A multi-scenario user-side automobile spraying effect simulation system and method

Abstract

The present invention relates to the technical field of automobile spraying simulation, specifically a multi-scenario user-side automobile spraying effect simulation system and method, the system includes a performance monitoring module, a resource configuration module, a spraying preference analysis module and a rendering quality optimization module. The present invention, by analyzing user behavior data and device performance, realizes adaptive adjustment of resource configuration and real-time rendering parameter optimization, improves the dynamic response capability and service quality of the simulation system, adopts the ARIMA model to predict the user's future operation trend, not only adapts to the user's immediate needs, but also predicts future preference changes, making resource allocation more accurate and reasonable, dynamically adjusting rendering parameters such as resolution, shadow quality and texture details, ensuring rapid response and providing high-quality visual effects under various operating conditions, not only improving user experience, but also enhancing the system's energy efficiency ratio, ensuring that high-quality rendering performance can be maintained in different environments.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile spray painting simulation, and in particular to a multi-scenario user-side automobile spray painting effect simulation system and method. Background Art

[0002] The field of automobile spray painting simulation technology mainly focuses on using software and hardware tools to simulate the effects of the automobile painting process, so as to perform visual preview and parameter adjustment before actual painting. This technology includes color selection, paint surface effects, gloss simulation and other aspects. Through advanced graphics rendering and color matching algorithms, it can accurately display the painting effects under different lighting and angles, helping designers and consumers make judgments and choices.

[0003] Among them, the multi-scenario user-side car painting effect simulation system allows users to simulate the effect of car painting under a variety of preset or customized environmental settings. The system is usually deployed on user-side devices, such as personal computers or mobile devices. Users can interactively select different colors and effects, and the system renders the appearance of the car in various scenarios (such as different lighting conditions and backgrounds) in real time. Its main purpose is to provide consumers with an intuitive tool to preview and personalize car color options, thereby increasing car purchase willingness and satisfaction.

[0004] Existing technologies have obvious limitations in adapting to users' personalized needs and managing resources. Although basic visual previews and parameter adjustments can be performed, they lack the ability to predict dynamic changes in user behavior, resulting in an inability to provide sufficiently targeted services and optimized experiences. The rendering parameters of existing systems are mostly static settings, and do not support automatic adjustment based on real-time user operations. This leads to a lack of consistency in user experience in changing usage scenarios, and resource allocation is fixed and not fully utilized, especially when dealing with complex or resource-intensive spray simulation tasks, resulting in inefficiency and waste of resources, limiting the scope of application of the system, and failing to effectively meet the needs for high efficiency and high adaptability, affecting consumer satisfaction and purchasing intention. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a multi-scenario user-side automobile spraying effect simulation system and method.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: a multi-scenario user-side automobile spraying effect simulation system, the system comprising:

[0007] The performance monitoring module collects CPU, GPU and memory usage based on real-time monitoring data of user devices, calculates the real-time changes of each indicator, evaluates resource requirements based on the calculation results, and adjusts the resource allocation process to obtain resource demand indicators;

[0008] The resource configuration module analyzes the user's regular usage pattern based on the resource demand indicators, predicts the resource demand of the spray simulation, adjusts the CPU and GPU resource allocation, optimizes the memory usage, and updates the resource configuration of the user's device to generate the device configuration optimization state;

[0009] The spray preference analysis module refines the user's color and light and shadow preferences based on the device configuration optimization status and the user's spray operation data, and uses the ARIMA model to predict the user's future operation trend and generate user preference preset information;

[0010] The rendering quality optimization module renders the car painting effect in real time based on the user preference preset information to match the visual effect expected by the user, and adjusts the rendering parameters, including resolution, shadow quality and texture details, to optimize resource usage and response speed to obtain the rendering output result.

[0011] The present invention is improved in that the calculation steps of the real-time changes are specifically as follows:

[0012] Based on the real-time monitoring data of the user's device, the CPU, GPU and memory usage are continuously collected and updated once per second using the formula:

[0013] D(t)={cpu(t),gpu(t),mem(t)}

[0014] Get the real-time monitoring data set, where D(t) is the data set at time point t, cpu(t) is the CPU usage, gpu(t) is the GPU usage, mem(t) is the memory usage, and t is the time point;

[0015] Based on the real-time monitoring data set, the formula is adopted:

[0016] ΔD(t)=D(t)-D(t-1)

[0017] Calculate the usage change of each resource ΔD(t), where D(t) is the data set at time point t, and D(t-1) is the data set at time point t-1;

[0018] Based on the usage change of each resource, the resource change index is calculated using the formula:

[0019] V(t)=k1·|Δcpu(t)|+k2·|Δgpu(t)|+k3·|Δmem(t)|

[0020] The resource change index V(t) is obtained, where Δcpu(t) is the CPU change, Δgpu(t) is the GPU change, Δmem(t) is the memory change, and k1, k2, and k3 are the corresponding weight factors of the resource items.

[0021] The present invention is improved in that the step of obtaining the resource demand indicator is specifically as follows:

[0022] Based on the calculation results, the resource demand is evaluated according to the real-time utilization rate of the resources, using the formula:

[0023]

[0024] Generate the estimated resource requirement RT, where aw h represents the influence coefficient of resource h, bw h Represents the weight adjustment index of resource h, which is used to adjust the demand impact of different resource types, U h is the real-time utilization rate;

[0025] Based on the resource requirements after evaluation, adjust the resource allocation process, using the formula:

[0026] TA=f0(RT,ST)

[0027] The resource demand index TA is obtained, where RT is the resource demand after evaluation, ST is the initial resource configuration state, and f0 is the adjustment function.

[0028] The present invention is improved in that the analysis steps of the conventional usage mode are specifically as follows:

[0029] Based on the resource demand indicators, calculate the key variables that affect user usage patterns using the formula:

[0030]

[0031] Get the key impact factor A w , where a w 、b w and c w is a model parameter used to adjust the decomposition method of the resource demand indicator TA, where TA is the resource demand indicator;

[0032] Based on the key influencing factors and combined with the user's historical usage data, statistical correlation analysis is performed using the formula:

[0033] P w =k w ·A w 2 +m w ·D w (t)+n w

[0034] Get the user's regular usage pattern P w , where k w 、m w and n wis the model parameter, A w is the key influencing factor, D w (t) is the user's historical usage data;

[0035] Based on the regular usage patterns of the users, a nonlinear model is used to predict future resource requirements, and the formula is:

[0036]

[0037] Get the predicted resource demand R future , where ξ, ω, ψ and μ are the adjustment parameters of the model, which are used to match the changing user usage patterns, and Pw is the user's regular usage pattern.

[0038] The present invention is improved in that the step of obtaining the device configuration optimization state is specifically as follows:

[0039] Based on the regular usage pattern, evaluate the existing equipment configuration and identify the excess in resource configuration using the formula:

[0040] C current =β g ·∑(P k -P avg ) 2

[0041] Get the difference C of the current configuration state current , where β g is the adjustment factor used to standardize configuration differences, P k Single usage data point, P avg is the average usage pattern data;

[0042] Based on the difference between the current configuration states, the required configuration adjustment is calculated using the formula:

[0043]

[0044] Generate adjusted device configuration C new , where R future is the predicted resource demand, α g and γ g is a model parameter used to adjust the configuration, C current is the difference of the current configuration state;

[0045] Execute the adjusted device configuration and monitor the device performance using the formula:

[0046]

[0047] Get the device configuration optimization status C optimized , where δ gand ∈ g is the adjusted configuration performance coefficient, which is used to evaluate the configuration execution effect. new is the adjusted device configuration.

[0048] The present invention is improved in that the steps of refining the color and light and shadow preference are specifically as follows:

[0049] Based on the optimized state of the equipment configuration, the user's spraying operation data is obtained using the formula:

[0050]

[0051] and

[0052]

[0053] Generate color using frequency F color and light and shadow usage frequency F shadow , where w i is a weight factor, assigned based on the frequency of operation or the strength of user preference, c i is the color value used in the i-th operation, s i is the light intensity used in the i-th operation, n is the number of operations;

[0054] Based on the frequency of use of the colors and shades, a non-linear transformation is applied to optimize the data's sensitivity to user preferences, using the formula:

[0055] P color =log(1+F color )

[0056] and

[0057] P shadow =log(1+F shadow )

[0058] Generate refined color preference P color and light and shadow preference P shadow , where F color is the frequency of use of the color, F shadow is the frequency of use of light and shadow;

[0059] Based on the refined color preference and light and shadow preference, combined with the user operation context, the formula is adopted:

[0060] H=α·P color +β·P shadow

[0061] Get the user's color and light preference configuration H, where α and β are the coefficients for adjusting the color and light effects, which are used to balance the weights of the two preferences in the user experience, Pcolor is the refined color preference, P shadow It is the refined light and shadow preference.

[0062] The present invention is improved in that the step of obtaining the user preference preset information is specifically as follows:

[0063] Based on the user's color and light preference, the user's operation frequency and preference strength are integrated and the formula is adopted:

[0064] WF=f a ×p b ÷(Δt+1)

[0065] Calculate the user preference weight WF, where f is the usage frequency, p is the preference strength, Δt is the time interval, and a and b are adjustment coefficients used to adjust the impact of frequency and preference strength on the weight;

[0066] Based on the user preference weights, combined with the ARIMA model, the future preference trend is predicted using the formula:

[0067] P u =ARIMA(WF, d, E, q, s)

[0068] Generate the prediction result Pu, where WF is the user preference weight, d, E and q are the ARIMA model parameters, s It is a seasonal cycle;

[0069] Based on the prediction results, combined with the current device configuration, the device's user interface and recommended information are adjusted to match the user's predicted preferences, using the formula:

[0070] S u =ConfigAdjust(P u , C, r)

[0071] Get user preference preset information S u , where P u is the prediction result, C is the device configuration parameter, and r is the adjustment coefficient, which is adjusted according to device performance or user feedback.

[0072] The present invention is improved in that the step of obtaining the rendering output result is specifically as follows:

[0073] Based on the user preference preset information, the user's spraying operation data is analyzed to identify the operation behavior trend, using the formula:

[0074]

[0075] Get behavioral trend data T p ,in, is the weight of the ith operation, oi is the amount of data or eigenvalue of the ith operation, cu i is the adjustment index of the i-th data, and n is the number of operations;

[0076] Based on the behavior trend data, the visual effect parameters of the rendering engine are matched using the formula:

[0077]

[0078] Generate visual effect data V p ,in, is the zth visual effect, is the adjustment parameter of the zth visual effect, Tp is the behavior trend data, n v is the total number of visual effects;

[0079] Based on the visual effect data, the rendering parameters including resolution, shadow quality and texture detail are optimized using the formula:

[0080]

[0081] Get the rendering output result R p , where V p is the visual effect data, r o 、s o and d o is the resolution, shadow quality and texture detail adjustment factor, and m is the environment adaptability parameter.

[0082] A multi-scenario user-side automobile spraying effect simulation method, the multi-scenario user-side automobile spraying effect simulation method is executed based on the multi-scenario user-side automobile spraying effect simulation system, comprising the following steps:

[0083] S1: Based on the real-time monitoring data of user devices, collect CPU, GPU and memory usage, calculate the change rate of each indicator, evaluate resource demand, adjust resource allocation, and generate resource demand indicators;

[0084] S2: Based on the resource demand indicators, analyze the user's regular usage pattern, predict the resources required for the spraying simulation, adjust the resource configuration of the CPU and GPU, optimize the memory usage, and generate the device configuration optimization state;

[0085] S3: Based on the optimized state of the equipment configuration, analyze the user's spraying operation data, refine the color and light and shadow preferences, apply the ARIMA model to predict future operation trends, and generate user preference preset information;

[0086] S4: based on the user preference preset information, rendering the car spraying effect in real time, adjusting rendering parameters, including resolution and shadow quality, to match the visual effect expected by the user, and generating a preliminary rendering result;

[0087] S5: Based on the preliminary rendering result, the rendering texture details are refined, resource usage and response speed are optimized, and a rendering output result is generated.

[0088] Compared with the prior art, the advantages and positive effects of the present invention are:

[0089] In the present invention, by analyzing user behavior data and device performance, adaptive adjustment of resource allocation and real-time rendering parameter optimization are achieved, the dynamic response capability and service quality of the simulation system are improved, and the ARIMA model is used to predict the user's future operation trend, which not only adapts to the user's immediate needs, but also predicts future preference changes, making resource allocation more accurate and reasonable, and dynamically adjusting rendering parameters such as resolution, shadow quality and texture details, ensuring rapid response and providing high-quality visual effects under various operating conditions, which not only improves the user experience, but also enhances the energy efficiency of the system, ensuring that high-quality rendering performance can be maintained in different environments, and improving users' car purchase willingness and satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0090] Figure 1 A module diagram of a multi-scenario user-side automobile spraying effect simulation system is proposed for the present invention;

[0091] Figure 2 is a calculation flow chart of real-time changes in the present invention;

[0092] Figure 3 A flowchart for obtaining resource demand indicators in the present invention;

[0093] Figure 4 is an analysis flow chart of the conventional usage mode in the present invention;

[0094] Figure 5 A flowchart for obtaining the optimized state of device configuration in the present invention;

[0095] Figure 6 A detailed flow chart of color and light and shadow preferences in the present invention;

[0096] Figure 7 A flowchart for obtaining user preference preset information in the present invention;

[0097] Figure 8 The figure is a flowchart for obtaining the rendering output result in the present invention. DETAILED DESCRIPTION

[0098] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0099] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating positions or positional relationships, are based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0100] Example

[0101] See also Figure 1 The present invention provides a technical solution: a multi-scenario user-side automobile spraying effect simulation system includes:

[0102] The performance monitoring module collects CPU, GPU and memory usage based on real-time monitoring data of user devices, calculates the real-time changes of each indicator, evaluates resource requirements based on the calculation results, and adjusts the resource allocation process to obtain resource demand indicators;

[0103] The resource configuration module analyzes the user's regular usage patterns based on resource demand indicators, predicts the resource demand of the spray simulation, adjusts the CPU and GPU resource allocation, optimizes memory usage, and updates the resource configuration of the user's device to generate the device configuration optimization status;

[0104] The spray preference analysis module refines the user's color and light and shadow preferences based on the device configuration optimization status and the user's spray operation data. It uses the ARIMA model to predict the user's future operation trends and generate user preference preset information.

[0105] The rendering quality optimization module renders the car painting effect in real time based on the user's preference preset information to match the user's expected visual effect, and adjusts the rendering parameters, including resolution, shadow quality and texture details, to optimize resource usage and response speed to obtain rendering output results.

[0106] Resource demand indicators include performance demand assessment results, resource utilization, and priority adjustment results. Equipment configuration optimization status includes configuration update rate, resource allocation process, and memory utilization optimization results. User preference preset information includes color preference mode, spray selection information, and light and shadow interaction preference information. Rendering output results include image resolution, rendering speed, and user interaction response delay.

[0107] See also Figure 2 , the calculation steps of real-time changes are as follows:

[0108] Based on the real-time monitoring data of the user's device, the CPU, GPU and memory usage are continuously collected and updated once per second using the formula:

[0109] D(t)={cpu(t),gpu(t),mem(t)}

[0110] Get the real-time monitoring data set, where D(t) is the data set at time point t, cpu(t) is the CPU usage, gpu(t) is the GPU usage, mem(t) is the memory usage, and t is the time point;

[0111] Based on the real-time monitoring data set, the formula is:

[0112] ΔD(t)=D(t)-D(t-1)

[0113] Calculate the usage change of each resource ΔD(t), where D(t) is the data set at time point t, and D(t-1) is the data set at time point t-1;

[0114] Based on the change in usage of each resource, calculate the resource change index using the formula:

[0115] V(t)=k1·|Δcpu(t)|+k2·|Δgpu(t)|+k3·|Δmem(t)|

[0116] The resource change index V(t) is obtained, where Δcpu(t) is the CPU change, Δgpu(t) is the GPU change, Δmem(t) is the memory change, and k1, k2, and k3 are the corresponding weight factors of the resource items.

[0117] Assume that at time t, the CPU usage is 50%, the GPU usage is 40%, and the memory usage is 8GB, then:

[0118] cpu(t)=50%

[0119] gpu(t)=40%

[0120] mem(t)=8GB

[0121] Insert the values ​​into the formula:

[0122] D(t)={50%, 40%, 8GB}

[0123] Assume that at time t-1, the CPU usage is 45%, the GPU usage is 35%, and the memory usage is 7.5GB.

[0124] D(t-1)={45%, 35%, 7.5GB}

[0125] Calculate the change:

[0126] ΔD(t)={50%-45%, 40%-35%, 8GB-7.5GB}

[0127] ΔD(t)={5%, 5%, 0.5GB}

[0128] Assuming weight factors k1 = 0.5, k2 = 0.3, k3 = 0.2, then:

[0129] V(t)=0.5·|5|+0.3·|5|+0.2·|0.5GB|

[0130] V(t)=0.5·5+0.3·5+0.2·0.5

[0131] V(t)=2.5+1.5+0.1

[0132] V(t)=4.1

[0133] The calculation result V(t)=4.1 means that after considering various resource weight factors, the resource change index is 4.1, indicating that there are certain resource fluctuations that need attention.

[0134] See also Figure 3 , the specific steps for obtaining resource demand indicators are:

[0135] Based on the calculation results, the resource demand is evaluated according to the real-time utilization rate of resources, using the formula:

[0136]

[0137] Generate the estimated resource requirement RT, where aw h represents the influence coefficient of resource h, bw h Represents the weight adjustment index of resource h, which is used to adjust the demand impact of different resource types, U h is the real-time utilization rate;

[0138] Based on the assessed resource requirements, adjust the resource allocation process using the formula:

[0139] TA=f0(RT,ST)

[0140] The resource demand index TA is obtained, where RT is the evaluated resource demand, ST is the initial resource configuration state, and f0 is the adjustment function, which dynamically optimizes resource allocation according to resource demand and current configuration.

[0141] Assume the following device configuration and real-time data:

[0142] CPU usage: 70%, capacity: 100% (U CPU =0.70)

[0143] GPU usage: 50%, capacity: 100% (U GPU =0.50)

[0144] Memory usage: 80%, capacity: 100% (U 内存 =0.80)

[0145] Assuming the coefficient aw h and bw h for:

[0146] aW CPU =1.5, aw GPU =1.2, aw 内存 =1.3

[0147] bw CPU =2, bw GPU =1,bw 内存 =1.5

[0148] Substituting the values ​​into the formula:

[0149] RT=1.5·0.70 2 +1.2 0.50 1 +1.3 0.80 1 . 5

[0150] RT=1.5·0.49+1.2·0.50+1.3·0.717

[0151] RT=0.735+0.60+0.9321

[0152] RT=2.2671

[0153] The value RT=2.2671 means that according to the current resource utilization rate and the weight of each resource's impact on the total demand, the resource demand level of the entire system is 2.2671, which can be used to evaluate whether it is necessary to increase resources or adjust the existing configuration.

[0154] Let ST = 1.0 (completely matching the requirements);

[0155] Substitute RT = 2.2671 into the formula and set the adjustment function f to be the ratio of resource demand to system configuration, that is:

[0156]

[0157] TA=2.2671

[0158] The adjusted resource allocation index of 2.2671 indicates the degree of resource allocation that needs to be adjusted under the current resource configuration, and a value higher than 1 indicates the need to increase resource input or optimize existing resource allocation to meet demand.

[0159] See also Figure 4 , the analysis steps of the general usage pattern are as follows:

[0160] Based on the resource demand indicators, calculate the key variables that affect user usage patterns using the formula:

[0161]

[0162] Get the key impact factor A w , where a w , b w and c w It is a model parameter used to adjust the decomposition method of the resource demand indicator TA and to analyze the impact on user behavior. TA is the resource demand indicator;

[0163] Based on the key influencing factors and combined with the user's historical usage data, statistical correlation analysis is performed using the formula:

[0164] P w =k w ·A w 2 +m w ·D w (t)+n w

[0165] Get the user's regular usage pattern P w , where kw, m w and n w is the model parameter, A w is the key influencing factor, D w (t) is the user's historical usage data;

[0166] Based on the user's regular usage pattern, a nonlinear model is used to predict future resource requirements. The formula is:

[0167]

[0168] Get the predicted resource demand R future, where ξ, ω, ψ and μ are adjustment parameters of the model to match the changing user usage patterns, P w It is the normal usage pattern of users.

[0169] Assume the following data:

[0170] TA = 100;

[0171] a w =0.5;

[0172] b w =2;

[0173] c w =1.

[0174] Do the calculation (using natural logarithms):

[0175]

[0176] log(TA)=log(100)=2

[0177] A w =0.5·10000+1·2=5002

[0178] Result A w =5002 represents the comprehensive score of key influencing factors calculated by resource demand indicators, which is used to further analyze user usage patterns.

[0179] Assume the following data:

[0180] A w =5002;

[0181] D w (t) = 5 (assuming the average user uses the app for 5 hours a day);

[0182] k w =0.01;

[0183] m w =3;

[0184] n w =50.

[0185] Calculation process:

[0186] A w 2 =5002 2 =25020004

[0187] k w ·A w 2=0.01·25020004=250200.04

[0188] P w =250200.04+3·5+50=250215.04

[0189] Results w =250215.04 represents the quantitative representation of user usage patterns based on the analysis of historical usage data and resource demand indicators.

[0190] Assume the following data:

[0191] P w =250215.04;

[0192] ξ=0.000001;

[0193] ω=1;

[0194] ψ=0.5;

[0195] μ=0.00001.

[0196] Calculation process:

[0197] p w 3 =250215.04 3 ≈1.5632×10 19

[0198]

[0199] Results future =2.199×10 18 It represents the predicted future resource demand based on the user's usage pattern and shows very large values, indicating that the model estimation needs further adjustment or parameter optimization to reflect the actual situation.

[0200] See also Figure 5 , the specific steps for obtaining the device configuration optimization status are:

[0201] Based on regular usage patterns, evaluate existing equipment configurations and identify excess resources in the configuration using the formula:

[0202] C current =β g ·∑(P k -P avg ) 2

[0203] Get the difference C of the current configuration state current , where β g is the adjustment factor used to standardize configuration differences, Pk Single usage data point, P avg is the average usage pattern data;

[0204] Based on the difference between the current configuration states, calculate the required configuration adjustments using the formula:

[0205]

[0206] Generate adjusted device configuration C new , where R future is the predicted resource demand, α g and γ g is a model parameter used to adjust the configuration, C current is the difference of the current configuration state;

[0207] Execute the adjusted device configuration and monitor the device performance using the formula:

[0208]

[0209] Get the device configuration optimization status C optimized , where δ g and ∈ g is the adjusted configuration performance coefficient, which is used to evaluate the configuration execution effect. new is the adjusted device configuration.

[0210] Assume there are three resource usage data points:

[0211] P k = [20, 30, 25];

[0212]

[0213] Compute the sum of squared deviations for each point:

[0214] ∑(P k -P avg ) 2 =(20-25) 2 +(30-25) 2 +(25-25) 2 =25+25+0=50

[0215] Select Beta g = 0.1 (assumed value) to standardize the difference:

[0216] C current =0.1·50=5

[0217] Result C current =5 represents the degree of deviation between the current configuration and the ideal state, and a smaller value means that the configuration is closer to the requirement.

[0218] Assume R future =10;

[0219] C current =5;

[0220] αg=2;

[0221] Y g =1.

[0222] Calculate new configuration requirements:

[0223]

[0224] Result C new =10 means that in order to meet future resource requirements, the configuration needs to be increased to a level of 10.

[0225] Assume the following data:

[0226] C new =10;

[0227] δg=0.5;

[0228] ∈g=2.

[0229] Calculate the optimized configuration state:

[0230]

[0231] Result C optimized =4.17 represents the final optimization status index after implementing the new configuration and performing performance monitoring. A lower value indicates that the configuration effect is significant and better matches the demand.

[0232] See also Figure 6 , the specific steps for refining color and light and shadow preferences are:

[0233] Based on the optimized state of equipment configuration, the user's spraying operation data is obtained using the formula:

[0234]

[0235] and

[0236]

[0237] Generate color using frequency F color and light and shadow usage frequency F shadow , where w i is a weight factor, assigned based on the frequency of operation or the strength of user preference, c i is the color value used in the i-th operation, s iis the light intensity used in the i-th operation, n is the number of operations;

[0238] Based on the frequency of use of colors and lighting, a nonlinear transformation is applied to optimize the data's sensitivity to user preferences using the formula:

[0239] P color =log(1+F color )

[0240] and

[0241] P shadow =log(1+F shadow )

[0242] Generate refined color preference P color and light and shadow preference P shadow , where F color is the frequency of use of the color, F shadow is the frequency of use of light and shadow;

[0243] Based on the refined color preference and light and shadow preference, combined with the user's operating context, the formula is adopted:

[0244] H=α·P color +β·P shadow

[0245] Get the user's color and light preference configuration H, where α and β are the coefficients for adjusting the color and light effects, which are used to balance the weights of the two preferences in the user experience, P color is the refined color preference, P shadow It is the refined light and shadow preference.

[0246] Suppose a user uses different color values ​​and light and shadow intensities in three operations, and assume that the weight of each operation is the same, that is:

[0247] W1=w2=w3=1;

[0248] Color values ​​c1=10, c2=20, c3=30;

[0249] Light and shadow values ​​s1=40, s2=50, s3=60.

[0250] Calculate the color usage frequency F color :

[0251]

[0252] Calculate the light and shadow intensity frequency F shadow :

[0253]

[0254] The results show an average color value of 20 and an average light and shadow value of 50.

[0255] Using the above result F color =20 and F shadow =50.

[0256] P color =log(1+20)=log(21)≈3.04

[0257] P shadow =log(1+50)=log(51)≈3.93

[0258] This means that the logarithmically transformed color preference index is approximately 3.04, and the light and shadow preference index is approximately 3.93.

[0259] Assume α=0.6 and β=0.4.

[0260] Calculation process:

[0261] H=0.6·3.04+0.4·3.93

[0262] H=1.824+1.572

[0263] H=3.396

[0264] It indicates that the user's comprehensive color and light preference index is 3.396.

[0265] See also Figure 7 , the specific steps for obtaining user preference preset information are:

[0266] Based on the user's color and light preference, the user's operation frequency and preference intensity are integrated, and the formula is used:

[0267] WF=f a ×p b ÷(Δt+1)

[0268] Calculate the user preference weight WF, where f is the usage frequency, p is the preference strength, Δt is the time interval, and a and b are adjustment coefficients used to adjust the impact of frequency and preference strength on the weight;

[0269] Based on the user preference weights, combined with the ARIMA model, the future preference trend is predicted using the formula:

[0270] P u =ARIMA(WF, d, E, q, s)

[0271] Generate the prediction result Pu, where WF is the user preference weight, d, E and q are the ARIMA model parameters, s It is a seasonal cycle;

[0272] Based on the prediction results, combined with the current device configuration, the device's user interface and recommended information are adjusted to match the user's predicted preferences using the formula:

[0273] S u =ConfigAdjust(P u , C, r)

[0274] Get user preference preset information S u , where P u is the prediction result, C is the device configuration parameter, and r is the adjustment coefficient, which is adjusted according to device performance or user feedback.

[0275] Assume that a user selects blue with a frequency of f=30 times, a preference strength of p=8 points, a time interval Δt=10 days since the last selection, and adjustment coefficients a=1.2 and b=0.8.

[0276] Calculate f a :

[0277] 30 1.2 ≈46.77

[0278] Calculate p b :

[0279] 8 0.8 ≈6.02

[0280] Calculate Δt+1:

[0281] 10+1=11

[0282] Calculate WF:

[0283]

[0284] The result WF=25.63 represents the user's comprehensive preference weight for blue.

[0285] Assumed parameters:

[0286] d = 1;

[0287] E=1:

[0288] q=0;

[0289] s=December.

[0290] Assuming that the weight data WF is calculated from the previous step, use the parameters to predict the preference trend for the next month.

[0291] Assume C = 50;

[0292] r=0.5.

[0293] If P u The forecast is a 20% increase, then:

[0294] S u =50+20×0.5=60

[0295] Results u =60 indicates the new configuration value of the device.

[0296] See also Figure 8 , the specific steps for obtaining the rendering output results are:

[0297] Based on the user's preference preset information, analyze the user's spraying operation data, identify the operation behavior trend, and use the formula:

[0298]

[0299] Get behavioral trend data T p ,in, is the weight of the ith operation, o i is the amount of data or eigenvalue of the ith operation, cu i is the adjustment index of the i-th operation data, and n is the number of operations;

[0300] Based on the behavior trend data, the visual effect parameters of the rendering engine are matched using the formula:

[0301]

[0302] Generate visual effect data V p ,in, is the zth visual effect, is the adjustment parameter of the zth visual effect, Tp is the behavior trend data, n v is the total number of visual effects;

[0303] Based on the visual effect data, the rendering parameters including resolution, shadow quality and texture details are optimized using the formula:

[0304]

[0305] Get the rendering output result R p , where V p is the visual effect data, r o 、s o and d o is the resolution, shadow quality and texture detail adjustment factor, and m is the environment adaptability parameter.

[0306] Assume there are three operations:

[0307] o = [10, 20, 15];

[0308] a p = [0.5, 0.3, 0.2];

[0309] cu=[1,2,1.5].

[0310] Calculate the contribution of each operation:

[0311] 0.5·10 1 =5

[0312] 0.3·20 2 =120

[0313] 0.2·15 1.5 ≈8.66

[0314] Accumulate the contribution of all operations to get T p :

[0315] T p =5+120+8.66=133.66

[0316] Results p =133.66 represents the weighted sum of user behaviors, where both the frequency and intensity of operations are taken into account.

[0317] Suppose there are three data:

[0318] V p =[50,70,90];

[0319] k p =[0.1,0.2,0.1];

[0320] T p =133.66.

[0321] Calculate the adjustment value for each visual effect parameter:

[0322]

[0323] Calculate the average matching visual effect V p :

[0324]

[0325] Results p =8269 represents the average visual effect parameter adjusted according to user behavior.

[0326] Assume the following data:

[0327] V p =8269;

[0328] ro =1.5;

[0329] s o =1.2;

[0330] d o =0.8;

[0331] m=0.5.

[0332] Calculation process:

[0333]

[0334] R p ≈7938.24

[0335] Results n ≈7938.24 represents the rendering effect, taking into account the adjustment of visual parameters and environmental factors.

[0336] A method for simulating automobile spraying effects on a user side in multiple scenarios, comprising the following steps:

[0337] S1: Based on the real-time monitoring data of user devices, collect CPU, GPU and memory usage, calculate the change rate of each indicator, evaluate resource demand, adjust resource allocation, and generate resource demand indicators;

[0338] S2: Based on the resource demand indicators, analyze the user's regular usage patterns, predict the resources required for spray simulation, adjust the resource configuration of CPU and GPU, optimize memory usage, and generate device configuration optimization status;

[0339] S3: Based on the equipment configuration optimization status, analyze the user's spraying operation data, refine the color and light and shadow preferences, apply the ARIMA model to predict future operation trends, and generate user preference preset information;

[0340] S4: Based on the user's preferred preset information, the car spraying effect is rendered in real time, and the rendering parameters, including resolution and shadow quality, are adjusted to match the user's expected visual effect, and a preliminary rendering result is generated;

[0341] S5: Based on the preliminary rendering results, refine the rendering texture details, optimize resource usage and response speed, and generate rendering output results.

[0342] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A multi-scenario user-side automobile spraying effect simulation system, characterized in that: The system comprises: The performance monitoring module collects CPU, GPU and memory usage based on real-time monitoring data of user devices, calculates the real-time changes of each indicator, evaluates resource requirements based on the calculation results, and adjusts the resource allocation process to obtain resource demand indicators; The resource configuration module analyzes the user's regular usage pattern based on the resource demand indicators, predicts the resource demand of the spray simulation, adjusts the CPU and GPU resource allocation, optimizes the memory usage, and updates the resource configuration of the user's device to generate the device configuration optimization state; The spray preference analysis module refines the user's color and light and shadow preferences based on the device configuration optimization status and the user's spray operation data, and uses the ARIMA model to predict the user's future operation trend and generate user preference preset information; The rendering quality optimization module renders the car painting effect in real time based on the user preference preset information to match the visual effect expected by the user, and adjusts the rendering parameters, including resolution, shadow quality and texture details, to optimize resource usage and response speed, and obtain the rendering output result; The steps for refining the color and light and shadow preferences are as follows: Based on the optimized state of the equipment configuration, the user's spraying operation data is obtained using the formula: and Generate color using frequency F color and light and shadow usage frequency F shadow , where w i is a weight factor, assigned based on the frequency of operation or the strength of user preference, c i is the color value used in the i-th operation, s i is the light intensity used in the i-th operation, n is the number of operations; Based on the frequency of use of the colors and shades, a non-linear transformation is applied to optimize the data's sensitivity to user preferences, using the formula: P color =log(1+F color ) P shadow =log(1+F shadow ) Generate refined color preference P color and light and shadow preference P shadow , where F color is the frequency of use of the color, F shadow is the frequency of use of light and shadow; Based on the refined color preference and light and shadow preference, combined with the user operation context, the formula is adopted: H=α·P color +β·P shadow Get the user's color and light preference configuration H, where α and β are the coefficients for adjusting the color and light effects, which are used to balance the weights of the two preferences in the user experience, P color is the refined color preference, P shadow It is the refined light and shadow preference; The steps for obtaining the user preference preset information are specifically as follows: Based on the user's color and light preference, the user's operation frequency and preference strength are integrated and the formula is adopted: WF=f a ×p b ÷(Δt+1) Calculate the user preference weight WF, where f is the usage frequency, p is the preference strength, Δt is the time interval, and a and b are adjustment coefficients used to adjust the impact of frequency and preference strength on the weight; Based on the user preference weights, combined with the ARIMA model, the future preference trend is predicted using the formula: P u =ARIMA(WF,d,E,q,s) Generate prediction result P u , where WF is the user preference weight, d, E and q are ARIMA model parameters, and s is the seasonal period; Based on the prediction results, combined with the current device configuration, the device's user interface and recommended information are adjusted to match the user's predicted preferences, using the formula: S u =ConfigAdjust(P u ,C,r) Get user preference preset information S u , where P u is the prediction result, C is the device configuration parameter, and r is the adjustment coefficient, which is adjusted according to device performance or user feedback.

2. The multi-scenario user-side automobile spraying effect simulation system according to claim 1 is characterized in that: The calculation steps of the real-time change are specifically as follows: Based on the real-time monitoring data of the user's device, the CPU, GPU and memory usage are continuously collected and updated once per second using the formula: D(t)={cpu(t),gpu(t),mem(t)} A real-time monitoring data set is obtained, where D(t) is the data set at time point t, cpu(t) is the CPU usage, gpu(t) is the GPU usage, mem(t) is the memory usage, and t is the time point.

3. The multi-scenario user-side automobile spraying effect simulation system according to claim 1 is characterized in that: The steps for obtaining the resource demand indicator are specifically as follows: Based on the calculation results, the resource demand is evaluated according to the real-time utilization rate of the resources, using the formula: Generate the estimated resource requirement RT, where aw h represents the influence coefficient of resource h, bw h Represents the weight adjustment index of resource h, which is used to adjust the demand impact of different resource types, U h is the real-time utilization rate; Based on the resource requirements after evaluation, adjust the resource allocation process, using the formula: TA=f0(RT,ST) The resource demand index TA is obtained, where RT is the resource demand after evaluation, ST is the initial resource configuration state, and f0 is the adjustment function.

4. The multi-scenario user-side automobile spraying effect simulation system according to claim 1 is characterized in that: The analysis steps of the conventional usage mode are specifically as follows: Based on the resource demand indicators, calculate the key variables that affect user usage patterns using the formula: Get the key impact factor A w , where a w 、b w and c w is a model parameter used to adjust the decomposition method of the resource demand indicator TA, where TA is the resource demand indicator; Based on the key influencing factors and combined with the user's historical usage data, statistical correlation analysis is performed using the formula: P w =k w ·A w 2 +m w ·D w (t)+n w Get the user's regular usage pattern P w , where k w 、m w and n w is the model parameter, A w is the key influencing factor, D w (t) is the user's historical usage data; Based on the regular usage patterns of the users, a nonlinear model is used to predict future resource requirements, and the formula is: Get the predicted resource demand R future , where ξ, ω, ψ and μ are adjustment parameters of the model to match the changing user usage patterns, P w It is the normal usage pattern of users.

5. The multi-scenario user-side automobile spraying effect simulation system according to claim 1 is characterized in that: The steps for obtaining the device configuration optimization state are specifically as follows: Based on the regular usage pattern, evaluate the existing equipment configuration and identify the excess in resource configuration using the formula: C current G. β g ·∑(P k -P avg ) 2 Get the difference C of the current configuration state current , where β g is the adjustment factor used to standardize configuration differences, P k Single usage data point, P avg is the average usage pattern data; Based on the difference between the current configuration states, the required configuration adjustment is calculated using the formula: Generate adjusted device configuration C new , where R future is the predicted resource demand, α g and γ g is a model parameter used to adjust the configuration, C current is the difference of the current configuration state; Execute the adjusted device configuration and monitor the device performance using the formula: Get the device configuration optimization status C optimized , where δ g and ε g is the adjusted configuration performance coefficient, which is used to evaluate the configuration execution effect. new is the adjusted device configuration.

6. The multi-scenario user-side automobile spraying effect simulation system according to claim 1 is characterized in that: The steps for obtaining the rendering output result are specifically as follows: Based on the user preference preset information, the user's spraying operation data is analyzed to identify the operation behavior trend, using the formula: Get behavioral trend data T p ,in, is the weight of the ith operation, o i is the amount of data or eigenvalue of the ith operation, cu i is the adjustment index of the i-th operation data, and n is the number of operations; Based on the behavior trend data, the visual effect parameters of the rendering engine are matched using the formula: Generate visual effect data V p ,in, is the zth visual effect, is the adjustment parameter of the zth visual effect, T p is the behavioral trend data, n v is the total number of visual effects; Based on the visual effect data, the rendering parameters including resolution, shadow quality and texture detail are optimized using the formula: Get the rendering output result R p , where V p is the visual effect data, r o 、s o and d o is the resolution, shadow quality and texture detail adjustment factor, and m is the environment adaptability parameter.

7. A multi-scenario user-side automobile spraying effect simulation method, characterized in that: The multi-scenario user-side automobile spraying effect simulation system according to any one of claims 1 to 6 is implemented, comprising the following steps: Based on the real-time monitoring data of user devices, collect CPU, GPU and memory usage, calculate the change rate of each indicator, evaluate resource demand, adjust resource allocation, and generate resource demand indicators; Based on the resource demand indicators, analyze the user's regular usage patterns, predict the resources required for spraying simulation, adjust the resource configuration of the CPU and GPU, optimize memory usage, and generate a device configuration optimization state; Based on the optimized state of the equipment configuration, analyze the user's spraying operation data, refine the color and light and shadow preferences, apply the ARIMA model to predict future operation trends, and generate user preference preset information; Based on the user preference preset information, real-time rendering of the car spraying effect, adjusting rendering parameters, including resolution and shadow quality, to match the user's desired visual effect, and generating a preliminary rendering result; Based on the preliminary rendering result, the rendering texture details are refined, resource usage and response speed are optimized, and a rendering output result is generated.

Images