Dynamic die pressure data analysis system for polyurethane pultrusion production

Abstract

The present invention relates to the field of data analysis technology, specifically to a dynamic mold pressure data analysis system for polyurethane pultrusion production. The system includes a mold pressure data capture module, a pressure gradient trend calculation module, a mold abnormal pressure identification module, a mold pressure adjustment data module, and a multi-dimensional pressure state monitoring module. In the present invention, through data capture by multi-node pressure sensors, high-precision real-time monitoring of the internal pressure of the mold is achieved, data integrity is enhanced, the limitations of single data are avoided, pressure characteristics of different areas of the mold are identified, the accuracy of abnormal judgment is improved, normal fluctuations and abnormal accumulation are distinguished, abnormal areas are accurately locked, the adjustment response speed is improved, temperature and ultrasonic signal analysis are introduced, curing anomalies and bubble aggregation are identified, the monitoring range is expanded, and data acquisition accuracy, analysis depth and control capabilities are improved, so that abnormality detection is more scientific, control is more intelligent, resin flow is optimized, and the risk of quality fluctuations is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, in particular to a dynamic die pressure data analysis system for polyurethane pultrusion production. Background Art

[0002] The field of data analysis technology includes the collection, storage, processing and analysis of data from various production and manufacturing processes in order to optimize production processes, improve product quality and increase production efficiency. This technical field covers the application of data acquisition equipment, the selection of data processing methods, the establishment of statistical analysis models and the development of visualization tools. In industrial manufacturing, data analysis technology can achieve real-time monitoring of production parameters, and judge the changing trends of key parameters through algorithmic analysis, and then adjust production strategies. Especially in continuous production and manufacturing, data analysis technology can be used to monitor material properties, equipment operating status and dynamic changes in parameters, providing data support for precise control and optimization.

[0003] Among them, the dynamic mold pressure data analysis system for polyurethane pultrusion production refers to a data processing system used to collect, monitor and analyze the changes in pressure inside the mold during the pultrusion process. The system uses high-precision pressure sensors to collect pressure at multiple positions in the mold in real time, and combines time series data analysis technology to model and analyze pressure change trends. The system uses data fitting methods to analyze the relationship between parameters such as different pultrusion speeds, resin rheological properties, and fiber impregnation, thereby extracting key pressure change patterns. The system uses multivariate regression analysis technology to correlate the mold opening and closing state with pressure distribution, establish a pressure control strategy, and provide data support for dynamic mold design.

[0004] Existing technologies for collecting internal mold pressure data rely primarily on single sensor data. Limited by the number and location of local measurement points, they struggle to accurately reflect the mold's global pressure distribution, leading to the easy overlooking of localized pressure anomalies. Data analysis primarily relies on trend modeling of time-series data, but fails to effectively incorporate dynamic factors such as track spacing, resin flow, and kneading wheel speed. Lack of multivariable interaction analysis prevents accurate analysis of the impact of different production parameters on pressure distribution. Identifying abnormal pressures primarily relies on static pressure thresholds, making it difficult to accurately identify abnormal pressure gradient change rates. This leads to lags in anomaly detection and the risk of misjudgment. Regarding control strategies, existing methods lack precise adjustment mechanisms for specific regional anomalies, making it difficult to achieve coordinated optimization of track spacing adjustment and resin flow control. Control strategies are empirically based and lack data-driven intelligent analysis, impacting the accuracy of production control. Regarding pressure state monitoring, existing technologies fail to fully consider the impact of temperature and material curing state on pressure distribution, making it difficult to effectively identify pressure anomalies caused by curing anomalies or bubble accumulation. This makes it difficult to accurately control material flow during production and increases product quality instability. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a dynamic die pressure data analysis system for polyurethane pultrusion production.

[0006] In order to achieve the above-mentioned object, the present invention adopts the following technical solution: A dynamic die pressure data analysis system for polyurethane pultrusion production includes:

[0007] The mold pressure data capture module extracts the mold track opening and closing gap, kneading wheel linear speed, and resin injection flow rate based on the real-time pressure data of multiple pressure sensor nodes inside the polyurethane pultrusion mold. It then determines the signal validity of the pressure measurement value of each node to obtain a mold pressure spatiotemporal dataset.

[0008] The pressure gradient trend calculation module identifies the distribution intervals between adjacent nodes based on the mold pressure spatiotemporal data set, analyzes the local pressure gradient, extracts the pressure change data of the resin flow area, screens the abnormal areas, and obtains the local pressure gradient trend of the mold;

[0009] The mold abnormal pressure identification module analyzes the gradient change characteristics of the abnormal area based on the local pressure gradient trend of the mold, determines whether there is a trend of continuous pressure increase and decrease in the abnormal area, determines the scope of the abnormal area and the pressure change, and obtains the mold abnormal pressure distribution area;

[0010] The mold pressure adjustment data module calls the abnormal pressure distribution area of ​​the mold, analyzes the influence of the track opening and closing spacing on the local pressure change based on the pressure gradient trend of the abnormal area, identifies the injection rate adjustment range, and obtains the mold pressure control data.

[0011] As a further solution of the present invention, the mold pressure spatiotemporal data set includes pressure sensor node data, track opening and closing gap data, kneading wheel linear speed data, and resin injection flow data; the local pressure gradient trend of the mold includes pressure gradient value, pressure change rate, and resin flow area pressure characteristics; the abnormal pressure distribution area of ​​the mold includes abnormal area position, pressure continuous rising area, and pressure continuous falling area; the mold pressure control data includes track spacing adjustment direction, resin injection rate adjustment amplitude, and pressure balance change trend.

[0012] As a further solution of the present invention, the mold pressure data capturing module includes:

[0013] The pressure signal verification submodule uses real-time pressure data from multiple pressure sensor nodes inside the polyurethane pultrusion die to identify the difference in pressure measurement values ​​between a single node and adjacent nodes, determine whether the difference exceeds the pressure anomaly threshold, filter out abnormal node data, and obtain a valid pressure signal set;

[0014] The dynamic feature calculation submodule calls the effective pressure signal set, combines the track opening and closing gap, the kneading wheel linear speed and the resin injection flow rate, identifies the pressure fluctuation, signal change rate and pressure gradient between nodes in each node time series, and uses the formula:

[0015]

[0016] Obtain the pressure dynamic characteristic intensity distribution;

[0017] Where D represents the dynamic characteristic intensity of pressure, P i represents the pressure measurement value of the i-th node, P i-1 Represents the pressure measurement value of the i-1th node, V i represents the linear speed of the rolling wheel at the i-th node, S i represents the track opening and closing gap of the i-th node, Q i represents the resin injection flow rate of the i-th node, and n represents the total number of nodes;

[0018] The spatiotemporal data generation submodule calibrates the spatiotemporal position relationship of node pressure according to the pressure dynamic characteristic intensity distribution and the pressure signal time sequence of the corresponding time node to obtain the mold pressure spatiotemporal data set.

[0019] As a further solution of the present invention, the pressure gradient trend calculation module includes:

[0020] The local gradient identification submodule calculates the pressure difference between adjacent nodes based on the mold pressure spatiotemporal data set, identifies the local pressure gradient in combination with the distribution interval, performs directional analysis, screens out invalid data, and obtains local pressure gradient distribution data;

[0021] The gradient change rate analysis submodule calls the local pressure gradient distribution data and uses the formula:

[0022]

[0023] Calculate the gradient change rate, analyze the pressure gradient change trend with regional distribution, and obtain the gradient change trend data set;

[0024] Where G represents the gradient change rate, ΔP k Represents the pressure difference between the kth adjacent nodes, ΔP k-1 Represents the pressure difference between the k-1th adjacent nodes, Δd k Represents the distribution interval between the kth adjacent nodes, Δd k-1 represents the distribution interval between the k-1th adjacent nodes, and m represents the number of adjacent nodes in the calculation area;

[0025] The abnormal gradient area screening submodule calls the gradient change trend data set, analyzes the gradient change rate in the area, sets the pressure gradient abnormality threshold, screens the area exceeding the threshold, identifies the abnormal gradient change area, and obtains the local pressure gradient trend of the mold.

[0026] As a further solution of the present invention, the mold abnormal pressure identification module includes:

[0027] The pressure gradient identification submodule identifies the pressure change rate of adjacent nodes based on the local pressure gradient trend of the mold, and uses the formula:

[0028]

[0029] Calculate and obtain the rate of change of local pressure gradient;

[0030] Where L represents the rate of change of the local pressure gradient, ΔB j represents the pressure change value between the jth adjacent nodes, m represents the number of adjacent nodes in the calculation area, T max Represents the maximum pressure value in the area, T min Represents the minimum pressure value in the area;

[0031] The abnormal pressure determination submodule calls the local pressure gradient change rate, screens the area with abnormal gradient change rate according to the preset pressure change threshold, determines whether the area has the characteristics of continuous pressure change, and obtains the abnormal pressure area;

[0032] The abnormal pressure trend analysis submodule calls the abnormal pressure area, analyzes the pressure change trend in the area, identifies the situation of continuous pressure increase and decrease, and obtains the abnormal pressure distribution area of ​​the mold based on the pressure distribution in the area.

[0033] As a further solution of the present invention, the mold pressure adjustment data module includes:

[0034] The abnormal pressure area identification submodule calls the abnormal pressure distribution area of ​​the mold, analyzes the pressure gradient trend of each point in the area, screens the local pressure gradient mutation points, screens the core area of ​​abnormal pressure, identifies the imbalance of pressure distribution in the core area, and obtains the core pressure imbalance;

[0035] The track direction determination submodule identifies the track opening and closing spacing change rate based on the core pressure imbalance, extracts the pressure balance change under differentiated direction adjustment, and determines the optimal track adjustment direction;

[0036] The injection rate adjustment amplitude identification submodule calls the optimal track adjustment direction, identifies the pressure response change in the corresponding area, analyzes the impact of the track adjustment on the local pressure, and uses the formula based on the pressure gradient trend in the area:

[0037]

[0038] Calculate the injection rate adjustment range and analyze the change trend of pressure balance after adjustment to obtain mold pressure control data;

[0039] Among them, F represents the injection rate adjustment range, R x Represents the local area pressure value of the x-th pressure sampling point, R avg Represents the average surrounding pressure, G x represents the pressure gradient change rate of the xth pressure sampling point, and M represents the total number of pressure sampling points in the calculation area.

[0040] As a further solution of the present invention, the system also includes a multi-dimensional pressure state monitoring module:

[0041] The multi-dimensional pressure status monitoring module calls the mold pressure control data, extracts the temperature data of the infrared thermometer and the ultrasonic signal reflection characteristic data, analyzes the impact of temperature changes on the pressure distribution, determines whether there are abnormal curing and pressure anomalies caused by bubble accumulation, and combines the pressure gradient change and track spacing adjustment data to obtain the mold pressure dynamic status assessment results;

[0042] The mold pressure dynamic state evaluation results include temperature impact analysis, curing anomaly identification, and bubble aggregation identification.

[0043] As a further solution of the present invention, the multi-dimensional pressure state monitoring module includes:

[0044] The pressure data extraction submodule calls the mold pressure control data, extracts the track spacing adjustment data, the infrared thermometer temperature data and the ultrasonic signal reflection characteristic data, screens the pressure matching parameters, and obtains the pressure matching parameter set;

[0045] The temperature and pressure correlation analysis submodule calls the pressure matching parameter set, analyzes the relationship between temperature data and pressure gradient changes, analyzes the impact of temperature on pressure distribution, and determines abnormal solidification and bubble aggregation pressure using the formula:

[0046]

[0047] Calculate the temperature-affected pressure error value, filter out abnormal pressure distribution, and obtain the temperature-affected pressure distribution result;

[0048] Among them, E represents the pressure error value affected by temperature, U a Represents the track spacing adjustment data of area a, W a Represents the temperature data of the infrared thermometer in area a, V a Represents the ultrasonic signal reflection characteristic data of area a, Y a Represents the pressure gradient change data of the ath region, Z represents the total number of calculation regions, and a represents the region index;

[0049] The dynamic pressure assessment submodule calls the temperature-affected pressure distribution result, combines it with the track spacing adjustment data, analyzes the dynamic pressure gradient, identifies the impact of temperature changes and track adjustment amplitude on mold pressure stability, and obtains the mold pressure dynamic state assessment result.

[0050] Compared with the prior art, the advantages and positive effects of the present invention are:

[0051] In the present invention, high-precision real-time monitoring of the internal pressure of the mold is achieved through data capture of multi-node pressure sensors. Combined with cross-analysis of data from the track opening and closing gap, kneading wheel linear speed, and resin injection flow rate, the integrity and timeliness of the data are enhanced, avoiding the limitations of single pressure data. Based on the analysis of the changing trend of the pressure gradient, the pressure characteristics of different areas of the mold are identified, making the judgment of local anomalies more accurate. By extracting the pressure change rate, it is possible to distinguish normal pressure fluctuations from abnormal pressure accumulation, improve the identification ability of abnormal areas, and ensure the accurate locking of abnormal distribution areas. Combined with the dynamic regulation of the track spacing adjustment direction and the resin injection rate, a precise intervention mechanism for abnormal pressure areas is constructed, improving the response speed and accuracy of pressure regulation. By introducing temperature data and ultrasonic signal reflection characteristics, pressure anomalies caused by curing anomalies and bubble accumulation are identified, making pressure state monitoring cover a wider range of influencing factors, providing more comprehensive data support for pressure control in the production process. The combination of multiple innovative means improves the collection accuracy, analysis depth, and regulation capability of mold pressure data, making the detection and intervention of pressure anomalies in the production process more scientific and real-time, improving the intelligent level of mold pressure management, optimizing the resin flow state, and reducing the risk of quality fluctuations. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 is a system flow chart of the present invention;

[0053] Figure 2 This is a flow chart of the mold pressure data capture module in the present invention;

[0054] Figure 3 This is a flow chart of the pressure gradient trend calculation module in the present invention;

[0055] Figure 4 This is a flow chart of the mold abnormal pressure identification module in the present invention;

[0056] Figure 5 This is a flow chart of the mold pressure adjustment data module in the present invention;

[0057] Figure 6 This is a flow chart of the multi-dimensional pressure state monitoring module in the present invention. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0059] In the description of the present invention, it should be understood that the terms "length," "width," "up," "down," "front," "back," "left," "right," "vertical," "horizontal," "top," "bottom," "inside," "outside," and the like, indicating positions or relationships, are based on the positions or relationships shown in the accompanying drawings and are intended only to facilitate the description of the present invention and simplify the description. They do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limiting the present invention. Furthermore, in the description of the present invention, "plurality" means two or more, unless otherwise expressly and specifically defined.

[0060] See also Figure 1 The dynamic die pressure data analysis system for polyurethane pultrusion production includes:

[0061] The mold pressure data capture module extracts the mold track opening and closing gap, kneading wheel linear speed, and resin injection flow rate based on real-time pressure data from multiple pressure sensor nodes inside the polyurethane pultrusion mold. It then determines the signal validity of the pressure measurement value at each node, filters out abnormal data, and obtains a mold pressure spatiotemporal dataset.

[0062] The pressure gradient trend calculation module, based on the mold pressure spatiotemporal dataset, identifies the local pressure gradient of a single segment according to the distribution interval between adjacent nodes, extracts the pressure change data in the resin flow area, analyzes the changing trend of the pressure gradient in adjacent areas, identifies the pressure distribution characteristics of the mold track contact area and the resin injection area, and screens areas with abnormal gradient change rates to obtain the local pressure gradient trend of the mold.

[0063] The mold abnormal pressure identification module extracts the pressure change rate data of adjacent nodes based on the local pressure gradient trend of the mold, analyzes the gradient change characteristics of the abnormal area, determines whether there is a trend of continuous pressure increase and decrease in the abnormal area, determines the scope of the abnormal area and the pressure change situation, and obtains the abnormal pressure distribution area of ​​the mold;

[0064] The mold pressure adjustment data module calls the mold abnormal pressure distribution area, analyzes the impact of the track opening and closing spacing on the local pressure change based on the pressure gradient trend of the abnormal area, determines the adjustment direction of the regional track, identifies the injection rate adjustment range, analyzes the change trend of the pressure balance after adjustment, and obtains the mold pressure control data;

[0065] The multi-dimensional pressure status monitoring module calls the mold pressure control data, extracts the temperature data of the infrared thermometer and the ultrasonic signal reflection characteristic data, analyzes the impact of temperature changes on the pressure distribution, and determines whether there are abnormal pressure conditions caused by curing abnormalities and bubble aggregation. Combined with the pressure gradient change and track spacing adjustment data, the dynamic state evaluation results of the mold pressure are obtained.

[0066] The mold pressure spatiotemporal dataset includes pressure sensor node data, track opening and closing gap data, kneading wheel linear speed data, and resin injection flow data. The local pressure gradient trend of the mold includes the pressure gradient value, pressure change rate, and pressure characteristics of the resin flow area. The abnormal pressure distribution area of ​​the mold includes the abnormal area location, the area with continuous pressure increase, and the area with continuous pressure decrease. The mold pressure control data includes the track spacing adjustment direction, the resin injection rate adjustment amplitude, and the pressure balance change trend. The mold pressure dynamic state evaluation results include temperature impact analysis, curing anomaly identification, and bubble aggregation identification.

[0067] See also Figure 2 , the mold pressure data capture module includes:

[0068] The pressure signal verification submodule uses real-time pressure data from multiple pressure sensor nodes inside the polyurethane pultrusion die to identify the difference in pressure measurement values ​​between a single node and adjacent nodes, determine whether the difference exceeds the pressure anomaly threshold, filter out abnormal node data, and obtain a valid pressure signal set;

[0069] By comparing the difference in pressure measurement values ​​between a single node and its adjacent nodes, for example, in an actual production scenario, if the measurement value of node 5 suddenly exceeds the average value of its adjacent nodes 4 and 6 by more than 10%, it is preliminarily determined to be an anomaly. This difference is achieved by measuring adjacent nodes and calculating the difference in their pressure values. Each node performs this comparison calculation to screen out normal pressure data. This process avoids the errors caused by direct reliance on a single node data, ensures the authenticity and accuracy of the data, and ultimately obtains a valid pressure signal set that contains all node data verified to be normal.

[0070] The dynamic feature calculation submodule calls the effective pressure signal set, combines the track opening and closing gap, the kneading wheel linear speed and the resin injection flow rate, and identifies the pressure fluctuation, signal change rate and pressure gradient between nodes in each node time series using the formula:

[0071]

[0072] Obtain the pressure dynamic characteristic intensity distribution;

[0073] Where D represents the dynamic characteristic intensity of pressure, P i represents the pressure measurement value of the i-th node, P i-1 Represents the pressure measurement value of the i-1th node, V i represents the linear speed of the rolling wheel at the i-th node, S i represents the track opening and closing gap of the i-th node, Q i represents the resin injection flow rate of the i-th node, and n represents the total number of nodes;

[0074] Combined with the production parameters such as track opening and closing gap, kneading wheel linear speed and resin injection flow rate, the dynamic characteristic strength calculation is carried out for the pressure data of each node in the polyurethane pultrusion process. The effective pressure data of nodes 1 to 5 are selected. It is assumed that the pressure measurement values ​​of each node are P1 = 50, P2 = 55, P3 = 52, P4 = 60, P5 = 58kPa, and the kneading wheel linear speed V i They are V1=0.8, V2=0.9, V3=0.85, V4=0.88, V5=0.9m / s, and the track opening and closing clearance S i They are S1=1.2, S2=1.1, S3=1.3, S4=1.0, S5=1.2 mm, and the resin injection flow rate Q i They are Q1=15, Q2=16, Q3=14, Q4=17, Q5=15L / min respectively;

[0075] Among them, P0 = 50kPa (set as the reference value of the 0th node), calculate the molecular part of each node in turn

[0076]

[0077] Node 1:

[0078] Node 2:

[0079] Node 3:

[0080] Node 4:

[0081] Node 5:

[0082] Then divide the numerator of each node by the corresponding Q i +1 to calculate D:

[0083] Node 1: D1 = 1.095 / (15 + 1) = 0.068;

[0084] Node 2: D2 = 5.548 / (16+1) = 0.326;

[0085] Node 3: D3 = 3.69 / (14+1) = 0.246;

[0086] Node 4: D4 = 8.04 / (17+1) = 0.447;

[0087] Node 5: D5 = 2.895 / (15 + 1) = 0.181;

[0088] The final calculated pressure dynamic characteristic intensities of the five nodes are 0.068, 0.326, 0.246, 0.447, and 0.181, respectively, forming a complete pressure dynamic characteristic intensity distribution, which reflects the pressure dynamic characteristics of each node during the production process.

[0089] The benefit of the formula is that by combining the four key participating items of pressure change rate, kneading wheel linear speed, track opening and closing gap, and resin injection flow rate, it comprehensively reflects the dynamic characteristics of node pressure in the pultrusion process. It can describe the node pressure behavior in multiple dimensions without relying on a single variable. The pressure dynamic characteristic intensity of node 4, 0.447, is the largest among all nodes, reflecting that the pressure fluctuations at this node are more severe during the production process, indicating that this position is a pressure-sensitive area in actual operation and requires key monitoring.

[0090] The spatiotemporal data generation submodule calibrates the spatiotemporal position relationship of node pressure based on the pressure dynamic characteristic intensity distribution and the pressure signal time series of the corresponding time node to obtain the mold pressure spatiotemporal data set;

[0091] The key to combining data with time series and establishing a multidimensional data matrix is ​​to match each moment of the time series with the pressure data of each node. For example, in a specific operation, if the dynamic characteristic intensity of the pressure displayed by node 10 is 20 at the 5th second, this will be reflected in the multidimensional data matrix as (10, 5 seconds, 20). Through this method, the pressure status of each node at each time point can be clearly seen, which provides a complete spatiotemporal view for analyzing the pressure changes in the entire production process. Finally, the spatiotemporal data set of mold pressure is obtained, which provides a basis for further data analysis and production process monitoring.

[0092] See also Figure 3 , the pressure gradient trend calculation module includes:

[0093] The local gradient identification submodule calculates the pressure difference between adjacent nodes based on the mold pressure spatiotemporal dataset, identifies the local pressure gradient based on the distribution interval, performs directional analysis, and eliminates invalid data to obtain the local pressure gradient distribution data.

[0094] On a polyurethane production line, the process of calculating the pressure difference between adjacent nodes involves the use of real-time monitoring equipment. The equipment can continuously collect pressure data between two points on the mold. After de-noising and calibration, the data is recorded at fixed time intervals. Real-time data monitoring is useful for predicting mold wear or blockage. By comparing and calculating the pressure difference between adjacent nodes, the operating parameters of the production line can be dynamically adjusted to optimize production efficiency and product quality. By aggregating differential data and applying directional analysis, engineers can identify areas with abnormal pressure distribution, thereby pre-adjusting production parameters or performing necessary maintenance to prevent equipment failure or product quality degradation, and ultimately obtain local pressure gradient distribution data.

[0095] The gradient change rate analysis submodule calls the local pressure gradient distribution data and uses the formula:

[0096]

[0097] Calculate the gradient change rate, analyze the pressure gradient change trend with regional distribution, and obtain the gradient change trend data set;

[0098] Where G represents the gradient change rate, ΔP k Represents the pressure difference between the kth adjacent nodes, ΔP k-1 Represents the pressure difference between the k-1th adjacent nodes, Δd k Represents the distribution interval between the kth adjacent nodes, Δd k-1 represents the distribution interval between the k-1th adjacent nodes, and m represents the number of adjacent nodes in the calculation area;

[0099] Parameter analysis of the formula:

[0100] G t : Gradient change rate, which measures the change in mold pressure gradient and helps monitor the uniformity of resin flow in the mold;

[0101] ΔP k : The pressure difference between the kth adjacent nodes (unit: Pa), measured by the pressure sensor;

[0102] Δd k : The distance between the kth adjacent nodes (unit: cm), determined according to the mold design;

[0103] m: the number of adjacent nodes in the calculation area, that is, the number of data points;

[0104] Assume that the pressure data measured by adjacent pressure sensors in a certain area of ​​the mold are as follows:

[0105] P1=5000Pa, P2=5100Pa, P3=5050Pa, P4=5200Pa;

[0106] The corresponding node distances are: Δd1 = Δd2 = Δd3 = 10 cm;

[0107] Calculate the pressure gradient between adjacent nodes:

[0108]

[0109] Calculate the gradient change between adjacent nodes:

[0110]

[0111] Gradient change rate calculation:

[0112] During the polyurethane pultrusion production process, fluctuations in the pressure gradient directly affect the flow stability and curing quality of the resin. Real-time calculation of the pressure gradient change rate G is of great significance. The above calculation process shows that when the G value in a certain area is large, for example 17.5Pa / cm, it means that the pressure gradient in this area changes rapidly, which means that the resin flow is obstructed or the mold surface is contaminated, resulting in uneven flow rate. This situation will lead to local uneven curing, thereby affecting the mechanical properties of the final product. In practical applications, engineers need to set a reasonable threshold. For example, if G exceeds 15Pa / cm, it is necessary to check whether there is contamination on the mold surface or adjust the resin injection rate to maintain flow stability. On an automated production line, the pressure sensor can be connected through software so that the calculation process is carried out in real time. When the set threshold is exceeded, an alarm is automatically triggered or the control parameters are adjusted to reduce production instability. Ultimately, a gradient change trend data set is obtained to provide a basis for production decision-making.

[0113] The abnormal gradient area screening submodule calls the gradient change trend data set, analyzes the gradient change rate within the area, sets the pressure gradient abnormality threshold, screens the areas exceeding the threshold, identifies the abnormal gradient change areas, and obtains the local pressure gradient trend of the mold;

[0114] A pressure gradient abnormality threshold is set. The threshold is derived based on historical data and statistical analysis. It can quantitatively indicate what level of pressure gradient change rate is within the normal range and what is abnormal, because abnormal pressure gradients indicate potential production problems, such as mold blockage or poor resin curing. By setting a reasonable threshold and conducting in-depth analysis of areas exceeding the threshold, process parameters can be adjusted or equipment maintenance can be performed in a timely manner to ensure product quality and production safety. Identifying areas of abnormal gradient changes helps to optimize the stability and efficiency of the entire production process and obtain the local pressure gradient trend of the mold.

[0115] See also Figure 4 , the mold abnormal pressure identification module includes:

[0116] The pressure gradient identification submodule identifies the pressure change rate of adjacent nodes based on the local pressure gradient trend of the mold. By comparing the pressure values ​​of adjacent nodes, the formula is used:

[0117]

[0118] Calculate and obtain the rate of change of local pressure gradient;

[0119] Where L represents the rate of change of the local pressure gradient, ΔB j represents the pressure change value between the jth adjacent nodes, m represents the number of adjacent nodes in the calculation area, T max Represents the maximum pressure value in the area, T min Represents the minimum pressure value in the area;

[0120] Multiple sensors are used to collect real-time pressure data at different nodes. The data comes from various key points of the mold in order to analyze the dynamic changes in pressure within the mold. After obtaining the pressure data of each adjacent node, the pressure change rate between adjacent nodes is calculated. For example, on a 3×3 node grid, assume the initial pressure data is as follows (unit: MPa):

[0121] After the pressure value of each node is recorded by the sensor, the pressure change rate of the adjacent nodes is calculated. That is, the calculation is performed for each pair of adjacent nodes. For example, in the first row, the pressure changes between the nodes are:

[0122] ΔB1=B2-B1=2.1-2.0=0.1;

[0123] ΔB2=B3-B2=2.2-2.1=0.1;

[0124] Calculate the second row:

[0125] ΔB3=B5-B4=2.5-2.3=0.2;

[0126] ΔB4=B6-B5=2.6-2.5=0.1;

[0127] Calculate the third row:

[0128] ΔB5=B8-B7=2.9-2.7=0.2;

[0129] ΔB6=B9-B8=3.0-2.9=0.1;

[0130] Then, take the absolute value of the pressure change of all adjacent nodes and sum them:

[0131]

[0132] Assuming the total number of nodes m = 6, the first part of the local pressure change rate is calculated as follows:

[0133] At the same time, the maximum pressure T in the calculation area max =3.0MPa, minimum pressure T min =2.0MPa, then calculate the pressure adjustment factor for the second part:

[0134] The final calculated local pressure gradient change rate is: L = 0.1333 × 0.2 = 0.0267;

[0135] This calculated value shows that the overall change rate of the pressure gradient in this area is 0.0267. If this value exceeds the set threshold (for example, 0.05), an abnormality occurs in this area. Further analysis of the pressure change trend can be performed to obtain the local pressure gradient change rate as the basic data for subsequent analysis.

[0136] The abnormal pressure determination submodule calls the local pressure gradient change rate, screens the area with abnormal gradient change rate according to the preset pressure change threshold, determines whether the area has the characteristics of continuous pressure change, and obtains the abnormal pressure area;

[0137] Analyze data to determine whether there are abnormal pressure areas in the mold. During the specific execution process, the local pressure gradient change rate is compared with the set pressure change threshold. For example, if the threshold is set to 0.05, all areas with gradient change rates higher than this value are considered abnormal. The area indicates failure or damage to the mold. This method can timely identify and respond to potential risks to protect production safety and equipment integrity. For the specific example of this area, if the gradient change rates of three consecutive nodes in a specific area of ​​the mold exceed 0.05, the area will be marked as an abnormal pressure area, and then detailed monitoring will be carried out in the area for further analysis and disposal. This monitoring and identification mechanism is a key step in preventing mold failures and optimizing the production process, and ultimately obtaining abnormal pressure areas.

[0138] The abnormal pressure trend analysis submodule calls the abnormal pressure area, analyzes the pressure change trend in the area, identifies the continuous increase and decrease of pressure, and combines the pressure distribution in the area to obtain the abnormal pressure distribution area of ​​the mold;

[0139] A detailed analysis of regional pressure change trends is performed to determine whether there is a continuous upward or downward trend, and the risk and fault type are judged accordingly. For example, if the pressure in a certain area continues to rise during consecutive production cycles, this indicates that a part of the mold is experiencing excessive pressure or is about to be damaged. This analysis helps engineers take timely measures, such as adjusting the mold's operating parameters or performing necessary maintenance to avoid accidents or equipment damage. In a specific embodiment, if an area marked as abnormal continues to show an upward pressure trend within the next five production cycles, a detailed inspection is prompted and shutdown maintenance is recommended. The core of this analysis process is to effectively predict and prevent potential risks by continuously tracking pressure data, thereby maintaining the stable operation of the production line and ultimately obtaining the abnormal pressure distribution area of ​​the mold.

[0140] See also Figure 5 , the mold pressure adjustment data module includes:

[0141] The abnormal pressure area identification submodule calls the mold abnormal pressure distribution area, analyzes the pressure gradient trend of each point in the area, screens the local pressure gradient mutation points, screens the pressure abnormal core area, identifies the imbalance of the pressure distribution in the core area, and obtains the core pressure imbalance;

[0142] By accurately calculating the pressure gradient trend at each point, local anomalies can be identified. For example, in a production cycle, the sensor records the pressure data of each point in the mold. The data is analyzed to determine which areas have pressure gradients exceeding the normal operating range. The area is the core area of ​​the pressure anomaly. The average pressure value of the abnormal core area and the average pressure value of the surrounding normal areas are calculated, and the difference between the two is compared to see if it exceeds the preset pressure anomaly judgment threshold. If so, it is confirmed as the core area of ​​pressure anomaly. This method can monitor and adjust the pressure distribution in the production process in real time, ensure the stability of product quality, and obtain the core pressure imbalance.

[0143] The track direction judgment submodule identifies the track opening and closing spacing change rate based on the core pressure imbalance, extracts the pressure balance change under differentiated direction adjustment, and determines the optimal track adjustment direction;

[0144] First, based on the core pressure imbalance provided by the abnormal pressure area identification submodule, the impact of different track opening and closing spacing on pressure balance is evaluated. In actual operation, for example, after adjusting the spacing of a certain track, the pressure distribution diagram before and after the adjustment is calculated using simulation software to obtain the track opening and closing spacing change rate. By calculating the change in pressure balance corresponding to different adjustment directions, the track adjustment direction that is most conducive to pressure balance is determined by comparing the changes. The precise execution of the process can significantly improve the product surface quality and structural integrity, and determine the optimal track adjustment direction.

[0145] The injection rate adjustment amplitude identification submodule calls the optimal track adjustment direction, identifies the pressure response change in the corresponding area, analyzes the impact of track adjustment on local pressure, and uses the formula based on the pressure gradient trend in the area:

[0146]

[0147] Calculate the injection rate adjustment range and analyze the change trend of pressure balance after adjustment to obtain mold pressure control data;

[0148] Among them, F represents the injection rate adjustment range, R x Represents the local area pressure value of the x-th pressure sampling point, R avg Represents the average surrounding pressure, G x represents the pressure gradient change rate of the x-th pressure sampling point, and M represents the total number of pressure sampling points in the calculation area;

[0149] Adjust the injection rate to optimize the pressure distribution. The process relies on accurate calculations and experimental data. The optimal track adjustment direction is called, and the track opening and closing spacing is adjusted based on this direction. The pressure change values ​​in different areas inside the mold are monitored in real time. High-precision sensors are used to record pressure data. Each area has 5 pressure measurement points. The current regional pressure distribution is obtained through data acquisition, and the local pressure response change R is calculated. x and the average pressure R in the surrounding area avg For example, if the pressure data of a certain area ranges from 250kPa to 400kPa, the calculated average pressure R avg is 320kPa, analyze the effect of track adjustment on local pressure, and calculate the pressure gradient change rate G in this area x , assuming the gradient range is between 0.5 kPa / mm and 1.8 kPa / mm, calculate the pressure compensation required for injection rate adjustment;

[0150] Substitute specific values ​​for calculation:

[0151] Set the number of pressure measurement points n=5;

[0152] Select the actual values ​​of some pressure measurement points (unit: kPa):

[0153] R1=250, R2=270, R3=300, R4=350, R5=400;

[0154] Corresponding gradient change rate (unit: kPa / mm):

[0155] G1=0.8, G2=1.2, G3=1.5, G4=0.9, G5=1.1;

[0156] Calculate the molecular part:

[0157] Calculate the denominator:

[0158] Calculate the final result:

[0159] Negative values ​​indicate that the injection rate needs to be reduced to reduce localized excessive pressure, prevent product deformation, and prevent uneven material distribution. Based on the calculation results, the injection rate should be adjusted by approximately 5.37 kPa. Further analysis of the mold pressure equilibrium trend is then performed to obtain mold pressure control data.

[0160] By comparing the local pressure value with the average pressure value of the surrounding area and incorporating the square root of the pressure gradient as a weighting factor, the calculated adjustment range is more consistent with the actual situation of mold pressure changes. Compared with the traditional simple averaging method, this method can more accurately evaluate local high or low pressure points, thereby more precisely adjusting the injection rate to optimize the pressure distribution inside the mold and improve product quality.

[0161] See also Figure 6 , the multi-dimensional pressure status monitoring module includes:

[0162] The pressure data extraction submodule calls the mold pressure control data, extracts the track spacing adjustment data, infrared thermometer temperature data and ultrasonic signal reflection characteristic data, screens the pressure matching parameters, and obtains the pressure matching parameter set;

[0163] The pressure data of each key position of the mold is extracted. The process involves real-time monitoring and recording of data from each production cycle from the control system, and quantitative analysis of the pressure distribution characteristics of each part inside the mold through real-time data. Through this method, the pressure state of the mold at different production stages and the direct relationship between the state and product quality can be clarified. For example, when producing polyurethane foam boards, pressure data can indicate whether the material is evenly distributed, thereby preventing density unevenness problems in the product. The reflection characteristic data of the infrared thermometer and ultrasonic signal are also collected synchronously. The data reflects the temperature and curing state of the material, which in turn affects the pressure distribution of the mold. Matching and analyzing the data with the adjustment data of the track spacing can optimize the production process and ensure that the quality standards of each batch of products are consistent. By integrating the data, a pressure matching parameter set is obtained. The results help to further optimize production parameters and improve product quality.

[0164] The temperature and pressure correlation analysis submodule calls the pressure matching parameter set, analyzes the relationship between temperature data and pressure gradient changes, analyzes the impact of temperature on pressure distribution, and determines abnormal solidification and bubble accumulation pressure using the formula:

[0165]

[0166] Calculate the temperature-affected pressure error value, filter out abnormal pressure distribution, and obtain the temperature-affected pressure distribution result;

[0167] Among them, E represents the pressure error value affected by temperature, U a Represents the track spacing adjustment data of area a, W a Represents the temperature data of the infrared thermometer in area a, V a Represents the ultrasonic signal reflection characteristic data of area a, Y a Represents the pressure gradient change data of the ath region, Z represents the total number of calculation regions, and a represents the region index;

[0168] Analyze the correlation between pressure data and temperature data to obtain track spacing adjustment data U a , the temperature data W of the infrared thermometer a , ultrasonic signal reflection characteristic data V a And the pressure gradient change data Y a , divide the data according to the production area, and calculate the pressure gradient changes under different temperature ranges to determine the impact of temperature changes on pressure distribution. During the production process, the temperature distribution inside the mold is not uniform, but presents a certain gradient with the curing of the material, track adjustment and changes in ambient temperature. For example, in the polyurethane pultrusion process, the temperature gradually drops from 100°C at the inlet to 40°C at the outlet, resulting in different curing rates of the material at different locations. If the temperature is too high, it will cause the material to cure too quickly locally, forming internal stress and affecting the strength of the final product. If the temperature is too low, it will lead to insufficient curing and affect the durability of the product. Therefore, it is necessary to establish a mathematical model to quantify the degree of influence of temperature on pressure and to determine whether there are problems of curing abnormalities and bubble aggregation;

[0169] Set the calculation parameters as follows:

[0170] In a certain area a=1, track spacing adjustment data U1=0.5mm, infrared thermometer temperature W1=90°C, ultrasonic signal reflection characteristic data V1=2.3, pressure gradient change data Y1=1.2MPa / m;

[0171] In another area a = 2, track spacing adjustment data U2 = 0.7 mm, temperature W2 = 85°C, ultrasonic signal V2 = 2.1, and pressure gradient change Y2 = 1.5 MPa / m;

[0172] The total number of calculation areas Z = 2;

[0173] Substitute into the formula to calculate:

[0174]

[0175] The calculated temperature-affected pressure error value E=36.68. The value shows that under the current temperature and pressure distribution, there is an obvious temperature-affected pressure abnormal area. This value can be further used to set a reasonable temperature adjustment plan to optimize the heating range of the mold and adjust the track spacing to avoid defects caused by local overheating or insufficient cooling. By calculating the E value of different areas, a dynamic curve of the temperature-affected pressure distribution can be established, and abnormal areas can be screened out. Finally, the temperature-affected pressure distribution result is obtained to guide the subsequent mold pressure control optimization.

[0176] The dynamic pressure assessment submodule uses the temperature-affected pressure distribution results, combines them with the track spacing adjustment data, analyzes the dynamic pressure gradient, identifies the impact of temperature changes and track adjustment amplitude on mold pressure stability, and obtains the mold pressure dynamic state assessment results;

[0177] Combined with the track spacing adjustment data, the dynamic gradient of mold pressure was analyzed. The analysis involved the pressure changes of the mold under different working conditions, especially in the material curing and molding stages. Different temperature and pressure gradients will directly affect the structure and performance of the product. For example, during the production process, if the pressure gradient in a certain area is detected and does not match the data after the track spacing adjustment, it is necessary to re-evaluate the material distribution or the temperature control setting of the mold. Through real-time monitoring and analysis of the data, the production parameters can be adjusted in time to avoid problems such as uneven structure or unstable performance of the finished product. The results of the temperature effect on pressure distribution are obtained. The results provide important feedback information for the production process, ensuring the consistency of product quality and the optimization of production efficiency.

[0178] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. Dynamic die pressure data analysis system for polyurethane pultrusion production, characterized by: The system comprises: The mold pressure data capture module extracts the mold track opening and closing gap, kneading wheel linear speed, and resin injection flow rate based on the real-time pressure data of multiple pressure sensor nodes inside the polyurethane pultrusion mold. It then determines the signal validity of the pressure measurement value of each node to obtain a mold pressure spatiotemporal dataset. The pressure gradient trend calculation module identifies the distribution intervals between adjacent nodes based on the mold pressure spatiotemporal data set, analyzes the local pressure gradient, extracts the pressure change data of the resin flow area, screens the abnormal areas, and obtains the local pressure gradient trend of the mold; The mold abnormal pressure identification module analyzes the gradient change characteristics of the abnormal area based on the local pressure gradient trend of the mold, determines whether there is a trend of continuous pressure increase and decrease in the abnormal area, determines the scope of the abnormal area and the pressure change, and obtains the mold abnormal pressure distribution area; The mold pressure adjustment data module calls the abnormal pressure distribution area of ​​the mold, analyzes the influence of the track opening and closing spacing on the local pressure change based on the pressure gradient trend of the abnormal area, identifies the injection rate adjustment range, and obtains the mold pressure control data.

2. The dynamic die pressure data analysis system for polyurethane pultrusion production according to claim 1, characterized in that: The mold pressure spatiotemporal dataset includes pressure sensor node data, track opening and closing gap data, kneading wheel linear speed data, and resin injection flow data. The local pressure gradient trend of the mold includes the pressure gradient value, pressure change rate, and resin flow area pressure characteristics. The abnormal pressure distribution area of ​​the mold includes the abnormal area position, the pressure continuous rising area, and the pressure continuous falling area. The mold pressure control data includes the track spacing adjustment direction, the resin injection rate adjustment amplitude, and the pressure balance change trend.

3. The dynamic die pressure data analysis system for polyurethane pultrusion production according to claim 1, characterized in that: The mold pressure data capture module includes: The pressure signal verification submodule uses real-time pressure data from multiple pressure sensor nodes inside the polyurethane pultrusion die to identify the difference in pressure measurement values ​​between a single node and adjacent nodes, determine whether the difference exceeds the pressure anomaly threshold, filter out abnormal node data, and obtain a valid pressure signal set; The dynamic feature calculation submodule calls the effective pressure signal set, combines the track opening and closing gap, the kneading wheel linear speed and the resin injection flow rate, identifies the pressure fluctuation, signal change rate and pressure gradient between nodes in each node time series, and uses the formula: Obtain the pressure dynamic characteristic intensity distribution; Where D represents the dynamic characteristic intensity of pressure, P i represents the pressure measurement value of the i-th node, P i-1 Represents the pressure measurement value of the i-1th node, V i Represents the linear speed of the rolling wheel at the i-th node, S i represents the track opening and closing gap of the i-th node, Q i represents the resin injection flow rate of the i-th node, and n represents the total number of nodes; The spatiotemporal data generation submodule calibrates the spatiotemporal position relationship of node pressure according to the pressure dynamic characteristic intensity distribution and the pressure signal time sequence of the corresponding time node to obtain the mold pressure spatiotemporal data set.

4. The dynamic die pressure data analysis system for polyurethane pultrusion production according to claim 3, characterized in that: The pressure gradient trend calculation module includes: The local gradient identification submodule calculates the pressure difference between adjacent nodes based on the mold pressure spatiotemporal data set, identifies the local pressure gradient in combination with the distribution interval, performs directional analysis, screens out invalid data, and obtains local pressure gradient distribution data; The gradient change rate analysis submodule calls the local pressure gradient distribution data and uses the formula: Calculate the gradient change rate, analyze the pressure gradient change trend with regional distribution, and obtain the gradient change trend data set; Where G represents the gradient change rate, ΔP k Represents the pressure difference between the kth adjacent nodes, ΔP k-1 Represents the pressure difference between the k-1th adjacent nodes, Δd k Represents the distribution interval between the kth adjacent nodes, Δd k-1 represents the distribution interval between the k-1th adjacent nodes, and m represents the number of adjacent nodes in the calculation area; The abnormal gradient area screening submodule calls the gradient change trend data set, analyzes the gradient change rate in the area, sets the pressure gradient abnormality threshold, screens the area exceeding the threshold, identifies the abnormal gradient change area, and obtains the local pressure gradient trend of the mold.

5. The dynamic die pressure data analysis system for polyurethane pultrusion production according to claim 4, characterized in that: The mold abnormal pressure identification module includes: The pressure gradient identification submodule identifies the pressure change rate of adjacent nodes based on the local pressure gradient trend of the mold, and uses the formula: Calculate and obtain the rate of change of local pressure gradient; Where L represents the rate of change of the local pressure gradient, ΔB j represents the pressure change value between the jth adjacent nodes, m represents the number of adjacent nodes in the calculation area, T max Represents the maximum pressure value in the area, T min Represents the minimum pressure value in the area; The abnormal pressure determination submodule calls the local pressure gradient change rate, screens the area with abnormal gradient change rate according to the preset pressure change threshold, determines whether the area has the characteristics of continuous pressure change, and obtains the abnormal pressure area; The abnormal pressure trend analysis submodule calls the abnormal pressure area, analyzes the pressure change trend in the area, identifies the situation of continuous pressure increase and decrease, and obtains the abnormal pressure distribution area of ​​the mold based on the pressure distribution in the area.

6. The dynamic die pressure data analysis system for polyurethane pultrusion production according to claim 5, characterized in that: The mold pressure adjustment data module includes: The abnormal pressure area identification submodule calls the abnormal pressure distribution area of ​​the mold, analyzes the pressure gradient trend of each point in the area, screens the local pressure gradient mutation points, screens the core area of ​​abnormal pressure, identifies the imbalance of pressure distribution in the core area, and obtains the core pressure imbalance; The track direction determination submodule identifies the track opening and closing spacing change rate based on the core pressure imbalance, extracts the pressure balance change under differentiated direction adjustment, and determines the optimal track adjustment direction; The injection rate adjustment amplitude identification submodule calls the optimal track adjustment direction, identifies the pressure response change in the corresponding area, analyzes the impact of the track adjustment on the local pressure, and uses the formula based on the pressure gradient trend in the area: Calculate the injection rate adjustment range and analyze the change trend of pressure balance after adjustment to obtain mold pressure control data; Among them, F represents the injection rate adjustment range, R x Represents the local area pressure value of the x-th pressure sampling point, R avg Represents the average surrounding pressure, G x represents the pressure gradient change rate of the xth pressure sampling point, and M represents the total number of pressure sampling points in the calculation area.

7. The dynamic die pressure data analysis system for polyurethane pultrusion production according to claim 1, characterized in that: The system also includes a multi-dimensional pressure status monitoring module: The multi-dimensional pressure status monitoring module calls the mold pressure control data, extracts the temperature data of the infrared thermometer and the ultrasonic signal reflection characteristic data, analyzes the impact of temperature changes on the pressure distribution, determines whether there are abnormal curing and pressure anomalies caused by bubble accumulation, and combines the pressure gradient change and track spacing adjustment data to obtain the mold pressure dynamic status assessment results; The mold pressure dynamic state evaluation results include temperature impact analysis, curing anomaly identification, and bubble aggregation identification.

8. The dynamic die pressure data analysis system for polyurethane pultrusion production according to claim 7, characterized in that: The multi-dimensional pressure state monitoring module includes: The pressure data extraction submodule calls the mold pressure control data, extracts the track spacing adjustment data, the infrared thermometer temperature data and the ultrasonic signal reflection characteristic data, screens the pressure matching parameters, and obtains the pressure matching parameter set; The temperature and pressure correlation analysis submodule calls the pressure matching parameter set, analyzes the relationship between temperature data and pressure gradient changes, analyzes the impact of temperature on pressure distribution, and determines abnormal solidification and bubble aggregation pressure using the formula: Calculate the temperature-affected pressure error value, filter out abnormal pressure distribution, and obtain the temperature-affected pressure distribution result; Among them, E represents the pressure error value affected by temperature, U a Represents the track spacing adjustment data of area a, W a Represents the temperature data of the infrared thermometer in area a, V a Represents the ultrasonic signal reflection characteristic data of area a, Y a Represents the pressure gradient change data of the ath region, Z represents the total number of calculation regions, and a represents the region index; The dynamic pressure assessment submodule calls the temperature-affected pressure distribution result, combines it with the track spacing adjustment data, analyzes the dynamic pressure gradient, identifies the impact of temperature changes and track adjustment amplitude on mold pressure stability, and obtains the mold pressure dynamic state assessment result.

Images