Cement pole intelligent production manufacturing management system and method

Abstract

The present application relates to the technical field of cement pole production, in particular to a cement pole intelligent production and manufacturing management system and method, comprising a mold state acquisition module, which is used for digitally acquiring the inner diameter ovality of the mold from the self-supporting end to the spigot end and the joint misalignment amount after the completion of the centrifugal process, and outputting the mold geometric state parameter set; a wall thickness prediction module, which is used for acquiring the geometric state parameter sequence and the wall thickness distribution measured data of the mold after demolding in the historical production cycle of the mold used for the current batch, establishing a wall thickness coupling mapping model between the geometric state parameter set and the wall thickness distribution deviation, and determining the mixing ratio adjustment amount and the centrifugal forming parameter adjustment amount according to the wall thickness distribution deviation prediction value respectively, generating the mixing ratio correction instruction and the centrifugal forming correction instruction to be sent to the mixing station and the centrifugal forming equipment, so as to control the mixing station and the centrifugal forming equipment to perform mixing, pouring and centrifugal forming according to the corrected parameters, and then send the formed product into the steam curing kiln.

Description

Technical Field

[0001] This invention relates to the field of cement pole production technology, specifically to an intelligent production management system and method for cement poles. Background Technology

[0002] The manufacturing process of cement poles consists of concrete mixing, pouring into molds, centrifugal molding, steam curing, and demolding. In the centrifugal molding process, the ovality of the mold socket and spigot and the misalignment of the joint directly affect the uniformity of the concrete distribution space, thus determining the wall thickness distribution. In the steam curing process, the temperature history and mix proportion parameters jointly determine the compressive strength development rate. When demolding, it is necessary to comprehensively evaluate the structural uniformity and strength reserve to avoid damage to the pole.

[0003] In current production management, the geometric state of molds is generally inspected manually after machine shutdown, with data primarily based on paper records, limiting their involvement in online process control. Wall thickness control often relies on fixed centrifugal parameters and mix proportion process cards, with adjustments made based on experience. The causal relationship between mold ellipticity, joint misalignment, and wall thickness distribution lacks quantifiable methods. The steam curing process is executed according to a set time-temperature regime, and demolding timing is based on empirical values ​​of the curing cycle. The impact of temperature control fluctuations and mix proportion differences on strength development is difficult to quantify online. Demolding scheduling is usually arranged manually according to the kiln entry sequence, and the coupling relationship between wall thickness uniformity, mold roundness deterioration, and strength reserve is insufficiently reflected in decision-making.

[0004] The above production method has the following limitations: mold geometry data and wall thickness quality control data belong to different stages, historical data is difficult to trace by pairing the same mold, and the quantitative impact of mold deterioration on wall thickness deviation has been lacking for a long time; demolding judgment focuses on strength as a single dimension, and the stress concentration risk caused by excessive ellipticity or uneven wall thickness distribution is easily masked, and cracking or deformation of the rod often occurs. Therefore, the present invention provides an intelligent production and manufacturing management system and method for cement poles. Summary of the Invention

[0005] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0006] The technical solution adopted by this invention to solve its technical problem is: an intelligent production and manufacturing management system for cement poles, comprising: The mold status acquisition module is used to digitally acquire the inner diameter ovality and the misalignment of the joint from the socket end to the spigot end of the mold after the centrifugation process, and output the set of geometric status parameters of the mold. The wall thickness prediction module is used to obtain the sequence of geometric state parameters of the mold in the historical production cycle of the mold and the measured data of the wall thickness distribution of the corresponding pole after demolding for the current batch of molds. It establishes a wall thickness coupling mapping model between the geometric state parameter set and the wall thickness distribution deviation. Based on the mold geometric state parameter set and the wall thickness coupling mapping model, it calculates the predicted value of the wall thickness distribution deviation for the current batch. The process compensation module is used to determine the mix ratio adjustment amount and the centrifugal molding parameter adjustment amount according to the predicted value of the wall thickness distribution deviation, generate a mix ratio correction command and send it to the mixing station and the centrifugal molding correction command, so as to control the mixing station and the centrifugal molding equipment to perform mixing, casting and centrifugal molding according to the corrected parameters, and then send the molding into the steam curing kiln. The strength determination module is used to obtain the actual temperature control curve and mix ratio parameters of each pole in the steam curing kiln, calculate the equivalent curing age based on the actual temperature control curve, and call the preset strength-maturity relationship curve based on the mix ratio parameters and the equivalent curing age to obtain the current compressive strength. The demolding decision module is used to input the predicted value of the wall thickness distribution deviation, the current compressive strength, the minimum compressive strength required for the demolding structure safety, and the inner diameter ellipticity into a multi-dimensional coupled evaluation model, perform segmented coupled calculations on the structural uniformity risk and strength reserve risk, output a demolding quality level index that characterizes the degree of demolding damage risk, and generate a priority sequence. The scheduling and execution module is used to schedule the demolding devices to perform demolding operations sequentially according to the priority sequence; The backtracking calibration module is used to collect the actual wall thickness detection value of the pole after demolding, compare it with the predicted value of the wall thickness distribution deviation, and correct the coefficients of the wall thickness coupling mapping model in reverse and update it synchronously.

[0007] Preferably, the method for obtaining the set of geometric state parameters of the mold is as follows: S201. After the centrifugation process is completed, select three equally spaced sections along the mold axis from the socket end to the insert end as geometric condition inspection sections. S202. Perform non-contact scanning on the inner contour of each of the geometric state detection sections to obtain point cloud data of the inner contour of each section. S203. Perform least-squares ellipse fitting on the point cloud data of the inner contour of each cross section, extract the major axis and minor axis of each cross section, and take the ratio of the difference between the major axis and minor axis relative to the standard inner diameter as the ellipticity of the cross section. S204. Divide the point cloud data of each geometric state detection section into upper half mold point cloud and lower half mold point cloud according to the joint orientation, perform circle fitting to obtain two fitting circle centers, and calculate the offset between the two fitting circle centers as the joint misalignment of the section. S205. Integrate the ellipticity and misalignment of each section into the set of geometric state parameters of the mold according to the time sequence and output them.

[0008] Preferably, the specific method for establishing the wall thickness coupling mapping model is as follows: The wall thickness distribution deviation sequence is obtained by subtracting the measured wall thickness distribution data of each test section from the target wall thickness of the corresponding section. Using the ellipticity and joint misalignment of the mold geometric state parameter set as independent variables and the wall thickness distribution deviation sequence as the dependent variable, multiple linear regression is used to estimate the regression coefficients, and the wall thickness coupling mapping model is established. The set of mold geometric state parameters for the current batch is input into the wall thickness coupling mapping model to calculate the predicted value of the wall thickness distribution deviation for the current batch.

[0009] Preferably, the process compensation module is used for: The predicted wall thickness distribution deviation values ​​on the three axial detection sections are compared with the preset distribution pattern library to identify the current deviation distribution pattern. The deviation distribution pattern includes the overall offset pattern and the local deviation pattern. In the overall offset mode, the mean value of the predicted wall thickness distribution deviation of the three detection sections is calculated. Based on the preset amplitude range in which the mean value is located, the corresponding water-cement ratio adjustment amount and sand ratio adjustment amount are queried from the mix ratio compensation rule table to generate the mix ratio correction instruction. At the same time, the corresponding centrifugal speed adjustment amount and centrifugal duration adjustment amount are queried from the centrifugal molding compensation rule table to generate the centrifugal molding correction instruction. In the local bias mode, only the centrifugal molding compensation rule table is queried to generate the centrifugal molding correction instruction, and the mix ratio correction instruction maintains the default value.

[0010] Preferably, the current compressive strength is calculated as follows: S501. Obtain the surface temperature and ambient temperature of each electric rod in the steam curing kiln according to the preset sampling cycle to form the actual temperature control curve, and at the same time obtain the water-cement ratio in the mixing ratio parameter of the electric rod. S502. Based on the temperature data in the actual temperature control curve, the equivalent curing coefficient for each sampling period is calculated using the Arrhenius equation, and the equivalent curing age is obtained by integrating over the sampling time. S503. Based on the water-cement ratio and the equivalent curing age, the current compressive strength is calculated by calling the preset strength-maturity relationship curve.

[0011] Preferably, the multi-dimensional coupled evaluation model includes a structural homogeneity sub-model and a strength reserve sub-model, wherein: The structural uniformity sub-model is used to calculate the first relative ratio of the maximum value of the ellipticity of the three detection sections to the preset ellipticity warning benchmark value, and the second relative ratio of the absolute value of the predicted wall thickness distribution deviation to the preset wall thickness deviation warning benchmark value. The maximum value of the first relative ratio and the second relative ratio is taken as the structural uniformity risk value. The strength reserve sub-model is used to divide the minimum compressive strength required for the demolding structure by the current compressive strength to obtain the strength reserve ratio, and the value after subtracting 1 from the strength reserve ratio is used as the strength reserve risk value.

[0012] Preferably, the output characterizing the demolding quality level of demolding damage risk and generating a priority sequence specifically includes: First, determine whether the strength reserve risk value is greater than zero; When the value is greater than zero, the strength reserve risk value, the structural uniformity risk value, and the preset penalty constant are summed to obtain the demolding quality grade index. When the value is less than or equal to zero, the structural uniformity risk value is directly used as the demolding quality grade index; The demolding quality grade index of each pole in the steam curing kiln is sorted from smallest to largest. The scheduling execution module performs the demolding operation on the pole with the smallest value first, and then completes the demolding of the remaining poles in sequence.

[0013] Preferably, the scheduling execution module is used to read the pole identifiers sequentially according to the priority sequence, retrieve the demolding process parameters corresponding to the pole identifiers, and control the demolding equipment to perform lifting, demolding, and transfer operations based on the demolding process parameters.

[0014] Preferably, the backtracking calibration module performs the following steps: S901. Collect the actual wall thickness of each test section of the pole after demolding; S902. Subtract the actual wall thickness detection value of each section from the target wall thickness of the corresponding section to obtain the actual wall thickness distribution deviation. S903. The difference between the actual wall thickness distribution deviation and the predicted wall thickness distribution deviation is calculated to obtain the prediction residual. S904. Based on the predicted residuals, the regression coefficients of the wall thickness coupling mapping model are corrected in reverse using the gradient descent method. S905. Write the corrected regression coefficients back into the wall thickness coupling mapping model to replace the original coefficients.

[0015] A method for intelligent production and manufacturing management of cement poles includes the following steps: Step 1: After the centrifugation process is completed, collect the inner diameter ovality and joint misalignment of the mold socket and spigot, and output the set of mold geometric parameters. Step 2: Obtain the set of geometric state parameters and corresponding measured wall thickness data for the mold in the historical period, establish a wall thickness coupling mapping model, and calculate the predicted value of wall thickness distribution deviation based on the set of geometric state parameters of the mold. Step 3: Determine the mix proportion adjustment amount and centrifugal molding parameter adjustment amount based on the predicted wall thickness distribution deviation value, generate correction instructions and execute mixing, casting, centrifugation and then send it into the steam curing kiln; Step 4: Obtain the temperature control curve of each pole in the steam curing kiln and obtain the water-cement ratio of the pole. Calculate the equivalent curing age based on the temperature control curve, and call the preset strength-maturity relationship curve based on the water-cement ratio and the equivalent curing age to obtain the current compressive strength. Step 5: Input the predicted wall thickness distribution deviation, the current compressive strength, the minimum compressive strength required for demolding structure safety, and the inner diameter ellipticity into the multi-dimensional coupled evaluation model, perform piecewise coupled calculations of structural uniformity risk and strength reserve risk, and generate demolding quality level index and priority sequence. Step 6: Perform the demolding operation sequentially according to the priority sequence; Step 7: Collect the actual wall thickness detection value of the pole after demolding, compare it with the predicted value of the wall thickness distribution deviation to obtain the prediction residual, and correct the regression coefficient of the wall thickness coupling mapping model in reverse.

[0016] This invention provides an intelligent production management system and method for cement poles. It has the following beneficial effects: 1. By transforming the traditional manual offline sampling inspection of the inner diameter ovality and misalignment of the mold socket and spigot into digital acquisition after the process, the error of human reading is eliminated, providing a standardized and reproducible input data basis for predicting wall thickness distribution deviation, and enabling the mold geometry to be incorporated into the online control closed loop of the production process for the first time.

[0017] 2. By identifying the overall offset mode and local deviation mode of wall thickness distribution deviation in the axial direction, the process compensation is upgraded from a single fixed strategy to a precise compensation based on the type of deviation root cause: In the overall offset mode, the systemic deviation is corrected in a coordinated manner from both the mixing ratio and centrifugal molding dimensions; in the local deviation mode, only the centrifugal parameters are adjusted to keep the mixing ratio unchanged, thus avoiding material waste and product quality fluctuations caused by global adjustment of local problems. 3. Heterogeneous indices are normalized into dimensionless risk values ​​through structural uniformity sub-model and strength reserve sub-model respectively. A segmented coupling mechanism is adopted to realize hierarchical decision-making between strength baseline constraints and structural optimization constraints: when the strength does not meet the standard, priority demolding is prevented by risk value accumulation. After the strength meets the standard, the damage risk is assessed only by structural uniformity. This avoids the strength risk dilution problem caused by fixed weighting and provides clear quantitative basis for the demolding scheduling of multiple poles. Attached Figure Description

[0018] Figure 1 This is a system flowchart of the present invention; Figure 2 This is a flowchart of the process compensation deviation distribution pattern recognition of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] like Figures 1 to 2 As shown, the present invention proposes an intelligent production and manufacturing management system for cement poles, including: a mold status acquisition module, which is used to digitally acquire the inner diameter ellipticity and joint misalignment of the mold from the socket end to the spigot end after the previous batch of poles has been demolded and before the current batch is poured, and output a set of mold geometric status parameters. The wall thickness prediction module is used to obtain the sequence of geometric state parameters of the mold in the historical production cycle of the mold and the measured data of the wall thickness distribution of the corresponding pole after demolding for the current batch of molds. It establishes a wall thickness coupling mapping model between the geometric state parameter set and the wall thickness distribution deviation. Based on the mold geometric state parameter set and the wall thickness coupling mapping model, it calculates the predicted value of the wall thickness distribution deviation for the current batch. The process compensation module is used to determine the mix ratio adjustment amount and the centrifugal molding parameter adjustment amount according to the predicted value of the wall thickness distribution deviation, generate a mix ratio correction command and send it to the mixing station, generate a centrifugal molding correction command and send it to the centrifugal molding equipment, so as to control the mixing station and the centrifugal molding equipment to perform mixing, casting and centrifugal molding according to the corrected parameters, and then send the molded product into the steam curing kiln. The strength determination module is used to obtain the actual temperature control curve and mix ratio parameters of each pole in the steam curing kiln, calculate the equivalent curing age based on the actual temperature control curve, and call the preset strength-maturity relationship curve based on the mix ratio parameters and the equivalent curing age to obtain the current compressive strength. The demolding decision module is used to input the predicted value of the wall thickness distribution deviation, the current compressive strength, the minimum compressive strength required for the demolding structure safety, and the inner diameter ellipticity into a multi-dimensional coupled evaluation model, perform segmented coupled calculations on the structural uniformity risk and strength reserve risk, output a demolding quality level index that characterizes the degree of demolding damage risk, and generate a priority sequence. The scheduling and execution module is used to schedule the demolding devices to perform demolding operations sequentially according to the priority sequence; The backtracking calibration module is used to collect the actual wall thickness detection value of the pole after demolding, compare it with the predicted value of the wall thickness distribution deviation, and correct the coefficients of the wall thickness coupling mapping model in reverse and update it synchronously.

[0021] As an optional embodiment, the manufacturing process of cement poles is as follows: mixing, pouring, centrifugal molding, steam curing and demolding; among them, the centrifugal molding process relies on the high-speed rotation of the mold to throw the concrete onto the mold wall to form a dense pole body; during long-term cyclic use, the socket and spigot of the mold will first undergo elliptical deformation and misalignment due to the periodic dynamic load and the lateral pressure of the concrete.

[0022] The aforementioned geometric deterioration directly alters the concrete placement space within the mold cavity, leading to uneven wall thickness distribution and consequently, cracks or structural damage during steam curing and demolding. In existing technologies, mold inspection is mostly conducted offline through manual sampling; the inspection data fails to establish a quantitative correlation with wall thickness deviations and cannot be integrated into the online control loop of the production process. Therefore, it is necessary to perform post-process digital acquisition of the mold's geometric state after the centrifugation process to provide standardized input data for the wall thickness coupling mapping model of the wall thickness prediction module. Specifically, cement poles are typically tapered, so the mold is cone-shaped, with the socket and spigot located at opposite ends of the mold along the axial direction. During demolding, the pole emerges from the socket.

[0023] S201. After the previous batch of poles is demolded and before the current batch is poured, select three equally spaced sections along the mold axis from the socket end to the spigot end as geometric condition detection sections. The cement pole mold is formed by the upper half and the lower half of the mold being fastened together at the joint surface. During long-term cyclic use, the geometric deterioration of the mold starts from the socket end and the spigot end and develops towards the middle. Selecting three equally spaced sections from the socket end to the spigot end ensures that the socket end, the middle of the mold, and the spigot end are all covered. This can reflect the geometric condition of the main sections of the mold axis while controlling the data acquisition time. S202. Perform non-contact scanning (such as laser triangulation or structured light scanning) on ​​the inner contour of each inspection section to obtain point cloud data of the inner contour of each section. Using a non-contact method avoids secondary damage to the inner surface of the mold caused by contact gauges, while also meeting the online inspection requirements after high-temperature demolding. S203. Perform least-squares circle fitting on the point cloud data of the inner contour of each cross section to extract the major axis of each cross section. With short axis (Unit: mm), calculate the ellipticity of the cross section using the following formula. ,in The standard inner diameter of the mold at this cross-section, in mm; ellipticity. , dimensionless, represents the degree of deviation of the inner contour of the cross section from the ideal circle; S204. Divide the point cloud data of each detection section into upper half mold point cloud and lower half mold point cloud according to the joint orientation, and perform least squares circle fitting on each to obtain the center of the upper half simulated joint circle. And the lower half of the simulated center of the circle ; Calculate the joint misalignment of this section using the following formula. , where i is the section number, and is 1, 2, 3; Let be the coordinates of the fitted circle center of the point cloud of the upper half of the i-th cross section, in mm; Let be the coordinates of the fitted circle center of the point cloud of the lower half of the i-th cross section, in mm; The misalignment of the joint at the i-th cross section is expressed in mm. It represents the degree of offset between the fitting centers of the inner contours of the upper and lower half molds at the same cross section, reflecting the misalignment state of the two half molds at the joint surface. S205. Integrate the ellipticity and joint misalignment of each cross section into a set of mold geometric state parameters according to the time sequence. The output is then displayed, where t is the production cycle number, which is dimensionless and facilitates the backtracking calibration module to perform model comparison and correction according to the time series.

[0024] As an optional embodiment, during the production process of cement poles, after the same mold undergoes thousands of centrifugal cycles, the ellipticity of its socket and spigot gradually increases, the misalignment of the joint gradually accumulates, and the effective space inside the mold cavity changes non-uniformly, resulting in the deviation of the concrete throwing trajectory, which ultimately manifests as uneven wall thickness distribution. Therefore, it is necessary to establish a mapping model between the geometric state parameters and the wall thickness distribution deviation for each mold to achieve early prediction of wall thickness deviation based on the mold state. Specifically, obtain the set of mold geometric state parameters from the most recent N production cycles. And the measured wall thickness distribution data of the three test sections after the poles were demolded for each cycle. (i=1,2,3 are section numbers); the measured wall thickness distribution data refers to the wall thickness values ​​measured at equal angular intervals along the circumference of the annular section of the pole, not the section diameter. The cement pole is a hollow annular component, and its wall thickness refers to the radial thickness from the outer contour to the inner contour. The wall thickness distribution deviation sequence is obtained by subtracting the measured wall thickness distribution data of each test section from the target wall thickness of the corresponding section. ,and Positive deviation indicates that the wall thickness is too thick, and negative deviation indicates that the wall thickness is too thin. The measured wall thickness distribution data of each test section is taken as the arithmetic mean of the measured wall thickness values ​​at multiple points along the circumference of the section, which is used as the representative value of the wall thickness of the section. The wall thickness distribution deviation sequence is obtained by subtracting the representative value of the wall thickness of each section from the target wall thickness of the corresponding section. Calculations were performed on the three sections separately to generate a sequence of wall thickness distribution deviations. , Then, using the ellipticity and joint misalignment of the mold geometric state parameter set as independent variables and the wall thickness distribution deviation sequence as the dependent variable, multiple linear regression is used to estimate the regression coefficients, thus establishing the wall thickness coupling mapping model. Let the independent variable matrix be X (composed of N periods of ellipticity and joint misalignment), and the dependent variable vector be... The wall thickness coupling mapping model is then: in For the regression coefficient vector, The residual vector is estimated using the least squares method. This minimizes the sum of squared residuals from the historical data.

[0025] Finally, the set of mold geometric state parameters for the current batch is substituted into the wall thickness coupling mapping model to calculate the predicted value of the wall thickness distribution deviation for the current batch.

[0026] It should be noted that the ellipticity and joint misalignment were chosen as predictors of wall thickness deviation based on the following physical mechanism: increased ellipticity means that the inner cavity of the mold deviates from the ideal circle, and the centrifugal force of the concrete during centrifugal rotation is unevenly distributed in the circumferential direction, resulting in thinner wall thickness in the long axis direction and thicker wall thickness in the short axis direction; increased joint misalignment indicates that the two halves of the mold are not aligned, and the flow of concrete in the joint area is obstructed, resulting in local wall thickness abnormalities.

[0027] Both affect the wall thickness distribution from two dimensions: "overall roundness degradation" and "local docking deviation," respectively. They have a clear physical causal logic and are suitable as regression independent variables.

[0028] The reason for using the multiple linear regression method is that the sample size of industrial field data is limited and needs to be updated quickly online. Linear regression has low computational cost, strong interpretability, and is easy to embed into the production line control system for real-time operation. The reason for further adopting single mold independent modeling instead of multi-mold mixed modeling is that different molds have different service years, centrifugation frequency and maintenance history, and the baseline geometry is significantly different. Mixed data will introduce heterogeneous noise and reduce prediction accuracy.

[0029] As an optional embodiment, the distribution characteristics of wall thickness distribution deviation on different cross sections along the axial direction of the rod directly reflect the root cause of the deviation. If the deviation directions of the three test sections are consistent and the amplitudes are similar, it indicates that the deviation originates from improper overall concrete mix proportions or insufficient global centrifugal compaction, which is a systemic problem; if only a few cross sections show significant deviations while other cross sections are normal, it indicates that the deviation originates from axial non-uniformity of centrifugal attachment or local geometric deterioration of the mold, which is a local problem. Specifically, the predicted value of the wall thickness distribution deviation for the current batch (i=1,2,3 correspond to three axial detection sections) are compared with a preset distribution pattern library to identify the current deviation distribution pattern, which includes an overall offset pattern and a local deviation pattern.

[0030] The overall offset mode refers to the situation where the predicted wall thickness distribution deviation values ​​of the three detection sections are all located on the same side of zero, and the difference between the maximum absolute deviation and the minimum absolute deviation does not exceed the preset discrete value γ. In this mode, the wall thickness deviation exhibits a globally consistent offset, which is caused by the overall deviation of the concrete mix proportion from the design value or insufficient global centrifugal compaction, requiring simultaneous adjustment of the material ratio and molding process.

[0031] The local deviation mode refers to a situation where the predicted wall thickness distribution deviation of at least one detection section is opposite in direction to the deviation of other sections, or the difference between the maximum absolute deviation and the minimum absolute deviation exceeds the discrete value γ. In this mode, the wall thickness deviation exhibits a locally non-uniform distribution, which is caused by uneven axial flow of the slurry during centrifugal deposition or abrupt changes in the local ellipticity / joint misalignment of the mold. This can be specifically corrected by adjusting the centrifugal forming parameters.

[0032] It should be noted that the preset discrete value γ is used to quantify the dispersion of wall thickness deviations among the three cross-sections. Its value is determined as follows: Statistical samples of wall thickness distribution deviations across the three cross-sections are retrieved from the historical production cycles of the mold. The average difference between the maximum and minimum absolute deviations in each cycle is calculated, and this average is multiplied by an empirical coefficient to obtain the preset discrete value γ. The empirical coefficient ranges from 0.8 to 1.2. This preset discrete value γ divides the axial deviation distribution into two categories: a consistent distribution and a discrete distribution, serving as the decision boundary for pattern recognition.

[0033] Furthermore, in the overall offset mode, the mean of the predicted values ​​of the wall thickness distribution deviations of the three sections is calculated. As a representative quantity of overall deviation: in accordance with Within the preset range, the corresponding water-cement ratio adjustment amount and sand ratio adjustment amount are queried from the mix ratio compensation rule table to generate a mix ratio correction instruction and send it to the mixing plant. At the same time, the corresponding centrifugal speed adjustment amount and centrifugal duration adjustment amount are queried from the centrifugal molding compensation rule table to generate a centrifugal molding correction instruction.

[0034] The adjustment amount in two dimensions is based on The sign and absolute value are determined independently by looking up a table. When the sign is positive, the downward adjustment amount is retrieved; when the sign is negative, the upward adjustment amount is retrieved.

[0035] In the local deviation mode, the predicted value is based solely on the wall thickness distribution deviation of each section. The absolute value is used to query the centrifugal speed adjustment and centrifugal duration adjustment of the corresponding section from the centrifugal forming compensation rule table to generate the centrifugal forming correction instruction. The mix ratio correction instruction maintains the default value.

[0036] Specifically, this can be explained using the following example: The preset amplitude range refers to the total deviation represented by the preset amplitude range. The absolute value is divided into several consecutive numerical intervals, each interval corresponding to a fixed set of adjustment values. Both the mixing ratio compensation rule table and the centrifugal molding compensation rule table are arranged according to this interval structure.

[0037] Assuming the target wall thickness is 30mm, Divided into three amplitude ranges: Low amplitude range, Greater than zero and not exceeding 2mm. Within this range, the mix proportion correction instructions are: water-cement ratio fine-tuning ±0.02 and sand ratio fine-tuning ±1%; centrifugal molding correction instructions are: rotation speed fine-tuning ±5r / min and duration fine-tuning ±10s.

[0038] Medium amplitude range The thickness is greater than 2mm but not exceeding 5mm. Within this range, the mix proportion correction instructions are to adjust the water-cement ratio by ±0.05 and the sand ratio by ±2%; the centrifugal molding correction instructions are to adjust the rotation speed by ±15r / min and the duration by ±30s.

[0039] High amplitude range, Greater than 5mm. Within this range, the mix proportion correction instructions are to significantly adjust the water-cement ratio by ±0.10 and the sand ratio by ±4%; the centrifugal molding correction instructions are to significantly adjust the rotation speed by ±30r / min and the duration by ±60s.

[0040] in, When the value is greater than zero, the adjustment amount is reduced, that is, the water-cement ratio and sand ratio are adjusted downward, and the centrifugal speed and duration are adjusted downward; When the value is less than zero, the adjustment amount is increased, that is, the water-cement ratio and sand ratio are adjusted upwards, and the centrifugal speed and duration are adjusted upwards.

[0041] The boundary values ​​of the intervals and the adjustment amounts of each interval can be calibrated based on actual production data and written into the mix ratio compensation rule table and the centrifugal molding compensation rule table, which are then regularly maintained and updated by process management personnel.

[0042] As an optional embodiment: Cement poles undergo three stages in the steam curing kiln: heating, constant temperature, and cooling. The cement hydration reaction rate changes significantly with temperature. Traditional demolding decisions are based on a fixed curing period, failing to consider the impact of actual temperature control curves and mix proportion differences on the strength growth rate. This leads to varying strength development states among different poles in the same kiln, easily resulting in insufficient strength or over-curing at demolding. Therefore, it is necessary to estimate the real-time compressive strength of each pole online based on the actual temperature control curves and mix proportion parameters, using a strength estimation model based on equivalent curing ages, to provide an accurate strength basis for demolding decisions.

[0043] Specifically, S501, temperature control data acquisition: The surface temperature of each pole is collected by temperature sensors arranged inside the steam curing kiln according to a preset sampling period (unit: min). and ambient temperature This forms the actual temperature control curve. Simultaneously, the water-cement ratio from the mix proportion parameters of the pole is obtained from the process database. (Dimensionless). The surface temperature reflects the actual heating state of the rod, and the ambient temperature reflects the temperature of the atmosphere inside the kiln. The arithmetic mean of the two is taken as the representative temperature of the rod in this sampling period. S502. Calculation of Equivalent Curing Coefficient and Integralization of Equivalent Age: The cement hydration reaction rate has an exponential relationship with temperature; the higher the temperature, the faster the hydration, and vice versa. To convert the actual curing time at different temperatures into an equivalent cumulative amount under a unified standard, the Arrhenius equation is used to calculate the equivalent curing coefficient β(T) for each sampling period.

[0044] The Arrhenius equation is a well-known model in the field describing the effect of temperature on the rate of chemical reactions. Its physical meaning is: the cumulative hydration effect achieved by curing at temperature T for one unit of time is equivalent to curing at a reference temperature for β(T) units of time.

[0045] ;in The activation energy of cement hydration reaction is expressed in J / mol. For ordinary Portland cement, the commonly accepted value in this field is 41,500 J / mol. The universal gas constant is taken as 8.314 J / (mol⋅K); The reference temperature is 293K (i.e., 20℃), which is the standard curing temperature in this field; The absolute temperature at the sampling time is taken as the arithmetic mean of the surface temperature and the ambient temperature within that period. Integrating the sampling time yields the equivalent maintenance age. ,in The total number of sampling periods is dimensionless. The absolute temperature during the k-th sampling period; The actual duration of the k-th sampling period, in hours; It is the equivalent maintenance coefficient in the kth sampling period; Equivalent curing age, in hours, represents the time required to achieve the same level of hydration accumulation under standard curing conditions at 20°C.

[0046] S503, Estimated Strength: The current compressive strength This refers to the estimated cubic compressive strength of the concrete pole at the end of steam curing, expressed in MPa, used for comparison with the minimum strength required for the safety of demolded structures. A lower water-cement ratio results in higher cement paste strength; a longer equivalent curing period leads to more complete hydration and higher strength. Based on the stated water-cement ratio... and the equivalent maintenance age mentioned above The preset intensity-maturity relationship curve is invoked for calculation, i.e. ,in The final compressive strength under this water-cement ratio, in MPa, refers to the ultimate compressive strength that concrete can achieve after full hydration under the current mix proportion conditions. It is an asymptotic value determined by fitting the above equation with the measured data of the compressive strength of test blocks under at least three different equivalent curing ages for this mix proportion. The maturity constant, in units of h, reflects the rate of strength development of the mix proportion. It is determined by fitting the above equation with the measured compressive strength data of the test blocks of the mix proportion at at least three different equivalent curing ages.

[0047] As an optional implementation, demolding decisions involve several heterogeneous indicators—ellipticity characterizes the geometric deterioration of the mold, wall thickness distribution deviation characterizes the uniformity of concrete adhesion, and compressive strength characterizes the structural bearing capacity. These three indicators have different dimensions and physical meanings, and cannot be directly compared or simply added together.

[0048] If the timing of demolding is determined based on a single indicator, the following misjudgments are likely to occur: when the strength meets the standard but the ellipticity is severely deteriorated, demolding may cause circumferential cracks due to stress concentration; when the ellipticity is good but the strength has just reached the critical value, the demolding impact force may cause structural damage.

[0049] Therefore, it is necessary to construct a multi-dimensional coupled assessment model to normalize heterogeneous indicators into risk values ​​of the same dimension, and then perform unified calculations.

[0050] The multi-dimensional coupled evaluation model includes a structural homogeneity sub-model and a strength reserve sub-model, which are constructed as follows: Structural homogeneity sub-model: calculates the degree of deviation of the ellipticity index and the predicted value of the wall thickness distribution deviation from their respective warning benchmarks.

[0051] First relative ratio Characterizes the degree of ellipticity deviation relative to the warning baseline: ,in The maximum value of the ellipticity among the three detection cross sections, dimensionless; The preset ellipticity warning reference value is dimensionless and is determined by reducing it with a safety factor within the allowable range of GB / T4623 standard based on the mold model. Second relative ratio Characterizes the degree of deviation of the wall thickness deviation from the warning benchmark: ,in The value is the maximum absolute value among the predicted wall thickness distribution deviations of the three cross sections, in mm. The preset wall thickness deviation warning benchmark value, in mm, is determined based on the allowable deviation range of the pole's designed wall thickness. Pick and The maximum value in the range is used as the structural homogeneity risk value. ; It should be noted that the reason for using the maximum value instead of the weighted average is that the mechanisms of ellipticity deterioration and uneven wall thickness distribution on demolding damage are different. The former leads to stress concentration during demolding, while the latter leads to insufficient local bearing capacity. Either factor exceeding its warning benchmark may independently cause demolding damage. Therefore, using the less unfavorable factor to control the risk of overall structural uniformity is in line with the "take the larger" principle in engineering safety design.

[0052] Strength reserve sub-model: Calculates the reserve of current compressive strength relative to the demolding safety boundary.

[0053] Strength Reserve Ratio Minimum compressive strength requirement Compared with current compressive strength The ratio of the minimum compressive strength requirement The minimum compressive strength required for the safety of the demolded structure, expressed in MPa, is determined according to industry standards of ≥20MPa for reinforced concrete poles and ≥30MPa for prestressed poles.

[0054] Finally, the strength reserve risk value will be determined. Calculated as Subtract 1, that is ; It should be noted that the strength reserve risk value Defined as The reason for reducing it by 1 is: structural uniformity risk value Using zero as the lower bound (non-negative), to ensure a unified safety assessment benchmark for the two risk values, the strength reserve status also needs to be mapped to an indicator with zero as the critical point. After this definition, Equal to 0 corresponds exactly The critical state of achieving the standard; A value greater than 0 indicates that the current compressive strength is lower than the demolding safety requirement; A value less than 0 indicates that the current compressive strength is higher than the demolding safety requirements and has a margin of safety.

[0055] This definition sets both the strength reserve risk value and the structural uniformity risk value at "zero" as their safety boundary in the numerical dimension, facilitating subsequent piecewise coupling calculations. Using whether the strength is greater than zero as a criterion for determining whether the strength meets the standard avoids relying on... When 1 is the dividing point and Inconsistent benchmarks lead to cumbersome calculations.

[0056] As an optional embodiment, although the structural uniformity risk value and the strength reserve risk value have been transformed into dimensionless values ​​through the sub-model, the nature of their constraints on the demolding decision is fundamentally different. The strength reserve is a bottom-line safety constraint - demolding when the compressive strength does not meet the standard will directly lead to the breakage or severe deformation of the rod, at which point the influence of structural uniformity becomes secondary. Structural uniformity is a quality optimization constraint - it only affects the stress concentration and local cracking probability during the demolding process under the premise that the strength meets the standard.

[0057] If a fixed weight is used to simply sum the two, the high-risk value generated when the strength is severely insufficient may be diluted by the smaller value of the structural uniformity, causing poles that do not meet the strength standard to be incorrectly placed at the front of the sequence due to their low total score, which may lead to demolding accidents. Therefore, it is necessary to perform segmented coupling based on the positive and negative values ​​of the strength reserve risk value to ensure that the strength bottom line constraint is given priority.

[0058] Specifically, the following steps are taken: First, determine the risk value of the strength reserve. Is it greater than zero?

[0059] when A value greater than 0 indicates the current compressive strength. The minimum compressive strength is lower than the safety requirements for demolding structures. This indicates that the strength has not met the required standard. If demolding is forced at this point, the rod may break or severely deform due to insufficient load-bearing capacity, making the impact on structural uniformity secondary. Therefore, the aforementioned strength reserve risk value is... With respect to the structural uniformity risk value The demolding quality grade index Q is obtained by summing the results: This summation operation ensures that the risk of insufficient strength is fully accumulated in the demolding quality grade index, ensuring that poles that do not meet the strength standard are ranked lower in the order, and avoiding them being mistakenly judged as priority demolding objects due to their small structural uniformity risk value.

[0060] when When the value is 0, it indicates the current compressive strength. Exactly equal to The structure is within the safety boundary. While the strength meets the minimum requirement, it has no margin, and the impact load during demolding may still induce localized damage through structural inhomogeneity. Therefore, the structural homogeneity risk value is... It is directly used as the demolding quality grade index Q.

[0061] when When it is less than 0, it indicates the current compressive strength. Higher than There is a strength surplus. At this point, the strength reserve is no longer a constraint, and the risk of demolding damage is entirely determined by structural uniformity. Therefore, the structural uniformity risk value is also... It is directly used as the demolding quality grade index Q; Right now ,in A pre-defined penalty constant, dimensionless, is used, with a value greater than the historical maximum value of the structural uniformity risk value under the strength surplus state. This ensures that the Q value of any pole with substandard strength is greater than the Q value of all poles with strength surplus. The values ​​of all poles with strength surplus from historical production cycles are retrieved. The sample is used, and its maximum value is multiplied by a margin factor of 1.1 to 1.5 as the margin. .

[0062] In summary, the demolding quality level index Q of each pole in the steam curing kiln is sorted from smallest to largest, generating a priority sequence and outputting it to the scheduling execution module. A smaller Q value indicates better demolding conditions for that pole: Q takes a smaller value when strength is sufficient and the structure is uniform, and a larger value when strength is insufficient or the structure is severely deteriorated. The scheduling execution module performs the demolding operation first on the pole with the smallest Q value, and then completes the demolding of the remaining poles in sequence.

[0063] As an optional embodiment, multiple electric poles within the steam curing kiln are sorted by the demolding decision module to form a priority sequence. However, the demolding operation parameters differ for poles of different specifications—differences in pole length, tip diameter, reinforcement specifications, and wall thickness determine that the lifting point spacing, upper limit of lifting speed, demolding tilt angle, and transfer bracket specifications all need to be configured differently. If a uniform demolding action is used without distinguishing pole specifications during scheduling, improper lifting point spacing can easily lead to excessive bending moment in the pole body, or excessive demolding speed can cause cracking in thin-walled sections. Therefore, based on the priority sorting, it is necessary to retrieve specific process parameters for each individual pole to control the entire demolding process.

[0064] The scheduling execution module is executed in the following manner: First, the pole identifiers are read sequentially according to the priority sequence. Each pole identifier is a unique identification code generated by binding the mold with an RFID tag or QR code before the casting process. This identifier is transmitted throughout the process between the mold status acquisition module, wall thickness prediction module, strength determination module, and demolding decision module, ensuring that the data from each module corresponds one-to-one with the physical pole.

[0065] Then, using the pole identification as an index, the demolding process parameters corresponding to the pole's specifications and model are retrieved from the database. These demolding process parameters include: lifting point spacing (unit: m), calculated based on the pole length and reinforcement ratio to ensure that the maximum bending moment of the pole body during lifting does not exceed the allowable value; lifting speed (unit: m / min), determined based on the predicted wall thickness distribution deviation output by the wall thickness coupling mapping model, with the upper speed limit adjusted for thinner wall sections; demolding tilt angle θ (unit: °), determined based on the pole taper and the friction coefficient of the mold inner wall; and the transfer bracket model, selected to match the pole tip diameter and root diameter.

[0066] Finally, the demolding equipment is controlled to perform lifting, demolding, and transfer operations according to the demolding process parameters. The lifting involves controlling the lifting device to position and clamp the pole at the specified lifting point spacing, and lifting the pole at a uniform lifting speed. The demolding involves maintaining the demolding tilt angle during the lifting process, so that the pole slides smoothly out along the mold axis. The transfer involves placing the demolded pole on a matching transfer bracket and moving it to the quality inspection station.

[0067] As an optional embodiment, the wall thickness coupling mapping model is established based on historical data. However, the geometric deterioration trend of the mold during continuous service is not constant—the ellipticity growth rate of the socket and spigot may change due to material fatigue, maintenance intervals, or changes in the hardness of concrete aggregates. If the model coefficients are fixed for a long time, the prediction accuracy will gradually decrease as the actual deterioration trajectory of the mold deviates from the historical statistical law, and the feedforward adjustment of the process compensation module will also become inaccurate. Therefore, it is necessary to collect the actual wall thickness data after each demolding, compare it with the predicted value to obtain the residual, and use this residual to correct the model coefficients in reverse, so that the wall thickness coupling mapping model always tracks the latest deterioration state of the mold. Specifically, S901, Actual Wall Thickness Acquisition: After the pole is demolded and transported to the quality inspection station, multiple wall thickness values ​​are measured at equal angular intervals along the circumference of each test section using an ultrasonic thickness gauge or mechanical caliper. The arithmetic mean is taken as the actual wall thickness detection value of that section, denoted as S901. (i=1,2,3 are section numbers, in mm); S902, Actual Deviation Calculation: Calculate the actual wall thickness measured at each cross-section. The actual wall thickness distribution deviation is obtained by subtracting the target wall thickness from the corresponding cross-section. This deviation reflects the actual wall thickness offset of this batch of poles after process compensation and steam curing, and the unit is mm.

[0068] S903. The difference between the actual wall thickness distribution deviation and the predicted wall thickness distribution deviation output by the wall thickness prediction module is calculated to obtain the prediction residual. It should be noted that the physical meaning of the prediction residual is: the deviation of the predicted value of the wall thickness deviation of the cross section by the wall thickness coupling mapping model from the measured value during this production cycle. A residual of zero indicates that the prediction is completely accurate, a positive residual indicates that the predicted value is too low (the actual thickness is greater than the prediction), and a negative residual indicates that the predicted value is too high (the actual thickness is less than the prediction).

[0069] S904, Coefficient Correction: Based on the predicted residuals, the regression coefficient vector of the wall thickness coupling mapping model is corrected in reverse using the gradient descent method. Specifically, the regression coefficients are updated along the negative gradient direction using the sum of squared residuals as the cost function: ;in This is the current regression coefficient vector; This is the corrected regression coefficient vector; The preset learning rate is (dimensionless, ranging from 0.01 to 0.1, determined by the model convergence speed calibration); X is the matrix of independent variables composed of the set of mold geometric state parameters for the current batch; and e is the prediction residual vector. The norm of the independent variable matrix is ​​used; the gradient descent update adjusts the regression coefficients in the direction of reducing the prediction residuals, and the correction amount is proportional to the size of the residuals, with larger residuals resulting in larger corrections.

[0070] S905, Model Update: The corrected regression coefficients... Synchronously integrate into the wall thickness coupling mapping model, replacing the original coefficients. This allows the wall thickness prediction module to call the updated model for prediction in the next production cycle.

[0071] A method for intelligent production and manufacturing management of cement poles includes the following steps: Step 1: After the centrifugation process is completed, collect the inner diameter ovality and joint misalignment of the mold socket and spigot, and output the set of mold geometric parameters. Step 2: Obtain the set of geometric state parameters and corresponding measured wall thickness data for the mold in the historical period, establish a wall thickness coupling mapping model, and calculate the predicted value of wall thickness distribution deviation based on the set of geometric state parameters of the mold. Step 3: Determine the mix proportion adjustment amount and centrifugal molding parameter adjustment amount based on the predicted wall thickness distribution deviation value, generate correction instructions and execute mixing, casting, centrifugation and then send it into the steam curing kiln; Step 4: Obtain the temperature control curve of each pole in the steam curing kiln and obtain the water-cement ratio of the pole. Calculate the equivalent curing age based on the temperature control curve, and call the preset strength-maturity relationship curve based on the water-cement ratio and the equivalent curing age to obtain the current compressive strength. Step 5: Input the predicted wall thickness distribution deviation, the current compressive strength, the minimum compressive strength required for demolding structure safety, and the inner diameter ellipticity into the multi-dimensional coupled evaluation model, perform piecewise coupled calculations of structural uniformity risk and strength reserve risk, and generate demolding quality level index and priority sequence. Step 6: Perform the demolding operation sequentially according to the priority sequence; Step 7: Collect the actual wall thickness detection value of the pole after demolding, compare it with the predicted value of the wall thickness distribution deviation to obtain the prediction residual, and correct the regression coefficient of the wall thickness coupling mapping model in reverse.

[0072] In summary: By digitally acquiring the post-process data of mold geometry and using a wall thickness coupling mapping model, a quantitative correlation between mold deterioration and wall thickness deviation was established, enabling early prediction of wall thickness distribution deviation. Sub-mode process compensation based on deviation axial distribution pattern recognition avoids local adjustments for overall problems or global disturbances for local problems. Online strength estimation, combining temperature control curves and a strength estimation model based on equivalent curing age, ensures demolding decisions are based on actual strength reserves rather than fixed cycles. Segmented coupling evaluation of structural uniformity and strength reserves addresses the risk of demolding damage caused by single-dimensional decision-making. Residual feedback and gradient descent coefficient correction after demolding ensure the model continuously tracks mold deterioration trends, forming a closed-loop management system of acquisition, prediction, compensation, judgment, and backtracking.

[0073] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A smart manufacturing management system for cement poles, characterized in that: include: The mold status acquisition module is used to digitally acquire the inner diameter ovality and the misalignment of the joint from the socket end to the spigot end of the mold after the centrifugation process, and output the set of geometric status parameters of the mold. The wall thickness prediction module is used to obtain the sequence of geometric state parameters of the mold in the historical production cycle of the mold and the measured data of the wall thickness distribution of the corresponding pole after demolding for the current batch of molds. It establishes a wall thickness coupling mapping model between the geometric state parameter set and the wall thickness distribution deviation. Based on the mold geometric state parameter set and the wall thickness coupling mapping model, it calculates the predicted value of the wall thickness distribution deviation for the current batch. The process compensation module is used to determine the mix ratio adjustment amount and the centrifugal molding parameter adjustment amount according to the predicted value of the wall thickness distribution deviation, generate a mix ratio correction command and send it to the mixing station and the centrifugal molding correction command, so as to control the mixing station and the centrifugal molding equipment to perform mixing, casting and centrifugal molding according to the corrected parameters, and then send the molding into the steam curing kiln. The strength determination module is used to obtain the actual temperature control curve and mix ratio parameters of each pole in the steam curing kiln, calculate the equivalent curing age based on the actual temperature control curve, and call the preset strength-maturity relationship curve based on the mix ratio parameters and the equivalent curing age to obtain the current compressive strength. The demolding decision module is used to input the predicted value of the wall thickness distribution deviation, the current compressive strength, the minimum compressive strength required for the demolding structure safety, and the inner diameter ellipticity into a multi-dimensional coupled evaluation model, perform segmented coupled calculations on the structural uniformity risk and strength reserve risk, output a demolding quality level index that characterizes the degree of demolding damage risk, and generate a priority sequence. The scheduling and execution module is used to schedule the demolding devices to perform demolding operations sequentially according to the priority sequence; The backtracking calibration module is used to collect the actual wall thickness detection value of the pole after demolding, compare it with the predicted value of the wall thickness distribution deviation, and correct the coefficients of the wall thickness coupling mapping model in reverse and update it synchronously.

2. The intelligent production and manufacturing management system for cement poles according to claim 1, characterized in that: The method for obtaining the set of geometric state parameters of the mold is as follows: S201. After the centrifugation process is completed, select three equally spaced sections along the mold axis from the socket end to the insert end as geometric condition inspection sections. S202. Perform non-contact scanning on the inner contour of each of the geometric state detection sections to obtain point cloud data of the inner contour of each section. S203. Perform least-squares ellipse fitting on the point cloud data of the inner contour of each cross section, extract the major axis and minor axis of each cross section, and take the ratio of the difference between the major axis and minor axis relative to the standard inner diameter as the ellipticity of the cross section. S204. Divide the point cloud data of each geometric state detection section into upper half mold point cloud and lower half mold point cloud according to the joint orientation, perform circle fitting to obtain two fitting circle centers, and calculate the offset between the two fitting circle centers as the joint misalignment of the section. S205. Integrate the ellipticity and misalignment of each section into the set of geometric state parameters of the mold according to the time sequence and output them.

3. The intelligent production and manufacturing management system for cement poles according to claim 2, characterized in that: The specific method for establishing the wall thickness coupling mapping model is as follows: The wall thickness distribution deviation sequence is obtained by subtracting the measured wall thickness distribution data of each test section from the target wall thickness of the corresponding section. Using the ellipticity and misalignment of the joint in the set of geometric parameters of the mold as independent variables and the wall thickness distribution deviation sequence as dependent variable, the regression coefficients are estimated by multiple linear regression to establish the wall thickness coupling mapping model. The set of mold geometric state parameters for the current batch is input into the wall thickness coupling mapping model to calculate the predicted value of the wall thickness distribution deviation for the current batch.

4. The intelligent manufacturing management system for cement poles according to claim 3, characterized in that: The process compensation module is used for: The predicted wall thickness distribution deviation values ​​on the three axial detection sections are compared with the preset distribution pattern library to identify the current deviation distribution pattern. The deviation distribution pattern includes the overall offset pattern and the local deviation pattern. In the overall offset mode, the mean value of the predicted wall thickness distribution deviation of the three detection sections is calculated. Based on the preset amplitude range in which the mean value is located, the corresponding water-cement ratio adjustment amount and sand ratio adjustment amount are queried from the mix ratio compensation rule table to generate the mix ratio correction instruction. At the same time, the corresponding centrifugal speed adjustment amount and centrifugal duration adjustment amount are queried from the centrifugal molding compensation rule table to generate the centrifugal molding correction instruction. In the local bias mode, only the centrifugal molding compensation rule table is queried to generate the centrifugal molding correction instruction, and the mix ratio correction instruction maintains the default value.

5. The intelligent manufacturing management system for cement poles according to claim 4, characterized in that: The current compressive strength is calculated as follows: S501. Obtain the surface temperature and ambient temperature of each electric rod in the steam curing kiln according to the preset sampling cycle to form the actual temperature control curve, and at the same time obtain the water-cement ratio in the mixing ratio parameter of the electric rod. S502. Based on the temperature data in the actual temperature control curve, the equivalent curing coefficient for each sampling period is calculated using the Arrhenius equation, and the equivalent curing age is obtained by integrating over the sampling time. S503. Based on the water-cement ratio and the equivalent curing age, the current compressive strength is calculated by calling the preset strength-maturity relationship curve.

6. The intelligent production and manufacturing management system for cement poles according to claim 5, characterized in that: The multi-dimensional coupled evaluation model includes a structural homogeneity sub-model and a strength reserve sub-model, wherein: The structural uniformity sub-model is used to calculate the first relative ratio of the maximum value of the ellipticity of the three detection sections to the preset ellipticity warning benchmark value, and the second relative ratio of the absolute value of the predicted wall thickness distribution deviation to the preset wall thickness deviation warning benchmark value. The maximum value of the first relative ratio and the second relative ratio is taken as the structural uniformity risk value. The strength reserve sub-model is used to divide the minimum compressive strength required for the demolding structure by the current compressive strength to obtain the strength reserve ratio, and the value after subtracting 1 from the strength reserve ratio is used as the strength reserve risk value.

7. The intelligent production and manufacturing management system for cement poles according to claim 6, characterized in that: The output represents the demolding quality level index, which characterizes the degree of demolding damage risk, and generates a priority sequence, specifically as follows: First, determine whether the strength reserve risk value is greater than zero; When the value is greater than zero, the strength reserve risk value, the structural uniformity risk value, and the preset penalty constant are summed to obtain the demolding quality grade index. When the value is less than or equal to zero, the structural uniformity risk value is directly used as the demolding quality grade index; The demolding quality grade index of each pole in the steam curing kiln is sorted from smallest to largest. The scheduling execution module performs the demolding operation on the pole with the smallest value first, and then completes the demolding of the remaining poles in sequence.

8. The intelligent manufacturing management system for cement poles according to claim 7, characterized in that: The scheduling and execution module is used to read the pole identifiers sequentially according to the priority sequence, retrieve the demolding process parameters corresponding to the pole identifiers, and control the demolding equipment to perform lifting, demolding, and transfer operations based on the demolding process parameters.

9. The intelligent manufacturing management system for cement poles according to claim 3, characterized in that: The retrospective calibration module performs the following steps: S901. Collect the actual wall thickness of each test section of the pole after demolding; S902. Subtract the actual wall thickness detection value of each section from the target wall thickness of the corresponding section to obtain the actual wall thickness distribution deviation. S903. The difference between the actual wall thickness distribution deviation and the predicted wall thickness distribution deviation is calculated to obtain the prediction residual. S904. Based on the predicted residuals, the regression coefficients of the wall thickness coupling mapping model are corrected in reverse using the gradient descent method. S905. Write the corrected regression coefficients back into the wall thickness coupling mapping model to replace the original coefficients.

10. A method for intelligent production and manufacturing management of cement poles, characterized in that, This method employs an intelligent manufacturing management system for cement poles as described in any one of claims 1-9, comprising the following steps: Step 1: After the centrifugation process is completed, collect the inner diameter ovality and joint misalignment of the mold socket and spigot, and output the set of mold geometric parameters. Step 2: Obtain the set of geometric state parameters and corresponding measured wall thickness data for the mold in the historical period, establish a wall thickness coupling mapping model, and calculate the predicted value of wall thickness distribution deviation based on the set of geometric state parameters of the mold. Step 3: Determine the mix proportion adjustment amount and centrifugal molding parameter adjustment amount based on the predicted wall thickness distribution deviation value, generate correction instructions and execute mixing, casting, centrifugation and then send it into the steam curing kiln; Step 4: Obtain the temperature control curve of each pole in the steam curing kiln and obtain the water-cement ratio of the pole. Calculate the equivalent curing age based on the temperature control curve, and call the preset strength-maturity relationship curve based on the water-cement ratio and the equivalent curing age to obtain the current compressive strength. Step 5: Input the predicted wall thickness distribution deviation, the current compressive strength, the minimum compressive strength required for demolding structure safety, and the inner diameter ellipticity into the multi-dimensional coupled evaluation model, perform piecewise coupled calculations of structural uniformity risk and strength reserve risk, and generate demolding quality level index and priority sequence. Step 6: Perform the demolding operation sequentially according to the priority sequence; Step 7: Collect the actual wall thickness detection value of the pole after demolding, compare it with the predicted value of the wall thickness distribution deviation to obtain the prediction residual, and correct the regression coefficient of the wall thickness coupling mapping model in reverse.

Images