptability. Whether it is for the processing of new parts or the improvement of existing processes, it can provide effective optimization solutions. DETAILED DESCRIPTION
[0017] The technical solutions in the embodiments of the present invention will be described clearly and completely below.
[0018] In order to facilitate understanding of the above technical solutions of the present invention, the above technical solutions of the present invention are described in detail below through specific usage methods.
[0019] A CNC machining optimization algorithm based on knowledge graph and depth-first search includes the following steps: S1 Data Collection: Collect CNC machining process data, including machining equipment parameters, machining quality inspection data, process route, tool information, etc.; S2 data collation: S2.1: Cleaning and preprocessing the CNC machining process data, removing noise and erroneous data, and unifying the data format and unit; S2.2: Establish corresponding data standards for the CNC machining process data, and associate the CNC machining process data with evaluation indicators, wherein the evaluation indicators include machining quality, efficiency, cost, process stability, etc.; S3 constructs a knowledge graph: extracts a number of entities from the CNC machining process data, and determines the relationship between different entities; merges the newly extracted entities with the existing knowledge graph, removes duplicate and conflicting information, and thus completes the construction of the knowledge graph; S4 prepares optimization conditions: presets the search strategy and target conditions required for the search, wherein the search strategy includes specifying an entity corresponding to the part as a starting point; S5 depth-first search for process optimization: S5.1 Search: In the knowledge graph, possible processing paths are searched based on the search strategy; during the search, the processing paths are judged and screened according to the target conditions; S5.2 Optimization: If a processing path that meets the target condition is found, the process plan corresponding to the processing path is recorded, including the processing method, tool selection, cutting parameters, etc., and S5.3 is continued; otherwise, the search strategy is adjusted or the target condition is relaxed, so the target condition after relaxation is reset, and S5.1 is returned; S5.3 Summary: If the knowledge graph has been completely searched and all the recorded process plans are summarized, then continue to S5; otherwise, it means that there may be other processing paths that meet the current target conditions, which have not been searched yet, so return to S5.3; S6: Output all the process plans recorded in summary, that is, the optimized process plan, including specific processing methods, processing sequence, tool selection, cutting parameter settings, etc.; S7: Analyze the optimized process plan to evaluate whether it meets the actual production requirements. If so, end the operation; otherwise, further adjust the search strategy or relax the target conditions, and return to S5.
[0020] In a specific embodiment, in S1, the CNC machining process data is collected from multiple channels such as machining equipment control systems, quality inspection departments, process documents, tool suppliers, and operators.
[0021] In a specific embodiment, S2 comprises the following steps: S2.1 Entity recognition and extraction: extracting a number of entities from the CNC machining process data, including parts, equipment, tools, process parameters, etc.; S2.2 Determine entity relationships: Determine the relationships between different entities, such as "parts - used tools", "equipment - processed parts", etc. S2.3 Knowledge fusion: The newly extracted entities are integrated with the existing knowledge graph to construct a knowledge graph, thereby removing duplicate and conflicting information; S2.4 Knowledge storage: storing the knowledge graph to facilitate quick query and reasoning.
[0022] In a specific embodiment, in S2.4, a graph database is used to store the knowledge graph.
[0023] In a specific embodiment, in S5, a depth-first search algorithm is used to search for possible processing paths in the knowledge graph.
[0024] In a specific embodiment, in S7, the optimized process solution is analyzed in terms of quality, efficiency, cost, etc.
[0025] In a specific embodiment, the search strategy is preferably a path query.
[0026] In a specific embodiment, the target conditions include processing quality requirements, efficiency requirements, cost requirements, etc.
[0027] The CNC machining optimization algorithm can be understood according to the following working principles: (a) By constructing a knowledge graph to integrate various knowledge and experience in the field of CNC machining technology, a depth-first search algorithm is used to quickly find a process solution suitable for processing specific parts in the knowledge graph.
[0028] (II) The reasoning process of the tool and processing parameters for processing a certain part is as follows: Assume that our knowledge graph contains part nodes, tool nodes, processing parameter nodes, etc., as well as edges representing various relationships between them, such as "part-use tool", "tool-applicable processing parameters", etc. The starting node is the specific part node to be processed. We first mark the part node as visited and push it into the stack. Pop the top node of the stack (which is the part node at this time), and then traverse its adjacent nodes, which are the tool nodes related to the part. For each unvisited tool node: mark it as visited and push it into the stack.
[0029] (III) Assuming that the top node of the stack is a tool node, after popping it, traverse its adjacent nodes, that is, the processing parameter nodes applicable to the tool. Similarly, mark and push the unvisited processing parameter nodes into the stack.
[0030] (IV) In this process, we continuously check whether the popped-up nodes meet our requirements for the tool and processing parameters suitable for processing the part (such as whether the material of the tool, the value range of the processing parameters, etc. meet expectations). If they do, we have found a suitable path; if the stack is empty and no node that meets the conditions is found, it means that no suitable path has been found under the current knowledge graph structure and search conditions.
[0031] (V) In depth-first search, the following formula can be used to represent the search process: Let V be the set of nodes in the graph, E be the set of edges, s be the starting node, and t be the target node. Use visited[v] to represent whether node v has been visited. Initially, visited[s]=true, and the visited value of other nodes is false. Use stack to represent the stack. Initially, stack={s}.
[0032] When stack ≠ Ø, do the following: currentNode = stack.pop() If currentNode = t, the target node is found and the search ends For each unvisited neighbor of currentNode visited[neighbor]=true stack.push(neighbor).
[0033] In summary, through the above unique technical solutions, the present disclosure has the following advantages: (A). Improves the scientificity and systematicity of process optimization, uses knowledge graphs to integrate knowledge and experience in the field of CNC machining, avoids the limitations of traditional process optimization methods that rely on experience, and makes the process optimization process more scientific and systematic. Through the depth-first search algorithm, the process scheme suitable for the processing of specific parts can be quickly found in the knowledge graph, which improves the efficiency and accuracy of process optimization. (B). Effectively integrates multi-source CNC machining data, can collect CNC machining process data from multiple channels, and clean, pre-process and merge them, and fully explore the potential knowledge in multi-source data. This helps to improve the quality and reliability of the process scheme, and also provides more comprehensive data support for the decision-making of enterprises. (C). Quickly and accurately determine the process parameters and processing paths. Through the depth-first search algorithm, it can quickly search for possible processing paths in the knowledge graph, and screen them according to the preset target conditions, so as to quickly and accurately determine the process parameters and processing paths suitable for the processing of specific parts. This greatly shortens the time of process optimization and improves production efficiency. (D). Improves flexibility and adaptability, can adjust the search strategy and target conditions according to actual production needs, and has strong flexibility and adaptability. Whether it is the processing of new parts or the improvement of existing processes, it can provide effective optimization solutions.
[0034] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. A CNC machining optimization algorithm based on knowledge graph and depth-first search, characterized in that: The following steps are involved: S1 data collection: collecting CNC machining process data, which includes machining equipment parameters, machining quality inspection data, process route, and tool information; S2 data collation: S2.1: Cleaning and preprocessing the CNC machining process data, removing noise and erroneous data, and unifying the data format and unit; S2.2: Establishing corresponding data standards for the CNC machining process data, and associating the CNC machining process data with evaluation indicators, wherein the evaluation indicators include machining quality, efficiency, cost and process stability; S3 constructs a knowledge graph: extracts a number of entities from the CNC machining process data, and determines the relationship between different entities; merges the newly extracted entities with the existing knowledge graph, removes duplicate and conflicting information, and thus completes the construction of the knowledge graph; S4 prepares optimization conditions: presets the search strategy and target conditions required for the search, wherein the search strategy includes specifying an entity corresponding to the part as a starting point; S5 depth-first search for process optimization: S5.1 Search: In the knowledge graph, possible processing paths are searched based on the search strategy; during the search, the processing paths are judged and screened according to the target conditions; S5.2 Optimization: If a processing path that meets the target condition is found, the process plan corresponding to the processing path is recorded, including the processing method, tool selection, and cutting parameters, and S5.3 is continued; otherwise, the search strategy is adjusted or the target condition is relaxed, so the target condition after relaxation is reset, and S5.1 is returned; S5.3 Summary: If the knowledge graph has been completely searched and all the recorded process plans are summarized, then continue to S5; otherwise, it means that there may be other processing paths that meet the current target conditions, which have not been searched yet, so return to S5.3; S6: Output all the process plans recorded in summary, that is, the optimized process plan, including specific processing methods, processing sequence, tool selection, and cutting parameter settings; S7: Analyze the optimized process plan to evaluate whether it meets the actual production requirements. If so, end the operation; otherwise, further adjust the search strategy or relax the target conditions, and return to S5.
2. The NC machining optimization algorithm according to claim 1, characterized in that: In S1, the CNC machining process data is collected from the machining equipment control system, quality inspection department, process documents, tool suppliers and operators.
3. The numerical control machining optimization algorithm according to claim 1, characterized in that: The S2 comprises the following steps: S2.1 Entity recognition and extraction: extracting a number of entities from the CNC machining process data, including parts, equipment, tools, and process parameters; S2.2 Determine entity relationships: Determine the relationships between different entities, such as "parts - tool used", "equipment - processing parts"; S2.3 Knowledge fusion: The newly extracted entities are integrated with the existing knowledge graph to construct a knowledge graph, thereby removing duplicate and conflicting information; S2.4 Knowledge storage: storing the knowledge graph to facilitate quick query and reasoning.
4. The numerical control machining optimization algorithm according to claim 3, characterized in that: In S2.4, a graph database is used to store the knowledge graph.
5. The numerical control machining optimization algorithm according to claim 1, characterized in that: In S5, a depth-first search algorithm is used to search for possible processing paths in the knowledge graph.
6. The numerical control machining optimization algorithm according to claim 1, characterized in that: In S7, the optimized process solution is analyzed in terms of quality, efficiency and cost.
7. The numerical control machining optimization algorithm according to claim 1, characterized in that: The search strategy is path query.
8. The numerical control machining optimization algorithm according to claim 1, characterized in that: The target conditions include processing quality requirements, efficiency requirements, and cost requirements.