66 results about "Polynomial regression model" patented technology

The invention belongs to the field of phase change heat storage, and discloses a phase change heat exchange fin parameter optimization method and a related device.According to the method, a three-dimensional parameterized model of a tree-shaped fractal fin containing the length ratio, the width fractal dimension and the fin number is established, representative parameter sample points are efficiently generated through Box-Behnken experimental design, and the representative parameter sample points are optimized; and a liquid phase rate and a total heat storage capacity response value are obtained by combining numerical simulation of phase change heat transfer, and multi-target parameter global optimization is realized based on a second-order polynomial regression model. By the adoption of the method, optimization efficiency and precision are remarkably improved, on the premise that a large number of repeated experiments or simulations are not needed, the optimal fin structure parameters capable of cooperatively strengthening heat conduction, promoting melting of the phase-change materials and maximizing the heat storage capacity are rapidly obtained, and a reliable design basis is provided for engineering application of the fractal fin strengthening phase-change heat storage technology.

The invention discloses an establishing method, a monitoring method and an adding method of a monitoring model for the concentration of hydrochloric acid in an etching solution. The establishing method of the monitoring model for the concentration of hydrochloric acid in the etching solution comprises the following steps: acquiring sample data; calculating each coefficient value in a pre-established multivariate polynomial regression model according to the sample data to obtain a regression model; obtaining verification data, taking the temperature, the oxidation-reduction potential, the hydrochloric acid concentration and the Cu < 2 + > concentration in the verification data as the input of the regression model, and outputting the predicted value of the conductivity; calculating the closeness degree of the predicted value of the conductivity and the actual value of the conductivity in the verification data; when the calculated closeness degree is smaller than a preset threshold value, recalculating each coefficient value in the regression model until the closeness degree is greater than or equal to the preset threshold value; and when the calculated closeness degree is greater than or equal to a preset threshold value, taking the obtained regression model as a hydrochloric acid concentration monitoring model.

The invention provides a process parameter optimization method and system for a functionally graded material and a medium, which can be applied to the technical field of process parameter optimization. According to the method, after corresponding technological parameters and forming response data under different component gradients are obtained through an orthogonal test, a quadratic term polynomial regression model used for representing the mapping relation between the technological parameters and the forming response data in the different component gradients is constructed according to the technological parameters and the forming response data. A quadratic term polynomial regression model is used as a virtual simulation environment, meanwhile, after a multi-dimensional continuous strengthening model is constructed based on a double-delay depth deterministic strategy gradient algorithm, based on the multi-dimensional continuous strengthening model, through dynamic interaction of an intelligent agent and the virtual simulation environment, an optimal process parameter combination is searched in different component gradients; and thus, cross-component gradient process parameter optimization is realized. And finally, verifying the optimal process parameter combination through a physical experiment, and determining the accuracy of model prediction.

The invention discloses a bioretention facility design method based on numerical simulation and multi-objective optimization, and belongs to the technical field of rainwater management, and the method comprises the following steps: S1, obtaining a special background demand of a bioretention facility of a target area; s2: according to a filler selection layering model constructed by an analytic hierarchy process, inputting a special background demand into the filler selection layering model to obtain an internal optimal filler combination of the bioretention facility; s3, constructing a DRAINMOD hydrological model, and establishing a parameter space through BOX-Behnken experimental design to generate a plurality of groups of virtual working conditions; s4, simulating the multiple groups of virtual working conditions through a DRAINMOD hydrological model to obtain a simulation result; s5, on the basis of a sensitivity analysis method, key regulation and control parameters are determined according to simulation results; and S6, based on a simulation result, optimizing the key regulation and control parameters through a second-order polynomial regression model to obtain an optimal design parameter combination.

The invention relates to a subway station large passenger flow early warning method and system based on real-time data, and the method comprises the steps: obtaining the real-time mobile phone signaling data of a subway station and the surrounding, and carrying out the preprocessing of the real-time data; performing short-time passenger flow prediction and long-time passenger flow prediction based on the processed mobile phone signaling data; wherein the short-time passenger flow prediction adopts an ARIMA model to predict the passenger flow of each time window in the future set hours; according to long-time passenger flow prediction, an addition time sequence decomposition model is adopted to decompose a passenger flow sequence into a trend item, a season item and a residual item, the trend item is predicted through a polynomial regression model, the season item is predicted through a model based on Fourier series, and a long-time passenger flow prediction result is obtained based on the predicted trend item and the predicted season item; and performing early warning judgment according to the passenger flow prediction result and an early warning threshold dynamically updated based on the real-time passenger flow data. Compared with the prior art, the accuracy, timeliness and adaptability of large passenger flow early warning are effectively improved, and powerful support is provided for subway operation safety.

The invention provides a preparation method and device of a composite matrix for sewage treatment and a server, and relates to the technical field of water treatment material.The preparation method comprises the steps that center combination treatment is conducted on a ternary composite matrix for sewage treatment, and a test formula set is generated; obtaining test data corresponding to the test formula set, and taking the removal rate corresponding to each matrix as a response value to obtain target test data; substituting the target test data into a quadratic polynomial regression model of the ternary composite matrix to carry out regression analysis processing to obtain a model predicted value, and carrying out model verification updating processing on the quadratic polynomial regression model by utilizing the model predicted value and the target test data to determine a target quadratic polynomial regression model; and through the target quadratic polynomial regression model, optimizing the ratio of the ternary composite matrix, and determining the target mass ratio of the ternary composite matrix. According to the invention, the pollutant removal rate of synergistic decontamination of various pollutants in the sewage treatment process can be obviously improved.

The invention relates to an index prediction method and device for a regional economy space-time knowledge graph, and the method comprises the steps: constructing the regional economy space-time knowledge graph; performing knowledge representation learning on the spatial-temporal knowledge graph of the regional economy by using a preset relational graph convolutional network; determining target targeting indexes influencing regional economic development according to the cosine similarity; comparing index prediction degrees of a preset R-GCNs + XGBoost model, a preset random forest and a preset polynomial regression model for regional economic development; and predicting at least one key index of regional economy based on a preset R-GCNs + XGBoost model and the comparison data. According to the method, regional economy and the knowledge graph technology are creatively and deeply fused, inherent dynamic nature and space-time characteristics of economic development in a real scene are fully considered, and powerful technical support is provided for regional economy targeted index analysis and future index prediction by using powerful information organization and interpretation capabilities of the knowledge graph.

The invention discloses a sight tracking method and system based on cascade deep learning and distance correction, and a storage medium. The method comprises the following steps: processing an eye image through a cascade deep learning model, and obtaining pupil center coordinates and cornea reflection light spot coordinates of two eyes; calculating a binocular pixel distance as a distance correction factor based on the pupil center coordinates, and calculating a relative position difference vector based on the pupil and light spot coordinates; establishing a three-dimensional nonlinear polynomial regression model taking the relative position difference vector and the distance correction factor as input; obtaining calibration data through a multi-distance sampling strategy and solving a model coefficient; and in the real-time tracking stage, calculating the coordinates of the fixation point of the screen by using the solved model. According to the method, the distance correction factor and the three-dimensional mapping model are introduced, so that geometric errors caused by forward and backward movement of the head of the user are effectively eliminated, meanwhile, the feature point positioning precision is ensured through cascaded deep learning detection, and the precision and robustness of a sight line tracking system are remarkably improved.

The invention provides a direct drinking water equipment water quality data analysis and accurate control method based on an intelligent algorithm, and relates to the technical field of intelligent control of direct drinking water equipment.The method comprises the steps that water quality data including the total dissolved solid content, the pH value, the oxidation-reduction potential, the residual chlorine concentration and the temperature are collected in real time through a multi-sensor module; performing dynamic standardization preprocessing on the data to generate a multidimensional feature vector containing a timestamp; the water quality state is evaluated based on a water quality dynamic deviation index WDDI and a system response efficiency coefficient SRE, the WDDI quantifies the parameter dynamic deviation degree through a piecewise function, and the SRE represents the equipment response efficiency through a polynomial regression model; an NSGA-II algorithm and LSTM prediction are fused to generate a dynamic control strategy set; an optimal strategy is screened according to the real-time SRE value, and filter element mode switching, backwashing adjustment and sterilization power adjustment are driven; algorithm parameters are optimized through closed-loop feedback, and abnormal detection and safety mode switching are achieved in combination with an isolated forest algorithm.

The invention relates to the technical field of earphone equipment, in particular to a Bluetooth earphone non-contact type dynamic control system and method based on gesture recognition, and the system comprises an induction signal collection module which obtains the total number of clock cycles and generates an original count value, a signal standardization module which carries out the operation of the original count value to generate a standardization intensity value, and a control module. The spatial feature distinguishing module is used for generating a spatial curvature gradient coefficient based on the standardized intensity numerical value and screening an effective intensity numerical value; and the linear mapping control module is used for performing nonlinear correction mapping on the effective numerical value, generating a vertical distance coordinate numerical value and generating a volume gain control parameter based on a change rate. According to the method, environment interference is eliminated through standardized data differential operation, effective fingertip control gestures and large-area skin contact are distinguished through space curvature gradient coefficients, a polynomial regression model is called to establish the corresponding relation between signal intensity and distance coordinates, non-contact hovering interaction is achieved, and ear canal compression and mechanical abrasion are avoided.

The invention discloses an inflatable rescue boat self-righting structure optimization method based on RSM-GA, and belongs to the technical field of ship engineering and optimization design, and the method comprises the steps: S1, employing computer aided design software to establish an inflatable rescue boat three-dimensional model, employing simulation software to carry out static stability analysis, and obtaining a three-dimensional model of the inflatable rescue boat; s2, establishing a second-order polynomial regression model of the self-righting capability and design parameters of the inflatable rescue boat by using a response surface method, quantifying interaction of the parameters, and determining an optimization target, and S3, performing multi-target optimization on the second-order polynomial regression model by combining a genetic algorithm, and obtaining optimal design parameters of a self-righting structure when engineering constraint conditions are met at the same time; according to the method for optimizing the self-righting structure of the inflatable rescue boat based on the RSM-GA, the stability and the safety of the rescue boat can be guaranteed, meanwhile, the sailing resistance is effectively reduced, the sailing efficiency is improved, and the self-righting capacity and the sailing performance of the rescue boat are remarkably improved.

The present application relates to a kind of shale oil carbon dioxide fracturing production dynamic intelligent prediction method, the method is applied to the improved AHP-GRA-EWM method quantitatively calculate the influence weight of different categories factors to different stage carbon dioxide fracturing-throughput shale oil well production dynamic parameter, obtain the main control influence factor of different production stage production dynamic parameter;The main control influence factor of different production parameters of different production stages is respectively introduced into different artificial neural network system, nonlinear polynomial regression model system and different percolation mathematical model, establishes prediction model, selects the model with highest prediction accuracy to correct improvement, combination generates the production dynamic intelligent prediction model of the entire production cycle of target shale reservoir block carbon dioxide fracturing-throughput shale oil well;Based on the intelligent prediction model, the different production parameter values of different production stages of new well are calculated, and the production dynamic curve is drawn.The method designed by the present application can realize the fine and efficient prediction of carbon dioxide fracturing-throughput shale oil well production dynamics, provide basis for the development of development plan and optimization, improve the single-well production and recovery ratio of target shale reservoir block.

The present application relates to the technical field of prosthesis lattice design, in particular to a kind of orthopedic implant lattice design method, system and application, by establishing parameterized lattice model sample set, and the polynomial regression model of elastic modulus and permeability coefficient is combined with the bayesian optimization framework with comprehensive score objective function as optimization goal, so that it can be iteratively optimized in parameter range efficiently, finally output optimal design parameter to make the mechanical property and mass transfer performance reach the best balance, fundamentally solve the problem that existing PEEK lattice implant design lacks double performance collaborative optimization, further improve lattice prosthesis postoperative recovery effect.

The invention discloses a method for evaluating the influence of data-driven distributed photovoltaic grid connection on feeder line voltage fluctuation. According to the method, the thought of combining simulation data modeling and actual measurement data correction is comprehensively utilized, and accurate quantitative evaluation can be carried out on node voltage fluctuation characteristics in the absence of comprehensive monitoring conditions; a voltage variation polynomial regression model is constructed through a simulation sample, regression model parameter correction is carried out in combination with actually measured historical data, and finally distributed photovoltaic output is input to obtain a whole-network voltage variation sequence. And accurate prediction of node voltage fluctuation amplitude and frequency index caused by distributed photovoltaic output under the condition of incomplete monitoring data is realized through data processing. The method is high in accuracy and high in engineering applicability, is particularly suitable for being applied to a medium and low voltage power distribution system which comprises a plurality of photovoltaic access points and is relatively high in fluctuation characteristic, and has a good popularization prospect.

The invention relates to the technical field of intelligent beekeeping, and discloses a mature honey standardized production control system based on strong group multiple box bodies, which comprises a dielectric spectrum sensing array integrated on a comb foundation, a data acquisition module, a central processing module and a user interface module. According to the system, the micro interdigital capacitor unit is used for independently detecting electromagnetic characteristics of substances in cells, the central processing module obtains complex impedance spectroscopy and temperature data, a content classification model and a polynomial regression model containing temperature cross terms are operated in parallel, the types of the contents in the cells are recognized, and the water content of honey is accurately calculated. The system establishes data mapping between a super production area and a hive breeding area, respectively calculates a mature honey harvest index and a breeding health degree score, and automatically generates a honey extraction recommendation or a health alarm. According to the invention, lossless, in-situ and high-resolution monitoring of the internal state of the beehive is realized, and the problems that the bee colony is interfered by manual box opening and the honey maturity is lack of quantitative standards are solved.

This invention discloses an optimized process for drying rainbow trout skin using vacuum microwave combined with airflow puffing, belonging to the field of food processing technology. Addressing the problems of low crispness and lack of systematic optimization of process parameters in existing rainbow trout skin puffing processes, this invention investigates the effects of pre-drying moisture content, puffing temperature, and vacuum drying time on the puffing rate, hardness, and crispness of the product through single-factor experiments. Based on this, using crispness (deformation distance) as the response value, a quadratic polynomial regression model is established using Box-Behnken design, and the optimal combination of process parameters is determined through response surface methodology. The regression model is: Y = 0.2832 + 0.0837A + 0.0241B + 0.0186C + 0.0025AB + 0.0148AC + 0.0040BC + 0.2480A 2 +0.1190B 2 +0.0286C 2 The optimal process parameters obtained by this invention are: pre-drying moisture content of 30%, puffing temperature of 85℃, and vacuum drying time of 30 min. Under these conditions, the puffed rainbow trout skin product exhibits a deformation distance of 0.25–0.31 mm, a puffing rate exceeding 200%, significantly improved crispness, and stable quality. The regression model established in this invention shows a good fit (R²). 2 =0.9948), with high prediction accuracy, it provides a clear technical parameter window for the industrial production of rainbow trout skin puffed snack food, and has good prospects for industrial application.

The application provides a configuration method and device of a composite substrate for sewage treatment and a server, relates to the technical field of water treatment material preparation, and comprises the following steps: performing central combination treatment on a ternary composite substrate for sewage treatment to generate a test formula set; obtaining test data corresponding to the test formula set, taking the removal rates of various substrates as response values, and obtaining target test data; performing regression analysis processing on the target test data in a quadratic polynomial regression model of the ternary composite substrate to obtain model prediction values, and performing model verification and update processing on the quadratic polynomial regression model by using the model prediction values and the target test data to determine a target quadratic polynomial regression model; and performing optimization processing on the proportioning of the ternary composite substrate by using the target quadratic polynomial regression model to determine a target mass proportioning of the ternary composite substrate. The application can significantly improve the pollutant removal rate of the collaborative pollutant removal of various pollutants in the sewage treatment process.

The invention provides a collaborative optimization method for a magnetic-thermal composite field assisted laser cladding process. According to the method, a magnetic-thermal composite field-assisted laser cladding device is designed based on an energy field-assisted laser cladding process principle and is used for improving the structure and performance of a coating; the method comprises the following steps: generating a quadratic polynomial regression model on the basis of an RSM method, magnetic-thermal composite field assisted laser cladding process parameters and nickel-based composite coating quality evaluation indexes, performing multi-target optimization on the coating quality evaluation indexes on the basis of the quadratic polynomial regression model and an IMOGWO algorithm, determining a magnetic-thermal composite field assisted laser cladding target process parameter combination, and determining the quality of the nickel-based composite coating. And score calculation is carried out on the target process parameter combination based on an EWM-TOPSIS method to obtain a comprehensive evaluation sequence of the target process parameter combination, so that collaborative optimization of the magnetic-thermal composite field assisted laser cladding process is realized, and the performance optimization efficiency and quality of the nickel-based composite coating are improved.

This invention provides a method and system for controlling the stability of steel slag aggregate, belonging to the technical field of steel slag aggregate stability control. The method determines the degree of specimen damage under different combinations of maximum steel slag particle size and dosage through autoclaving accelerated testing. The degree of damage is quantified into a continuous stability index according to a preset grading standard. A quadratic polynomial regression model is established between the stability index and the maximum steel slag particle size and dosage, which also includes interaction terms for particle size and dosage. Based on the model and a preset engineering acceptable threshold, the maximum safe dosage or maximum allowable particle size of steel slag that meets the threshold requirement is calculated. The mix design of road base materials is then performed based on the calculation results. This invention achieves continuous quantitative characterization of the influence of steel slag particle size and dosage on stability, overcoming the shortcomings of existing technologies that rely on single-factor empirical limits and lack engineering back-calculation methods. It can improve the resource utilization rate of steel slag while ensuring stability.

The invention discloses a gangue concrete filling material ratio improving method based on a multi-objective optimization algorithm, which comprises the following steps: step 1, determining raw material composition and key influence factors of a gangue concrete filling material, step 2, carrying out test design and performance test based on RSM-BBD, the method comprises the following steps of 1, establishing a multi-objective optimization model, 2, establishing a quadratic polynomial regression model and carrying out reliability analysis, 3, determining a multi-objective optimization objective, a constraint condition and an optimization algorithm, 5, solving the multi-objective optimization algorithm and screening an optimal ratio, 6, carrying out performance verification and engineering application of the optimal ratio, and 7, carrying out dynamic updating and iteration of an optimization model. According to the method, the RSM-BBD test design and the multi-objective optimization algorithm are fused, the comprehensive influence and interaction of the mass fraction of the rubber material, the mixing amount of the alkali activator and the mass fraction of the solid material on the bleeding rate, the strength and the slump of the filling material are comprehensively considered, the limitation of a traditional single-factor experiment method is overcome, and the scientificity and the reliability of a matching optimization result are improved.

A lane line detection method and apparatus, and a device and a medium. The method comprises: acquiring three-dimensional radar point cloud data of a traveling direction environment of a vehicle (S110); performing grid division on the three-dimensional radar point cloud data, so as to obtain grid feature data (S120); inputting the grid feature data into a deep polynomial regression model, so as to obtain at least one set of output data (S130), wherein each set of output data comprises polynomial parameters in a preset univariate polynomial function and first coordinate data of a lane line target point, and the first coordinate data is coordinate data of the lane line target point in a traveling direction; for each set of output data, obtaining a target polynomial function on the basis of the polynomial parameters and the preset univariate polynomial function in the output data (S140); and on the basis of the target polynomial function, and the first coordinate data of the lane line target point in the output data, obtaining a target lane line corresponding to each set of output data (S150).

The present disclosure provides an image color correction method and apparatus, a color correction robot, and a storage medium. The method comprises: acquiring an image to be processed; extracting color values of pixels in a plurality of color blocks in the image to be processed to obtain a color value list; setting a weight coefficient for each color value in the color value list; using the color value list having the weight coefficients to fit a polynomial regression model to obtain a fitted polynomial regression model; using the fitted polynomial regression model to generate a color conversion matrix, wherein the color conversion matrix is used for representing a mapping relationship between the color values in the image to be processed and standard color values in a color chart; and using the color conversion matrix to correct the color values of the pixels in the image to be processed, to obtain a corrected image. The present embodiment improves the color accuracy of an image and ensures the reliability and authenticity of image information, thereby providing reliable data support for subsequent intelligent analysis and assisted diagnosis.

This application discloses a voltage control method and system for a solar power generation and energy storage system. The method includes: (i) a preparation step; (ii) a test calculation and virtual power compensation step; (iii) a machine learning step: the training input data is the voltage value of each grid-connected node, and the output is the virtual power compensation amount of its grid-connected node; a cubic polynomial regression controller with the same number of energy storage systems is trained separately. Each controller uses the results of autonomous compensation of each energy storage system in the main transformer area of ​​the secondary substation as the source of training data. The operating scenario is that the secondary side voltage of the main transformer is 1.0 pu, the energy storage system is discharging, and the load is from 0.1% to 100%. Load condition data is captured in 0.1% increments. Before training, the data is normalized; (iv) an application implementation step: the cubic polynomial regression model of the solar power and energy storage system is obtained. The required virtual power compensation amount is obtained by inputting the voltage data of the grid nodes and the voltage control requirements, and the voltage of the system is controlled.

The invention discloses a method for predicting the maximum displacement of a dredging ship gauge pile. The method comprises the following steps: constructing a trolley-gauge pile-soil body coupling numerical model based on ABAQUS software; establishing a cubic polynomial regression model of the maximum displacement of the gauge pile with respect to the three influence factors by applying a response surface method; data samples are stochastic in a regression model parameter range, and a training set, a verification set and a prediction set are divided; constructing a BP neural network model, and training and testing the BP neural network model by using the training set and the verification set to obtain an optimized prediction model; and inputting influence factors of the prediction set into the trained BP neural network for prediction, performing dual verification on a prediction result by using a relative error and a determination coefficient, and outputting the prediction result when the prediction result meets a verification standard. According to the method, high-precision real-time prediction of the displacement of the gauge pile is realized, and an important decision basis is provided for safe construction of a dredging ship.

The invention discloses a metering method, device and system for setting production by sales, and solves the problems of warehouse expansion and production halt caused by full-load production of surface mines. The method comprises the following steps: firstly, collecting and calibrating data such as inventory, sales volume and productivity; then, based on the inventory available capacity and the sales volume predicted through a quadratic polynomial regression model, combining productivity constraints, and taking a minimum value to determine the maximum production plan quantity of single-specification finished products; under the constraint of the finished product proportion, optimizing the production capacity proportion of each specification by using a linear programming algorithm, and calculating the total planned quantity and the production duration; and finally, issuing a production instruction and performing real-time monitoring, and when the inventory is close to the upper limit or the sales volume suddenly exceeds the threshold value, re-triggering the plan calculation process to form a closed loop of data acquisition, calculation, execution and feedback. The scheme can dynamically match the sales volume and the productivity, avoids warehouse expansion, improves the resource utilization rate, reduces the production cost, and is suitable for production scenes of multi-specification finished products.

The invention discloses a variable-frequency driving access system suitable for a coal mine high-voltage power grid and a control method, and particularly relates to the technical field of access control of a coal mine power system. The method comprises the steps of firstly obtaining a voltage grade, a power grid topological structure, a grounding mode, system impedance, a power grid load change curve and a bus transient stability coefficient of a coal mine high-voltage power grid, and constructing a dynamic stability model; the output voltage, the current harmonic component, the impact current, the voltage change rate and the carrier frequency of the variable frequency driving equipment are obtained and serve as disturbance input for multi-scene simulation, and a bus response index is extracted; a polynomial regression model is utilized to predict an interference level coefficient, the interference level coefficient is compared with a safety threshold value, a target access parameter group is screened, an access path and a buffer modulation strategy are determined, a control instruction is generated and transmitted to a main control unit, and stable access of equipment in a high-voltage power grid is achieved; according to the invention, the adaptability and safety of frequency conversion equipment access are improved, and the method is suitable for dynamic control requirements in a complex power grid environment.

The invention discloses a power consumption management method, device and equipment for a desulfurization plant, a storage medium and a program product. The method comprises the following steps: acquiring current uncontrollable working condition parameters influencing power consumption of a target desulfurization plant and an actual power consumption rate; according to the uncontrollable working condition parameters, predicting by using the trained polynomial regression model to obtain a theoretical power consumption rate; and determining the power consumption rate deviation between the actual power consumption rate and the theoretical power consumption rate, so as to evaluate the regulation and control effect based on the controllable working condition parameters influencing the power consumption of the target desulfurization plant according to the power consumption rate deviation. The power consumption rate is predicted by using the polynomial regression model, a nonlinear relation can be accurately described, so that the prediction precision is improved, meanwhile, the regulation and control effect of the controllable working condition parameters is evaluated through the deviation between the theoretical power consumption rate obtained through prediction based on the uncontrollable working condition parameters and the corresponding actual power consumption rate, uncontrollable factors are well stripped, and the method is suitable for large-scale popularization and application. Therefore, the actual regulation and control capability of the operation team on the controllable factors can be evaluated more objectively and fairly.

The present invention provides a deep learning-based traffic flow and charging station load forecasting method, which aims to improve the accuracy of traffic flow forecasting and accurately predict the load of electric vehicle charging stations. The method includes the following steps: First, a spatial-temporal graph convolutional network model based on the attention mechanism is used to predict future traffic flow and traffic speed. Secondly, based on the probability distribution of the vehicle's daily travel distance, a hybrid model and an expectation-maximization algorithm are used to calculate the charging demand of electric vehicles, and the energy consumption of electric vehicles is predicted using a polynomial regression model. Finally, based on the driving behavior of electric vehicles, an electric vehicle queuing model is constructed. The model comprehensively considers behaviors such as charging station capacity limitations, long waiting times, and impatience to leave, and calculates the load forecast of the charging station using a random Markov chain analysis method. This method can effectively capture the spatiotemporal characteristics of traffic flow and electric vehicle charging demand, and improve the accuracy of charging station load forecasting.

The present invention provides a method for optimizing process parameters for selective laser melting (SLM) based on statistical analysis, which belongs to the field of additive manufacturing. More specifically, the method includes the following steps: preparing samples using a selective laser melting process and performing performance tests on them to obtain experimental data on the performance corresponding to different process parameters; establishing polynomial regression models for different performances based on the experimental data; optimizing the polynomial regression models for each performance using a stepwise regression analysis method, and adjusting the thresholds in the stepwise regression analysis method using a one-dimensional search method to obtain prediction models for different performances, which are then used to optimize process parameters. The present invention provides a fast and accurate method for predicting the performance of SLM products, which achieves rapid optimization of process parameters for target performance products manufactured by SLM. It can quickly, efficiently, and cost-effectively explore the optimal SLM process parameter range, effectively reducing scrap rates, improving the quality of formed parts, and increasing production efficiency.

This invention relates to an automatic focusing method for laser galvanometers based on a polynomial regression algorithm, including variable definition, data collection of sample data within different power ranges from 10% to 80%, and data preprocessing; employing a piecewise polynomial regression model, and adding an L2 regularization term to the model to prevent overfitting; and minimizing the observed value D. i The model parameters are estimated using the sum of squared errors between the predicted and actual values. The model is used to predict values ​​in real time. When the absolute value of the deviation ∈ between the predicted and actual roundness values ​​D exceeds the threshold Th, the galvanometer height H or laser power P is dynamically adjusted to make D approach the target value D. target =0; the system is based on the optimal height H * The Z-axis position of the galvanometer is adjusted in real time to focus the laser beam at the optimal position; the deviation between the actual value D and the predicted value ∈ is periodically collected to update the regression model coefficients. This invention enables precise focusing of the laser beam, improving product consistency and yield.