Electric toothbrush injection molding intelligent monitoring method and system based on sensing feedback

Abstract

The application discloses a method and system for intelligent monitoring of injection molding of an electric toothbrush based on sensing feedback, and relates to the technical field of intelligent monitoring. The method comprises the following steps: dividing node regions of an injection molding machine of the electric toothbrush to obtain N injection molding process node regions, and deploying a multi-modal sensor group; collecting N node injection molding process data streams, and transmitting data between the multi-modal sensor group and an injection molding control center through RS485 communication network; performing molding prediction simulation and abnormal defect identification by the injection molding control center to determine injection molding abnormal defect parameters; performing injection molding adaptive monitoring feedback to obtain injection molding abnormal feedback parameters, and performing molding optimization control. The technical problem of unstable quality of injection molding of the electric toothbrush in the prior art is solved, intelligent identification and adaptive feedback control of molding abnormalities are realized through multi-modal sensing monitoring of the injection molding process, and the technical effect of improving the stability of the injection molding quality is achieved.

Description

Technical Field

[0001] This invention relates to the field of intelligent monitoring technology, specifically to an intelligent monitoring method and system for injection molding of electric toothbrushes based on sensor feedback. Background Technology

[0002] In the production of electric toothbrushes, injection molding is a crucial process determining the precision and surface quality of the outer shell structure. Current electric toothbrush injection molding processes typically rely on manual experience or fixed process parameters for control, lacking the ability to monitor and dynamically adjust key process parameters such as temperature, pressure, and flow rate in real time. Abnormal temperature or pressure field distribution within the mold cavity can easily lead to molding defects such as product deformation, shrinkage, silver streaks, and bubbles, resulting in reduced yield and unstable quality. Although some injection molding equipment has incorporated basic monitoring modules, its data acquisition is limited and response delays are significant, making it difficult to achieve multi-dimensional state identification and feedback control of the molding process. This also hinders the effective handling of the impact of complex material properties and environmental disturbances on injection molding quality. Summary of the Invention

[0003] This application provides a method and system for intelligent monitoring of electric toothbrush injection molding based on sensor feedback, which solves the technical problem of unstable injection molding quality of electric toothbrushes in the prior art.

[0004] The first aspect of this application provides a smart monitoring method for injection molding of electric toothbrushes based on sensor feedback, the method comprising:

[0005] The electric toothbrush injection molding machine is divided into N injection process node areas according to the electric toothbrush injection molding process. A multimodal sensor group is deployed on each of these N injection process node areas. The multimodal sensor group monitors and collects the injection process data streams from the N nodes, and the data is transmitted between the multimodal sensor group and the injection molding control center via an RS485 communication network. The N node injection process data streams are then transmitted to the injection molding control center for molding prediction simulation and abnormal defect identification to determine injection molding abnormal defect parameters. Based on these parameters, adaptive monitoring and feedback of the injection molding process are performed to obtain injection molding abnormal feedback parameters, which are then used for molding optimization control.

[0006] A second aspect of this application provides an intelligent monitoring system for injection molding of electric toothbrushes based on sensor feedback, the system comprising:

[0007] The module consists of several modules: **Region Division Module:** The electric toothbrush injection molding machine is divided into N injection process node regions according to the injection molding process of an electric toothbrush. A multimodal sensor group is deployed on each of these N injection process node regions. **Communication Transmission Module:** The multimodal sensor group monitors and collects the injection process data streams from the N nodes. This data is then transmitted via an RS485 communication network between the multimodal sensor group and the injection molding control center. **Defect Identification Module:** The N node injection process data streams are transmitted to the injection molding control center for molding prediction simulation and abnormal defect identification, determining abnormal defect parameters in the injection molding process. **Optimization Control Module:** Based on the abnormal defect parameters, adaptive monitoring and feedback of the injection molding process are performed to obtain abnormal feedback parameters. These parameters are then used for molding optimization control.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0009] First, the electric toothbrush injection molding machine is divided into N injection process node areas according to the electric toothbrush injection molding process. Multimodal sensor groups are then deployed in these N injection process node areas. Next, the multimodal sensor groups monitor and collect the injection process data streams from the N nodes, and transmit the data between the multimodal sensor groups and the injection molding control center via an RS485 communication network. Then, the N node injection process data streams are transmitted to the injection molding control center for molding prediction simulation and abnormal defect identification, determining the parameters of abnormal injection molding defects. Finally, based on the abnormal injection molding defect parameters, adaptive monitoring and feedback of the injection molding process are performed to obtain injection molding abnormality feedback parameters, which are then used for molding optimization control. This solves the technical problem of unstable injection molding quality in existing electric toothbrush technologies. By using multimodal sensing monitoring of the injection molding process, intelligent identification and adaptive feedback control of molding abnormalities are achieved, resulting in improved injection molding quality stability. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 A schematic diagram of the intelligent monitoring method for injection molding of an electric toothbrush based on sensor feedback provided in an embodiment of this application;

[0012] Figure 2This is a schematic diagram of the structure of an intelligent monitoring system for injection molding of an electric toothbrush based on sensor feedback, provided in an embodiment of this application.

[0013] Explanation of reference numerals in the attached diagram: Area division module 11, communication transmission module 12, defect identification module 13, optimization control module 14. Detailed Implementation

[0014] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0015] Example 1, as Figure 1 As shown, this application provides a smart monitoring method for injection molding of electric toothbrushes based on sensor feedback, wherein the method includes:

[0016] The electric toothbrush injection molding machine is divided into node regions according to the electric toothbrush injection molding process to obtain N injection molding process node regions, and a multimodal sensor group is deployed on the N injection molding process node regions.

[0017] In this embodiment, based on the standard process flow for injection molding of electric toothbrush shells, the main process steps of the injection molding machine are decomposed sequentially, including the raw material melting zone, injection filling zone, pressure holding steady-state zone, cooling and shaping zone, and demolding release zone. For the structural layout and functional characteristics of each process step, the corresponding mechanical components, mold cavity structure, and thermodynamic parameter control parts are extracted as process node identification objects. Based on the process node identification results, the injection molding machine space is divided into multiple functional sub-regions, corresponding to N injection molding process node regions. Each node region represents a specific molding stage and key process control point. On the N injection molding process node regions, according to the process characteristics and monitoring requirements of each node, a multimodal sensor group is deployed. The multimodal sensor group includes temperature sensors, pressure sensors, strain gauges, displacement sensors, and infrared imaging sensors, etc.

[0018] Furthermore, N injection molding process node regions are obtained, including:

[0019] The process stages of the electric toothbrush injection molding process are extracted to obtain a set of toothbrush injection molding process stages, which includes material melting, mold filling, pressure holding, cooling, and demolding. Process nodes are extracted from each of the toothbrush injection molding process stages to obtain a set of related nodes for each injection molding process stage. Based on the quality control requirements of the electric toothbrush, the set of related nodes for each injection molding process stage is screened and merged to obtain a set of key injection molding process nodes. Based on the set of key injection molding process nodes, the electric toothbrush injection molding machine is divided into node regions to obtain the N injection molding process node regions.

[0020] First, the process stages of the electric toothbrush injection molding process are extracted. Based on the injection molding machine control program and the specific molding steps of the electric toothbrush shell, the entire molding process is divided into multiple consecutive process stages, forming a set of toothbrush injection molding process stages. This set includes five stages: material melting, mold filling, pressure holding, cooling, and demolding. For each process stage, by analyzing the mechanical actions, temperature changes, pressure changes, and polymer flow characteristics within the mold cavity, process node extraction is performed. Key operational nodes characterizing the process state of that stage are extracted, such as the heating zone node in the melting stage, the gate node in the mold filling stage, the mold locking node in the pressure holding stage, the mold wall node in the cooling stage, and the ejector pin node in the demolding stage. This yields a set of associated nodes for the injection molding process stages. Subsequently, based on the quality control requirements of electric toothbrush products, the set of related nodes in the injection molding process was screened and merged. Specifically, according to the quality evaluation indicators of electric toothbrush molding, including shell dimensional stability, surface finish, and sealing accuracy, the process sensitivity of each node was analyzed. Redundant nodes or nodes with little impact on quality were eliminated, and nodes with significant impact on quality control were retained and merged to obtain the set of key injection molding process nodes. Finally, based on the set of key injection molding process nodes, the electric toothbrush injection molding machine was spatially divided. Physical mapping relationships were established for corresponding parts such as heating, injection, mold cavity, cooling, and demolding, forming N injection molding process node regions. Each injection molding process node region serves as an independent sensing and control unit, providing a regional basis for subsequent multimodal sensor deployment and molding process monitoring.

[0021] Furthermore, a multimodal sensor array is deployed in the N injection molding process node regions, including:

[0022] Based on the N injection molding process node regions, determine the injection molding process objectives for each of the N node regions; perform monitoring requirement analysis on the injection molding process objectives for each of the N node regions to obtain the process monitoring requirement parameters for the N node regions; based on the process monitoring requirement parameters for the N node regions, perform sensor selection and deployment location analysis on the N injection molding process node regions to obtain the sensor type parameters and sensor location parameters for the N node regions; deploy multimodal sensor groups on the N injection molding process node regions according to the sensor type parameters and sensor location parameters for the N node regions.

[0023] Specifically, for the melting zone nodes, the injection molding process objectives are uniform plasticization of raw materials and stable temperature distribution; for the filling zone nodes, the injection molding process objectives are injection pressure, flow rate, and integrity of mold cavity filling; for the holding zone nodes, the injection molding process objectives are residual stress in the cavity and control of holding time; for the cooling zone nodes, the injection molding process objectives are uniform temperature gradient and stable cooling rate; and for the demolding zone nodes, the injection molding process objectives are smooth demolding and integrity of the part surface.

[0024] A monitoring requirements analysis is conducted for N injection molding process nodes. Considering potential defect types at each stage, such as bubbles, shrinkage, warpage, and cracks, the types of process monitoring signals and measurement accuracy requirements for each node are determined, resulting in process monitoring requirement parameters for the N node regions. These parameters include temperature measurement range, pressure range, sampling frequency, response time, and signal output interface type. After obtaining these parameters, sensor selection and deployment analysis are performed for the N injection molding process node regions based on the process characteristics and equipment installation conditions of each node. Specifically, this includes: selecting K-type thermocouples or infrared temperature sensors for temperature measurement; selecting piezoelectric pressure sensors or thin-film strain gauges for pressure monitoring; selecting laser displacement sensors or Hall effect position sensors for displacement and motion state monitoring; and selecting miniature infrared imaging sensing units for monitoring mold cavity surface temperature and fill distribution. Through finite element simulation or process experiments, the optimal deployment positions of each sensor within the node regions are determined, obtaining sensor type parameters and sensor location parameters for the N node regions. Finally, according to the sensor type and location parameters mentioned above, the physical installation and calibration of the multimodal sensor group are completed in each node region, enabling it to achieve multidimensional synchronous acquisition of temperature field, pressure field and flow state.

[0025] The multimodal sensor group monitors and collects injection molding process data streams from N nodes, and connects to an RS485 communication network for data communication and transmission between the multimodal sensor group and the injection molding control center.

[0026] After deploying sensors in each node area, the synchronous acquisition mode of the multimodal sensor group is activated, enabling each node area sensor to collect multi-source signals such as temperature, pressure, displacement, vibration, and infrared radiation intensity during the injection molding process in real time according to a preset sampling frequency. The output signals of each sensor are filtered, amplified, and converted by an A / D converter through a local signal conditioning module, forming a node injection molding process data stream. Subsequently, data communication between the multimodal sensor group and the injection molding control center is achieved through a communication network built via an RS485 bus. Preferably, the RS485 communication adopts a master-slave bus structure, with the injection molding control center as the master station and each sensor node as a slave station. During the communication initialization phase, the master station assigns addresses and synchronizes baud rates for each sensor node to ensure the timing consistency of multi-node data frames. The acquired node process data stream is formatted and encapsulated according to the Modbus RTU protocol, including sensor number, node number, sampling timestamp, and monitoring parameter value fields, and is uploaded to the injection molding control center periodically via the RS485 bus.

[0027] Furthermore, the RS485 communication network is used for data communication and transmission between the multimodal sensor group and the injection molding control center, including:

[0028] The data transmission requirements and data transmission environment are used as communication transmission constraint parameters; the RS485 communication network is optimized and analyzed based on the communication transmission constraint parameters according to the data transmission target, and a communication network parameter optimization strategy is constructed; the data communication transmission between the multimodal sensor group and the injection molding control center is controlled based on the communication network parameter optimization strategy.

[0029] Specifically, data transmission requirements and data transmission environment are used as communication transmission constraint parameters; the data transmission requirements include node data upload rate, real-time requirements, data frame length, number of communication nodes, and network load capacity; the data transmission environment includes the electromagnetic interference intensity, temperature and humidity conditions, wiring length, and equipment grounding status of the production workshop where the electric toothbrush injection molding machine is located.

[0030] Based on the data transmission objectives, the RS485 communication network is optimized by analyzing the communication transmission constraints. Specifically, the matching relationship between bus transmission baud rate, communication distance, and signal attenuation is determined; the optimal baud rate range is determined through simulation calculation and field verification; the data frame interval time is set in combination with the number of nodes and the transmission rate to prevent bus collisions; appropriate differential signal amplification coefficient and terminal matching resistor value are selected according to the electromagnetic interference level to improve anti-interference capability; and the communication timing tolerance is adjusted according to the temperature and humidity environment compensation algorithm to form a communication network parameter optimization strategy.

[0031] The data communication transmission between the multimodal sensor group and the injection molding control center is controlled based on the communication network parameter optimization strategy. This includes: periodically broadcasting synchronization signals at the injection molding control center to correct the sampling clock of each sensor node; enabling an adaptive inter-frame delay control mechanism at the slave station to eliminate link delay differences; implementing data integrity detection through differential verification and CRC redundancy verification; and automatically triggering retransmission and node reconnection processes when communication anomalies (such as packet loss, delay exceeding limits, or interference distortion) are detected to ensure the continuity and reliability of the transmission link.

[0032] The injection molding process data streams of the N nodes are transmitted to the injection molding control center for molding prediction simulation and abnormal defect identification, and the abnormal defect parameters of injection molding are determined.

[0033] Furthermore, determining the parameters of abnormal defects in injection molding includes:

[0034] Based on the structural design information of the electric toothbrush injection molding machine, a 3D model is generated to create a 3D model of the electric toothbrush injection molding machine. An injection molding prediction driving model is constructed and coupled to the 3D model of the electric toothbrush injection molding machine to establish an electric toothbrush injection molding twin. Based on the electric toothbrush injection molding twin, molding prediction simulation and abnormal defect identification are performed on the injection molding process data stream of the N nodes to determine the injection molding abnormal defect parameters.

[0035] A 3D model is created based on the structural design information of an electric toothbrush injection molding machine. Specifically, key structural dimensions, mold cavity geometry parameters, gate and cooling channel layout, heating element location, and drive system installation location are obtained. A precise geometric model of the injection molding machine is then constructed using 3D modeling software. This model includes core components such as the injection unit, clamping unit, mold system, heating system, and cooling system. The spatial configuration and thermal characteristics of the injection molding machine are accurately reproduced to a realistic scale, generating a 3D model of the electric toothbrush injection molding machine.

[0036] A predictive driving model for injection molding is constructed. This model, based on historical injection molding process data and molding mechanism, is established using a combination of time series prediction and finite element thermo-fluid-structure interaction (FEM) simulation. It is used to dynamically predict the evolution of the thermal flow field, melt pressure distribution, and residual stress changes during the injection molding process. Specifically, it involves using multi-dimensional process parameters such as temperature, pressure, filling rate, and cooling rate as inputs, and part molding quality indicators (including dimensional stability, surface finish, and density uniformity) as outputs. Multiple sets of historical samples are used for predictive training to form a molding predictive driving model with real-time response capabilities.

[0037] By coupling the injection molding prediction-driven model to the 3D model of an electric toothbrush injection molding machine, a virtual simulation system integrating entity and data is formed, creating an injection molding twin of the electric toothbrush. This injection molding twin can dynamically map the operating state of the real injection molding machine in the virtual environment and receive real-time node process data streams collected from a multimodal sensor array, enabling synchronous simulation calculation and feature deduction of the physical equipment state.

[0038] Based on a twin of an electric toothbrush injection molding process, the system performs molding prediction simulation and anomaly / defect identification on N-node injection molding process data streams. By analyzing the differences between the simulation prediction results and real-time monitoring data, possible anomaly patterns and defect types are identified, such as under-injection, localized overheating, uneven mold cavity filling, or abnormal cooling shrinkage. The system extracts corresponding anomaly feature values ​​based on the identified anomaly patterns, including temperature anomaly thresholds, pressure fluctuation amplitudes, time lag, and positional offsets, and synthesizes these into injection molding anomaly / defect parameters.

[0039] Furthermore, constructing a prediction-driven model for injection molding includes:

[0040] A historical dataset of electric toothbrush injection molding is collected, cleaned, and arranged chronologically to obtain an electric toothbrush injection molding sequence dataset. An injection molding machine prediction task set is preset, and the electric toothbrush injection molding sequence dataset is trained based on the injection molding machine prediction task set to construct an injection molding prediction-driven model.

[0041] Specifically, a historical dataset of electric toothbrush injection molding is collected. This dataset originates from the process monitoring and quality inspection systems of the injection molding machine during long-term production operation, covering key parameters such as raw material heating temperature, melt pressure curve, mold filling rate, holding pressure, cooling time, mold temperature, part deformation rate, and surface defect information. The dataset undergoes data cleaning, anomaly removal, and time-series reconstruction. Specifically, a sliding window method is used to smooth out abnormal peaks, and linear interpolation or temporal proximity values ​​are used to compensate for missing points. Then, based on the injection molding cycle, the multi-source data is aligned and rearranged in segments according to timestamps to form a continuous and consistent electric toothbrush injection molding sequence dataset.

[0042] A pre-defined set of injection molding machine prediction tasks is provided, including tasks for predicting melt flow uniformity, mold cavity pressure distribution, cooling rate and temperature gradient, and part deformation trends. For each prediction task, corresponding input-output mapping relationships and objective functions are defined. For example, the flow prediction task uses injection pressure, screw speed, and mold temperature as input variables, and melt front velocity and filling completeness as output objectives; the cooling prediction task uses mold cavity heat flux density and cooling water flow rate as input variables, and temperature gradient uniformity index as the output objective.

[0043] The injection molding prediction model is trained on an electric toothbrush injection molding sequence dataset based on an injection molding machine prediction task set. Preferably, a time-series-based neural network structure, such as a Long Short-Term Memory (LSTM) network or a Temporal Convolutional Network (TCN), is used for training to capture the temporal coupling and dynamic nonlinear relationships between multiple process parameters. Cross-validation and sliding window evaluation mechanisms are used to verify and optimize model performance, improving prediction accuracy and generalization ability. After training, the generated injection molding prediction-driven model can dynamically output the molding state change trend and potential defect probability distribution based on real-time input process parameters, enabling early prediction and intelligent diagnosis of the injection molding process.

[0044] Furthermore, based on the injection molding machine prediction task set, the electric toothbrush injection molding sequence dataset is used for prediction-driven training to construct an injection molding prediction-driven model, including:

[0045] Based on the injection molding machine prediction task set, the electric toothbrush injection molding sequence dataset is associated and split to obtain a task-associated injection molding sequence dataset; a time series model is used to perform prediction-driven training, validation, optimization and integration on the task-associated injection molding sequence dataset to construct the injection molding prediction-driven model.

[0046] The electric toothbrush injection molding sequence dataset is correlated and split based on the injection molding machine prediction task set. Specifically, the multidimensional process parameters contained in the electric toothbrush injection molding sequence dataset are feature-associated and divided according to the input requirements of different prediction tasks. For example, for the melt flow prediction task, features related to injection pressure, temperature gradient, and mold filling rate are extracted; for the mold cavity pressure prediction task, features of clamping force, holding pressure, and injection time are extracted; and for the cooling and deformation prediction task, features of mold temperature, cooling water flow rate, and part surface temperature gradient are extracted. Through feature importance analysis and principal component analysis, a feature-task association matrix is ​​established, thereby splitting the original sequence data and obtaining multiple task-associated injection molding sequence datasets.

[0047] A time series model is employed to drive prediction training, validation, and optimization on a dataset of injection molding sequence data associated with specific tasks. Preferably, a Long Short-Term Memory (LSTM) network, gated recurrent units (GRUs), or a temporal convolutional network (TCN) is used to capture the temporal dependencies and nonlinear coupling characteristics between various process parameters. During training, the target variables for each task (such as peak cavity pressure, temperature distribution uniformity index, and cooling shrinkage) are used as supervisory signals, and mean squared error is used as the loss function. Parameter optimization is achieved through backpropagation and gradient descent algorithms. After training, the model is validated and optimized: first, the model's prediction accuracy and stability are evaluated using an independent validation set, and overfitting is prevented through dynamic learning rate adjustment, dropout regularization, and early stopping strategies; then, the feature weights output by different task models are weighted and fused with the error feedback to form a unified injection molding prediction-driven model.

[0048] Based on the abnormal defect parameters of injection molding, adaptive monitoring and feedback of injection molding are performed to obtain abnormal feedback parameters of injection molding, and molding optimization control is performed through the abnormal feedback parameters of injection molding.

[0049] When the injection molding control center receives abnormal defect parameters from the digital twin, it first matches and locates the corresponding process links based on the abnormality type, occurrence node, and deviation magnitude to determine the injection node area affected by the abnormality. Subsequently, the control center activates the adaptive monitoring feedback module, which uses the abnormal defect parameters as input to dynamically adjust the sampling frequency, signal resolution, and communication update rate of the monitoring nodes to achieve enhanced monitoring of the abnormal area.

[0050] During the adaptive monitoring and feedback process, the system triggers a local adaptive networking mechanism via the RS485 communication network to prioritize bandwidth allocation and signal synchronization correction for affected multimodal sensor nodes. Simultaneously, an online data fusion algorithm is used to perform differential analysis on the real-time monitoring data of abnormal nodes and historical baseline characteristics, extracting abnormal trend parameters and recovery rate parameters to comprehensively form injection molding anomaly feedback parameters. These parameters describe the dynamic evolution characteristics of the abnormal region, including indicators such as temperature recovery rate, pressure stability coefficient, cooling compensation deviation, and changes in mold cavity balance.

[0051] After obtaining the abnormal feedback parameters of injection molding, the control center enters the molding optimization control stage. Specifically, based on the mapping relationship between the abnormal feedback parameters and the process control model, real-time process parameter corrections are performed: when an abnormal melt temperature is detected, the heating belt power output and screw speed are automatically adjusted; when insufficient or excessive filling pressure is identified, the injection pressure curve and holding time are dynamically corrected; when uneven cooling or abnormal demolding is detected, the cooling water flow rate and mold opening and closing rate are adjusted.

[0052] While performing optimized control, the system continuously collects node data before and after optimization, compares and evaluates the trend of molding quality changes, and achieves adaptive parameter updates through feedback closed loop, enabling the control model to have continuous learning and dynamic self-correction capabilities.

[0053] Furthermore, the parameters for obtaining injection molding anomaly feedback include:

[0054] Based on the injection molding abnormality defect parameters and the N injection molding process node regions, the injection molding abnormality node region set is determined; based on the injection molding abnormality node region set, the RS485 communication network is activated to perform adaptive networking and injection molding monitoring feedback to obtain injection molding abnormality feedback parameters.

[0055] Specifically, the abnormal node numbers, abnormal types, and spatial coordinate information in the injection molding abnormal defect parameters are compared and analyzed with the node region mapping table. Based on the physical location and process stage attributes of the abnormal event, the corresponding injection molding machine structural parts and process nodes are determined. Through dual verification of spatial geometric matching and process timing matching, all nodes directly or indirectly related to the abnormal defects are screened out, forming an injection molding abnormal node region set.

[0056] Based on the set of abnormal injection molding node areas, the RS485 communication network is activated for adaptive networking and injection molding monitoring feedback. Specifically, the control center issues adaptive networking commands to the multimodal sensor groups within the abnormal node area set, adjusting the topology and bandwidth allocation strategy of the communication network to achieve priority communication and high-frequency sampling mode for abnormal nodes. The communication priority and inter-frame interval of each sensor node are dynamically modified through the master station scheduling mechanism to ensure the real-time performance and integrity of data in the abnormal area. Simultaneously, the system uses a communication status monitoring module to detect the signal quality and link latency of each node, automatically activating backup communication channels or redundant nodes to prevent data interruption caused by single-point failures. In adaptive networking mode, sensors within the abnormal node area continuously collect high-precision temperature, pressure, vibration, and flow state data, which are then denoised and corrected using a real-time fusion algorithm. The control center compares the characteristic change trends before and after the anomaly, calculates dynamic parameters such as the anomaly repair rate, pressure recovery coefficient, temperature gradient compensation, and energy consumption deviation rate, and performs time-weighted and normalized processing to comprehensively form injection molding anomaly feedback parameters.

[0057] Furthermore, the molding optimization control through the injection molding anomaly feedback parameters includes:

[0058] The abnormal feedback parameters of the injection molding are used to locate the root cause of the abnormality and determine the root cause of the abnormality in the electric toothbrush injection molding. Based on the root cause of the abnormality in the electric toothbrush injection molding, the process parameters of the electric toothbrush injection molding machine are optimized and the injection molding is controlled.

[0059] The injection molding control center, based on multi-dimensional indicators such as temperature gradient change rate, pressure fluctuation amplitude, energy consumption deviation rate, and recovery rate included in the feedback parameters, invokes the anomaly causal analysis module to calculate the correlation between anomaly characteristics and process parameters. Preferably, a causal inference algorithm based on Bayesian networks or grey relational analysis is employed to probabilistically model and weight the interaction between different process variables (such as screw speed, melt temperature, holding time, cooling flow rate, and mold cavity temperature difference) and anomalies (such as shrinkage, short shots, warpage, and bubbles). By calculating the contribution of each parameter change to the anomaly indicators, the dominant control variable with the greatest impact on the anomaly is selected, and the anomaly source path is determined.

[0060] Based on the causal reasoning results, the system further conducts a time-series backtracking analysis of the process status within the abnormal node region. By comparing the parameter trends and fluctuation characteristics before and after the anomaly, the accuracy of the root cause inference is verified, and finally, the root cause results of the electric toothbrush injection molding anomaly are output.

[0061] Based on the root cause analysis of abnormalities in electric toothbrush injection molding, process parameters and molding control of the electric toothbrush injection molding machine were optimized. The injection molding control center performed dynamic process corrections on the injection molding machine according to the key control parameters identified in the root cause analysis. For example: when the melt zone temperature was detected to be too high, the heating band power was reduced or the screw back pressure was adjusted; when insufficient mold filling was identified, the peak injection pressure was increased or the holding time was extended; when uneven cooling caused deformation, the cooling water flow rate and mold temperature difference compensation were adjusted; when excessive residual stress in the mold cavity was detected, the cooling cycle was extended or the pressure distribution during the holding stage was optimized.

[0062] During the optimization control process, the system continuously monitors the response results of the corrected process parameters through the feedback loop and compares the quality indicators of the parts before and after optimization, such as dimensional deviation, surface finish, and density distribution. If the improvement effect reaches the preset threshold, the parameter group is locked as the new optimal process configuration; if it does not meet the standard, a second adaptive optimization loop is triggered until the abnormal correction converges.

[0063] In summary, the embodiments of this application have at least the following technical effects:

[0064] First, the electric toothbrush injection molding machine is divided into N injection process node areas according to the electric toothbrush injection molding process. Multimodal sensor groups are then deployed in these N injection process node areas. Next, the multimodal sensor groups monitor and collect the injection process data streams from the N nodes, and transmit the data between the multimodal sensor groups and the injection molding control center via an RS485 communication network. Then, the N node injection process data streams are transmitted to the injection molding control center for molding prediction simulation and abnormal defect identification, determining the parameters of abnormal injection molding defects. Finally, based on the abnormal injection molding defect parameters, adaptive monitoring and feedback of the injection molding process are performed to obtain injection molding abnormality feedback parameters, which are then used for molding optimization control. This solves the technical problem of unstable injection molding quality in existing electric toothbrush technologies. By using multimodal sensing monitoring of the injection molding process, intelligent identification and adaptive feedback control of molding abnormalities are achieved, resulting in improved injection molding quality stability.

[0065] Example 2 is based on the same inventive concept as the sensor-feedback-based intelligent monitoring method for electric toothbrush injection molding in the foregoing examples, such as... Figure 2 As shown, this application provides an intelligent monitoring system for electric toothbrush injection molding based on sensor feedback, wherein the system includes:

[0066] Region Division Module 11: Divides the electric toothbrush injection molding machine into N injection process node regions according to the electric toothbrush injection molding process, and deploys a multimodal sensor group on the N injection process node regions; Communication Transmission Module 12: Monitors and collects the injection process data stream of the N nodes through the multimodal sensor group, and connects to the RS485 communication network to carry out data communication transmission between the multimodal sensor group and the injection molding control center; Defect Identification Module 13: Transmits the injection process data stream of the N nodes to the injection molding control center for molding prediction simulation and abnormal defect identification, and determines the injection molding abnormal defect parameters; Optimization Control Module 14: Performs injection molding adaptive monitoring feedback based on the injection molding abnormal defect parameters, obtains injection molding abnormal feedback parameters, and performs molding optimization control through the injection molding abnormal feedback parameters.

[0067] Furthermore, the region division module 11 is used to perform the following method:

[0068] The process stages of the electric toothbrush injection molding process are extracted to obtain a set of toothbrush injection molding process stages, which includes material melting, mold filling, pressure holding, cooling, and demolding. Process nodes are extracted from each of the toothbrush injection molding process stages to obtain a set of related nodes for each injection molding process stage. Based on the quality control requirements of the electric toothbrush, the set of related nodes for each injection molding process stage is screened and merged to obtain a set of key injection molding process nodes. Based on the set of key injection molding process nodes, the electric toothbrush injection molding machine is divided into node regions to obtain the N injection molding process node regions.

[0069] Furthermore, the region division module 11 is used to perform the following method:

[0070] Based on the N injection molding process node regions, determine the injection molding process objectives for each of the N node regions; perform monitoring requirement analysis on the injection molding process objectives for each of the N node regions to obtain the process monitoring requirement parameters for the N node regions; based on the process monitoring requirement parameters for the N node regions, perform sensor selection and deployment location analysis on the N injection molding process node regions to obtain the sensor type parameters and sensor location parameters for the N node regions; deploy multimodal sensor groups on the N injection molding process node regions according to the sensor type parameters and sensor location parameters for the N node regions.

[0071] Furthermore, the communication transmission module 12 is used to perform the following method:

[0072] The data transmission requirements and data transmission environment are used as communication transmission constraint parameters; the RS485 communication network is optimized and analyzed based on the communication transmission constraint parameters according to the data transmission target, and a communication network parameter optimization strategy is constructed; the data communication transmission between the multimodal sensor group and the injection molding control center is controlled based on the communication network parameter optimization strategy.

[0073] Furthermore, the defect identification module 13 is used to perform the following method:

[0074] Based on the structural design information of the electric toothbrush injection molding machine, a 3D model is generated to create a 3D model of the electric toothbrush injection molding machine. An injection molding prediction driving model is constructed and coupled to the 3D model of the electric toothbrush injection molding machine to establish an electric toothbrush injection molding twin. Based on the electric toothbrush injection molding twin, molding prediction simulation and abnormal defect identification are performed on the injection molding process data stream of the N nodes to determine the injection molding abnormal defect parameters.

[0075] Furthermore, the defect identification module 13 is used to perform the following method:

[0076] A historical dataset of electric toothbrush injection molding is collected, cleaned, and arranged chronologically to obtain an electric toothbrush injection molding sequence dataset. An injection molding machine prediction task set is preset, and the electric toothbrush injection molding sequence dataset is trained based on the injection molding machine prediction task set to construct an injection molding prediction-driven model.

[0077] Furthermore, the defect identification module 13 is used to perform the following method:

[0078] Based on the injection molding machine prediction task set, the electric toothbrush injection molding sequence dataset is associated and split to obtain a task-associated injection molding sequence dataset; a time series model is used to perform prediction-driven training, validation, optimization and integration on the task-associated injection molding sequence dataset to construct the injection molding prediction-driven model.

[0079] Furthermore, the optimization control module 14 is used to perform the following method:

[0080] Based on the injection molding abnormality defect parameters and the N injection molding process node regions, the injection molding abnormality node region set is determined; based on the injection molding abnormality node region set, the RS485 communication network is activated to perform adaptive networking and injection molding monitoring feedback to obtain injection molding abnormality feedback parameters.

[0081] Furthermore, the optimization control module 14 is used to perform the following method:

[0082] The abnormal feedback parameters of the injection molding are used to locate the root cause of the abnormality and determine the root cause of the abnormality in the electric toothbrush injection molding. Based on the root cause of the abnormality in the electric toothbrush injection molding, the process parameters of the electric toothbrush injection molding machine are optimized and the injection molding is controlled.

[0083] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for intelligent monitoring of electric toothbrush injection molding based on sensor feedback, characterized in that, The method includes: The electric toothbrush injection molding machine is divided into node regions according to the electric toothbrush injection molding process to obtain N injection molding process node regions. A multimodal sensor group is deployed on the N injection molding process node regions. The multimodal sensor group monitors and collects injection molding process data streams from N nodes, and connects to the RS485 communication network for data communication and transmission between the multimodal sensor group and the injection molding control center. The injection molding process data streams of the N nodes are transmitted to the injection molding control center for molding prediction simulation and abnormal defect identification, and the parameters of abnormal defects in injection molding are determined. Based on the injection molding abnormality defect parameters, adaptive monitoring and feedback of injection molding is performed to obtain injection molding abnormality feedback parameters, and molding optimization control is performed through the injection molding abnormality feedback parameters. Determine the parameters of abnormal defects in injection molding, including: Based on the structural design information of the electric toothbrush injection molding machine, a three-dimensional model is created to generate a three-dimensional model of the electric toothbrush injection molding machine. Construct an injection molding prediction driving model, and couple the injection molding prediction driving model to the three-dimensional model of the electric toothbrush injection molding machine to establish an electric toothbrush injection molding twin; Based on the electric toothbrush injection molding twin, the injection molding process data stream of the N nodes is used to perform molding prediction simulation and abnormal defect identification to determine the injection molding abnormal defect parameters. Constructing a prediction-driven model for injection molding, including: We collected historical data on electric toothbrush injection molding, cleaned and sorted the data in chronological order to obtain a dataset of electric toothbrush injection molding sequence. A preset injection molding machine prediction task set is used. Based on the injection molding machine prediction task set, the electric toothbrush injection molding sequence dataset is trained for prediction-driven training to construct an injection molding prediction-driven model, including: Based on the injection molding machine prediction task set, the electric toothbrush injection molding sequence dataset is associated and split to obtain the task-associated injection molding sequence dataset. A time series model is used to perform prediction-driven training, validation, and optimization on the task-related injection molding sequence dataset to construct the injection molding prediction-driven model.

2. The intelligent monitoring method for electric toothbrush injection molding based on sensor feedback as described in claim 1, characterized in that, We obtain N injection molding process node regions, including: The process stages of the electric toothbrush injection molding process are extracted to obtain a set of toothbrush injection molding process stages, which includes material melting, mold filling, pressure holding, cooling and demolding. Process nodes are extracted from the toothbrush injection molding process stage set to obtain the injection molding process stage associated node set. Based on the quality control requirements of electric toothbrushes, the set of related nodes in the injection molding process stages is screened and merged to obtain the set of key injection molding process nodes; Based on the set of key injection molding process nodes, the electric toothbrush injection molding machine is divided into node regions to obtain the N injection molding process node regions.

3. The intelligent monitoring method for electric toothbrush injection molding based on sensor feedback as described in claim 1, characterized in that, A multimodal sensor array is deployed in the N injection molding process node areas, including: Based on the N injection molding process node regions, determine the injection molding process objectives for the N node regions; The monitoring requirements for the injection molding process targets in the N node regions are analyzed to obtain the process monitoring requirements parameters for the N node regions. Based on the process monitoring requirements parameters of the N node regions, sensor selection and deployment location analysis are performed on the N injection molding process node regions to obtain sensor type parameters and sensor location parameters for the N node regions. Multimodal sensor groups are deployed on the N injection molding process node regions according to the sensor type parameters and sensor position parameters of the N node regions.

4. The intelligent monitoring method for electric toothbrush injection molding based on sensor feedback as described in claim 1, characterized in that, Accessing the RS485 communication network enables data communication and transmission between the multimodal sensor group and the injection molding control center, including: Data transmission requirements and data transmission environment are used as communication transmission constraint parameters; Based on the communication transmission constraint parameters according to the data transmission target, the RS485 communication network is analyzed for parameter optimization, and a communication network parameter optimization strategy is constructed. The data communication transmission between the multimodal sensor group and the injection molding control center is controlled based on the communication network parameter optimization strategy.

5. The intelligent monitoring method for electric toothbrush injection molding based on sensor feedback as described in claim 1, characterized in that, Obtain injection molding anomaly feedback parameters, including: Based on the injection molding abnormality defect parameters and the N injection molding process node regions, the set of injection molding abnormality node regions is determined. Based on the set of abnormal injection molding node regions, the RS485 communication network is activated to perform adaptive networking and injection molding monitoring feedback, thereby obtaining injection molding abnormality feedback parameters.

6. The intelligent monitoring method for electric toothbrush injection molding based on sensor feedback as described in claim 1, characterized in that, Molding optimization control is performed using the injection molding anomaly feedback parameters, including: The abnormal feedback parameters of the injection molding were used to locate the root cause of the abnormality and determine the root cause of the abnormality in the injection molding of the electric toothbrush. Based on the root cause results of the abnormal injection molding of the electric toothbrush, the process parameters of the electric toothbrush injection molding machine are optimized and the injection molding process is controlled.

7. An intelligent monitoring system for electric toothbrush injection molding based on sensor feedback, characterized in that, The system is used to implement the intelligent monitoring method for injection molding of electric toothbrushes based on sensor feedback as described in any one of claims 1-6, the system comprising: Region division module: Divide the electric toothbrush injection molding machine into node regions according to the electric toothbrush injection molding process to obtain N injection molding process node regions, and deploy a multimodal sensor group on the N injection molding process node regions; Communication transmission module: Monitors and collects injection molding process data streams from N nodes through the multimodal sensor group, and connects to the RS485 communication network to conduct data communication and transmission between the multimodal sensor group and the injection molding control center; Defect identification module: Transmits the injection molding process data streams of the N nodes to the injection molding control center for molding prediction simulation and abnormal defect identification, and determines the abnormal defect parameters of injection molding; Optimization control module: Based on the injection molding abnormality defect parameters, it performs adaptive monitoring and feedback of injection molding to obtain injection molding abnormality feedback parameters, and performs molding optimization control through the injection molding abnormality feedback parameters.

Images