Roughness optimization method for additive manufacturing flow channel of aerospace propulsion chamber

Abstract

The invention discloses an aerospace propulsion chamber additive manufacturing flow channel roughness optimization method, which comprises the following steps of: setting a manufacturing parameter disturbance strategy, executing a plurality of groups of additive manufacturing experiments, obtaining a roughness forming sequence of the inner wall of a channel, generating a roughness change trend by utilizing a grey prediction model # imgabs0 #, and calculating the roughness of the inner wall of the channel. Combining roughness distribution data of multiple positions of the wall surface, guiding the Dirichlet process to establish a spatial distribution model of the channel inner wall roughness, fusing trend parameters and spatial characteristics, collecting real-time roughness data in the manufacturing process, and constructing an error feedback mechanism to realize linkage correction of a roughness prediction result; and finally, an instruction for controlling the manufacturing path is generated and used for guiding the machining process, so that fine control and optimization of the roughness of the inner wall of the flow channel of the aerospace propulsion chamber are achieved. According to the method, linkage optimization of roughness evolution law modeling, spatial distribution identification and manufacturing path cooperative adjustment is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of additive manufacturing process control, and in particular to a method for optimizing the roughness of a flow channel in additive manufacturing of an aerospace propulsion chamber. Background Art

[0002] With the widespread application of additive manufacturing technology in the aerospace field, the manufacturing accuracy and functional stability of the complex flow channel structure inside the aerospace propulsion chamber have become key issues of engineering concern. Especially in the propulsion channel structure working under high heat flux and high pressure coupling conditions, the surface roughness of the inner wall directly affects the fuel flow behavior and heat transfer performance, which in turn is related to the propulsion efficiency and thermal safety redundancy of the entire machine.

[0003] In existing technologies, the inner wall surfaces of flow channels formed by additive manufacturing processes such as metal selective laser melting often have uncontrollable microscopic roughness structures. Current process control methods mainly rely on empirical parameter settings and static error modeling, making it difficult to achieve dynamic understanding and feedforward prediction of the roughness formation mechanism. Specifically, existing technologies have obvious shortcomings in the following aspects: 1. Lack of trend prediction capability: Current methods rely on a large amount of experimental data and fixed models to predict roughness changes, and lack a dynamic modeling mechanism suitable for small sample and highly coupled manufacturing environments.

[0004] 2. Model response rigidity: Traditional statistical regression or fitting methods have limited ability to characterize the spatial distribution of roughness and are unable to capture local abnormal behaviors during the multi-peak heterogeneous evolution of roughness.

[0005] 3. Lagging feedback mechanism: The existing control system fails to establish an efficient prediction error feedback path, resulting in the manufacturing process being unable to dynamically correct and optimize the control path based on real-time roughness deviations.

[0006] 4. Insufficient parameter coupling processing: The formation of roughness is jointly affected by parameter disturbances such as laser power and scanning speed. The existing technology fails to establish a linkage mechanism between high-dimensional disturbances and roughness behavior.

[0007] Therefore, how to provide a method for optimizing the roughness of flow channels in additive manufacturing of aerospace propulsion chambers is an urgent problem that needs to be solved by those skilled in the art. Summary of the Invention

[0008] One purpose of the present invention is to propose a roughness optimization method for the flow channel of the additive manufacturing of aerospace propulsion chamber. The present invention collects roughness data by manufacturing parameter perturbation experiment and constructs a gray prediction model based on the roughness of the flow channel. The roughness evolution trend model is combined with the Dirichlet process to generate the roughness spatial distribution model. A real-time roughness monitoring and residual feedback mechanism is introduced for linkage optimization control, and finally, manufacturing path control instructions are generated, which have the advantages of high prediction accuracy, strong spatial adaptability, timely manufacturing feedback, and fine roughness regulation.

[0009] A method for optimizing the roughness of the additive manufacturing flow channel of an aerospace propulsion chamber according to an embodiment of the present invention includes the following steps: Set the manufacturing parameter perturbation strategy, perform additive forming experiments under multiple perturbation groups, and collect the roughness formation sequences of the corresponding channel inner walls for constructing the roughness input-response relationship; Construct a grey prediction model based on the collected roughness formation sequences , generate a roughness time evolution model of the roughness changing with the perturbation conditions in the form of a differential equation. The roughness time evolution model is used to characterize the trend change behavior of the additive manufacturing parameter perturbation on the inner wall roughness of the aerospace propulsion chamber flow channel, and output the roughness trend parameters and the roughness prediction value sequence; Collect the roughness distribution data at different positions on the channel wall surface, and jointly use the roughness trend parameters as prior inputs to guide the Dirichlet process to establish a spatial probability distribution model of the inner wall roughness of the aerospace propulsion chamber flow channel; Establish a fusion mechanism between the roughness trend parameters and the spatial probability distribution model, and output the roughness trend parameters and the roughness spatial probability distribution characteristics; Collect real-time roughness monitoring data during the additive manufacturing process, compare the real-time roughness monitoring data with the roughness prediction value sequence, and calculate the prediction residual. The prediction residual is used to measure the error between the roughness prediction and the actual formation; Construct a residual feedback mechanism based on the prediction residual, and introduce the residual feedback mechanism into the fusion mechanism to drive the linkage optimization process; Generate manufacturing path control instructions based on the roughness trend parameters and the roughness spatial probability distribution characteristics output by the fusion mechanism. The manufacturing path control instructions are used to dynamically adjust the scanning path density or energy input strategy in the aerospace propulsion chamber flow channel area during the additive manufacturing process to achieve the target optimization control of the inner wall roughness of the flow channel in terms of spatial position.

[0010] Optionally, the steps of setting the manufacturing parameter perturbation strategy, performing the additive forming experiment, and collecting the roughness formation sequence of the channel inner wall include: Select the laser power, scanning speed, scanning spacing, and forming layer thickness as the manufacturing parameter perturbation dimensions, and construct a manufacturing parameter perturbation combination matrix, which is the composition content of the manufacturing parameter perturbation strategy; According to the manufacturing parameter perturbation combination matrix, additive manufacturing experiments are sequentially performed on a selective laser melting additive manufacturing equipment for metals. The additive manufacturing experiments generate specimens containing the flow channel structure of the aerospace propulsion chamber for different parameter combinations. After each group of additive manufacturing experiments is completed, a surface topography measurement device is used to measure the roughness at multiple positions on the inner wall of the flow channel of the aerospace propulsion chamber of the specimen. The roughness measurement area covers multiple characteristic positions in the intake section, the middle section, and the outlet section. The manufacturing parameter perturbation combinations used in each group of additive manufacturing experiments are paired with the corresponding roughness measurement results of the inner wall of the flow channel of the aerospace propulsion chamber to construct a roughness formation sequence. The roughness formation sequence serves as the input data basis for the modeling process of the grey prediction model GM(1,1).

[0011] Optionally, constructing a grey prediction model based on the collected roughness formation sequence , generating a roughness time evolution model of the roughness varying with the perturbation conditions in the form of a differential equation, and outputting the roughness trend parameters and the roughness prediction value sequence specifically including: Construct the roughness formation sequence into the original time series data of the roughness of the inner wall of the flow channel of the aerospace propulsion chamber, and based on the preset time window division rule, perform a segmented adjustable cumulative generation operation to form multiple roughness grey generation subsequences. Based on each roughness grey generation subsequence, establish the corresponding grey prediction sub-model , and generate a set of roughness trend characteristic parameters for trend analysis in each grey prediction sub-model . In each grey prediction sub-model , construct a non-uniform weight background value sequence based on the roughness grey generation subsequence: ; where is the th non-uniform weight background value sequence, is the roughness grey generation subsequence, is the weight coefficient, and the value range is , is the sequence index; Based on the non-uniform weight background value sequence, construct the differential prediction equation of each grey prediction sub-model : ; where is the roughness grey generation subsequence, and are the parameters to be estimated of the grey prediction sub-model GM(1,1), and parameter identification is performed by the least squares method. Summarize the roughness prediction value sequences output by each grey prediction sub-model as the roughness prediction input data for the residual feedback mechanism; Summarize the roughness prediction value sequences output by each grey prediction sub-model as the roughness prediction input data for the residual feedback mechanism; Jointly integrate the roughness trend characteristic parameters extracted by each grey prediction sub-model to form roughness trend parameters, which are used as prior input variables for the trend in the Dirichlet process roughness modeling module. Jointly integrate the roughness trend characteristic parameters extracted by each grey prediction sub-model to form roughness trend parameters, which are used as prior input variables for the trend in the Dirichlet process roughness modeling module.

[0012] Optionally, the steps of collecting the roughness distribution data at different positions on the inner wall of the flow channel of the space propulsion chamber and using the roughness trend parameters as priors to guide the Dirichlet process to establish a spatial probability distribution model of the roughness on the inner wall of the flow channel of the space propulsion chamber include: Divide the inner wall of the flow channel of the space propulsion chamber into multiple spatial measurement regions. Respectively collect three types of roughness indexes, namely arithmetic mean roughness, maximum profile height, and root mean square roughness, in each spatial measurement region, and record the three-dimensional spatial position coordinates corresponding to each group of roughness indexes to construct the roughness distribution data on the inner wall of the flow channel of the space propulsion chamber; Use the roughness distribution data on the inner wall of the flow channel of the space propulsion chamber as the input for non-parametric Bayesian modeling. The Dirichlet process uses a multi-dimensional Gaussian distribution as the base distribution, and sets the concentration parameter to control the non-parametric prior strength of the number of clusters; In the Dirichlet process modeling, the initial state is an empty cluster set. During the sampling process, each new data point uses the current three-dimensional spatial position coordinates and roughness indexes as inputs, and calculates the likelihood values of the distribution parameters of each existing cluster. The distribution parameters of the cluster include the mean vector and covariance matrix of the roughness indexes; According to the distribution parameters, likelihood values, and corresponding cluster sizes of each cluster, jointly calculate the membership probability with the concentration parameter, perform cluster membership sampling, and determine whether a new roughness sub-distribution cluster is generated; In each sampling round, introduce the roughness trend parameters output by the grey prediction model GM(1,1) as prior adjustment variables to update the cluster distribution weights, so that the roughness evolution trend and the spatial distribution structure are adjusted synchronously; After the sampling converges, merge all the roughness sub-distribution clusters and output the spatial probability distribution model of the roughness on the inner wall of the flow channel of the space propulsion chamber, which is used to describe the multi-peak distribution state and trend guiding behavior of the roughness in different spatial regions.

[0013] Optionally, the specific implementation of the fusion mechanism between the roughness trend parameter and the spatial probability distribution model includes establishing a mapping relationship between the roughness trend parameter and each roughness sub-distribution cluster in the spatial probability distribution model one by one. In each mapping pair, calculate the residual between the roughness trend parameter and the mean value of the corresponding roughness sub-distribution cluster, and construct a fusion weight adjustment function based on the residual to dynamically correct the probability weights of each roughness sub-distribution cluster in the spatial probability distribution model of the inner wall roughness of the aerospace propulsion chamber flow channel. At the same time, use the residual as a feedback variable to iteratively update the roughness trend parameter, so as to realize the two-way dynamic coupling fusion between the grey prediction model GM(1,1) and the spatial probability distribution model of the inner wall roughness of the aerospace propulsion chamber flow channel.

[0014] Optionally, the specific implementation of collecting real-time roughness monitoring data during the additive manufacturing process, comparing the real-time roughness monitoring data with the roughness prediction value sequence, and calculating the prediction residual includes: During the additive manufacturing process of the aerospace propulsion chamber flow channel, based on the preset scanning path and time interval, use a confocal laser measurement device to perform on-line measurement on the inner wall of the aerospace propulsion chamber flow channel, collect three types of roughness indexes including arithmetic mean roughness, maximum profile height and root mean square roughness, and attach the sampling time index and three-dimensional spatial position coordinates to form the real-time roughness monitoring data of the inner wall of the aerospace propulsion chamber flow channel. Reorganize the real-time roughness monitoring data of the inner wall of the aerospace propulsion chamber flow channel into a time series structure according to the time index, and correspond it to the roughness prediction value sequence of the inner wall of the aerospace propulsion chamber flow channel in chronological order to complete the matching of the three types of predicted roughness indexes and the three types of measured roughness indexes at each sampling moment. At each sampling time point, calculate the numerical difference between the predicted values and the measured values of the three types of roughness indexes, namely arithmetic mean roughness, maximum profile height and root mean square roughness, as the prediction residual at the corresponding time point. Construct the prediction residual data set of the inner wall roughness of the aerospace propulsion chamber flow channel from the prediction residuals calculated at all sampling time points. The roughness prediction residual data set is used as the input in the fusion mechanism to drive the dynamic correction and update of the parameters of the grey prediction model GM(1,1) and the spatial probability distribution model of the inner wall roughness of the aerospace propulsion chamber flow channel.

[0015] Optionally, the specific implementation of constructing a residual feedback mechanism based on the prediction residual and introducing the residual feedback mechanism into the fusion mechanism to drive the linkage optimization process includes: On the basis of obtaining the real-time roughness monitoring data of the inner wall of the aerospace propulsion chamber flow channel and the sequence of predicted values of the roughness of the inner wall of the aerospace propulsion chamber flow channel, pair the predicted roughness values and the real-time roughness monitoring values point by point according to the time index, calculate the arithmetic mean roughness residual, the maximum profile height residual and the root mean square roughness residual at each time point, and generate a residual time series; Segment the residual time series with a sliding window, statistically calculate the mean value, standard deviation and trend slope of the residual sequence within each window, and construct a residual feedback vector; Use the residual feedback vector as the input of the residual feedback mechanism, and respectively act on the roughness trend parameter generation process and the iterative process of the spatial probability distribution model of the roughness of the inner wall of the aerospace propulsion chamber flow channel to form a dual-channel adjustment mechanism; In the roughness trend parameter generation process, the residual feedback mechanism is used to adjust the weight distribution structure of the non-uniform weight background value construction method in the grey prediction model GM(1,1) and the iterative update step size of the model parameters; In the spatial probability distribution model of the roughness of the inner wall of the aerospace propulsion chamber flow channel, the residual feedback mechanism is used to adjust the sampling of each roughness sub-distribution cluster and the threshold for generating new clusters, and dynamically correct the cluster distribution weight of the spatial distribution; Through the dual-path injection method of the residual feedback mechanism, construct a linkage optimization process, so that the roughness trend parameters and the spatial probability distribution characteristics of the roughness of the inner wall of the aerospace propulsion chamber flow channel are updated synergistically according to the roughness prediction residuals;

[0016] Optionally, the steps of generating a manufacturing path control instruction based on the roughness trend parameters and spatial probability distribution characteristics output by the fusion mechanism include: Expand the roughness trend parameters by time index, extract the roughness change rate, fluctuation amplitude and local evolution trend corresponding to different manufacturing time periods, and form the time dynamic regulation factor in the manufacturing path control instruction; Decompose the spatial probability distribution characteristics of the roughness of the inner wall of the aerospace propulsion chamber flow channel by region, extract the spatial position center, covariance matrix and roughness distribution weight of each roughness sub-distribution cluster, and form the spatial distribution regulation factor in the manufacturing path control instruction; Respectively based on the time dynamic regulation factor and the spatial distribution regulation factor, construct a manufacturing path control instruction parameter set, and the manufacturing path control instruction includes: Scanning path structure parameters, including the three-dimensional coordinate sequence of the scanning segment, the scanning path direction vector and the scanning segment jump strategy; Laser power control parameters, including the roughness trend regulation coefficient and the laser power base value and adjustment amplitude under the spatial distribution mapping; Scanning speed control parameters, including a scanning speed function constructed based on the change rate of roughness trend, defining the starting scanning speed, maximum speed, and the form of speed gradient change; Scanning spacing control parameters, including the scanning overlap rate, interlayer offset, and path spacing adjustment amount corresponding to the change in the spatial density of roughness; Jointly allocate each parameter in the set of manufacturing path control instruction parameters according to the cross-region of the roughness trend parameters and the spatial probability distribution characteristics of the inner wall roughness of the flow channel of the space propulsion chamber, construct a fusion control mapping model, and generate a set of manufacturing path control instructions for different manufacturing regions and time periods; Convert the set of manufacturing path control instructions into a manufacturing control file format recognizable by the additive manufacturing equipment for the flow channel of the space propulsion chamber, and transmit it to the control system to complete manufacturing execution.

[0017] The beneficial effects of the present invention are as follows: (1) By constructing a combination of manufacturing parameter perturbations and a roughness formation sequence, the present invention uses the grey prediction model GM(1,1) to establish a roughness time evolution trend model, and introduces a piecewise adjustable cumulative sum and non-uniform weighted background value construction mechanism, realizing an interpretable modeling of the roughness evolution law and significantly improving the prediction ability of the roughness change trend.

[0018] (2) By collecting multi-region and multi-dimensional roughness index data of the inner wall of the flow channel of the space propulsion chamber, guiding the Dirichlet process to establish a spatial probability distribution model, and introducing roughness trend parameters for trend prior regulation, the present invention realizes multi-modal modeling and trend guidance of the roughness spatial distribution, significantly enhancing the adaptability of the model to complex spatial roughness states.

[0019] (3) By designing a fusion mechanism, dynamically combining the roughness time trend parameters and the spatial probability distribution model, on this basis, introducing real-time roughness monitoring data and a sequence of roughness prediction values to calculate the residual difference, constructing a residual feedback mechanism and introducing it into the trend prediction and spatial modeling processes, the present invention realizes a two-way feedback regulation mechanism for the manufacturing process, improving the real-time response ability of the system and the model self-adaptive ability.

[0020] (4) Finally, the present invention takes the fused roughness evolution trend and spatial distribution characteristics as inputs, combines with the manufacturing requirements of the space propulsion chamber structure to generate manufacturing path control instructions. The instructions include spatial position indexes, manufacturing parameter fine-tuning items, and process adjustment weights, realizing refined regulation of the additive manufacturing process and effectively improving the roughness control accuracy and consistency of the final product. Description of the Drawings

[0021] The accompanying drawings are used to provide a further understanding of the present invention and form a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation to the present invention. In the drawings: Figure 1 It is a flowchart of a method for optimizing the roughness of the flow channel in the additive manufacturing of an aerospace propulsion chamber proposed by the present invention. Detailed implementation manners

[0022] Now, the present invention will be further described in detail with reference to the accompanying drawings. These drawings are all simplified schematic diagrams, only illustrating the basic structure of the present invention in a schematic manner, so they only show the components related to the present invention.

[0023] Reference Figure 1 , a method for optimizing the roughness of the flow channel in the additive manufacturing of an aerospace propulsion chamber, includes the following steps: Set the manufacturing parameter perturbation strategy, perform additive manufacturing experiments under multiple perturbation groups, and collect the roughness formation sequences of the corresponding channel inner walls for constructing the roughness input-response relationship; Construct a grey prediction model based on the collected roughness formation sequences , generate a roughness time evolution model of the change of roughness with perturbation conditions in the form of a differential equation. The roughness time evolution model is used to characterize the trend change behavior of the additive manufacturing parameter perturbation on the inner wall roughness of the aerospace propulsion chamber flow channel, and output the roughness trend parameters and the roughness prediction value sequences; Collect the roughness distribution data at different positions on the channel wall surface, and jointly use the roughness trend parameters as prior inputs to guide the Dirichlet process to establish a spatial probability distribution model of the inner wall roughness of the aerospace propulsion chamber flow channel; Establish a fusion mechanism between the roughness trend parameters and the spatial probability distribution model, and output the roughness trend parameters and the roughness spatial probability distribution characteristics; Collect real-time roughness monitoring data during the additive manufacturing process, compare the real-time roughness monitoring data with the roughness prediction value sequences, and calculate the prediction residuals. The prediction residuals are used to measure the error between the roughness prediction and the actual formation; Construct a residual feedback mechanism based on the prediction residuals, and introduce the residual feedback mechanism into the fusion mechanism to drive the linkage optimization process; Generate a manufacturing path control instruction based on the roughness trend parameters and the roughness spatial probability distribution characteristics output by the fusion mechanism. The manufacturing path control instruction is used to dynamically adjust the scanning path density or energy input strategy in the aerospace propulsion chamber flow channel area during the additive manufacturing process to achieve the target optimization control of the inner wall roughness of the flow channel in terms of spatial position.

[0024] In this embodiment, the steps of setting the manufacturing parameter perturbation strategy, performing the additive manufacturing experiment, and collecting the roughness formation sequence of the channel inner wall include: Select the laser power, scanning speed, scanning spacing, and forming layer thickness as the manufacturing parameter perturbation dimensions, and construct a manufacturing parameter perturbation combination matrix, which is the composition content of the manufacturing parameter perturbation strategy; According to the manufacturing parameter perturbation combination matrix, perform additive manufacturing experiments in sequence on the selective laser melting additive manufacturing equipment for metals. The additive manufacturing experiments generate samples containing the flow channel structure of the aerospace propulsion chamber for different parameter combinations; After each group of additive manufacturing experiments is completed, use a surface topography measurement device to measure the roughness at multiple positions on the inner wall of the flow channel of the aerospace propulsion chamber of the sample. The roughness measurement area covers multiple characteristic positions such as the intake section, middle section, and outlet section; Pair the manufacturing parameter perturbation combinations used in each group of additive manufacturing experiments with the corresponding roughness measurement results of the inner wall of the flow channel of the aerospace propulsion chamber to construct a roughness formation sequence, which serves as the input data basis for the modeling process of the grey prediction model GM(1,1).

[0025] In this embodiment, by selecting the laser power, scanning speed, scanning spacing, and forming layer thickness as the perturbation dimensions to construct a manufacturing parameter perturbation combination matrix, and implementing the forming experiments group by group on the selective laser melting additive manufacturing equipment for metals, samples with the flow channel structure of the aerospace propulsion chamber are prepared. Using a surface topography measurement device, the roughness of the inner wall of the sample is measured at multiple characteristic positions such as the intake section, middle section, and outlet section, and multi-dimensional indexes such as the arithmetic mean roughness are extracted to construct the corresponding relationship between the manufacturing parameters and the roughness data. Finally, a roughness formation sequence is formed as the basic data for the grey prediction model modeling, effectively establishing the mapping structure between the perturbation parameters and the roughness response.

[0026] In this embodiment, constructing the grey prediction model based on the collected roughness formation sequence , generating a roughness time evolution model of the roughness changing with the perturbation conditions in the form of a differential equation, and outputting the roughness trend parameters and the roughness prediction value sequence specifically includes: Construct the roughness formation sequence into the original time series data of the roughness of the inner wall of the flow channel of the aerospace propulsion chamber, and based on the preset time window division rule, perform a segmented adjustable cumulative generation operation to form multiple roughness grey generation subsequences; Establish corresponding grey prediction sub-models based on each roughness grey generation subsequence , and generate a set of roughness trend characteristic parameters for trend analysis in each grey prediction sub-model ; In each grey prediction sub-model Based on the roughness grey generated subsequence, construct the non-uniform weighted background value sequence: ; Wherein, is the th non-uniform weighted background value sequence, is the roughness grey generated subsequence, is the weight coefficient, and its value range is , is the sequence index; This formula is used to construct the non-uniform weighted background value sequence in the grey prediction model. Its core idea is to perform weighted averaging on adjacent data points in the roughness grey generated subsequence to form a background value sequence that better conforms to the actual change trend. Different from the traditional uniform weighting process, this formula introduces a weight coefficient to perform non-equilibrium combination on the values at the current moment and the previous moment, enabling the background value to reflect the dynamic change characteristics of the time series in model training. This method enhances the response ability of the model to trend changes by increasing the sensitivity to internal changes in the sequence, reduces the dependence of the model on data stationarity, and has strong versatility and adaptability. In terms of creativity, this method adjusts the modeling sensitivity by introducing a non-uniform weighting mechanism, realizes the flexible expression of the roughness evolution law in the time series modeling process, overcomes the problem of the decline in prediction accuracy of traditional grey modeling under complex disturbance conditions, and has a significant technical breakthrough.

[0027] Construct each grey prediction sub-model based on the non-uniform weighted background value sequence ; Wherein, is the roughness grey generated subsequence, and are the parameters to be estimated of the grey prediction sub-model GM(1,1), and parameter identification is performed by the least squares method; This formula represents the differential prediction form of the grey prediction sub-model constructed based on the non-uniform weighted background value sequence. Its principle is to describe the dynamic process of the roughness generation subsequence changing over time by constructing a first-order linear differential equation. The two parameters to be estimated in the model represent the attenuation factor of system evolution and the external driving term. The parameters are accurately fitted through the least squares method, enabling the model to capture the change trend and perturbation response behavior of the data. This method not only continues the grey system's ability to model with a small number of samples but also introduces a differential prediction structure on this basis, realizing continuous modeling and fine-grained characterization of the roughness change process, and enhancing the prediction resolution and dynamic adaptability of the model under perturbation drive. In terms of creativity, by introducing non-uniform weighted background values into grey differential modeling, the transformation from static weighted estimation to dynamic evolution modeling is realized, enhancing the generalization ability and prediction stability of the model in complex manufacturing scenarios.

[0028] Sum up the roughness prediction value sequences output by each grey prediction sub-model as the roughness prediction input data for the residual feedback mechanism; Jointly integrate the roughness trend characteristic parameters extracted from each grey prediction sub-model to form roughness trend parameters, which are used as the trend prior input variables in the Dirichlet process roughness modeling module.

[0029] In this embodiment, the roughness formation sequence of the inner wall of the flow channel of the space propulsion chamber is constructed as time series data, and a segmented adjustable accumulation operation is performed under a preset time window to generate multiple grey generation subsequences of roughness. Grey prediction sub-models GM(1,1) are established respectively. The non-uniform weighted background value construction method is introduced into each model to establish a differential prediction equation, so as to realize refined modeling and dynamic tracking of the roughness evolution trend. The roughness prediction values output by each sub-model are used for the subsequent residual feedback mechanism, and the trend characteristic parameters are integrated to form roughness trend parameters and input into the Dirichlet process modeling process. This method improves the ability of the prediction model to describe the roughness evolution dynamics under changing perturbation conditions through multi-segment modeling, and enhances the trend recognition accuracy of the system for uncertain manufacturing perturbations.

[0030] In this embodiment, the steps of collecting the roughness distribution data at different positions on the inner wall of the flow channel of the space propulsion chamber and using the roughness trend parameters as prior inputs to guide the Dirichlet process to establish a spatial probability distribution model of the roughness on the inner wall of the flow channel of the space propulsion chamber include: Divide the inner wall of the flow channel of the space propulsion chamber into multiple spatial measurement regions. Respectively collect three roughness indexes, namely arithmetic mean roughness, maximum profile height, and root mean square roughness, in each spatial measurement region, and record the three-dimensional spatial position coordinates corresponding to each group of roughness indexes to construct the roughness distribution data of the inner wall of the flow channel of the space propulsion chamber; Take the roughness distribution data of the inner wall of the aerospace propulsion chamber flow channel as the input of non-parametric Bayesian modeling. The Dirichlet process uses a multi-dimensional Gaussian distribution as the base distribution, and sets a concentration parameter to control the non-parametric prior strength of the number of clusters; In the Dirichlet process modeling, the initial state is an empty cluster set. During the sampling process, each new data point uses the current three-dimensional space position coordinates and roughness index as the input, and calculates the likelihood value of the distribution parameters of each existing cluster. The distribution parameters of the cluster include the mean vector and covariance matrix of the roughness index; According to the distribution parameters, likelihood values and corresponding cluster sizes of each cluster, jointly calculate the attribution probability with the concentration parameter, perform cluster attribution sampling, and judge whether a new roughness sub-distribution cluster is generated; In each sampling round, introduce the roughness trend parameter output by the grey prediction model GM(1,1) as a prior adjustment variable to update the cluster distribution weight, so that the roughness evolution trend and the spatial distribution structure are adjusted synchronously; After sampling convergence, merge all roughness sub-distribution clusters, and output the roughness spatial probability distribution model of the inner wall of the aerospace propulsion chamber flow channel, which is used to describe the multi-peak distribution state and trend guiding behavior of roughness in different spatial regions.

[0031] In this embodiment, the inner wall of the aerospace propulsion chamber flow channel is divided into multiple spatial measurement regions, multi-dimensional indexes including arithmetic mean roughness, maximum profile height and root mean square roughness are collected, and roughness distribution data is constructed by combining three-dimensional space coordinates. The Dirichlet process is used to introduce a multi-dimensional Gaussian base distribution and a concentration parameter for non-parametric Bayesian modeling to establish an adaptive clustering model of roughness. During the continuous sampling process, the cluster attribution is dynamically adjusted according to the likelihood relationship between the current data point and the existing distribution clusters and the prior control factor, and the trend parameter generated by the grey prediction model GM(1,1) is introduced in each iteration to synchronously adjust the cluster weight. Finally, a roughness spatial probability distribution model that integrates trend and spatial structure is formed, realizing the accurate characterization of the multi-peak distribution and dynamic evolution behavior of the inner wall roughness of the flow channel, and improving the model's ability to identify local roughness anomalies in complex manufacturing environments.

[0032] In this embodiment, the specific steps of establishing the fusion mechanism between the roughness trend parameter and the spatial probability distribution model include: establishing a one-to-one mapping relationship between the roughness trend parameter and each roughness sub-distribution cluster in the spatial probability distribution model, calculating the residual between the roughness trend parameter and the mean value of the corresponding roughness sub-distribution cluster in each mapping pair, and constructing a fusion weight adjustment function based on the residual to dynamically correct the probability weights of the roughness sub-distribution clusters in the spatial probability distribution model of the inner wall of the flow channel of the aerospace propulsion chamber. At the same time, using the residual as a feedback variable to iteratively update the roughness trend parameter, so as to realize the two-way dynamic coupling fusion between the grey prediction model GM(1,1) and the spatial probability distribution model of the inner wall roughness of the flow channel of the aerospace propulsion chamber.

[0033] In this embodiment, a one-to-one mapping relationship is established between the roughness trend parameter and each roughness sub-distribution cluster in the spatial probability distribution model of the inner wall roughness of the flow channel of the aerospace propulsion chamber. The residual between the trend parameter and the sub-cluster mean value in each mapping pair is calculated, and a fusion weight adjustment function is constructed. The dynamic correction of the probability weight of the sub-distribution cluster is driven by the residual. At the same time, the residual is fed back for the iterative update of the trend parameter, realizing the two-way regulation and coupling between the grey prediction model and the roughness spatial modeling, so as to establish a dynamic consistency relationship between the roughness evolution trend and the spatial characteristics, and improve the accuracy of roughness modeling and the response sensitivity of manufacturing control.

[0034] In this embodiment, the steps of collecting real-time roughness monitoring data during the additive manufacturing process, comparing the real-time roughness monitoring data with the roughness prediction value sequence, and calculating the prediction residual specifically include: During the additive manufacturing process of the flow channel of the aerospace propulsion chamber, based on the preset scanning path and time interval, use a confocal laser measurement device to perform on-line measurement on the inner wall of the flow channel of the aerospace propulsion chamber, collect three types of roughness indexes including arithmetic mean roughness, maximum profile height, and root mean square roughness, and attach the sampling time index and three-dimensional spatial position coordinates to form the real-time roughness monitoring data of the inner wall of the flow channel of the aerospace propulsion chamber; Reorganize the real-time roughness monitoring data of the inner wall of the flow channel of the aerospace propulsion chamber into a time series structure according to the time index, and correspond it to the roughness prediction value sequence of the inner wall of the flow channel of the aerospace propulsion chamber in chronological order to complete the matching of the three types of predicted roughness indexes and the three types of measured roughness indexes at each sampling moment; At each sampling time point, calculate the numerical difference between the three types of roughness prediction values and the measured values of arithmetic mean roughness, maximum profile height, and root mean square roughness respectively as the prediction residual at the corresponding time point; Construct a prediction residual data set of the inner wall roughness of the flow channel in the space propulsion chamber from the prediction residuals calculated at all sampling time points. The roughness prediction residual data set is used as the input in the fusion mechanism to drive the dynamic correction and update of the parameters of the grey prediction model GM(1,1) and the space probability distribution model of the inner wall roughness of the flow channel in the space propulsion chamber.

[0035] In this embodiment, during the additive manufacturing process of the flow channel in the space propulsion chamber, a confocal laser measuring device is used to collect the inner wall roughness data of the channel in real time according to a preset path and time interval, and three types of indexes, namely arithmetic mean roughness, maximum profile height, and root mean square roughness, are extracted. After constructing a time series structure, the residuals are calculated by point-by-point matching with the predicted values to form a roughness prediction residual data set, which is further used as feedback input to drive the dynamic correction of the grey prediction model and the roughness space probability distribution model, realizing the real-time perception of the roughness change during the manufacturing process and model adjustment, significantly enhancing the system's response ability to manufacturing disturbances and prediction correction accuracy, and effectively improving the intelligent level of additive manufacturing roughness control and the consistency of the manufactured parts.

[0036] In this embodiment, a residual feedback mechanism is constructed based on the prediction residuals, and the residual feedback mechanism is introduced into the fusion mechanism to drive the linkage optimization process, which specifically includes: On the basis of obtaining the real-time roughness monitoring data of the inner wall of the flow channel in the space propulsion chamber and the sequence of predicted values of the inner wall roughness of the flow channel in the space propulsion chamber, the predicted roughness values and the real-time roughness monitoring values are paired point by point according to the time index, and the arithmetic mean roughness residual, maximum profile height residual, and root mean square roughness residual at each time point are calculated, and a residual time series is generated; Segment the residual time series by a sliding window, and statistically calculate the mean, standard deviation, and trend slope of the residual series within each window, and construct a residual feedback vector; Take the residual feedback vector as the input of the residual feedback mechanism, and respectively act on the roughness trend parameter generation process and the iterative process of the space probability distribution model of the inner wall roughness of the flow channel in the space propulsion chamber to form a dual-channel adjustment mechanism; In the roughness trend parameter generation process, the residual feedback mechanism is used to adjust the weight distribution structure of the non-uniform weight background value construction method in the grey prediction model GM(1,1) and the model parameter iteration update step size; In the space probability distribution model of the inner wall roughness of the flow channel in the space propulsion chamber, the residual feedback mechanism is used to adjust the sampling of each roughness sub-distribution cluster and the threshold for generating new clusters, and dynamically correct the cluster distribution weight of the space distribution; Through the dual-path injection method of the residual feedback mechanism, a linkage optimization process is constructed, so that the roughness trend parameters and the space probability distribution characteristics of the inner wall roughness of the flow channel in the space propulsion chamber are updated synergistically according to the roughness prediction residuals.

[0037] In this embodiment, by pairwise matching the real-time roughness monitoring data of the inner wall of the aerospace propulsion chamber flow channel with the roughness prediction value sequence, multiple types of residual time series are generated. A sliding window is used to statistically analyze the residual sequence to construct a residual feedback vector, which is further injected as feedback input into the grey prediction model GM(1,1) and the roughness spatial probability distribution modeling respectively, to dynamically adjust the construction strategy of the non-uniform weight background value and the model parameter update step size respectively, and to adaptively regulate the distribution cluster sampling mechanism and the generation threshold, thereby constructing a trend-space dual-path linkage optimization process for the manufacturing process disturbance environment, realizing the synchronous update of the roughness evolution trend and the spatial distribution structure, and significantly improving the robustness and regulation accuracy of the model in an uncertain manufacturing environment.

[0038] In this embodiment, the steps of generating a manufacturing path control instruction based on the roughness trend parameter and the spatial probability distribution feature output by the fusion mechanism include: Expand the roughness trend parameter by time index, extract the roughness change rate, fluctuation amplitude and local evolution trend corresponding to different manufacturing time periods, and form the time dynamic regulation factor in the manufacturing path control instruction; Perform regional decomposition on the spatial probability distribution feature of the inner wall roughness of the aerospace propulsion chamber flow channel, extract the spatial position center, covariance matrix and roughness distribution weight of each roughness sub-distribution cluster, and form the spatial distribution regulation factor in the manufacturing path control instruction; Based on the time dynamic regulation factor and the spatial distribution regulation factor respectively, construct a manufacturing path control instruction parameter set, and the manufacturing path control instruction includes: Scanning path structure parameters, including the three-dimensional coordinate sequence of the scanning segment, the scanning path direction vector and the scanning segment jump strategy; Laser power control parameters, including the roughness trend regulation coefficient and the laser power base value and adjustment amplitude under the spatial distribution mapping; Scanning speed control parameters, including a scanning speed function constructed based on the roughness trend change rate, defining the scanning start speed, maximum speed and speed gradient change form; Scanning spacing control parameters, including the scanning overlap rate, interlayer offset and path spacing adjustment amount corresponding to the roughness spatial density change; Jointly allocate each parameter in the manufacturing path control instruction parameter set according to the cross region of the roughness trend parameter and the spatial probability distribution feature of the inner wall roughness of the aerospace propulsion chamber flow channel, construct a fusion control mapping model, and generate a manufacturing path control instruction set for different manufacturing regions and time periods; Convert the manufacturing path control instruction set into a manufacturing control file format recognizable by the additive manufacturing equipment for the flow channel of the aerospace propulsion chamber, and transmit it to the control system to complete the manufacturing execution.

[0039] In this embodiment, the roughness trend parameter is expanded into a dynamic regulation factor with a time index, and the spatial probability distribution characteristics of the inner wall roughness of the aerospace propulsion chamber flow channel are analyzed into multiple sub-distribution clusters. The spatial center, distribution form, and weight parameters are extracted as spatial regulation factors, and a fusion control mapping model is constructed. The time and spatial regulation factors are respectively mapped to manufacturing path control parameters such as the scanning path, laser power, scanning speed, and spacing to form a manufacturing path control instruction set, which is then converted into a control file executable by the additive manufacturing equipment to achieve precise regulation of the manufacturing process. This method enables the manufacturing path to respond in real time to the roughness evolution trend and spatial distribution changes, thereby improving the surface consistency and forming accuracy of the key inner wall structure and laying a high-quality manufacturing foundation for the enhanced design of thermal-fluid coupling.

[0040] Example 1: To verify the feasibility of the present invention in implementation, the present invention is applied to an optimization project of the propulsion chamber structure carried out at the combustion system experimental base of a certain aerospace research institute. The research team selects the combustion chamber of a small liquid oxygen-kerosene rocket engine as the verification object, and attempts to apply the "manufacturing path regulation method based on the fusion of roughness trend and spatial distribution modeling" proposed by the present invention to the additive manufacturing link for the forming quality and surface roughness consistency problems of its internal cooling flow channels. The internal cooling channels of this type of engine combustion chamber are mostly spiral winding channel structures, with a cross-sectional size between 0.8 mm and 1.5 mm, long paths, frequent spatial variable cross-sections, and extremely high requirements for the evolution stability, overall consistency, and defect control of the wall surface roughness.

[0041] During the implementation process, first, a manufacturing parameter perturbation combination matrix including scanning speed, power, spacing, and layer thickness perturbation is established. Ten groups of channel structure specimens are respectively manufactured on IN718 powder materials using a selective laser melting additive manufacturing equipment for metals, and the roughness generation behavior under each combination parameter is recorded. Subsequently, a grey prediction model GM(1,1) is constructed, and the trend characteristics of the channel roughness formation sequence are extracted by means of piecewise accumulation and non-uniform weighted background modeling, and the roughness evolution behavior in different perturbation intervals is predicted to generate corresponding trend parameters and prediction sequences. Three types of roughness indexes are collected for multiple regions such as the inlet section, middle section, and outlet section of the specimen channels, and a three-dimensional spatial probability distribution model is constructed by combining the grey model trend parameters to guide the Dirichlet process, and a prior spatial model of the roughness multi-peak structure is established.

[0042] In the actual additive manufacturing process, a confocal laser on-line measurement device is deployed to collect the roughness index of the channel area during the building process in real time, and calculate the residuals at each time point between the roughness trend prediction sequence. In the residual feedback mechanism, these residuals are used to construct a residual feedback vector in a sliding window manner and introduced into the grey prediction model and Dirichlet modeling respectively, for dynamically adjusting the prediction accuracy and spatial modeling weights, so as to realize the closed-loop coupling control between the roughness evolution model and the actual manufacturing state.

[0043] The spatio-temporal information extracted from the roughness trend parameters and spatial probability distribution parameters is converted into manufacturing path control instructions. The control strategies include: generating the three-dimensional vector of the scanning path and the jump structure, constructing the laser power base value adjustment function, defining the speed function and the overlap rate regulation model, and finally generating a set of control instruction files that take into account the regional roughness density and evolution trend, and importing them into the device system for execution.

[0044] In terms of specific effects, the team conducted a comparative experiment on 10 channel specimens manufactured by the traditional strategy and the method of the present invention respectively, measured and recorded the distribution of three types of roughness indexes on the inner wall surface of the channels, the Rq stability and defect rate in the middle section. The specimen numbers A1 - A10 correspond to the samples of the traditional and the present invention manufacturing strategies respectively.

[0045] Table 1 Results of the comparative experiment on the roughness distribution of the propulsion chamber channels

[0046] From the data analysis of the "Comparison Experiment Results of Roughness Distribution in the Propulsion Chamber Passage", it can be seen that the samples manufactured by the method of the present invention are significantly superior to the traditional strategy in multiple key roughness indicators. First, in terms of the arithmetic mean roughness (Ra), the average values of the traditional samples A1–A5 are concentrated between 15.9 μm and 17.1 μm, while the samples A6–A10 manufactured by the method of the present invention are significantly reduced to 13.1 μm to 13.8 μm, indicating a lower and flatter overall roughness level. Second, the coefficient of variation of Ra is reduced from 0.18–0.22 of the traditional method to 0.07–0.09, reflecting that the roughness control in the manufacturing process of the new method is more stable and consistent. In terms of the maximum profile height (Rz), the maximum values of the traditional samples are generally above 80 μm, while the maximum Rz of the samples under the control of the present invention is stable at 65–68 μm, effectively suppressing local roughness mutations. The stability index of the middle section region of the root mean square roughness (Rq) is increased from about 0.76 of the traditional method to 0.91 under the present invention, indicating that this method can significantly improve the surface morphology stability of the middle section of the passage. In terms of the manufacturing defect rate, the average of the traditional method is 7.2%, while it is significantly reduced to about 2.5% under the control strategy of the present invention, reflecting the effective suppression ability of the present invention for surface anomalies and structural defects. Comprehensive analysis shows that the method of the present invention has achieved substantial improvements in roughness control accuracy, distribution consistency, and defect suppression, verifying the significant effect of the described manufacturing path control strategy in the additive manufacturing process of complex channels.

[0047] In summary, the roughness trend prediction and spatial modeling fusion path control method proposed by the present invention can effectively improve the manufacturing consistency and surface quality of the aerospace propulsion chamber passage, achieve stable roughness control under multi-position and multi-scale conditions, and significantly reduce the incidence of manufacturing defects, with high generalizability and engineering practical value.

[0048] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and all should be covered within the protection scope of the present invention.

Claims

1. An additive manufacturing flow channel roughness optimization method for a space propulsion chamber, characterized in that, The steps include: Set the manufacturing parameter perturbation strategy, perform additive manufacturing experiments under multiple perturbation groups, and collect the sequence of inner wall roughness of the corresponding channels. Construct a grey prediction model based on the sequence formed by the collected roughness , generate a roughness time evolution model of roughness varying with disturbance conditions in the form of a differential equation, and output the roughness trend parameters and the sequence of roughness prediction values; Collect the roughness distribution data at different positions on the channel wall surface, and jointly use the roughness trend parameters as prior inputs to guide the Dirichlet process to establish a spatial probability distribution model. Establish a fusion mechanism between the roughness trend parameters and the spatial probability distribution model, and output the roughness trend parameters and the roughness spatial probability distribution characteristics. Collect real-time roughness monitoring data during the additive manufacturing process, compare the real-time roughness monitoring data with the sequence of roughness prediction values, and calculate the prediction residuals. Construct a residual feedback mechanism based on the prediction residuals, and introduce the residual feedback mechanism into the fusion mechanism to drive the linkage optimization process. Generate manufacturing path control instructions based on the roughness trend parameters and the roughness spatial probability distribution characteristics output by the fusion mechanism.

2. The additive manufacturing flow channel roughness optimization method for a space propulsion chamber according to claim 1, wherein, The steps of setting the manufacturing parameter perturbation strategy, performing additive manufacturing experiments, and collecting the sequence of inner wall roughness of the channels include: Select the laser power, scanning speed, scanning spacing, and forming layer thickness as the manufacturing parameter perturbation dimensions, construct a manufacturing parameter perturbation combination matrix, and the manufacturing parameter perturbation combination matrix serves as the component content of the manufacturing parameter perturbation strategy. According to the manufacturing parameter perturbation combination matrix, sequentially perform additive manufacturing experiments on a selective laser melting additive manufacturing equipment for metals. The additive manufacturing experiments generate samples containing the flow channel structure of the aerospace propulsion chamber for different parameter combinations. After each group of additive manufacturing experiments, use a surface topography measurement device to measure the roughness at multiple positions on the inner wall of the flow channel of the aerospace propulsion chamber in the sample. The roughness measurement area covers multiple characteristic positions such as the intake section, middle section, and outlet section. Pair the manufacturing parameter perturbation combinations used in each group of additive manufacturing experiments with the corresponding roughness measurement results of the inner wall of the flow channel of the aerospace propulsion chamber to construct a roughness formation sequence.

3. The additive manufacturing flow channel roughness optimization method for a space propulsion chamber according to claim 1, characterized in that Construct a grey prediction model based on the collected roughness formation sequence , generate a roughness time evolution model of roughness varying with disturbance conditions in the form of a differential equation, and output the roughness trend parameters and the roughness prediction value sequence, specifically including: Construct the roughness formation sequence into the original time series data of the inner wall roughness of the flow channel of the aerospace propulsion chamber, and perform a piecewise adjustable cumulative generation operation based on a preset time window division rule to form multiple roughness grey generation subsequences. Based on each roughness grey generation subsequence, establish corresponding grey prediction sub-models respectively , and in each grey prediction sub-model generate a set of roughness trend characteristic parameters for trend analysis; In each grey prediction sub-model a non-uniform weighted background value sequence is constructed based on the roughness grey generation subsequence: ； Among them, is the th non-equal-weight background value sequence, is the roughness gray generation subsequence, is the weight coefficient, and its value range is , is the sequence index; Construct each grey prediction sub-model based on the non-uniform weighted background value sequence of the differential prediction equation; Summarize the roughness prediction value sequences output by each grey prediction sub-model ; Integrate the roughness trend feature parameters extracted by each grey prediction sub-model jointly to form the roughness trend parameters.

4. A method for optimizing the roughness of a flow channel in an aerospace propulsion chamber by additive manufacturing according to claim 1, characterized in that, The specific steps of collecting the roughness distribution data at different positions on the channel wall surface and jointly using the roughness trend parameters as prior inputs to guide the Dirichlet process to establish a spatial probability distribution model include: Divide the inner wall of the flow channel of the aerospace propulsion chamber into multiple spatial measurement regions, respectively collect three roughness indexes, namely the arithmetic mean roughness, the maximum profile height, and the root mean square roughness, in each spatial measurement region, and record the three-dimensional spatial position coordinates corresponding to each group of roughness indexes to construct the roughness distribution data of the inner wall of the flow channel of the aerospace propulsion chamber. Use the roughness distribution data of the inner wall of the flow channel of the aerospace propulsion chamber as the input for non-parametric Bayesian modeling. The Dirichlet process uses a multi-dimensional Gaussian distribution as the base distribution, and sets a concentration parameter to control the non-parametric prior strength of the number of clusters. In the Dirichlet process modeling, the cluster set is empty in the initial state. During the sampling process, each new data point takes the current three-dimensional spatial position coordinates and roughness index as inputs, and calculates the likelihood values of the distribution parameters of each existing cluster. The distribution parameters of the cluster include the mean vector and covariance matrix of the roughness index. According to the distribution parameters, likelihood values, and corresponding cluster sizes of each cluster, the concentration parameter is jointly calculated to obtain the membership probability, and the cluster membership sampling is performed, and it is judged whether a new roughness sub-distribution cluster is generated. In each sampling round, introduce the roughness trend parameter output by the grey prediction model as a prior adjustment variable to update the cluster distribution weights, so as to synchronously adjust the roughness evolution trend and the spatial distribution structure; After the sampling converges, all the roughness sub-distribution clusters are merged, and the spatial probability distribution model of the inner wall roughness of the aerospace propulsion chamber flow channel is output.

5. A method for optimizing the roughness of a flow channel in an aerospace propulsion chamber by additive manufacturing according to claim 1, characterized in that, The establishment of the fusion mechanism between the roughness trend parameter and the spatial probability distribution model specifically includes establishing a mapping relationship between the roughness trend parameter and each roughness sub-distribution cluster in the spatial probability distribution model one by one. In each mapping pair, the residual between the roughness trend parameter and the mean of the corresponding roughness sub-distribution cluster is calculated, and a fusion weight adjustment function is constructed based on the residual to dynamically correct the probability weights of each roughness sub-distribution cluster in the spatial probability distribution model of the inner wall roughness of the aerospace propulsion chamber flow channel. At the same time, the residual is used as a feedback variable to iteratively update the roughness trend parameter.

6. The additive manufacturing flow channel roughness optimization method for a space propulsion chamber according to claim 1, wherein The collection of real-time roughness monitoring data during the additive manufacturing process and the comparison of the real-time roughness monitoring data with the roughness prediction value sequence to calculate the prediction residual specifically include: During the additive manufacturing process of the aerospace propulsion chamber flow channel, based on the preset scanning path and time interval, a confocal laser measurement device is used to perform on-line measurement on the inner wall of the aerospace propulsion chamber flow channel, and three types of roughness indexes including arithmetic mean roughness, maximum profile height, and root mean square roughness are collected, and the sampling time index and three-dimensional spatial position coordinates are attached to form the real-time roughness monitoring data of the inner wall of the aerospace propulsion chamber flow channel. The real-time roughness monitoring data of the inner wall of the aerospace propulsion chamber flow channel is reorganized into a time series structure according to the time index, and is corresponding to the roughness prediction value sequence of the inner wall of the aerospace propulsion chamber flow channel one by one in chronological order to complete the matching of the three types of predicted roughness indexes and the three types of measured roughness indexes at each sampling moment. At each sampling time point, the numerical differences between the predicted values and measured values of the arithmetic mean roughness, maximum profile height, and root mean square roughness are calculated respectively as the prediction residuals at the corresponding time points.

7. A method for optimizing the roughness of a flow channel in an aerospace propulsion chamber by additive manufacturing according to claim 1, characterized in that, The construction of the residual feedback mechanism based on the prediction residuals and the introduction of the residual feedback mechanism into the fusion mechanism to drive the linkage optimization process specifically include: On the basis of obtaining the real-time roughness monitoring data of the inner wall of the aerospace propulsion chamber flow channel and the roughness prediction value sequence of the inner wall of the aerospace propulsion chamber flow channel, the roughness prediction values and real-time roughness monitoring values are paired point by point according to the time index, and the arithmetic mean roughness residual, maximum profile height residual, and root mean square roughness residual at each time point are calculated, and a residual time series is generated. The residual time series is segmented by a sliding window, the mean, standard deviation, and trend slope of the residual sequence in each window are statistically calculated, and a residual feedback vector is constructed. Taking the residual feedback vector as the input of the residual feedback mechanism, it acts on the roughness trend parameter generation process and the iterative process of the spatial probability distribution model of the inner wall roughness of the aerospace propulsion chamber flow channel respectively, forming a dual-channel adjustment mechanism; During the generation of roughness trend parameters, a residual feedback mechanism is used to adjust the weight distribution structure of the non-uniform weighted background value construction method and the model parameter iteration update step size in the grey prediction model; In the spatial probability distribution model of the inner wall roughness of the aerospace propulsion chamber flow channel, the residual feedback mechanism is used to adjust the sampling of each roughness sub-distribution cluster and the threshold for generating new clusters, and dynamically correct the cluster distribution weights of the spatial distribution.

8. A method for optimizing the roughness of a flow channel in an aerospace propulsion chamber by additive manufacturing according to claim 1, characterized in that The manufacturing path control instructions specifically include: Performing time-index expansion on the roughness trend parameters, extracting the roughness change rate, fluctuation amplitude, and local evolution trend corresponding to different manufacturing time periods, and constituting the time dynamic regulation factor in the manufacturing path control instructions; Performing regional decomposition on the spatial probability distribution characteristics of the inner wall roughness of the aerospace propulsion chamber flow channel, extracting the spatial position center, covariance matrix, and roughness distribution weight of each roughness sub-distribution cluster, and constituting the spatial distribution regulation factor in the manufacturing path control instructions; Based on the time dynamic regulation factor and the spatial distribution regulation factor respectively, constructing manufacturing path control instructions, and the manufacturing path control instructions include: Scanning path structure parameters, including the three-dimensional coordinate sequence of the scanning segment, the scanning path direction vector, and the scanning segment jump strategy; Laser power control parameters, including the roughness trend regulation coefficient and the laser power base value and adjustment amplitude under the spatial distribution mapping; Scanning speed control parameters, including the scanning speed function constructed based on the roughness trend change rate, defining the scanning start speed, maximum speed, and the form of speed gradient change; Scanning spacing control parameters, including the scanning overlap rate, interlayer offset, and path spacing adjustment amount corresponding to the change in the roughness spatial density.

Images