Manufacturing method of mobile phone structural parts based on microporous injection molding technology

Abstract

The present invention discloses a method for manufacturing a mobile phone structural part based on microporous injection molding technology, S1, using a morphological symbiotic control algorithm to manage the global parameters of injection molding and coordinate the flow of materials in the mold; S2, guiding the material flow path through a nonlinear vortex-driven material self-organization algorithm to form a microporous structure; S3, establishing a data exchange mechanism to achieve coordinated control of the morphological symbiotic control algorithm and the nonlinear vortex algorithm; S4, when detecting abnormal material flow, adjusting the global parameters through the data exchange mechanism to optimize the material path; S5, dynamically adjusting the injection path and molding parameters according to real-time feedback to maintain the stability of the microporous structure; S6, inputting molding data into a learning module to update and optimize molding parameters and path planning; S7, self-correcting through a closed-loop feedback mechanism to improve the consistency and quality of the microporous structure. The present invention improves the stability and consistency of the microporous structure molding process.

Description

Technical Field

[0001] The invention relates to the technical field of microporous injection, in particular to a method for manufacturing a mobile phone structural part based on the microporous injection molding technology. Background Art

[0002] With the continuous development of manufacturing technology, microporous injection molding, as a high-precision and high-efficiency molding process, has been widely used in electronic equipment and industrial applications. The formation of microporous structure in injection molding has a significant impact on the strength, heat dissipation performance and weight of the product. However, the existing microporous injection molding technology still has many problems and challenges in practical applications, especially in terms of material flow control and microporous structure consistency.

[0003] In the prior art, the flow path and parameter control methods of injection molding mostly rely on static injection models and simple control algorithms. These methods are difficult to cope with dynamic changes in the injection molding process, such as fluctuations in material flow, pressure fluctuations, and temperature changes. Specifically, the following are the main defects in the prior art:

[0004] Existing material flow control methods lack flexible adjustment and adaptive capabilities when dealing with much more complex parameter conditions. Traditional flow control technology mainly relies on fixed injection speed and pressure settings, and cannot be adjusted in real time according to small changes in material flow, which easily leads to uneven flow and ultimately affects the formation and quality of microporous structures. Especially in the injection molding process, uneven material flow can lead to pore size deviation and structural inconsistency in the product, seriously affecting the performance of the finished product.

[0005] The existing heat conduction and stress coupling processing methods are too simple to achieve effective regulation in high-precision molding. Heat conduction and stress characteristics have a direct impact on the flow behavior of materials, but current processes often treat them as separate factors and lack comprehensive consideration. This method cannot effectively coordinate when the material's stress and heat dissipation characteristics change, resulting in uneven material cooling speed during the molding process, local stress concentration, and affecting the consistency and stability of the microporous structure.

[0006] Traditional material parameter control algorithms are mostly based on linear control or simple feedback loops, lacking a more complex mechanism for coordinated parameter optimization. Linear control is inefficient in dealing with the dynamic behavior of materials and cannot quickly respond to the interaction of multiple variables in material flow, especially in microporous molding with complex molds. This problem is more prominent. The formation of microporous structures requires uniform flow of materials in the mold, but the existing technology lacks effective multi-parameter control methods, resulting in uneven distribution of flow paths and unstable pore formation in the products.

[0007] Existing feedback control systems have limited performance in real-time correction and adaptive learning. Existing feedback systems are usually adjusted based on a single variable and cannot provide comprehensive feedback based on multiple factors such as flow rate, pressure, viscosity, heat conduction and stress. In this way, flow deviations and pore size problems that occur during the molding process cannot be corrected quickly and effectively, affecting the production efficiency and quality consistency of the product. Therefore, how to provide a method for manufacturing mobile phone structural parts based on microporous injection molding technology is an urgent problem that technicians in this field need to solve. Summary of the invention

[0008] One purpose of the present invention is to propose a method for manufacturing mobile phone structural parts based on microporous injection molding technology. The present invention manages the global parameters in the injection molding process by combining the morphological symbiotic control algorithm and the nonlinear vortex-driven self-organization algorithm, and guides the flow path of the material in the mold to form a microporous structure. The present invention solves the problems of unstable material flow path, inconsistent microporous structure and difficulty in achieving multi-parameter coordinated optimization in the prior art.

[0009] According to an embodiment of the present invention, a method for manufacturing a mobile phone structural component based on microporous injection molding technology comprises the following steps:

[0010] S1. Use the morphological symbiotic control algorithm to manage the global parameters involved in the injection molding process and coordinate the flow of materials in the mold;

[0011] S2, guiding the material flow path in the mold through the nonlinear vortex-driven material self-organization algorithm to form the desired microporous structure;

[0012] S3. Establish a data exchange mechanism to achieve information transfer between the morphological symbiosis control algorithm and the nonlinear vortex-driven material self-organization algorithm, and combine global and local parameters for coordinated control;

[0013] S4. When anomalies are detected in the material flow, the global parameters are adjusted through the data exchange mechanism to ensure continuous optimization of the material flow path;

[0014] S5. According to the real-time feedback data, the injection path and molding parameters are dynamically adjusted to maintain the formation stability of the microporous structure;

[0015] S6, inputting the data in the molding cycle into the learning module to update and optimize the subsequent molding parameters and path planning;

[0016] S7, continuous self-correction through closed-loop feedback mechanism to optimize the consistency and quality of micropore structure.

[0017] Optionally, the S1 includes the following steps:

[0018] S11. Collect real-time material flow data during injection molding and construct a global parameter matrix P, where P = { p1,p2,…,p n} , p i Represents the initial parameters of the flow at different nodes in the mold, including flow rate, pressure, temperature and material properties;

[0019] S12. In the morphological symbiotic regulation algorithm, each material flow node is regarded as a symbiotic unit S i ={f i ,ν i ,κ i ,τ i} , where f i represents the flow characteristics, ν i is the material viscosity, κ i is the thermal conductivity, τ i is the dynamic stress response of the material;

[0020] S13. Establish a symbiotic relationship optimization model and define the symbiotic optimization function between units:

[0021]

[0022] Among them, F is the eigenvector, α i is the weight adjustment coefficient of the material unit, μ ij is the symbiosis coefficient between adjacent units, Ψ(f i ,f j ) is the symbiotic function: Where σ is the adjustment parameter of feature difference, represents the response gradient between units; Used to describe the synergy between viscosity, thermal conductivity, and stress;

[0023] S14, optimizing the symbiotic function by derivation, and obtaining a gradient expression for guiding the adjustment of material flow;

[0024] S15, adjusting the material flow characteristics according to the gradient update algorithm to generate a characteristic vector F';

[0025] S16, transmitting the optimized feature vector F' to the control module to control the key parameters in the injection molding process, including injection speed, pressure, and temperature, to form a consistent microporous structure;

[0026] S17, during the molding cycle, the feedback mechanism is used to dynamically adjust the symbiotic unit parameter α i , symbiosis coefficient μ ij and stress response coefficient τ i The optimization results are used to train the morphological symbiosis control algorithm to adapt to different materials and molding conditions.

[0027] Optionally, S2 includes the following steps:

[0028] S21. Collect real-time material flow data during injection molding and establish a local path matrix M, where M = { m1,m2,…,m n} , each m i Indicates the initial flow parameters of the path node, including the flow velocity v i , pressure p i , stress τ i and temperature T i ;

[0029] S22, input the local path matrix M into the nonlinear vortex-driven material self-organization algorithm to establish a vortex field with rotation drive as the core, and each path node is defined as U i = { v i ,p i ,θ i ,τ i ,ξ i} ; where θ i is the node angle parameter, ξ i represents the local rotational response coefficient of the material;

[0030] S23. Through the nonlinear interaction between materials, the vortex field control equation is defined:

[0031]

[0032] Among them, α i is the material self-regulation weight, γ ij is the coupling coefficient between nodes;

[0033] S24. By taking the derivative, the adjustment gradient of the material path is obtained to guide the optimization of the material flow path:

[0034]

[0035] S25, monitor the stress and temperature changes on the path nodes and adjust α in real time i , γ ij and the rotation response ξ i , maintain the consistency and stability of material flow;

[0036] S26. Output the optimized local path matrix M' for precise control and forming of the microporous structure.

[0037] Optionally, S3 includes the following steps:

[0038] S31, based on the establishment of global parameter matrix P and local path matrix M, introduce the two-way data exchange protocol D ex , so that real-time data transmission can be achieved between the morphological symbiosis control algorithm and the nonlinear vortex-driven material self-organization algorithm, and the data exchange vector E is defined as:

[0039] E={e ij |e ij =ζ(p i ,m j )·ω ij};

[0040] Among them, ζ(p i ,m j ) represents the interaction function between global parameters and local paths, ω ij is the weighting factor between nodes, controlling the intensity of data exchange;

[0041] S32, the morphological symbiosis control algorithm receives the data exchange vector E and adjusts the eigenvector F in the global parameter matrix P in real time;

[0042] S33, the nonlinear vortex-driven material self-organization algorithm uses the data exchange vector E to update the node state of the local path matrix M:

[0043]

[0044] Among them, τ ( e ij ) is the path self-regulation coefficient, Θ ( M ) is the vortex control field of the local path node;

[0045] S34, synchronize the real-time update of global and local parameters, and combine the feedback control mechanism to adjust the response parameter β of the morphological symbiosis regulation algorithm i Rotational response ξ with nonlinear vortex algorithm j Coupling to maintain the flow consistency and stability of the material in the mold:

[0046]

[0047] Among them, η is the learning rate, which controls the update amplitude;

[0048] S35. By iterating the above steps, optimize the data exchange protocol D ex The parameters work together under different environments and molding conditions to maintain the accuracy of the material flow path and the molding quality of the microporous structure;

[0049] S36, outputting the final optimized global parameter matrix P' and local path matrix M' for path control and microporous structure formation in the subsequent molding process.

[0050] Optionally, S4 includes the following steps:

[0051] S41. Real-time monitoring of the flow data of the material in the mold, identification of abnormalities in the flow path, and establishment of an abnormality detection matrix A = { a1,a2,…,a n} ; Each a i Indicates detected abnormal characteristic parameters, such as flow rate fluctuations, pressure imbalance or stress concentration;

[0052] S42, through the data exchange mechanism D ex The anomaly detection matrix A is passed to the morphological symbiosis control algorithm and the nonlinear vortex-driven material self-organization algorithm to trigger the dynamic adjustment of global and local parameters;

[0053] S43, the morphological symbiosis control algorithm receives the anomaly detection matrix A, and optimizes it by adjusting the global parameter matrix P using the global adjustment function;

[0054] S44, the nonlinear vortex-driven material self-organization algorithm adjusts the local path matrix M according to the abnormal data in the abnormal detection matrix A and updates the node status;

[0055] S45. Synchronize the adjustment of global and local parameters through feedback mechanisms to maintain the stability and consistency of material flow paths;

[0056] S46. During the parameter update process, the control factor β is adjusted according to the real-time data of material flow. i and the path response ψ ( a i ,m j ) Perform iterative optimization to make the adjustment of global and local paths collaborative and adaptive;

[0057] S47, outputting the optimized global parameter matrix P' and local path matrix M' to guide the material flow path adjustment and microporous structure forming accuracy improvement in the next stage.

[0058] Optionally, S5 includes the following steps:

[0059] S51. Collect real-time feedback data of material flow during injection molding, including flow rate, pressure, temperature and stress distribution, and construct feedback matrix F d = { f d1 ,f d2 ,…,f dn} , where each fdi Represents the real-time feedback parameters on different path nodes;

[0060] S52, the feedback matrix F d The input is used in the morphological symbiotic regulation algorithm and the nonlinear vortex-driven material self-organization algorithm to analyze and generate preliminary adjustment vectors of global and local parameters;

[0061] S53, combined with the feedback matrix F d , further optimize the local path parameters and update the node status in the path matrix M;

[0062] S54, after comprehensive global and local parameter adjustments, the real-time balance morphology symbiosis control algorithm and the nonlinear vortex-driven material self-organization algorithm are controlled to maintain the stability of the material flow path through iterative calculations;

[0063] S55, according to the updated global parameter matrix P' and local path matrix M', the adjustment factor ρ in the dynamic adjustment algorithm is adjusted in combination with real-time feedback ( f di) and λ;

[0064] S56. During the forming process, continuously monitor the feedback data on the path nodes to further optimize the parameter settings of the morphological symbiosis control algorithm to achieve adaptive regulation of the path and flow;

[0065] S57, outputting the optimized global parameter matrix P' and local path matrix M', which are used to guide the flow path of the material in the mold.

[0066] Optionally, the S6 includes the following steps:

[0067] S61, collect and integrate real-time data within the molding cycle, and construct the cycle data matrix C d =c d1 ,c d2 ,…,c dn} ; where each c di Represents the flow rate, pressure, stress, and temperature data recorded at a specific path node;

[0068] S62, the periodic data matrix C d Input learning module, perform global parameter optimization based on morphological symbiosis control algorithm and nonlinear vortex-driven material self-organization algorithm, and calculate global optimization vector ΔP;

[0069] S63, based on periodic data matrix C d , calculate the local path adjustment vector ΔM for the optimization of path nodes;

[0070] S64, updating the global parameter matrix P and the local path matrix M;

[0071] S65, verify the updated global parameter matrix P' and path matrix M', identify potential anomalies and perform fine-tuning;

[0072] S66, real-time adjustment of the adaptive adjustment coefficient σ according to the feedback data ( c di ) and λ ( c dj ) , so that the learning module gradually adapts to different molding conditions;

[0073] S67. Through iterative optimization, the final global parameter matrix P' and local path matrix M' are output for maintaining the consistency and stability of the material flow path and the microporous structure in subsequent molding operations.

[0074] The beneficial effects of the present invention are:

[0075] The present invention proposes a microporous injection molding control method combined with a morphological symbiotic control algorithm, which coordinates the flow of materials in the mold by dynamically collecting and managing global parameters. The morphological symbiotic control algorithm can optimize multiple parameters such as injection speed, pressure and temperature in real time to make the material flow path more stable. Compared with traditional fixed parameter control, this method can better adapt to the complex changes in material flow, reduce microporous structural defects caused by uneven flow, and improve the consistency and accuracy of molding.

[0076] The present invention introduces a nonlinear vortex-driven material self-organization algorithm, and overcomes the problems of material accumulation and uneven flow caused by complex mold structures in conventional technologies by dynamically adjusting the material flow path. The nonlinear vortex drive can form a self-regulating vortex structure in the material flow, guide the material flow path in real time, and make the micropore formation more uniform and stable. This method is particularly suitable for microporous structure molding under complex molds and variable production conditions, and effectively improves product quality and production stability.

[0077] The present invention realizes information transmission and coordinated regulation between the morphological symbiotic regulation algorithm and the nonlinear vortex algorithm by establishing a data exchange mechanism, so that the optimization interaction between the global parameters and the local paths is achieved. When a local anomaly in the material flow is detected, the system can trigger the adaptive adjustment of the global parameters, optimize the flow path in real time and update the injection conditions. This mechanism solves the problem of multi-parameter difficulty in coordinated optimization in the prior art, and improves the response speed and regulation effect of the system.

[0078] The present invention introduces a closed-loop feedback mechanism, combines the real-time monitoring data in the molding cycle with the historical production data, and continuously adjusts the optimization strategy through the adaptive correction function. After each production cycle, the system inputs the production data into the learning module for analysis, and updates the molding parameters and path planning for the next cycle. This closed-loop feedback and learning mechanism can continuously improve the molding process, significantly improve the stability and consistency of the microporous structure, reduce molding defects and scrap rates, and improve production efficiency and product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0080] Figure 1 This is a flow chart of the method for manufacturing a mobile phone structural component based on microporous injection molding technology proposed by the present invention;

[0081] Figure 2 This is a schematic diagram of the structure of the morphological symbiosis regulation algorithm proposed in the present invention;

[0082] Figure 3 This is a schematic diagram of a nonlinear vortex-driven self-organizing algorithm proposed in the present invention. DETAILED DESCRIPTION

[0083] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0084] refer to Figure 1-3 , a method for manufacturing a mobile phone structural part based on microporous injection molding technology, comprising the following steps:

[0085] S1. Use the morphological symbiotic control algorithm to manage the global parameters involved in the injection molding process and coordinate the flow of materials in the mold;

[0086] S2, guiding the material flow path in the mold through the nonlinear vortex-driven material self-organization algorithm to form the desired microporous structure;

[0087] S3. Establish a data exchange mechanism to achieve information transfer between the morphological symbiosis control algorithm and the nonlinear vortex-driven material self-organization algorithm, and combine global and local parameters for coordinated control;

[0088] S4. When anomalies are detected in the material flow, the global parameters are adjusted through the data exchange mechanism to ensure continuous optimization of the material flow path;

[0089] S5. According to the real-time feedback data, the injection path and molding parameters are dynamically adjusted to maintain the formation stability of the microporous structure;

[0090] S6, inputting the data in the molding cycle into the learning module to update and optimize the subsequent molding parameters and path planning;

[0091] S7, continuous self-correction through closed-loop feedback mechanism to optimize the consistency and quality of micropore structure.

[0092] In this implementation, S1 includes the following steps:

[0093] S11. Collect real-time material flow data during the injection molding process and construct a global parameter matrix P, where P = {p1, p2, ..., p n}, p i Represents the initial parameters of the flow at different nodes in the mold, including flow rate, pressure, temperature and material properties;

[0094] S12. In the morphological symbiotic regulation algorithm, each material flow node is regarded as a symbiotic unit S i ={f i ,ν i ,κ i ,τ i}, where f i represents the flow characteristics, ν i is the material viscosity, κ i is the thermal conductivity, τ i is the dynamic stress response of the material;

[0095] S13. Establish a symbiotic relationship optimization model and define the symbiotic optimization function between units:

[0096]

[0097] Among them, F is the eigenvector, α i is the weight adjustment coefficient of the material unit, μ ij is the symbiosis coefficient between adjacent units, Ψ ( f i ,f j ) is the symbiotic function: Where σ is the adjustment parameter of feature difference, represents the response gradient between units; Used to describe the synergy between viscosity, thermal conductivity, and stress;

[0098] S14, optimizing the symbiotic function by derivation, and obtaining a gradient expression for guiding the adjustment of material flow;

[0099] S15, adjusting the material flow characteristics according to the gradient update algorithm to generate a characteristic vector F';

[0100] S16, transmitting the optimized feature vector F' to the control module to control the key parameters in the injection molding process, including injection speed, pressure, and temperature, to form a consistent microporous structure;

[0101] S17, during the molding cycle, the feedback mechanism is used to dynamically adjust the symbiotic unit parameter α i , symbiosis coefficient μ ij and stress response coefficient τ i The optimization results are used to train the morphological symbiosis control algorithm to adapt to different materials and molding conditions.

[0102] In this implementation, S2 includes the following steps:

[0103] S21. Collect real-time material flow data during injection molding and establish a local path matrix M, where M = { m1,m2,…,m n} , each m i Indicates the initial flow parameters of the path node, including the flow velocity v i , pressure p i , stress τ i and temperature T i ;

[0104] S22, input the local path matrix M into the nonlinear vortex-driven material self-organization algorithm to establish a vortex field with rotation drive as the core, and each path node is defined as U i = { v i ,p i ,θ i ,τ i ,ξ i} ; where θ i is the node angle parameter, ξ i represents the local rotational response coefficient of the material;

[0105] S23. Through the nonlinear interaction between materials, the vortex field control equation is defined:

[0106]

[0107] Among them, α i is the material self-regulation weight, γ ij is the coupling coefficient between nodes;

[0108] S24. By taking the derivative, the adjustment gradient of the material path is obtained to guide the optimization of the material flow path:

[0109]

[0110] S25, monitor the stress and temperature changes on the path nodes and adjust α in real time i , γ ij and the rotation response ξ i , maintain the consistency and stability of material flow;

[0111] S26. Output the optimized local path matrix M' for precise control and forming of the microporous structure.

[0112] In this implementation, S3 includes the following steps:

[0113] S31, based on the establishment of global parameter matrix P and local path matrix M, introduce the two-way data exchange protocol D ex , so that real-time data transmission can be achieved between the morphological symbiosis control algorithm and the nonlinear vortex-driven material self-organization algorithm, and the data exchange vector E is defined as:

[0114] E={e ij |e ij =ζ(p i ,m j )·ω ij};

[0115] Among them, ζ(p i ,m j ) represents the interaction function between global parameters and local paths, ω ij is the weighting factor between nodes, controlling the intensity of data exchange;

[0116] S32, the morphological symbiosis control algorithm receives the data exchange vector E and adjusts the eigenvector F in the global parameter matrix P in real time;

[0117] S33, the nonlinear vortex-driven material self-organization algorithm uses the data exchange vector E to update the node state of the local path matrix M:

[0118]

[0119] Among them, τ(e ij ) is the path self-regulation coefficient, Θ(M) is the vortex regulation field of the local path node;

[0120] S34, synchronize the real-time update of global and local parameters, and combine the feedback control mechanism to adjust the response parameter β of the morphological symbiosis regulation algorithm i Rotational response ξ with nonlinear vortex algorithm j Coupling to maintain the flow consistency and stability of the material in the mold:

[0121]

[0122] Among them, η is the learning rate, which controls the update amplitude;

[0123] S35. By iterating the above steps, optimize the data exchange protocol D ex The parameters work together under different environments and molding conditions to maintain the accuracy of the material flow path and the molding quality of the microporous structure;

[0124] S36, outputting the final optimized global parameter matrix P' and local path matrix M' for path control and microporous structure formation in the subsequent molding process.

[0125] In this implementation, S4 includes the following steps:

[0126] S41. Real-time monitoring of the flow data of the material in the mold, identification of abnormalities in the flow path, and establishment of an abnormality detection matrix A = { a1,a2,…,a n} ; Each a i Indicates detected abnormal characteristic parameters, such as flow rate fluctuations, pressure imbalance or stress concentration;

[0127] S42, through the data exchange mechanism D ex The anomaly detection matrix A is passed to the morphological symbiosis control algorithm and the nonlinear vortex-driven material self-organization algorithm to trigger the dynamic adjustment of global and local parameters;

[0128] S43, the morphological symbiosis control algorithm receives the anomaly detection matrix A, and optimizes it by adjusting the global parameter matrix P using the global adjustment function;

[0129] S44, the nonlinear vortex-driven material self-organization algorithm adjusts the local path matrix M according to the abnormal data in the abnormal detection matrix A and updates the node status;

[0130] S45. Synchronize the adjustment of global and local parameters through feedback mechanisms to maintain the stability and consistency of material flow paths;

[0131] S46. During the parameter update process, the control factor β is adjusted according to the real-time data of material flow. i and the path response ψ ( a i ,m j ) Perform iterative optimization to make the adjustment of global and local paths collaborative and adaptive;

[0132] S47, outputting the optimized global parameter matrix P' and local path matrix M' to guide the material flow path adjustment and microporous structure forming accuracy improvement in the next stage.

[0133] In this implementation, S5 includes the following steps:

[0134] S51. Collect real-time feedback data of material flow during injection molding, including flow rate, pressure, temperature and stress distribution, and construct feedback matrix F d = { f d1 ,f d2 ,…,f dn} , where each f di Represents the real-time feedback parameters on different path nodes;

[0135] S52, the feedback matrix F d The input is used in the morphological symbiotic regulation algorithm and the nonlinear vortex-driven material self-organization algorithm to analyze and generate preliminary adjustment vectors of global and local parameters;

[0136] S53, combined with the feedback matrix F d , further optimize the local path parameters and update the node status in the path matrix M;

[0137] S54, after comprehensive global and local parameter adjustments, the real-time balance morphology symbiosis control algorithm and the nonlinear vortex-driven material self-organization algorithm are controlled to maintain the stability of the material flow path through iterative calculations;

[0138] S55, according to the updated global parameter matrix P' and local path matrix M', the adjustment factor ρ in the dynamic adjustment algorithm is adjusted in combination with real-time feedback ( f di) and λ;

[0139] S56. During the forming process, continuously monitor the feedback data on the path nodes to further optimize the parameter settings of the morphological symbiosis control algorithm to achieve adaptive regulation of the path and flow;

[0140] S57, outputting the optimized global parameter matrix P' and local path matrix M', which are used to guide the flow path of the material in the mold.

[0141] In this implementation, S6 includes the following steps:

[0142] S61, collect and integrate real-time data within the molding cycle, and construct the cycle data matrix C d =c d1 ,c d2 ,…,c dn} ; where each c di Represents the flow rate, pressure, stress, and temperature data recorded at a specific path node;

[0143] S62, the periodic data matrix C dInput learning module, perform global parameter optimization based on morphological symbiosis control algorithm and nonlinear vortex-driven material self-organization algorithm, and calculate global optimization vector ΔP;

[0144] S63, based on periodic data matrix C d , calculate the local path adjustment vector ΔM for the optimization of path nodes;

[0145] S64, updating the global parameter matrix P and the local path matrix M;

[0146] S65, verify the updated global parameter matrix P' and path matrix M', identify potential anomalies and perform fine-tuning;

[0147] S66, real-time adjustment of the adaptive adjustment coefficient σ according to the feedback data ( c di ) and λ ( c dj ) , so that the learning module gradually adapts to different molding conditions;

[0148] S67. Through iterative optimization, the final global parameter matrix P' and local path matrix M' are output for maintaining the consistency and stability of the material flow path and the microporous structure in subsequent molding operations.

[0149] Embodiment 1:

[0150] In modern microporous injection molding applications, especially in the manufacture of high-precision electronic components, how to ensure the consistency and stability of the microporous structure has always been a difficult problem that plagues the production process. Due to the inflexible parameter adjustment of traditional injection molding methods, the flow path of the material in the complex mold is often affected, resulting in uneven microporous structure and affecting the quality of the final product. In order to solve this problem, the present invention was tested in an advanced electronic component manufacturing company located in Shenzhen to verify the effectiveness of the present invention.

[0151] In this application, the manufacturing company selected a microporous injection molding process for the mesh structure of mobile phone speakers, requiring the micropore diameter to be accurate to 50 microns and the pore size distribution to be uniform to ensure the sound transmission effect and structural stability. The control and optimization are carried out by combining the morphological symbiotic regulation algorithm in the present invention with the nonlinear vortex-driven self-organization algorithm. The production line equipment includes a high-precision injection molding machine, a temperature control module and an online data monitoring system. The production environment temperature is 25°C and the humidity is 45%.

[0152] In specific production applications, the morphological symbiotic regulation algorithm of the present invention first collects global material flow data, establishes an initial parameter matrix, and dynamically manages parameters such as flow velocity, pressure, and temperature at different nodes in the mold. The nonlinear vortex-driven self-organization algorithm adjusts the material flow path in real time and guides the material flow in the mold to form a uniform microporous structure. When an abnormality or pressure fluctuation is detected in the material flow, the data exchange mechanism will be activated to transmit the information to the morphological symbiotic regulation algorithm for adaptive adjustment of global parameters, thereby achieving dynamic optimization of material flow.

[0153] In order to verify the effectiveness of the present invention, an experiment of 200 consecutive injection molding cycles was carried out. Experimental data show that after applying the method of the present invention, the standard deviation of the micropore diameter in each cycle was reduced from 3.2 microns in the traditional method to 1.1 microns, and the uniformity of micropore distribution was improved by more than 20%. During the production process, the changes in injection speed, pressure, and temperature were recorded by the data monitoring system, showing that the algorithm of the present invention can maintain the stability of parameters in real-time regulation, and the adjustment time is shortened from 3 seconds in the traditional method to less than 1 second. After each adjustment, the system performs self-correction through a closed-loop feedback mechanism to ensure the continuity and stability of subsequent production cycles.

[0154] In the on-site production environment, production efficiency has also been significantly improved. In the traditional method, 45 out of every 1,000 products were judged as defective due to inconsistent apertures or structural defects. After the application of the present invention, this number dropped to 12, and the defective rate was reduced by 73%. The entire production process was tested for two weeks at the production base in Shenzhen. The injection molding system of the present invention not only maintained production stability and consistency, but also reduced the frequency of manual intervention and improved the level of production automation.

[0155] Table 1 Production data

[0156] Test items Traditional method data Data of the method of the present invention Micropore diameter standard deviation 3.2 microns 1.1 micron Micropore distribution uniformity 78％ 98％ Parameter adjustment time Average 3 seconds Average 0.8 seconds Defective rate 4.5% (45 per 1,000) 1.2% (12 per 1,000) Adjusting response time Average 2.5 seconds Average 0.9 seconds

[0157] Table 2 Data table

[0158] Data Item Traditional method data Data of the method of the present invention Average production cycle time 15 seconds 12 seconds Energy consumption per injection 5.2 kWh 4.7 kWh Micropore diameter consistency deviation ±3.5％ ±1.2％ Injection pressure fluctuation range 15％ 6％ Material utilization 92％ 97％ Finished product qualification rate 91％ 98.50％ Number of manual interventions (daily) Average 8 times Average 2 times Production cost savings No significant changes About 12% reduction Die wear rate Maintenance 5 times per month Maintenance once a month

[0159] The application of this invention in microporous injection molding of high-precision electronic components solves the problems of inflexible parameter control, unstable flow path and uneven microporous structure in traditional technologies. By introducing the morphological symbiotic control algorithm and the nonlinear vortex-driven self-organization algorithm, real-time control and closed-loop optimization are achieved, making the production process more intelligent and significantly improving the quality of products.

[0160] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A photovoltaic power station intelligent operation and maintenance method based on multimodal time series prediction, characterized in that: The steps include: S1. Collect real-time data of photovoltaic power plants through distributed sensor networks, including photovoltaic panel temperature, humidity, light intensity, voltage and current, to form a multimodal time series data set; S2. Use the conditional generation contrast event network to generate events and extract causal features from multimodal time series data. Create virtual fault events and energy efficiency feature events through the conditional generation mechanism, and extract key causal features through contrastive learning; S3, based on the adaptive heterogeneous information flow transformation model, the causal feature data is transformed into heterogeneous information flows, and the dynamic association of information flows is achieved by adaptively decomposing and reorganizing data flows from different sources; S4, inputting the processed causal feature data into a multi-task time series prediction network to perform multi-task prediction on the equipment operation status, failure probability and power generation trend of the photovoltaic power station; S5. Based on the time series prediction results, the power output and load distribution strategy of the photovoltaic power station are optimized using the adaptive energy efficiency optimization algorithm, and the unit power generation power is dynamically adjusted; S6, when sudden environmental changes or fault warnings are detected, the rapid response module is triggered to dynamically adjust the operating parameters based on the fuzzy logic controller and real-time feedback information; S7. Upload the prediction and optimization data to the intelligent operation and maintenance platform to generate an operation and maintenance report containing equipment status assessment, energy efficiency optimization results and maintenance recommendations.

2. According to claim 1, a photovoltaic power station intelligent operation and maintenance method based on multimodal time series prediction is characterized in that: The S2 comprises the following steps: S21, receive and parse input features from multimodal time series data through conditional generation contrast event network, including photovoltaic panel temperature T(t), humidity H(t), light intensity I 光 (t), voltage V(t) and current A(t), and then normalize them to form the standardized input feature vector X(t) = {T(t), H(t), I 光 (t), V(t), A(t)}; S22. In the condition generation module, a multidimensional fault event generation matrix M is constructed based on the input feature X(t) and environmental factors. 故障 , generate a virtual fault event sample set E 故障 , each virtual fault event sample e i : Among them, d is the dimension of the input data, α k and γ k is the feature weight, θ k is the offset, β is the regularization parameter, ∈ i is the fault noise term; S23, construct a multidimensional energy efficiency event generation matrix M based on input features X(t) and energy efficiency conditions 能效 , generate a virtual energy efficiency feature event set E 能效 : Among them, δ k and k is the feature weight, η k is the offset parameter, λ is the regularization parameter, ξ j is the energy efficiency noise term; S24, the generated virtual fault event set E 故障 and energy efficiency event set E 能效 Compare with the actual collected data X(t) to construct the fault event comparison matrix ΔF 故障 And energy efficiency event comparison matrix ΔF 能效 : S25. Calculate the overall causal feature extraction factor ΔF based on the difference in the comparison matrix 因果 : Where, ΔF 因果 Represents the nonlinear causal differences between faults and energy efficiency events, providing deep causal analysis of multimodal data; S26, the contrastive learning module outputs a high-dimensional causal feature matrix F 因果 =[f1,f2,...,f k ], where each causal feature f k Represent the nonlinear relationship between faults and energy efficiency events for subsequent information flow transformation and energy efficiency optimization; S27, the causal feature matrix F 因果 The input is then fed into the heterogeneous information flow transformation model as the key causal feature input for the intelligent operation and maintenance of photovoltaic power plants.

3. The photovoltaic power station intelligent operation and maintenance method based on multimodal time series prediction according to claim 1 is characterized in that: The S3 comprises the following steps: S31, the high-dimensional causal feature matrix F in claim 3 因果 =[f1,f2,...,f k ] to perform dynamic feature deconstruction and obtain the local feature subset F 局部 and the global feature subset F 全局 : F 局部 ={f i ∈F 因果 |i≤p},F 全局 ={f j ∈F 因果 |j>p}; S32. Constructing adaptive weighted local information flow I 局部 (t), obtained by the following nonlinear combination: in, is the adaptive weight, which enhances the dynamic response of the local information flow through the combination of exponential and sinusoidal terms, and α, β, γ and θ are dynamic adjustment parameters; S33, according to the global feature subset F 全局 Building a Global Information Flow I 全局 (t): That δ, ζ, η and λ are characteristic response adjustment parameters; S34, through the nonlinear double composite recombination function g(I 局部 ,I 全局 ) Heterogeneously fuse local and global information flows to form heterogeneous information flow I 异构 (t), is defined as follows: Among them, ξ is the heterogeneous fusion control coefficient; S35. Heterogeneous Information Flow I 异构 (t) Perform multi-time window decomposition to construct dynamic information flow sequence I 动态 (t), the formula is as follows: I 动态 (t)={I 异构 (t-τ1),I 异构 (t-τ2),...,I 异构 (t-τ n )}; Among them, τ i Represent different time windows to capture the temporal dependencies of information flows; S36, dynamic information flow sequence I 动态 (t) Apply adaptive smoothing to generate a smooth information flow I 平滑 (t), the specific formula is as follows: in, ρ is the smoothing adjustment coefficient; S37, smoothing information flow I 平滑 (t) is transmitted to the multimodal association model as the basic information for subsequent energy efficiency optimization and fault prediction.

4. The method for intelligent operation and maintenance of a photovoltaic power station based on multimodal time series prediction according to claim 1, characterized in that: The S4 comprises the following steps: S41, the smooth information flow I obtained in claim 4 平滑 (t) Perform feature deconstruction to form an independent task feature subset: equipment operation status feature I 状态 (t), Failure probability characteristic I 故障 (t) and power generation trend characteristics I 发电 (t), each feature subset I 任务 (t) is expressed as: Among them, φ k (t') is the dynamic weight of the task feature, and d represents the dimension of the causal feature; S42, feature subset I 状态 (t), I 故障 (t) and I 发电 (t) Input to each task channel of the multi-task timing prediction network, using the task weighting function W 任务 Balance the importance of task features and calculate weighted prediction input in, γ 任务 is the weighted adjustment parameter; S43. Processing weighted input using recurrent neural network architecture To capture the temporal characteristics between multiple tasks; S44, from the hidden state h t The prediction output of each task is deconstructed into the equipment operation status prediction Failure probability prediction and power generation trend forecast S45. Define the multi-task loss function L 任务 The error feedback is given to the prediction results of each task, specifically: Among them, λ 任务 is the task loss weight, y 任务 (t) is the real task value; S46. Based on the loss function L 任务 Gradient feedback is used to optimize the weights and biases of the time series prediction network using the back-propagation algorithm; S47, optimize the multi-task prediction results and The data is transmitted to the intelligent operation and maintenance decision-making module as the basic data for status assessment, fault warning and power generation optimization of the photovoltaic power station.

5. The photovoltaic power station intelligent operation and maintenance method based on multi-modal time series prediction according to claim 1 is characterized in that: The S5 comprises the following steps: S51, predicting the device status output by the multi-task prediction network in claim 5 Failure probability prediction and power generation trend forecast Integrate and build an energy efficiency prediction matrix S52, based on energy efficiency prediction matrix Y 预测 (t) and the real-time output power P of the photovoltaic power station unit 当前 (t) and predicted load demand L 需求 (t), construct an adaptive optimization objective function to achieve a balance between power output and load demand: Among them, the first term represents the deviation between power output and load demand, the second term is a smoothing term, and the regularization coefficient α controls the volatility of power; S53, in the adaptive energy efficiency optimization algorithm, the unit power generation power P is iteratively adjusted by the gradient descent method 单元 (t), through the gradient of the loss function Update power value; S54, based on the optimized unit power generation P 单元 (t), dynamically generate load distribution weight ω j (t); S55, assigning weights ω to the loads of each unit j (t) is applied to the unit power to calculate the actual allocated power P of each unit 分配 (t); S56, P generated based on load distribution model 分配 (t), and the adaptive control module monitors its relationship with the actual load demand L 需求 (t) deviation, and dynamically adjust the allocated power using a feedback mechanism; S57, the adjusted load distribution data P 分配 (t) Output to the photovoltaic power station operation and maintenance system to provide real-time feedback for the actual unit power generation control.

6. The photovoltaic power station intelligent operation and maintenance method based on multi-modal time series prediction according to claim 1, characterized in that: The S6 comprises the following steps: S61, distribute the generated load data P 分配 (t) and energy efficiency prediction matrix Y 预测 (t) Conduct real-time monitoring and generate environmental change indicators C 环境 (t) and fault warning indicator C 故障 (t), is defined as follows: Among them, P 分配，k (t) represents the load distribution power of the kth unit, is the predicted value of failure probability; S62, when C is detected 环境 (t) or C 故障 (t) When the preset threshold is exceeded, the fuzzy logic controller is triggered and the fuzzy variable set F = {F 环境 ,F 故障 }; S63, based on the fuzzy logic control rule table, the fuzzy input variable F 环境 and F 故障 Mapping to control output ΔP 调整 (t); S64, control output ΔP 调整 (t) Perform defuzzification processing to generate the actual operating parameter adjustment value P 实际调整 (t); S65, defuzzify the result P 实际调整 (t) Applied to the real-time feedback module to update the operating parameters of the photovoltaic power station through the feedback mechanism; S66, monitor the adjusted power output P in the real-time feedback module 分配 (t)+P 实际调整 (t), if further environmental changes or failure trends are detected, steps S62 to S65 are repeated until the system is stable.

Images