Electric bicycle form design method and system, electronic equipment and storage medium

Abstract

The invention belongs to the field of vehicle industry design, and discloses an electric bicycle form design method and system, electronic equipment and a storage medium, and the method comprises the steps: mining user emotion vocabularies through online comments, constructing an emotion lexicon in combination with an improved word frequency-inverse document frequency algorithm and a D-S evidence theory, and screening out key perceptual vocabularies; constructing a convolutional long-short-term memory neural network model optimized by a snake swarm algorithm to realize a mapping model between customer sensibility and product morphological characteristics; carrying out subjective and objective comprehensive evaluation on the design scheme in combination with an eye movement experiment and subjective evaluation so as to screen out an optimal design scheme; an optimal scheme is selected and input into a generative AI platform for multi-angle visual rendering, ergonomic modeling and aerodynamics simulation are assisted, and the structural feasibility and performance are verified. The method provides systematic technical support for emotional value improvement and design optimization of industrial products.

Description

Technical Field

[0001] The present invention relates to the field of sustainable vehicle industrial design technology driven by big data, and in particular to an electric bicycle form design method, system, electronic device and storage medium that integrates perceptual engineering, deep learning and multi-source evaluation. Background Art

[0002] Driven by the trend toward green transportation and sustainable urban development, micro-mobility products have become practical tools for short-distance commuting. Electric-assisted bicycles, with their environmentally friendly, efficient, and health-promoting features, have become a key option for short-distance travel. Compared to traditional bicycles, electric-assisted bicycles, powered by batteries, not only improve riding comfort and efficiency but also alleviate urban traffic congestion and carbon emissions.

[0003] As production technology continues to improve, the electric bicycle market is becoming increasingly homogenized. Mainstream brands are finding it difficult to differentiate their products in features like battery life, speed, and weight. As a result, users are shifting their focus to the emotional satisfaction and personalization that products offer. This trend requires product design to not only meet functional requirements but also accurately respond to users' emotional needs to influence their purchasing decisions. However, during the product design process, companies struggle to listen to customers and accurately capture their true needs, leading to sales failures for new products. Therefore, using scientific methods to investigate user needs can improve R&D efficiency, shorten product development cycles, and increase the success rate of product launches.

[0004] Existing technologies for capturing user perceptions rely heavily on traditional questionnaires and interviews, which are subject to strong subjectivity, limited sample sizes, and low efficiency. During the design generation phase, these methods rely on the designer's experience and lack data-driven intelligent generation and optimization mechanisms. During the scheme evaluation phase, subjective evaluations are often disconnected from objective engineering verification. Therefore, there is an urgent need for an electric bicycle form design method that can systematically integrate user perceptions, intelligently generate design solutions, and conduct multi-dimensional scientific verification. Summary of the Invention

[0005] Based on this, the present invention aims to provide a systematic sensory-driven electric bicycle form design method, system, electronic device and storage medium, aiming to solve the problems in the existing electric-assisted bicycle form design process, such as inaccurate capture of user sensory needs, reliance on subjective experience in design scheme generation, and lack of comprehensiveness and scientificity in scheme evaluation.

[0006] In a first aspect, the present invention provides a method for designing an electric bicycle, comprising the following steps: Based on online user review data, we extract and identify key emotional words used to represent user emotional needs by combining a word frequency-inverse document frequency algorithm, weight correction of word position, part of speech, and category factors, and DS evidence theory. Constructing and training a convolutional neural network-long short-term memory network model optimized by a snake swarm algorithm, wherein the convolutional neural network-long short-term memory network model optimized by the snake swarm algorithm is used to establish a nonlinear mapping relationship between the key perceptual vocabulary and the morphological characteristics of the electric bicycle composed of multiple morphological component codes, and using the convolutional neural network-long short-term memory network model optimized by the snake swarm algorithm to predict and generate a corresponding optimal morphological design combination for each key perceptual vocabulary; Combining subjective questionnaire evaluation data based on the Likert scale with objective physiological data based on eye tracking experiments, the priority order method was used to comprehensively evaluate and rank the optimal morphological design combinations to screen out the optimal design scheme; The optimal design scheme is visually rendered from multiple angles, and is subjected to ergonomic simulation and computational aerodynamic simulation to verify its structural feasibility and performance.

[0007] As an optional implementation of the first aspect of the present application, the step of extracting and determining key emotional words used to characterize user emotional needs includes: collecting and preprocessing user comment texts to form a valid comment text set; using the word frequency-inverse document frequency algorithm to calculate the terms in the text set to obtain initial weights; dividing the comment text into three semantic structure areas: title, first sentence and body, and assigning preset weighting coefficients to them respectively, and correcting the initial weights to obtain corrected weights; using the normalized initial weights and the corrected weights as two independent sources of evidence, fusing them using the Dempster synthesis rule of DS evidence theory to generate a final comprehensive score, and sorting from high to low according to the comprehensive score to screen out the key emotional words.

[0008] As an optional implementation of the first aspect of the present application, the convolutional neural network-long short-term memory network model optimized by the snake swarm algorithm is used to establish a nonlinear mapping relationship between the key perceptual vocabulary and the electric bicycle morphological features composed of multiple morphological component encodings: the electric bicycle morphological features are composed of multiple morphological components including at least handlebars, seat, frame, fenders, pedals, wheels and sprocket sets, and the different forms of each morphological component are encoded as preset integers.

[0009] As an optional implementation scheme of the first aspect of the present application, the steps of constructing and training a convolutional neural network-long short-term memory network model optimized by the snake swarm algorithm include: using a one-dimensional convolutional neural network layer to perform convolution and pooling operations on the sequence composed of the morphological component encodings to extract local spatial features; inputting the local spatial features into a two-layer long short-term memory network to capture the sequential dependencies between the morphological component combinations; the snake swarm algorithm uses the mean square error of the convolutional neural network-long short-term memory network model on the validation set as the fitness function, and optimizes the number of hidden layer units of the long short-term memory network by iteratively updating the positions of the snake head, body and tail.

[0010] As an optional implementation manner of the first aspect of the present application, the priority order method is used to comprehensively evaluate and rank the optimal morphological design combination in combination with the subjective questionnaire evaluation data based on the Likert scale and the objective physiological data based on the eye tracking experiment to screen out the optimal design scheme: the objective physiological data are at least seven eye movement indicators obtained through the eye tracking experiment, including total gaze time, average gaze time, number of gazes, first gaze time, total visit time, average visit time and number of visits; the comprehensive evaluation is performed by assigning a weight of 50% to the weighted average of the subjective questionnaire evaluation data and the objective physiological data, and calculating the final score using the priority order method.

[0011] As an optional implementation manner of the first aspect of the present application, the optimal design scheme is subjected to multi-angle visual rendering, and ergonomic simulation and computational aerodynamic simulation are performed on it: the multi-angle visual rendering includes: using a sketch-aware rendering platform to convert the line drawing of the optimal design scheme into a basic rendering image, and then inputting the basic rendering image into a three-dimensional generation platform to obtain a multi-angle image, and finally inputting the multi-angle image into a stable diffusion model for refined rendering; the ergonomic simulation is performed using CATIA software, and the RULA score is used to evaluate and optimize the comfort of the riding posture; the aerodynamic simulation is performed using Fluent software, and the aerodynamic performance is evaluated by calculating the aerodynamic drag coefficient and analyzing the velocity field, pressure field and trajectory diagram.

[0012] As an optional implementation of the first aspect of the present application, the optimization of the ergonomic simulation includes: adjusting the geometric parameters of the optimal design scheme including the top tube length, handlebar height, seat height and seat tube angle until the RULA score is reduced to a preset reasonable range.

[0013] In a second aspect, an embodiment of the present application provides an electric bicycle form design system, comprising: The sentiment vocabulary extraction module is configured to extract and determine key sentiment words used to represent user sentiment needs based on online user review data, using a method that combines a word frequency-inverse document frequency algorithm, weight correction of word position, part of speech, and category factors, and DS evidence theory integration; a morphological mapping generation module configured to construct and train a convolutional neural network-long short-term memory network model optimized by a snake swarm algorithm, wherein the convolutional neural network-long short-term memory network model optimized by the snake swarm algorithm is used to establish a nonlinear mapping relationship between the key perceptual vocabulary and the morphological features of the electric bicycle composed of a plurality of morphological component codes, and to use the convolutional neural network-long short-term memory network model optimized by the snake swarm algorithm to predict and generate a corresponding optimal morphological design combination for each key perceptual vocabulary; a design scheme evaluation and screening module configured to combine subjective questionnaire evaluation data based on the Likert scale with objective physiological data based on the eye tracking experiment, and use a priority order method to comprehensively evaluate and sort the optimal form design combination to screen out the optimal design scheme; The scheme rendering and verification module is configured to perform multi-angle visual rendering of the optimal design scheme, and perform ergonomic simulation and computational aerodynamic simulation on it to verify its structural feasibility and performance.

[0014] In a third aspect, an embodiment of the present application provides an electronic device comprising a processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein when the program or instruction is executed by the processor, the steps of the method described in the first aspect are implemented.

[0015] In a fourth aspect, an embodiment of the present application provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented.

[0016] Compared with the prior art, the present invention has the following beneficial effects: 1. Precise demand capture: By integrating an improved TF-IDF algorithm and DS evidence theory, we objectively and efficiently mine core sentimental vocabulary from massive user reviews, avoiding the subjective biases and limitations of traditional research methods and ensuring that design inputs are more aligned with real market needs.

[0017] 2. Intelligent design generation: Using a convolutional neural network-long short-term memory network model optimized by the snake swarm algorithm, it can automatically learn and establish the complex nonlinear relationship between perceptual imagery and specific product form characteristics, realizing the intelligent generation from perceptual needs to design solutions, greatly improving design efficiency and innovation.

[0018] 3. Scientific evaluation system: Combining subjective questionnaires and objective eye movement experiments, a comprehensive evaluation system integrating subjective and objective factors was constructed, and quantitative ranking was carried out through the priority order method, making the solution screening process more comprehensive and reliable, effectively making up for the shortcomings of a single evaluation method.

[0019] 4. Integrated verification process: Combining generative AI visual rendering with professional CATIA ergonomics and Fluent aerodynamic simulation, we achieve an integrated verification loop from aesthetic expression to engineering feasibility, ensuring that the final solution combines emotional value, user comfort, and excellent physical performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a flow chart of a method for designing the shape of an electric bicycle according to an embodiment of the present invention; Figure 2 This is a flow chart of the convolutional neural network-long short-term memory network model (SO-CNN-LSTM) optimized by the snake swarm algorithm proposed in the present invention; Figure 3 It is a structural schematic diagram of an electric bicycle shape design system provided by an embodiment of the present invention.

[0021] The following specific embodiments will further illustrate the present invention in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0023] The terms "first", "second", etc. in the specification and claims of this application are used to distinguish similar objects, and are not used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the application can be implemented in a sequence other than those illustrated or described here. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally represents that the objects associated before and after are in a kind of "or" relationship. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically limited.

[0024] Example 1 See also Figure 1 , is a flow chart of a method for designing the shape of an electric bicycle provided by an embodiment of the present invention. The method may include the following steps: S1: Based on online user review data, a method combining the word frequency-inverse document frequency algorithm, weight correction of word position, part of speech, and category factors, and DS evidence theory is used to extract and determine the key emotional words used to represent user emotional needs.

[0025] Exemplarily, step S1 specifically includes the following steps S11 to S14: S11. Build a sentiment vocabulary We crawled user reviews of electric-assisted bicycles from major global e-commerce platforms between 2015 and 2025, collecting a total of 16,024 original texts. Using the Jieba word segmentation tool, we performed Chinese word segmentation and part-of-speech tagging. Furthermore, we performed preprocessing steps such as stop word filtering and low-frequency word removal to ultimately generate a valid review text set.

[0026] S12. Calculate word frequency-inverse document frequency modeling The improved word frequency-inverse document frequency model is used to perform weighted calculation on the original corpus to construct a basic perceptual word candidate set. Specifically, the word frequency-inverse document frequency weight consists of the following two parts: TF is used to measure the frequency of a term in a specific text: in, Indicates word frequency The number of times it appears in the document, is the total number of occurrences of all word frequencies in the document. represents the entire corpus, i.e. the collection of all collected review texts; represents a specific comment in the corpus, and .

[0027] IDF is used to measure the discrimination provided by a term in the entire corpus: in, is the total number of comments in the text collection, Indicates word frequency The number of documents is added to avoid the denominator being 0.

[0028] Finally, the term frequency-inverse document frequency score of the term It can be expressed as: In the actual calculation, the collected corpus of user reviews on electric-assisted bicycles is first segmented and tagged with parts of speech in Chinese. After filtering out stop words, the word frequency-inverse document frequency score of each word is calculated, and the words are sorted from high to low according to the score. The top few representative preliminary perceptual vocabulary candidate sets are screened out, laying the foundation for subsequent structural position information correction and weight fusion.

[0029] S13. Word Position, Part of Speech, and Category Factors (Word Position, Part of Speech, and Category Factors) Structural Modification and Position-Aware Modeling The word position, part of speech, and category factors are introduced to modify the traditional word frequency-inverse document frequency score to improve the context sensitivity of perceptual word extraction. Specifically, the word position, part of speech, and category factor method divides the review text into three semantic structure areas: title, first sentence, and body. Different weight coefficients are assigned to each area. , , , and build the following location-aware weighted model: in Sentimental words Traditional in the comment title area Score, Words expressing emotion In the first sentence of the comment Score, Words expressing emotion In the body of the comment Score.

[0030] in, In this embodiment, based on experience and verification, , , , highlighting the importance of the front term of the structure.

[0031] S14. DS Evidence Theory Fusion and Confidence Weighting Assume that the candidate set of perceptual words is , n represents the number of candidate perceptual words, then each evidence source For each term Assign a basic probability distribution function , which satisfies the following constraints: Where A represents any subset, Represents the empty set.

[0032] In this embodiment, the original score of word frequency-inverse document frequency and the score modified by word position, part of speech and category factors are normalized and used as two independent evidence sources. and Specifically, the normalized term weights are directly mapped to their confidence values ​​in each piece of evidence: in Words expressing emotion Tradition in the original text Score, Represents the weight after the introduction of word position, part of speech and category factor correction mechanism Score.

[0033] Then, the two basic probability distributions are fused according to Dempster's combination rule. The combination formula is: in, It represents the degree of conflict between two pieces of evidence and is defined as follows: Finally, based on the fusion Sort from high to low, and retain the perceptual words with the highest confidence as the core input for modeling and image generation in this embodiment.

[0034] S2: Construct and train a convolutional neural network-long short-term memory network model optimized by the snake swarm algorithm, wherein the convolutional neural network-long short-term memory network model optimized by the snake swarm algorithm is used to establish a nonlinear mapping relationship between the key perceptual vocabulary and the morphological features of the electric bicycle composed of multiple morphological component codes, and use the convolutional neural network-long short-term memory network model optimized by the snake swarm algorithm to predict and generate the corresponding optimal morphological design combination for each key perceptual vocabulary.

[0035] For example, step S2 specifically includes the following steps S21 to S22, which can be found in Figure 2 , which is a flowchart of the convolutional neural network-long short-term memory network model (SO-CNN-LSTM) optimized by the snake swarm algorithm proposed in the present invention.

[0036] S21. The physical structures of 100 commercially available electric bicycles (covering urban commuting, mountain biking, and other types) were morphologically deconstructed, identifying seven typical design components: handlebars, seat, frame, fenders, pedals, wheels, and sprockets. Each component category was coded as an integer from 1 to 7 based on its morphological differences, forming a unified dataset of electric bicycle morphological features. Based on this, a standardized morphological analysis table was constructed. Furthermore, to obtain the perceptual label data required for model training, a questionnaire based on a 7-point Likert scale was designed. 100 participants independently rated each electric bicycle model on the four dimensions of "lightness," "flexibility," "refinement," and "comfort." Each user evaluated the appearance images of each of the 100 electric bicycles without brand identification. The average score of the perceptual vocabulary was used as the final rating for each model, resulting in a perceptual evaluation matrix.

[0037] S22. Build a mapping model Among them, step S22 can be further divided into the following steps S221 to S223: S221, Snake Swarm Algorithm In this embodiment, the snake swarm algorithm is used to optimize the key parameters in the long short-term memory neural network time series model: (1) Individual initialization: Assume the search space dimension is , the population size is , then The initial position vector of each individual is: in and Respectively represent The upper and lower bounds of the dimension parameter.

[0038] (2) Fitness function design: The mean square error (MSE) of the convolutional neural network-long short-term memory network in the validation set is used as the fitness evaluation function: in Score for realism. is the model's predicted value.

[0039] (3) Snake head update (global guidance): The snake head performs global guidance based on the current optimal position, and its position update formula is: in represents a random variable Compliance interval The uniform distribution on represents the uniform distribution function, Represents the position vector of the snake head individual at the current iteration step t, is the step size factor, is the current global optimal individual.

[0040] (4) Snake body update (exploration enhancement): The body part simulates the collaborative update of the snake group, and its update method is: in Control the intensity of local disturbances, represents the standard normal distribution, Represents the position vector of the snake body at the current iteration step t.

[0041] (5) Snake tail update (local convergence): The snake tail simulates the local search process and follows the snake head to enhance convergence: in is the local convergence adjustment factor, Represents the position vector of the snake tail individual at the current step length t.

[0042] S222, Convolutional Neural Network (1) Convolution operation: Assume that the input feature sequence is , Indicates the last moment l The input is of length One-dimensional convolution kernel Perform local feature extraction. Convolution output at time It is given by: in, represents the convolution operation, is the bias term, is a nonlinear activation function, Indicates The convolution window data centered at .

[0043] (2) Pooling layer: To reduce the feature dimension and retain key information, the convolution output will be processed by maximum pooling. Let the pooling window size be , the result is expressed as: in, The pooling operation can effectively compress the feature length, improve computational efficiency and reduce overfitting.

[0044] Step S223: Long Short-Term Memory Network Feature sequence after pooling As a compressed representation of the time series, the input is fed into the long short-term memory network to capture dependencies over a longer period of time. The calculation steps of the long short-term memory network are as follows: (1) Calculation of candidate memory units: in, is the candidate memory content, is the current time step input, is the hidden state at the previous moment, , is the corresponding weight matrix, is the bias term.

[0045] (2) Input gate and forget gate control: in, and Represents the activation output of the input gate and the forget gate respectively, controlling the degree of writing of current information and the degree of retention of information at the previous moment. is the sigmoid function, Represents input The weight matrix between the input gate and Represents input The weight matrix between the forget gate and Represents the previous hidden state The weight matrix between the input gate and Represents the previous hidden state The weight matrix between the forget gate and represents the bias term of the input gate, Represents the bias term of the forget gate.

[0046] (3) Memory unit status update: This formula indicates that the current memory state is obtained by weighted combination of the memory state at the previous moment and the current input information. Indicates the memory state of the previous moment Passing the Gate of Oblivion Weighted retention, with current input information Input Gate Weighted writes are composed together.

[0047] (4) Output gate and hidden state update: The final hidden state The current state of the memory cell through After activation and control by the output gate, it is output as the long short-term memory network output of the current time step. represents the activation vector of the input gate, represents the hidden state vector at the current time step, Indicates the current input The weight matrix between the input gate and Represents the previous hidden state The weight matrix between the output gate, Represents the bias term of the output gate.

[0048] S3: Combining the subjective questionnaire evaluation data based on the Likert scale with the objective physiological data based on the eye tracking experiment, the priority order method is used to comprehensively evaluate and sort the optimal morphological design combinations to screen out the optimal design scheme.

[0049] Exemplarily, step S3 specifically includes the following steps S31 to S33: S31. Subjective evaluation In terms of subjective evaluation, this example invited 67 users to participate in the evaluation, 23 of whom had design backgrounds. Using a 7-point Likert scale, the top five electric-assisted bicycle design combinations were independently scored based on the four perceptual terms of "lightweight," "flexible," "refined," and "comfortable." To avoid visual interference, the electric-assisted bicycle combinations were evaluated before rendering. The specific formula is: but For the Users rated this design. is the total number of users who participated in the subjective evaluation, Represents the average subjective evaluation score of the design scheme.

[0050] S32. Objective Evaluation A visual preference experiment was conducted using Tobii Pro Glasses 3 eye-tracking equipment and a 27-inch computer monitor. Five graduate students with a design background were invited. The experiment was set up so that each observer browsed for about 6 minutes and prepared an eye-tracking experiment chart based on the screen size. Seven key indicators of the subject during the observation process were recorded: (1) total fixation time; (2) average fixation time; (3) number of fixations; (4) first fixation time; (5) total visit time; (6) average visit time; (7) number of visits. After the experiment, the hot spots and observation trajectories of 20 electric bicycles were exported using Tobii Pro Lab software. At the same time, accurate frame selection was performed based on the shape of each electric bicycle to export accurate attention data. All indicators were normalized by minimum-maximum value: but is the original value of the eye movement index, is the minimum value of this indicator among all design schemes. is the maximum value of this indicator among all design schemes. is the normalized index value.

[0051] S33. Comprehensive evaluation through priority order method In terms of weight setting, subjective scores and objective eye movement indicators each account for 50%, ensuring balanced representation of users' subjective perceptions and real visual behaviors.

[0052] 7 normalized eye movement indicators are integrated into an objective comprehensive score . Using weighted average: in For the The normalized value of the eye movement index, For the The weight of an indicator.

[0053] Calculate comprehensive evaluation score And according to the priority order method, the optimal design scheme is screened out. The specific calculation formula is: S4: Perform multi-angle visual rendering of the optimal design scheme, and perform ergonomic simulation and computational aerodynamic simulation on it to verify its structural feasibility and performance.

[0054] Exemplarily, step S4 specifically includes the following steps S41 to S43: S41. Generate high-quality renderings This method uses the Vizcom platform to perform sketch-aware rendering, then uses this as the initial image input into a 3D generation platform to generate multi-angle images. This image is then input into a stable diffusion model to optimize the rendering of these multi-angle images. The core concept is to achieve condition-guided high-fidelity image generation by gradually introducing Gaussian noise and then inversely reconstructing the original image through a learned denoising network. The specific method is as follows: (1) Forward process: In the diffusion stage, the original image Gaussian noise is gradually added to form a noise image sequence . The noise form of the step is: in, represents the cumulative retention factor, , is the predefined noise scheduling coefficient, (˙) represents Gaussian normal distribution, and I represents the unit covariance matrix, that is, isotropic Gaussian noise with equal variance.

[0055] (2) Reverse process: by training the neural network model , for any moment Denoising is performed to reconstruct the original image. The reverse modeling probability distribution is: Among them, the mean and variance Determines the reconstruction performance of the model. The mean calculation expression is as follows: (3) Initial image prediction and reconstruction: Based on the noise image at the current moment With estimated noise , the approximate estimate of the original image can be deduced , used for model reconstruction: (4) To control the stability of the denoising process, the variance term It can be set as a fixed constant or a learning parameter. A common practice is to set it to a scaled form of unit variance: in Usually based on It can be derived and can also be fine-tuned during training.

[0056] (5) Model optimization objective function The stable diffusion model learns the network parameters by minimizing the reconstruction error loss function, whose basic form is the mean square error of the prediction noise: in [˙] represents the original data sample ,noise and time steps joint expectations.

[0057] S42, CATIA human body simulation verification Using an adult male human model (175 cm tall, 70 kg) as a benchmark, a 3D solid model of the S13 solution was constructed using CATIA V5 R20, and a complete human-machine assembly was completed. Virtual assembly and posture simulation of the S13 solution were performed. A simulated urban commuting riding scenario (without additional load, Load = 0) was used to evaluate riding comfort using the RULA score and pressure distribution.

[0058] During the first simulation, the initial design resulted in a RULA score of 5 due to its short top tube (430mm), low handlebars (780mm), insufficient seat height (757mm), and steep seat tube angle (70°), indicating fatigue after prolonged riding. After optimization, the top tube was increased to 510mm, the handlebars were raised to 790mm, the seat was adjusted to 770mm, and the seat tube angle was increased to 72°. The RULA score was significantly reduced to 3, indicating that this configuration provides an ergonomically sound riding position and good muscle load distribution.

[0059] S43, aerodynamic simulation verification In order to evaluate the aerodynamic performance of electric bicycles during riding, the aerodynamic drag coefficient ( C d ) is as follows: in, Indicates aerodynamic force, is the air density, is the air velocity (set wind speed or average riding speed), is the frontal area of ​​the vehicle.

[0060] Example 2 See also Figure 3 , shown is a schematic diagram of the structure of an electric bicycle shape design system proposed in the second embodiment of the present application, which includes the following key modules: The sentiment vocabulary extraction module 100 is configured to extract and determine key sentiment vocabulary used to represent user sentiment needs based on online user review data by using a method that combines a word frequency-inverse document frequency algorithm, weight correction of word position, part of speech, and category factors, and DS evidence theory fusion; a morphological mapping generation module 200 configured to construct and train a convolutional neural network-long short-term memory network model optimized by a snake swarm algorithm, wherein the convolutional neural network-long short-term memory network model optimized by the snake swarm algorithm is used to establish a nonlinear mapping relationship between the key perceptual words and the morphological features of the electric bicycle composed of a plurality of morphological component codes, and to use the convolutional neural network-long short-term memory network model optimized by the snake swarm algorithm to predict and generate a corresponding optimal morphological design combination for each key perceptual word; The design scheme evaluation and screening module 300 is configured to combine the subjective questionnaire evaluation data based on the Likert scale with the objective physiological data based on the eye tracking experiment, and use a priority order method to comprehensively evaluate and sort the optimal morphological design combinations to screen out the optimal design scheme; The scheme rendering and verification module 400 is configured to perform multi-angle visual rendering on the optimal design scheme, and perform ergonomic simulation and computational aerodynamic simulation on it to verify its structural feasibility and performance.

[0061] In the embodiments of the present application, an electric bicycle form design system can be a device, or a component, integrated circuit, or chip in a terminal. The device can be a mobile electronic device or a non-mobile electronic device. For example, the mobile electronic device can be a mobile phone, tablet computer, laptop computer, PDA, in-vehicle electronic device, wearable device, ultra-mobile personal computer (UMPC), netbook, or personal digital assistant (PDA), etc. The non-mobile electronic device can be a server, network attached storage (NAS), personal computer (PC), etc., which are not specifically limited in the embodiments of the present application.

[0062] In the embodiment of the present application, an electric bicycle shape design system can be a device having an operating system. The operating system can be an Android operating system, an iOS operating system, or other possible operating systems, which are not specifically limited in the embodiment of the present application.

[0063] The electric bicycle shape design system provided in the embodiment of the present application can achieve Figure 1 To avoid repetition, each process of implementing the electric bicycle form design method in the method embodiment will not be described here.

[0064] Optionally, an embodiment of the present application also provides an electronic device, including a processor, a memory, and a program or instruction stored in the memory and executable on the processor. When the program or instruction is executed by the processor, each process of the above-mentioned electric bicycle form design method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0065] An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the various processes of the embodiment of the above-mentioned electric bicycle form design method are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0066] The processor is the processor in the electronic device described in the above embodiment. The readable storage medium includes a computer-readable storage medium, such as a computer read-only memory (ROM), random access memory (RAM), a magnetic disk, or an optical disk.

[0067] It should be noted that, in this article, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element. In addition, it should be noted that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in the opposite order according to the functions involved. For example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0068] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a more preferred embodiment. Based on this understanding, the technical solution of this application, or the part that contributes to the existing technology, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of this application.

[0069] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of this application, ordinary technicians in this field can also make many forms without departing from the purpose of this application and the scope of protection of the claims, all of which are within the protection of this application.

Claims

1. A method for designing the shape of an electric bicycle, characterized in that: The following steps are involved: Based on online user review data, we extract and identify key emotional words used to represent user emotional needs by combining a word frequency-inverse document frequency algorithm, weight correction of word position, part of speech, and category factors, and DS evidence theory. Constructing and training a convolutional neural network-long short-term memory network model optimized by a snake swarm algorithm, wherein the convolutional neural network-long short-term memory network model optimized by the snake swarm algorithm is used to establish a nonlinear mapping relationship between the key perceptual vocabulary and the morphological characteristics of the electric bicycle composed of multiple morphological component codes, and using the convolutional neural network-long short-term memory network model optimized by the snake swarm algorithm to predict and generate a corresponding optimal morphological design combination for each key perceptual vocabulary; Combining subjective questionnaire evaluation data based on the Likert scale with objective physiological data based on eye tracking experiments, the priority order method was used to comprehensively evaluate and rank the optimal morphological design combinations to screen out the optimal design scheme; The optimal design scheme is visually rendered from multiple angles, and is subjected to ergonomic simulation and computational aerodynamic simulation to verify its structural feasibility and performance.

2. The method according to claim 1, characterized in that The step of extracting and determining key perceptual words for representing the user's emotional needs includes: Collect and preprocess user comment texts to form a valid comment text set; The word frequency-inverse document frequency algorithm is used to calculate the terms in the text set to obtain the initial weights; Divide the review text into three semantic structure areas: title, first sentence, and body, assign preset weight coefficients to each area, and modify the initial weights to obtain modified weights; The normalized initial weight and the revised weight are used as two independent evidence sources, and are fused using the Dempster synthesis rule of the DS evidence theory to generate a final comprehensive score. The comprehensive scores are then sorted from high to low to screen out the key perceptual words.

3. The method according to claim 1 or 2, characterized in that In the step of using the convolutional neural network-long short-term memory network model optimized by the snake swarm algorithm to establish a nonlinear mapping relationship between the key perceptual vocabulary and the electric bicycle morphological features composed of multiple morphological component codes: The morphological features of the electric bicycle are composed of multiple morphological components including at least a handlebar, a seat, a frame, a mudguard, a pedal, a wheel and a sprocket set, and the different shapes of each morphological component are encoded as a preset integer.

4. The method according to claim 1, wherein The steps of constructing and training a convolutional neural network-long short-term memory network model optimized by the snake swarm algorithm include: Performing convolution and pooling operations on the sequence of morphological component codes using a one-dimensional convolutional neural network layer to extract local spatial features; Inputting the local spatial features into a two-layer long short-term memory network to capture the sequential dependencies between the morphological component combinations; The snake swarm algorithm uses the mean square error of the convolutional neural network-long short-term memory network model on the validation set as the fitness function, and optimizes the number of hidden layer units of the long short-term memory network by iteratively updating the positions of the snake head, body and tail.

5. The method according to claim 1, wherein Combining the subjective questionnaire evaluation data based on the Likert scale with the objective physiological data based on the eye tracking experiment, the priority order method is used to comprehensively evaluate and rank the optimal morphological design combinations to screen out the optimal design solution: The objective physiological data are at least seven eye movement indicators obtained through eye tracking experiments, including total fixation duration, average fixation duration, number of fixations, first fixation duration, total visit duration, average visit duration, and number of visits; The comprehensive evaluation is performed by assigning a weight of 50% to the weighted average of the subjective questionnaire evaluation data and the objective physiological data, and calculating the final score using a priority order method.

6. The method according to claim 1, characterized in that In the steps of performing multi-angle visual rendering on the optimal design solution and performing ergonomic simulation and computational aerodynamic simulation on the optimal design solution: The multi-angle visual rendering includes: using a sketch perception rendering platform to convert the line drawing of the optimal design solution into a basic rendering image, then inputting the basic rendering image into a 3D generation platform to obtain a multi-angle image, and finally inputting the multi-angle image into a stable diffusion model for refined rendering; The ergonomic simulation is performed using CATIA software, and the RULA score is used to evaluate and optimize the comfort of the riding posture; The aerodynamic simulation is performed using Fluent software, and the aerodynamic performance is evaluated by calculating the aerodynamic drag coefficient and analyzing the velocity field, pressure field, and trajectory diagram.

7. The method according to claim 6, characterized in that The optimization of the ergonomic simulation includes adjusting geometric parameters of the optimal design solution, including top tube length, handlebar height, seat height, and seat tube angle, until the RULA score is reduced to a preset reasonable range.

8. An electric bicycle shape design system, characterized in that: include: The sentiment vocabulary extraction module is configured to extract and determine key sentiment words used to represent user sentiment needs based on online user review data, using a method that combines a word frequency-inverse document frequency algorithm, weight correction of word position, part of speech, and category factors, and DS evidence theory integration; a morphological mapping generation module configured to construct and train a convolutional neural network-long short-term memory network model optimized by a snake swarm algorithm, wherein the convolutional neural network-long short-term memory network model optimized by the snake swarm algorithm is used to establish a nonlinear mapping relationship between the key perceptual vocabulary and the morphological features of the electric bicycle composed of a plurality of morphological component codes, and to use the convolutional neural network-long short-term memory network model optimized by the snake swarm algorithm to predict and generate a corresponding optimal morphological design combination for each key perceptual vocabulary; a design scheme evaluation and screening module configured to combine subjective questionnaire evaluation data based on the Likert scale with objective physiological data based on the eye tracking experiment, and use a priority order method to comprehensively evaluate and sort the optimal form design combination to screen out the optimal design scheme; The scheme rendering and verification module is configured to perform multi-angle visual rendering of the optimal design scheme, and perform ergonomic simulation and computational aerodynamic simulation on it to verify its structural feasibility and performance.

9. An electronic device, characterized in that: The method comprises a processor, a memory, and a program or instruction stored in the memory and executable on the processor. When the program or instruction is executed by the processor, the steps of the electric bicycle form design method as described in any one of claims 1 to 7 are implemented.

10. A readable storage medium, characterized in that: The readable storage medium stores a program or instruction, and when the program or instruction is executed by the processor, the steps of the electric bicycle form design method as described in any one of claims 1 to 7 are implemented.

Images