CN-115712968-B - Engine assembly process optimizing system

CN115712968BCN 115712968 BCN115712968 BCN 115712968BCN-115712968-B

Abstract

The invention relates to the technical field of digital assembly of engines, and discloses an engine assembly process optimization system which comprises an assembly characteristic extraction module, an assembly planning module, a part transfer abrasion model building module, an assembly analysis module, an assembly workshop information acquisition module and an assembly correction module. According to the invention, the feature extraction module is used for extracting assembly information from the three-dimensional model of the parts, meanwhile, tolerance analysis information among the parts is obtained, and the assembly planning module is used for establishing an assembly planning model based on the extracted assembly planning information, so that an assembly sequence and an assembly path are obtained, the optimal assembly of the parts with different sizes is achieved, the problem of out-of-tolerance assembly between the parts repaired by the remanufacturing technology and the original parts of the engine is avoided, and the assembly accuracy is improved.

Inventors

TANG XIUJIAN
WEI SHICHENG
WANG TAO
LIU QIAN
WANG LONG
DU SHOUXIN

Assignees

中国人民解放军陆军装甲兵学院

Dates

Publication Date: 20260512
Application Date: 20221213

Claims (5)

1. An engine assembly process optimization system, comprising: The assembly characteristic extraction module is used for extracting assembly information of the engine parts based on the engine three-dimensional model, wherein the assembly information comprises assembly planning information and tolerance analysis information; The assembly planning module is used for establishing an assembly planning model based on the assembly planning information extracted by the assembly characteristic extraction module, wherein the assembly planning comprises an assembly sequence planning and an assembly path planning; The component transfer wear model building module is used for building a wear model aiming at assembly position points of each component from a warehouse to an assembly workshop and obtaining probability wear data in the component transfer process; the assembly analysis module is used for generating a size chain based on the tolerance analysis information extracted by the assembly feature extraction module, combining a plurality of closed loops according to the size chain and the actual size and probability abrasion data detected during the warehouse entry of the parts, screening the closed loops meeting the requirements, and marking the size range of the parts in the closed loops and the closed loop part set; The assembly workshop information acquisition module is used for detecting the parts to be assembled before assembly, acquiring actual wear data, and inputting the wear data detected before assembly of the parts into the assembly correction module; the assembly correction module is used for comparing the size range of each part in the assembly analysis module with the actual size of the worn part, judging whether the actual size of the worn part is in the size range of the part, if so, assembling the part, otherwise, inputting the actual size of the worn part into the assembly analysis module; the assembly analysis module firstly automatically generates a dimension chain, marks tolerance parameters of each component ring and the closed ring in the dimension chain, and then performs assembly analysis; The assembly analysis process is as follows: step one, acquiring tolerance parameters of each component ring and a closed ring of the generated size chain, and acquiring all warehousing actual sizes of parts of each component ring; step two, selecting a reference part of each component ring, assembling each component ring and a closed ring according to three dimensional data of the reference part and the sum of the warehouse-in actual size and the maximum abrasion loss and the sum of the warehouse-in actual size and the minimum abrasion loss of other parts, measuring errors of each component ring and the closed ring according to an assembly sequence, and simultaneously marking the dimensional range of the parts of the later assembly sequence of each size part and part numbers meeting the dimensional range to form a closed ring part set; Step three, comparing whether the errors of the combined closed loop and each component loop are in the corresponding tolerance range, if not, removing the combination of the closed loop and each component loop, and deleting the closed loop part set; Step four, further screening the closed loop according to the maximum utilization rate of the existing parts, selecting the closed loop with the minimum residual quantity of the existing parts, and reserving the closed loop part set; Step five, if the actual size of a certain existing part does not meet the requirement of forming a complete closed loop, feeding back the supplementary information of the part; step six, selecting parts according to the part numbers in the reserved closed loop part set, transferring the parts to assembly position points of workshops for assembly, carrying out abrasion detection on each part before assembly, and storing abrasion data in a database after associating the abrasion data with product basic information of the part to be used as part transfer abrasion model training historical data; after the assembly analysis module receives the part information transmitted by the assembly correction module, the following steps are carried out: Step 1, positioning a component ring to which the part belongs according to the number of the part; Step 2, obtaining all sequences behind the component of the component ring and all existing rest parts of the assembly sequences of all component rings behind the component ring; step 3, according to the actual size of the worn parts, the warehouse-in actual size of the rest parts, the sum of the warehouse-in actual size and the maximum wear amount and the sum of the warehouse-in actual size and the minimum wear amount, assembling and reassembling each component ring and the sealing ring according to an assembling sequence; Step 4, performing the assembly analysis process step three to step five; And 5, evaluating the closed loop formed by the original assembly analysis and the newly formed closed loop, if the evaluation result is that the original closed loop evaluation is superior to the newly formed closed loop, rejecting the part, re-selecting the part meeting the size range of the part of the type, and if the part of the type is not available, feeding back the part supplementary information.
2. An engine assembly process optimization system as set forth in claim 1 wherein: the construction of the part transfer wear model is based on the specific operation condition of each assembly factory on the part transfer, and the input quantity of the part transfer wear model comprises transfer road condition information, transfer tool information, loading and unloading tool information and transfer personnel information; The output of the part transfer wear model comprises a wear result, a wear type and a wear amount, when the part is worn, the wear result value is 1, otherwise, the wear result value is 0, the wear type comprises size wear and shape wear, and the wear amount comprises a maximum wear amount and a minimum wear amount.
3. An engine assembly process optimization system as set forth in claim 2 wherein: The part transfer abrasion model training data are abrasion data of historical parts in the transfer process, the abrasion data are comparison values of detection data of the parts in storage and abrasion detection data of the parts before being assembled at an assembly position point, the maximum abrasion amount and the minimum abrasion amount are obtained according to the historical comparison values, and the part transfer abrasion model is optimized at fixed time according to continuously updated historical part abrasion data.
4. The system for optimizing an engine assembly process of claim 1, wherein said tolerance analysis information comprises product base information, dimensional tolerance information, and assembly model fit relationship information; The basic information of the product comprises a number, a part name, a size type, a size value, a maximum abrasion loss, a tolerance type, an upper deviation and a lower deviation; the dimensional tolerance information comprises an identification of an assembly part, a part name, a tolerance value, a reference coordinate system X coordinate of a model, a reference coordinate system y coordinate of the model and a reference coordinate system z coordinate of the model; The assembly model matching relation information comprises a number, an assembly part name, a driving part identifier, a driven part identifier, a driving part matching type and a driven part matching type.
5. The engine assembling process optimizing system according to claim 1, wherein the characteristic information extracted by the assembling characteristic extracting module includes engineering attribute information of the parts, geometric information of the parts, pose information, assembling constraint information, assembling level information, and the assembling characteristic information extracting step is as follows: S1, loading an assembly model of a product; s2, accessing all the features in the assembly model and storing the features into a feature array; S3, obtaining the types of the features item by item from the feature array; s4, judging whether the feature type is an element, if the feature type is the element, entering S5, and if the feature type is not the element, jumping to S3; S5, acquiring an element belonging to the feature type; S6, acquiring information such as component attribute information, coordination constraint information, pose information and the like of the element; s7, storing the related information into data; s8, judging whether the element is a sub-assembly or not, if so, entering S9, and if so, jumping to S2 to carry out recursion operation; And S9, judging whether the feature array is acquired, if so, stopping the algorithm, and if not, jumping to S3 to continue the cyclic operation.

Description

Engine assembly process optimizing system Technical Field The invention relates to the technical field of digital assembly of engines, in particular to an engine assembly process optimization system. Background The digital assembly technology is a technology for constructing a digital reality environment and an assembly digital model by using a digital reality technology, a computer graphics technology, an artificial intelligence technology, a simulation technology and the like, and in the assembly process of equipment, the assembly sequence, the assembly path, the assembly precision, the assembly performance and the like of products are subjected to interactive analysis, planning, simulation and optimization, so that the number of times of real object trial assembly of assembly is effectively reduced, and the assembly quality, the efficiency and the reliability of the equipment are improved. The technology can overcome the defects in the traditional assembly process, and has important significance in shortening the assembly period, reducing the assembly cost and improving the assembly quality. After a certain period of use, the size of the parts is out of tolerance, such as the cylinder diameter of the engine is enlarged, and the piston matched with the parts is smaller. The method has the advantages that after the remanufacturing technology is adopted, a large cylinder can be matched with a large-diameter piston, and a small cylinder is matched with a small-diameter piston, but if the large cylinder is adopted, the gap is out of tolerance, so that in the assembly process of the waste engine parts after remanufacturing and maintenance, the sizes of the related maintained parts are required to be selected according to the sizes of the current parts of the engine, and the assembly is reasonably carried out, so that the assembly accuracy is improved, particularly the assembly of the piston and the cylinder is carried out, and the assembly accuracy of the piston and the cylinder is used as a core part of the engine, and directly influences the performance of the engine. In addition, the actual size of the parts is generally measured only when the parts are put in storage, and normal collision and abrasion during the process of transferring the parts from the warehouse to the assembly workshop are ignored, so that the selection range of the parts is too large as a result of combining the parts by using the actual size when put in storage, and if the actual size of each part is detected again before assembly and the parts are combined again by using the detected actual size, a great amount of time is wasted, and the assembly efficiency is reduced. Disclosure of Invention Aiming at the defects of the prior engine assembly system in the use process, the invention provides an engine assembly process optimization system, which has the advantages of taking the transfer abrasion of parts into consideration in advance, reducing the selection range of the parts and improving the selection accuracy of the parts, and solves the problems in the prior art. The invention provides a technical scheme that an engine assembly process optimizing system comprises: The assembly characteristic extraction module is used for extracting assembly information of the engine parts based on the engine three-dimensional model, wherein the assembly information comprises assembly planning information and tolerance analysis information; The assembly planning module is used for establishing an assembly planning model based on the assembly planning information extracted by the assembly characteristic extraction module, wherein the assembly planning comprises assembly sequence planning and assembly path planning, and an algorithm adopted by the assembly planning model is an ant colony algorithm; the component transfer abrasion model building module is used for building an abrasion model aiming at assembly position points of each component from a warehouse to an assembly workshop based on a BP neural network algorithm, and obtaining probability abrasion data in the component transfer process; the assembly analysis module is used for generating a size chain based on the tolerance analysis information extracted by the assembly feature extraction module, combining a plurality of closed loops according to the size chain and the actual size and probability abrasion data detected during the warehouse entry of the parts, screening the closed loops meeting the requirements, and marking the size range of the parts in the closed loops and the closed loop part set; The assembly workshop information acquisition module is used for detecting the parts to be assembled before assembly, acquiring actual wear data, and inputting the wear data detected before assembly of the parts into the assembly correction module; And the assembly correction module is used for comparing the size range of each part in the assembly analysis module with the actual size of the worn pa