A cement pole centrifugal compactness online monitoring method based on multi-parameter fusion

Abstract

The present application relates to the technical field of cement pole manufacturing, in particular to a cement pole centrifugal compactness online monitoring method based on multi-parameter fusion, three types of signals of motor running current, mold vibration and acoustic emission are synchronously collected and stored in alignment during the centrifugal forming process; exponential decay fitting is carried out according to the centrifugal stage, and the steady time is extracted; the steady coupling coefficient is constructed by the ratio of the difference between the steady time of vibration and acoustic emission signals and the steady time of current and vibration signals, the compactness state is determined by combining the steady order and the qualified coefficient interval determined by statistics; when the order is inconsistent, the fault link is identified and the machine is stopped for alarm, when the coefficient is out of range, the rotating speed and the duration are adjusted according to the deviation direction or a transition centrifugal stage is inserted, the effectiveness is verified by re-determination and deviation distance convergence; after the forming is completed, a monitoring file is formed and the qualified coefficient interval is updated by reverse verification, the present application realizes online quantitative determination, fault positioning and process self-adjustment of the centrifugal compactness of the cement pole.

Description

Technical Field

[0001] This invention relates to the field of cement pole manufacturing technology, specifically to an online monitoring method for the centrifugal density of cement poles based on multi-parameter fusion. Background Technology

[0002] Cement poles are widely used foundation components in overhead transmission lines and distribution networks. Their molding generally adopts centrifugal technology: after the concrete mixture is fed into a longitudinally split steel mold, the centrifuge drives the mold to rotate around its own axis. Under the action of centrifugal force, the concrete adheres to the mold wall and is gradually squeezed and compacted, eventually forming a hollow ring-shaped pole structure. The centrifugal molding process is usually divided into several stages, such as low-speed material distribution, medium-speed transition, and high-speed compaction. The rotation speed and duration of each stage directly determine the degree of compaction of the concrete, and the degree of compaction is the core indicator that determines the demolding strength and finished product quality of the pole.

[0003] Currently, the process control of centrifugal molding of cement poles mainly relies on manual experience: the rotation speed and duration of each centrifugation stage are set by the operator based on experience according to the specifications of the poles, and whether the centrifugation process is dense is judged afterward by observing the operation status of the centrifuge, listening to the operating sound, or sampling and testing the strength after demolding.

[0004] The above methods have significant shortcomings: First, the determination of compaction lags behind the molding process, and insufficient compaction is often only discovered during demolding inspection, at which point the rods have become scrap, resulting in a waste of materials and time. Second, process parameters rely on manual experience for setting and adjustment; if the adjustment is too small, repeated adjustments are required, while if the adjustment is too large, it can easily cause over-vibration and segregation. Furthermore, inconsistencies in processes between different operators and different batches lead to fluctuations in product quality. Third, when equipment abnormalities such as mold loosening or sensor failure occur during centrifugation, existing monitoring methods cannot distinguish them from insufficient compaction. Fault location relies on troubleshooting after shutdown, resulting in long downtime.

[0005] Therefore, this invention provides an online monitoring method for the centrifugal density of cement poles based on multi-parameter fusion. Summary of the Invention

[0006] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0007] The technical solution adopted by this invention to solve its technical problem is: an online monitoring method for the centrifugal density of cement poles based on multi-parameter fusion, comprising: S1. Before centrifugal molding, establish the monitoring task for the current rod, and associate the rod's specification information, concrete mix ratio information, and rotation speed and duration of each centrifugation stage; during centrifugal molding, synchronously collect the operating current signal of the centrifuge drive motor, the vibration signal of the centrifuge mold, and the acoustic emission signal attached to the outer wall of the mold to obtain three types of signals, and store the three types of signals in alignment. S2. Divide the three types of signals into segments according to the centrifugal stage, and perform exponential decay fitting on the signal in each segment to obtain the decay time constant and the stabilization time. S3. The time difference obtained by subtracting the stabilization time of the acoustic emission signal from the stabilization time of the vibration signal is used as the first interval, and the time difference obtained by subtracting the stabilization time of the vibration signal from the stabilization time of the operating current signal is used as the second interval; when the first interval or the second interval is zero, it is determined that the current centrifugation stage is abnormally compacted and an alarm is triggered; when both the first interval and the second interval are not zero, the ratio of the first interval to the second interval is used as the stabilization coupling coefficient. S4. When the acoustic emission signal first stabilizes, the vibration signal second stabilizes, and the operating current signal finally stabilizes, and the stabilization coupling coefficient is within the qualified coefficient range, it is determined that the current centrifugation stage is in normal compaction. The qualified coefficient range is determined by statistical analysis of the stabilization coupling coefficient of the same centrifugation stage during the qualified centrifugation process of rods of the same specification, and each centrifugation stage corresponds to its own qualified coefficient range. When the stabilization order is inconsistent and both the first interval and the second interval are not zero, the fault link on the dense transmission chain is identified according to the order of the stabilization times of the three types of signals, the corresponding fault prompt information is output, and a shutdown alarm is triggered. When the stabilization sequence is consistent but the stabilization coupling coefficient exceeds the qualified coefficient range, it is determined that the compaction is insufficient: when the stabilization coupling coefficient is greater than the upper limit of the qualified coefficient range, the rotation speed of the current centrifugation stage is increased by the first rotation speed increment, and the duration is extended by the first delay duration; when the stabilization coupling coefficient is less than the lower limit of the qualified coefficient range, a transition centrifugation stage is inserted after the current centrifugation stage. After adjustment, repeat steps S2 to S4 for re-judgment, and compare the deviation distance of the current stabilizing coupling coefficient from the qualified coefficient range with the deviation distance of the previous judgment: if the current deviation distance is less than the previous deviation distance and the stabilizing coupling coefficient is still outside the qualified coefficient range, continue to adjust and re-judge; if the current deviation distance is greater than or equal to the previous deviation distance, determine that the adjustment is invalid, determine that the current centrifugal forming of the rod is unqualified, and stop the machine with an alarm. S5. After centrifugal molding is completed, the decay time constant, stabilization time, stabilization coupling coefficient, judgment result and adjustment record of each centrifugation stage are associated and stored with the identification information of the rod to form a centrifugal density monitoring file.

[0008] Preferably, in step S1, the determination of the rotation speed and duration of each centrifugation stage includes: Based on the specifications of the rod, query the historical centrifugal density monitoring records of rods of the same specifications; when a qualified historical centrifugal density monitoring record exists, retrieve the last adjusted rotation speed and duration recorded therein, and use them as the rotation speed and duration for each centrifugation stage of the current rod. When there is no qualified historical centrifugation density monitoring file, the preset initial process parameters are used as the rotation speed and duration of each centrifugation stage of the current rod, and the current rod is marked as the first rod. The stabilizing coupling coefficient obtained in this centrifugation process is used as the initial sample to establish the qualified coefficient range of the rod of this specification.

[0009] Preferably, in step S1, the alignment storage includes: The rotation pulse signal of the centrifuge mold is collected; whenever the rotation pulse signal indicates that the centrifuge mold has rotated through a preset angle interval, the sampling of the operating current signal, the vibration signal and the acoustic emission signal is triggered simultaneously. The sampled values ​​obtained from each sampling are associated with and stored with the corresponding angular positions, so that the sampling points of the three types of signals are aligned with the angular positions of the centrifugal mold. The average of consecutive sampled values ​​at the same diagonal position is superimposed, and the average value is used as the sampled value at that diagonal position.

[0010] Preferably, in step S2, the signal within each segment is processed as follows: S21. Obtain the actual rotational speed of the centrifuge in each segment; determine the moment when the actual rotational speed in the segment first reaches the rotational speed of the centrifugation stage as the transition end moment; when the actual rotational speed at the end of the segment still has not reached the rotational speed of the centrifugation stage, determine that the centrifuge speed is abnormal and stop the machine with an alarm; discard the sampled data between the start moment of the segment and the transition end moment, and take the remaining sampled data as the valid sampling interval. S22. Perform exponential decay fitting on the signal within the effective sampling interval to obtain the goodness of fit; S23. Shift the transition end time back by one sampling interval, redetermine the effective sampling interval, and re-perform exponential decay fitting; when the goodness of fit of the current exponential decay fitting is greater than the goodness of fit of the previous exponential decay fitting, repeat step S23. When the goodness of fit of the current exponential decay fitting is less than or equal to the goodness of fit of the previous exponential decay fitting, or when there is no sampled data in the effective sampling interval, step S23 is stopped. S24. When the goodness of fit of each exponential decay fitting is lower than the preset goodness of fit threshold, the current segment signal is determined to be abnormal, an abnormal signal quality prompt is output, and a shutdown alarm is triggered. The fit with the highest goodness among all exponential decay fits is taken as the target fit; the decay time constant obtained by solving the target fit is determined as the decay time constant of this segment, and the stabilization time is determined based on the fitting curve obtained by the target fit.

[0011] Preferably, in step S2, the stabilization time is determined as follows: Obtain the initial and stable amplitudes obtained from the exponential decay fitting; The absolute value of the difference between the signal amplitude and the stable amplitude of the fitted curve obtained by the target fitting at each time moment is calculated as the deviation value at each time moment; The product of the preset ratio and the initial amplitude is used as the stabilization threshold. The moment when the deviation value first becomes less than or equal to the stabilization threshold is determined as the stabilization moment.

[0012] Preferably, in step S4, when the stabilizing coupling coefficient is greater than the upper limit of the qualified coefficient range, the first speed increment and the first delay duration are determined as follows: Obtain adjustment records from the historical centrifugal compaction monitoring archives of rods of the same specification; the adjustment records include the first speed increment, the first delay duration, and the stabilizing coupling coefficient obtained by re-judgment after each adjustment. Select qualified adjustment records from the adjustment records after re-judgment, whose stabilizing coupling coefficient falls back to the qualified coefficient range; determine the average value of the first speed increment in each qualified adjustment record as the first speed increment of this adjustment, and determine the average value of the first delay duration in each qualified adjustment record as the first delay duration of this adjustment; When no qualified adjustment record exists, the preset initial speed increment is determined as the first speed increment of this adjustment, and the preset initial delay duration is determined as the first delay duration of this adjustment.

[0013] Preferably, in step S4: When the stabilizing coupling coefficient is less than the lower limit of the qualified coefficient range and the current centrifugation stage is not the last centrifugation stage, a transition centrifugation stage is inserted after the current centrifugation stage; the rotation speed of the transition centrifugation stage is the average of the rotation speed of the current centrifugation stage and the rotation speed of the next centrifugation stage. When the current centrifugation stage is the last centrifugation stage, the current rod is determined to be unqualified for centrifugation forming, and the machine is stopped and an alarm is issued; The duration of the transition centrifugation phase is half the duration of the current centrifugation phase; during the transition centrifugation phase, steps S2 to S4 are repeated. When the re-judgment result of the transition centrifugation stage is that the compaction is normal, the next centrifugation stage is entered; when the re-judgment result of the transition centrifugation stage is that the compaction is insufficient, the transition centrifugation stage is not continued, the current rod centrifugation is determined to be unqualified, and the machine is stopped and an alarm is issued.

[0014] Preferably, in step S4, when re-executing steps S2 to S4 after adjustment, the validity of the adjustment is verified in the following manner: When the stabilizing coupling coefficient is greater than the upper limit of the qualified coefficient range, the deviation distance is the difference between the stabilizing coupling coefficient and the upper limit; when the stabilizing coupling coefficient is less than the lower limit of the qualified coefficient range, the deviation distance is the difference between the lower limit and the stabilizing coupling coefficient. Compare the deviation distance of this re-judgment with the deviation distance of the previous judgment; if the deviation distance of this re-judgment is less than the deviation distance of the previous judgment, and the stabilizing coupling coefficient is still outside the qualified coefficient range, continue to adjust and re-judge according to the adjustment method in step S4; If the current deviation distance is greater than or equal to the previous deviation distance, the adjustment is deemed invalid, the current centrifugal forming of the rod is deemed unqualified, and the machine is stopped with an alarm.

[0015] Preferably, in step S4, identifying the faulty link in the dense transmission chain based on the sequential relationship of the stabilization times of the three types of signals includes: When the stabilization time of the operating current signal is earlier than the stabilization time of the vibration signal, it is determined that the force transmission between the mold and the concrete is interrupted, and a mold loosening fault prompt is output. When the stabilization time of the vibration signal is earlier than the stabilization time of the acoustic emission signal, it is determined that the acquisition of the acoustic emission signal has failed, and an acoustic emission sensor fault prompt is output. If the acoustic emission signal does not stabilize within the specified duration, and both the vibration signal and the operating current signal have stabilized, it is determined that the concrete feeding amount is insufficient, and an abnormal feeding prompt is output. When the stabilization order of the three types of signals does not conform to the above three situations and does not conform to the normal stabilization order, an abnormal output transmission chain prompt will be issued, and a shutdown alarm will be triggered.

[0016] Preferably, step S5 includes: Obtain the strength test results of the rod after demolding, and associate the strength test results with the centrifugal density monitoring file of the rod; When the judgment results of each centrifugation stage are all normal, the stabilizing coupling coefficients of each centrifugation stage of the rod are first added to the sample set of the corresponding centrifugation stage; when the strength test result is qualified, it is retained in the sample set of the corresponding centrifugation stage; when the strength test result is unqualified, the stabilizing coupling coefficients of each centrifugation stage of the rod are removed from the sample set of the corresponding centrifugation stage, and a monitoring criterion review prompt is output. Based on the sample set after addition or removal, the pass coefficient range for the corresponding centrifugation stage is re-statistically determined.

[0017] This invention provides an online monitoring method for the centrifugal compaction of cement poles based on multi-parameter fusion. It has the following beneficial effects: 1. This invention simultaneously collects three types of signals during the centrifugal molding process: motor operating current, mold vibration, and acoustic emission. The stabilization time of each signal is extracted through exponential decay fitting. A stabilization coupling coefficient is constructed based on the ratio of the stabilization time intervals. The density state is then compositely judged by combining the stabilization order and the statistically determined qualified coefficient range. The determination of density is shifted from post-demolding sampling to online quantitative determination during the centrifugation process. Insufficient density can be detected and dealt with during the molding process, avoiding the generation of batches of defective products and eliminating batch-to-batch quality fluctuations caused by manual experience judgment.

[0018] 2. This invention uses the transmission chain of the concrete compaction state along the material interior, mold, and drive system as the physical basis, and maps the disordered pattern of the three types of signal stabilization order to the fault links on the chain one by one. While judging the abnormal compaction molding, it outputs fault location prompts such as mold loosening, acoustic emission sensor failure, and insufficient feeding. This upgrades the abnormality handling from a single shutdown alarm to fault location and shutdown protection in parallel. Maintenance personnel can directly locate the fault location without stopping the machine to check item by item, thus shortening the fault diagnosis time.

[0019] 3. This invention performs differentiated adjustments based on the direction of the deviation of the stabilizing coupling coefficient from the qualified coefficient range, and automatically verifies the effectiveness of the adjustment through the convergence of the deviation distance. The adjustment amount is generated by statistical analysis of historical adjustment records that have been verified as valid through re-judgment. The qualified coefficient range is back-verified by the demolding strength test results and continuously updated, forming a self-evolving closed loop of judgment, adjustment, verification, and record accumulation. Process parameters and adjustment amounts are automatically corrected as production data accumulates, eliminating the need for manual setting based on experience and periodic calibration. Invalid adjustments are terminated the first time they occur, saving energy and reducing equipment wear and tear. Attached Figure Description

[0020] Figure 1 This is a flowchart of the online centrifugal density monitoring method of the present invention; Figure 2 This is a flowchart illustrating the closed-loop evolution of the pass coefficient interval statistics of the present invention. Detailed Implementation

[0021] 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.

[0022] like Figures 1 to 2 As shown, this invention proposes an online monitoring method for the centrifugal compaction of cement poles based on multi-parameter fusion, including: S1, before centrifugation, establishing the monitoring task for the current pole, associating the pole's specification information, concrete mix ratio information, and the rotation speed and duration of each centrifugation stage; during centrifugation, synchronously collecting the operating current signal of the centrifuge drive motor, the vibration signal of the centrifuge mold, and the acoustic emission signal attached to the outer wall of the mold, obtaining three types of signals, and aligning and storing the three types of signals; The acoustic emission sensor transmits the collected signals to the monitoring system via a wireless transmission module (or slip ring); S2. Divide the three types of signals into segments according to the centrifugal stage, and perform exponential decay fitting on the signal in each segment to obtain the decay time constant and the stabilization time. S3. The time difference obtained by subtracting the stabilization time of the acoustic emission signal from the stabilization time of the vibration signal is the first interval, and the time difference obtained by subtracting the stabilization time of the vibration signal from the stabilization time of the operating current signal is the second interval; when the first interval or the second interval is zero, it is determined that the current centrifugation stage is abnormally compacted and an alarm is triggered; when neither the first interval nor the second interval is zero, the ratio of the first interval to the second interval is used as the stabilization coupling coefficient. S4. When the acoustic emission signal first stabilizes, the vibration signal second stabilizes, and the operating current signal finally stabilizes, and the stabilization coupling coefficient is within the qualified coefficient range, it is determined that the current centrifugation stage is normal in terms of compaction and forming. The qualified coefficient range is determined by the statistical analysis of the stabilization coupling coefficient of the same centrifugation stage during the qualified centrifugation process of the same specification rod. Each centrifugation stage corresponds to its own qualified coefficient range. It should be noted that the physical basis for the above stabilization sequence is as follows: During centrifugal compaction, the rearrangement of particles and compaction of pores inside the concrete first cause the acoustic emission signal to stabilize; after the change in the compaction state of the material is transmitted through the mold, the vibration signal then stabilizes; after compaction, the concrete adheres to the wall and forms a ring, causing a change in load, and the operating current signal finally stabilizes; that is, the compaction state is transmitted sequentially along the path of the material interior, the mold, and the drive system, and the stabilization sequence of the three types of signals reflects the sequential response of this transmission chain. Therefore, during normal centrifugation, the acoustic emission signal stabilizes first, the vibration signal stabilizes second, and the operating current signal stabilizes last. When the stabilization order is incorrect and both the first and second intervals are not zero, the faulty link on the dense transmission chain is identified based on the stabilization time relationship of the three types of signals, the corresponding fault prompt information is output, and a shutdown alarm is triggered. When the stabilization sequence matches but the stabilization coupling coefficient exceeds the acceptable coefficient range, the compaction is deemed insufficient: when the stabilization coupling coefficient is greater than the upper limit of the acceptable coefficient range, the rotation speed of the current centrifugation stage is increased by the first rotation speed increment, and the duration is extended by the first delay duration; when the stabilization coupling coefficient is less than the lower limit of the acceptable coefficient range, a transition centrifugation stage is inserted after the current centrifugation stage. After adjustment, repeat steps S2 to S4 for re-judgment, and compare the deviation distance of the current stabilizing coupling coefficient from the qualified coefficient range with the deviation distance of the previous judgment: if the current deviation distance is less than the previous deviation distance and the stabilizing coupling coefficient is still outside the qualified coefficient range, continue to adjust and re-judge; if the current deviation distance is greater than or equal to the previous deviation distance, the adjustment is deemed invalid, the current centrifugal forming of the rod is deemed unqualified, and the machine is stopped and an alarm is issued. S5. After centrifugation molding is completed, the decay time constant, stabilization time, stabilization coupling coefficient, judgment results and adjustment records of each centrifugation stage are associated and stored with the identification information of the rod to form a centrifugation density monitoring file.

[0023] In some implementations, the first speed increment and the first delay duration are determined based on adjustment records in the historical centrifugal density monitoring archives of rods of the same specification: qualified adjustment records that have stabilized and fallen back to the qualified coefficient range after re-judgment are selected from the adjustment records, and the average of the first speed increment and the first delay duration in each qualified adjustment record is used as the first speed increment and the first delay duration for this adjustment, respectively; when there are no qualified adjustment records, a preset initial speed increment and a preset initial delay duration are used. Thus, the adjustment amount is statistically generated from the historical adjustment records that have been verified to be valid after re-judgment, and is automatically corrected as production data accumulates, without the need for manual setting based on experience, avoiding repeated adjustments caused by too small an adjustment amount and over-vibration segregation caused by too large an adjustment amount.

[0024] As an optional embodiment, in actual production, historical data is lacking when new specification rods are first put into production, making it impossible to statistically determine the range of pass coefficients, and thus the range determination in step S4 cannot be performed; moreover, the centrifugation process parameters rely on manual experience for setting, and inconsistencies in the process between different batches of rods of the same specification lead to fluctuations in compaction quality. Therefore, this embodiment adopts the following technical solution: Specifically, after receiving the monitoring task for the current member, the monitoring system uses the member's specifications as the search criteria to query the archives of members with the same specifications in the historical centrifugal density monitoring archive. The specifications of the member include at least the tip diameter, length, and wall thickness. Members with all three being the same specification are members with the same specifications.

[0025] When a qualified historical centrifugal compaction monitoring file is found, the most recent qualified file is retrieved, and the actual rotation speed and duration of each centrifugation stage are read, written into the current rod's monitoring task, and subsequent monitoring is executed.

[0026] It should be noted that a qualified historical centrifugation density monitoring file refers to a file in which the judgment results of each centrifugation stage are normal and no shutdown alarm occurs; the final actual speed and duration refer to the process parameters executed for the last time after adjustment and re-judgment, rather than the initial settings when the task was established. These parameters have been fully re-judged and verified, indicating that the process for this specification of rod has been confirmed to be feasible.

[0027] When no qualified historical centrifugation density monitoring file is found, the preset initial process parameters are read according to the current rod specification category, written into the monitoring task, and the current rod is marked as the first rod. The preset initial process parameters refer to the rotation speed and duration of each centrifugation stage, which are pre-stored in the monitoring system and set according to the rod specification category. The first rod refers to the first rod of a certain specification that is centrifuged according to the preset initial process parameters when no qualified historical monitoring file is available. After the first rod is centrifuged, the stabilizing coupling coefficients obtained from each centrifugation stage of this centrifugation process are written into the initial sample set of the corresponding centrifugation stage of the rod specification. In some implementations, after each successful centrifugation process, the stabilizing coupling coefficients of each centrifugation stage of the rod are added to the sample set of the corresponding centrifugation stage. Let M be the number of samples in the sample set of the j-th centrifugation stage, μ be the sample mean, and σ be the sample standard deviation. When M reaches a preset minimum sample size... At that time, establish the range of acceptable coefficients for members of this specification. Where j is the centrifugation stage number, , These are the sample mean and sample standard deviation for the j-th centrifugation stage, respectively. At that time, the lower limit of the qualified coefficient interval of the centrifugation stage is a preset minimum positive value, which is a pre-set value greater than 0. Since the stabilization coupling coefficient is the ratio of two time intervals, its value is always positive. If the lower limit is zero or negative, the interval has no physical meaning.

[0028] It should be noted that the purpose of establishing the acceptable coefficient range is to provide a judgment criterion for step S4: In step S4, the stabilizing coupling coefficient of the current member needs to be compared with the acceptable coefficient range to determine whether the compaction is normal or insufficient. Therefore, only by establishing this range can the above judgment be executed. (Minimum sample size preset) This refers to the minimum sample size required to ensure statistical stability of the sample mean μ and sample standard deviation σ; its value is predetermined. When the sample size is less than... When the sample size is small, σ fluctuates wildly, making the established upper and lower limits of the interval unreliable. During the accumulation phase, only samples are recorded, and interval determination is not performed until M reaches [a certain value]. Then the qualified coefficient range established by statistics is used.

[0029] It should also be noted that the records of rods that experience a shutdown alarm or are ultimately deemed unqualified during centrifugation are not included in the sample accumulation, but are only used for anomaly tracing to prevent abnormal data from causing the sample mean μ and sample standard deviation σ to deviate from the true statistical level of qualified rods.

[0030] Through the above methods, the centrifugal process parameters of rods of the same specification are automatically inherited and evolved with the qualified files, eliminating batch-to-batch process fluctuations caused by manual settings; at the same time, rods of new specifications are included in the monitoring system from the first piece, and the accumulation of samples and the establishment of the qualified coefficient range are completed automatically, so that the transition from the production of new specifications to the determination of stability can be achieved without manual intervention.

[0031] As an optional embodiment, the rotational speeds differ in each stage of centrifugal molding, and there are rotational speed fluctuations within the same stage due to load changes. When sampling at equal time intervals, the distribution of sampling points for the three types of signals on the mold circumference drifts with the rotational speed, causing distortion of the periodic features related to the mold angular position in the vibration and acoustic emission signals, which in turn affects the extraction accuracy of the decay time constant and the stabilization moment. Therefore, this embodiment uses the rotational angle of the centrifugal mold as a unified benchmark to synchronously sample and align the three types of signals for storage. Specifically, the monitoring system collects the rotation pulse signal of the centrifuge mold. It should be noted that the rotation pulse signal refers to the electrical signal that outputs pulses in fixed angular units as the centrifuge mold rotates. It can be generated by an encoder or Hall sensor installed on the rotating shaft of the centrifuge mold. Each pulse corresponds to the centrifuge mold rotating through a fixed angular unit.

[0032] The monitoring system counts the rotation pulse signals. Whenever the accumulated number of pulses indicates that the centrifuge mold has rotated through a preset angle interval, it synchronously triggers a sampling of the operating current signal, vibration signal, and acoustic emission signal. Among them, the preset angle interval refers to the angle through which the centrifugal mold rotates between two adjacent samplings. Its value is preset and remains unchanged throughout the centrifugal forming process; and synchronous triggering means that the sampling of the three types of signals is started by the same trigger signal, so that the sampling time of the three types of signals is the same in each sampling.

[0033] In some implementations, the preset angle interval is determined as follows: Let P be the number of pulses output per revolution of the centrifuge mold, and N be the number of sampling points per revolution. Then, whenever the cumulative... A sampling is triggered once every pulse; specifically, the value of N should be such that the number of sampling points in each revolution meets the analysis requirements of vibration signal and acoustic emission signal. For example, N=64, that is, a sampling is triggered once every 5.625° revolution of the centrifuge mold.

[0034] After each sampling, the monitoring system associates and stores the sampled value with the angular position of the centrifuge mold when the sampling was triggered; The angular position is determined based on the preset zero position on the centrifuge mold and the cumulative angle rotated. The sampling points of the three types of signals are thus indexed by the angular position, and any sampling point corresponds to the same angular position in the three types of signals, thereby aligning the three types of signals according to the angular position of the centrifuge mold. It should be noted that this alignment method is different from the method of aligning by timestamp after equal-time sampling: under equal-time sampling, the alignment reference of the three types of signals is the sampling time, and the angular position of the sampling point is inconsistent with the rotation speed. In this application, the sampling is directly triggered by the rotation pulse signal, and the alignment reference of the three types of signals is the angular position itself, which is independent of the rotation speed.

[0035] In the subsequent step S2, when performing segmented and exponentially decayed fitting, the signal sequences within each segment are arranged in angular position order, and the angular difference between adjacent sampling points is always equal to the preset angular interval.

[0036] It should be noted that the number of revolutions in the superimposed average of multiple revolutions is much smaller than the total number of revolutions contained in each centrifugation stage; the monitoring system takes the sampled value sequence obtained by superimposing and averaging each consecutive multiple revolution as the midpoint of the consecutive multiple revolutions and arranges them in chronological order according to the timestamps to form a signal sequence that changes with time in each segment; the subsequent step S2 extracts the signal amplitude point by point from the signal sequence to obtain the amplitude sequence that decays with time, which is used for exponential decay fitting.

[0037] In this way, regardless of how the speed is switched in each centrifugation stage or how the speed fluctuates within the same stage, the sampling points of the three types of signals are always in the same position on the circumference of the mold. The periodic characteristics related to the angular position will not be distorted due to the change in speed, thereby improving the extraction accuracy of the decay time constant and the stabilization moment. At the same time, the sampling of the three types of signals is started by the same trigger signal, without relying on the external synchronization clock between each acquisition channel, which reduces the hardware requirements for synchronous acquisition.

[0038] As an optional embodiment, the speed increases during the switching of centrifugation stages. The three types of signals at the beginning of each stage reflect the speed transition process rather than the concrete compaction decay process. Furthermore, the impact oscillation may continue for a period of time after the speed reaches the set value, and the length of the transition segment varies with different speed settings. Insufficient or excessive rejection will reduce the accuracy of the exponential decay fitting. Therefore, this embodiment processes the signals in each segment as follows: Specifically, the monitoring system acquires the actual rotational speed of the centrifuge in each segment. The actual rotational speed refers to the rotational speed indicated by the speed feedback signal of the centrifuge drive motor, with the unit being r / min, which changes continuously over time. The rotational speed recorded in each centrifugation stage in the monitoring task is the set target value, while the actual rotational speed is the measured process value. There is a difference between the two during the speed ramp-up stage.

[0039] The monitoring system determines the moment when the actual rotational speed within a segment first reaches the rotational speed of that centrifugation stage as the transition end moment. This moment signifies the end of the speed climb and the entry of the speed into a stable operating state. The sampled data between the start moment of the segment and the transition end moment are discarded, and the remaining sampled data is the valid sampling interval. If the actual rotational speed still does not reach the rotational speed of that centrifugation stage at the end of the segment, it indicates that the centrifuge cannot reach the set speed. The cause may be a drive motor failure or a transmission system abnormality. The monitoring system determines that the centrifuge speed is abnormal and stops the machine with an alarm. The reason for using the speed reaching moment to locate the boundary of the transition segment is that the length of the transition segment is determined by the speed climb process, and the moment when the speed climb ends can be directly read from the speed feedback signal without the need for manually setting rejection parameters.

[0040] Before performing exponential decay fitting on the signal within the effective sampling interval, the signal amplitude at each sampling point must be obtained. It should be noted that the signal amplitude is a scalar that characterizes the signal strength at the sampling point. The original sampling data of the three types of signals are waveform sequences that oscillate over time, and they must be converted into point-by-point amplitude sequences before fitting can be performed. Specifically, for the operating current signal, the effective value of the current at each sampling point is used as the signal amplitude, with the unit being A; for the vibration signal, the vibration velocity envelope value at each sampling point is used as the signal amplitude, with the unit being mm / s; for the acoustic emission signal, the acoustic emission envelope amplitude at each sampling point is used as the signal amplitude, with the unit being mV. Here, the envelope value or envelope amplitude refers to the numerical value that characterizes the waveform contour intensity obtained after extracting the envelope from the waveform sequence. The envelope extraction can be performed using the Hilbert transform.

[0041] The reason for choosing the envelope value as the amplitude of the vibration signal and the acoustic emission signal is that the instantaneous values ​​of these two types of signals oscillate around zero, making direct fitting meaningless. The envelope value, however, decays monotonically with the compaction process, which can characterize the signal intensity change trend. The running current signal is a unipolar signal, and its effective value itself characterizes the signal intensity, so the effective value is used directly.

[0042] Subsequently, the signal amplitude sequence was fitted with an exponential decay model. It should be noted that the exponential decay model was chosen because during centrifugal compaction, the amplitudes of all three types of signals monotonically decay and tend to stabilize as the concrete gradually compacts, representing a typical relaxation process. The exponential decay model is a commonly used model to describe relaxation processes, and its model parameters have clear physical meanings. The fitted model is as follows: ; Where t is the time taken as the origin from the start time of the effective sampling interval, in seconds; Let be the signal amplitude at time t. C is the initial amplitude, which is the initial value of the attenuable part of the signal amplitude at t=0, and C is the stable amplitude, which is the constant value that the signal amplitude tends to after attenuation ends. , The units of C, C, and C are consistent with the signal amplitude of the corresponding signals; e is the natural constant, which is the base of the natural logarithm. Its value is approximately 2.718 and it is dimensionless. Let τ be the attenuation factor, representing the proportion of the attenuation amplitude at time t to the initial attenuation amplitude, where both t and τ are in seconds, and their ratio t / τ is dimensionless. It is a dimensionless number, and its relation to Dimensions after multiplication Consistent, the dimensions on both sides of the formula are the same.

[0043] τ is the decay time constant, in seconds, which characterizes the rate at which the signal decays to a stable state: when t=τ, That is, after time τ, the signal attenuation amplitude decreases to approximately 36.8% of the initial attenuation amplitude; as t approaches infinity, As the amplitude approaches 0, the signal amplitude decays to a stable amplitude C; therefore, the smaller τ is, the faster the signal stabilizes, meaning the concrete compacts and stabilizes more quickly. It should be noted that the signal amplitude dimensions of the three types of signals are different, but when fitting each type of signal individually... , Since the units of C and τ are self-consistent and the dimension of τ is unified as s, the three types of signals can be processed using the same fitting process, and there is no problem of inconsistent dimensions. The exponential decay fitting uses the least squares method, which aims to minimize the sum of squared residuals between the fitted curve and the signal amplitude at each sampling point. The least squares method was chosen because it is a classic method for curve fitting, which can suppress random noise in the sampled data, and the solution results are stable and the calculation process is deterministic.

[0044] After each fit, the goodness of fit is calculated. In some implementations, the coefficient of determination is used to determine the goodness of fit. It should be noted that the coefficient of determination is used. The reason is: As a dimensionless quantity, its value is between 0 and 1. Its value does not change with the signal dimension and amplitude. Therefore, the same goodness-of-fit evaluation standard can be used for three types of signals with different dimensions, and there is no need to set evaluation parameters for different signals. The closer the value is to 1, the better the fit between the fitted curve and the sampled data.

[0045] In some implementations, the monitoring system continues to refine the effective sampling interval based on the transition end time. Specifically, the transition end time is shifted backward by one sampling interval, where the sampling interval is the time interval between two adjacent samples, measured in seconds. This interval is the inherent minimum time unit of the sampled data. Using the sampling interval as the shift step size allows the boundary correction to reach the highest accuracy allowed by the sampled data. Each shift re-determines the effective sampling interval and re-performs exponential decay fitting; when the determination coefficient of this fitting... Greater than the coefficient of determination of the previous fit When the coefficient of determination for this fit is reached, continue shifting to the next step; when the coefficient of determination for this fit is reached... Less than or equal to the coefficient of determination of the previous fit If no sampling data exists within the effective sampling interval, the shift will stop.

[0046] It should be noted that a gradual shift was adopted and... The basis for determining the stopping time based on the comparison results is that: after the rotational speed reaches the target, the impact oscillation may still continue; moving the stop time further can eliminate residual oscillation data. This increases accordingly; however, excessive shifting will discard sampled data that normally reflects the dense attenuation process, resulting in a reduced amount of remaining data and an incomplete attenuation process. It then declined; therefore The point where the increase turns into a decrease is the boundary where the degree of elimination is most appropriate.

[0047] It should be noted that the coefficient of determination for each exponential decay fit is... When all values ​​are below the preset goodness-of-fit threshold, it indicates that the signal of the current segment does not conform to the exponential decay law. This could be due to sensor malfunction, signal transmission interruption, or the concrete not forming a normal compaction decay process. In this case, even if the largest R² value is used for the fit, the resulting decay time constant and stabilization time are unreliable. Therefore, the monitoring system determines that the current segment signal is abnormal, outputs a signal quality anomaly warning, and stops the system with an alarm, ceasing subsequent extraction of the stabilization time and calculation of the stabilization coupling coefficient. The preset goodness-of-fit threshold is pre-set, for example, to 0.8.

[0048] After completing the above processing, the coefficients of determination in each exponential decay fit are... The largest fit is taken as the target fit, and the τ obtained from solving the target fit is determined as the decay time constant of this segment. The stabilization time is determined based on the signal after the target fit.

[0049] Specifically, the stabilization time is determined by the formula. Confirmed, among which For the time to stabilize, The signal amplitude at time t is the fitted curve obtained by fitting the target. C is the initial amplitude, C is the stable amplitude, and η is the preset ratio. The preset ratio η is set in advance, for example, 5%. That is, when the difference between the amplitude of the fitted curve and the stable amplitude first shrinks to within 5% of the initial amplitude, the signal is considered to have stabilized.

[0050] The reason for using a fitted curve instead of the original sampled amplitude to determine the stabilization time is that the original sampled amplitude contains random noise, which may occasionally meet the judgment condition due to noise fluctuations in the early stage of decay, leading to an premature misjudgment of the stabilization time; the fitted curve decays monotonically, and the judgment result is stable and unique. The stabilization time read from the fitted curve is taken as the origin of the start time of the effective sampling interval, and the time difference between the start time of the effective sampling interval and the start time of the centrifugation stage is added. That is, it is converted into the stabilization time with the start time of the centrifugation stage as the timing origin, which is used for the subsequent calculation of the first and second intervals.

[0051] In this way, the boundary of the transition section is automatically located at the actual speed when the target speed is reached, and is gradually corrected by the goodness of fit feedback with the sampling interval as the step size. It can accommodate the situation where the length of the transition section is different at different speed gears without the need to manually set the rejection parameters. At the same time, the feature output is the one with the largest goodness of fit, which avoids feature distortion caused by insufficient rejection and data incompleteness caused by excessive rejection, and improves the extraction accuracy of decay time constant and stabilization time.

[0052] As an optional embodiment, the three types of signals come from three independent acquisition channels. The system clocks of each channel may differ, and the start times of each centrifugation stage are different. If the stabilization time uses the clock readings of each channel, the time base difference will mix into the calculation results of the first and second intervals, making the intervals unable to truly reflect the stabilization sequence of the three types of signals. In addition, the compaction process is transmitted along the path of the material interior, mold, and drive system, and there will inevitably be a time lag in the stabilization of adjacent links. If any interval is zero, it indicates that the transmission chain has been broken. Therefore, this embodiment makes the following limitations on the calculation of the stabilization coupling coefficient: Specifically, in step S3, each stabilization time uses the start time of its respective centrifugation stage as the timing origin. Specifically, the stabilization time read from the fitted curve is added to the time difference between the start time of the effective sampling interval and the start time of the respective centrifugation stage to obtain the stabilization time with the start time of the respective centrifugation stage as the timing origin. The timing origin refers to the zero point of the time coordinate when calculating the stabilization time. The stabilization times of the three types of signals are all re-timed with the start time of the same centrifugation stage as the zero point, so that the stabilization times of the three types of signals are in the same time coordinate system. The difference between the first interval and the second interval is only determined by the actual stabilization order of the three types of signals and does not include the time base difference between each acquisition channel.

[0053] Before calculating the stabilizing coupling coefficient, the monitoring system first checks whether the first and second intervals are zero. The physical basis for checking the zero interval is as follows: During the centrifugal compaction process, the rearrangement of particles and the compaction of pores inside the concrete first cause the acoustic emission signal to stabilize; after the change in the compaction state of the material is transmitted through the mold, the vibration signal then stabilizes; after compaction, the concrete adheres to the wall and forms a ring, causing a change in load, and the operating current signal finally stabilizes. The above transmission process must have a time lag, so during normal centrifugation, both the first and second intervals are greater than zero.

[0054] If the first interval is zero, meaning both the acoustic emission signal and the vibration signal stabilize simultaneously, it indicates an abnormal transmission relationship between the internal compaction event of the material and the mechanical response of the mold. If the second interval is zero, meaning both the vibration signal and the operating current signal stabilize simultaneously, it indicates an abnormal transmission relationship between the mechanical response of the mold and the load response of the drive system. These abnormalities can be caused by mold loosening, sensor failure, or transmission system malfunction.

[0055] In some implementations, when the first or second interval is zero, the monitoring system does not calculate the stabilizing coupling coefficient, but directly determines that the current centrifugation stage is abnormally compacted and issues a shutdown alarm. It should be noted that the reason for not using fault tolerance processing and calculating normally here is that the stabilizing coupling coefficient is the ratio of the first interval to the second interval. When the second interval is zero, this ratio is meaningless, and when the first interval is zero, this ratio is always zero. In both cases, the calculation results cannot reflect the compaction state. Moreover, the zero interval itself is a manifestation of the destruction of the conduction chain and has independent abnormal indication significance. Therefore, the zero interval is directly used as the abnormality criterion, rather than the boundary fault tolerance case of numerical calculation.

[0056] When both the first and second intervals are non-zero, the stabilization coupling coefficient is calculated based on the ratio of the first interval to the second interval, and subsequent order and interval determinations are performed. It should be noted that the zero-interval check and the order and interval determinations in step S4 are sequential: after the monitoring system unifies the timing origin at the stabilization moment, it first performs the zero-interval check; only when both the first and second intervals are non-zero is the stabilization coupling coefficient calculated and the determination in step S4 performed. Thus, the case where the order is consistent but a certain interval is zero is intercepted before the stabilization coupling coefficient is calculated and will not flow into the subsequent interval determination.

[0057] As an optional embodiment, in centrifugal molding adjustment, the first speed increment and the first delay duration directly affect the adjustment effect: if the values ​​are too small, insufficient compaction cannot be eliminated and repeated adjustments are required; if the values ​​are too large, it may cause excessive vibration and segregation. If these two parameters are set manually based on experience, the adjustment effect depends on the operator's experience level, and the adjustment amounts are inconsistent between different batches of rods of the same specification. To solve the above problems, this application determines the first speed increment and the first delay duration based on the historical adjustment records of rods of the same specification; Specifically, the centrifugal density monitoring file formed in step S5 records the first speed increment, the first delay duration, and the stabilization coupling coefficient obtained after adjustment for each adjustment. These three together constitute an adjustment record. The first speed increment, the first delay duration, and the stabilization coupling coefficient obtained after adjustment in the adjustment record correspond to each other and can be used to determine whether each adjustment is effective.

[0058] Before the monitoring system performs step S4, it acquires adjustment records from the historical centrifugal density monitoring archives of members of the same specification and selects qualified adjustment records from them. Qualified adjustment records refer to those whose stabilizing coupling coefficient falls back to the qualified coefficient range after re-judgment, i.e., the adjustment records that have been verified as valid after re-judgment. Adjustment records whose stabilizing coupling coefficient is still outside the qualified coefficient range after re-judgment are invalid adjustment records, and their adjustment parameters have not been verified as valid and are not included in subsequent statistics.

[0059] After selecting qualified adjustment records, the first speed increment and the first delay duration for this adjustment are determined using the following formula: ; Where K is the total number of qualified adjustment records, which is dimensionless; This is the first speed increment used in the i-th qualified adjustment record, in r / min; The first delay duration used in the i-th qualified adjustment record, in seconds; This is the first speed increment in this adjustment, in r / min; This is the first delay duration in this adjustment, measured in seconds.

[0060] It should be noted that the above formula is an arithmetic mean formula, with the same units on both sides, r / min or s. The reason for using the arithmetic mean is that the degree of insufficient compaction corresponding to each qualified adjustment record is different, and the adjustment amount of a single adjustment record is random. The arithmetic mean of K effective adjustment amounts represents the typical adjustment amount required to eliminate insufficient compaction of the rod of this specification, which can reduce the influence of random deviation of a single record. When there are no qualified adjustment records in the historical archives (K=0), the above formula cannot be calculated. In this case, the preset initial speed increment is determined as the first speed increment of this adjustment, and the preset initial delay duration is determined as the first delay duration of this adjustment. It should be noted that the preset initial speed increment and preset initial delay duration refer to the initial adjustment parameters that are pre-stored in the monitoring system and set according to the pole specification category. Their values ​​correspond to a small adjustment range and are used for the first adjustment when there is no effective adjustment experience for the pole specification. After the first adjustment, if it is deemed effective, this adjustment becomes a qualified adjustment record, and subsequent adjustments will use the above arithmetic average. The preset initial speed increment and preset initial delay duration will no longer be used.

[0061] In this way, the first speed increment and the first delay duration are generated by statistical analysis of historical adjustment records that have been verified and validated. The adjustment amount is automatically corrected as production data accumulates, without the need for manual setting based on experience. At the same time, the use of statistical averages instead of single records reduces the random deviation caused by differences in the degree of insufficient compaction, ensuring that the adjustment amount is consistent between different batches of rods of the same specification. This avoids repeated adjustments caused by excessively small adjustment amounts and excessive vibration segregation caused by excessively large adjustment amounts.

[0062] If the stabilization sequence is consistent, it indicates that the compaction transmission chain is complete and the changes in the compaction state can be transmitted normally along the material interior, mold, and drive system, which can rule out faults such as mold loosening, acoustic emission sensor failure, and insufficient feeding. At this time, the stabilization coupling coefficient exceeds the qualified coefficient range, which indicates that the compaction rhythm itself deviates from the statistical level of qualified rods, so it is judged that the compaction is insufficient.

[0063] The fact that the stabilization coupling coefficient is greater than the upper limit of the acceptable coefficient range indicates that the first interval is relatively large compared to the second interval. That is, after the acoustic emission signal stabilizes, the vibration signal fails to stabilize for a long time, and the transmission time from the compaction event inside the concrete to the mechanical response of the mold is too long, reflecting that the concrete compaction speed is too slow and the degree of compaction is insufficient. Increasing the rotation speed of the current centrifugal stage can increase the centrifugal force on the concrete, and extending the duration can increase the compaction time. Both can accelerate the compaction process and shorten the first interval, so that the stabilization coupling coefficient falls back to the acceptable coefficient range. Therefore, the adjustment method of increasing the rotation speed and extending the duration is adopted for this situation.

[0064] As an optional embodiment, if the stabilization coupling coefficient is less than the lower limit of the qualified coefficient range, it indicates that the vibration signal stabilizes too quickly, the material is locked too early, and the compaction rhythm is out of balance. If the next centrifugation stage is directly entered at this time, the speed will increase abruptly, which will impact the concrete that has not yet been fully compacted, causing the coarse and fine aggregates to separate and exacerbating the uneven compaction. If the current centrifugation stage is repeated, the concrete has already stabilized at the current speed, and simply repeating it will not change the compaction state.

[0065] Therefore, this embodiment inserts a transition centrifugation stage between the current centrifugation stage and the next centrifugation stage, allowing the concrete to continue to compact at the connecting speed: Specifically, the rotational speed of the transitional centrifugation stage is the average of the rotational speed of the current centrifugation stage and the rotational speed of the next centrifugation stage. It should be noted that the reason for using the average rotational speed is that the role of the transitional centrifugation stage is to form a connection between the two rotational speeds. The average rotational speed is higher than the rotational speed of the current centrifugation stage, which can provide new centrifugal force increments to the concrete and promote its continued compaction; while it is lower than the rotational speed of the next centrifugation stage, so as to avoid the impact of a speed jump on the insufficiently compacted concrete.

[0066] Let the rotational speed during the current centrifugation stage be... The rotation speed in the next centrifugation stage is The units are all r / min, so the rotation speed during the transition centrifugation stage is... The unit is r / min, and the units on both sides of the formula must be consistent. For example, if the rotation speed of the current centrifugation stage is 400 r / min and the rotation speed of the next centrifugation stage is 700 r / min, then the rotation speed of the transition centrifugation stage is 550 r / min.

[0067] The duration of the transition centrifugation stage is half the duration of the current centrifugation stage. It should be noted that the reason for using half the duration of the current centrifugation stage is that the transition centrifugation stage is a connecting stage, and its duration should match the compaction rhythm of the current centrifugation stage. If the duration is too short, the centrifugal force increment will not have been fully utilized before the speed is switched, and the compaction state cannot be improved. If the duration is too long, it will be close to the current centrifugation stage itself, lose the meaning of transition, and easily lead to over-vibration segregation.

[0068] Take half the duration of the current centrifugation phase, somewhere in between, and this value is directly determined by the parameters already recorded in the monitoring task, without the need for manual setting.

[0069] During the transition centrifugation phase, steps S2 to S4 are repeated, which involves acquiring, segmenting, fitting, calculating the stabilizing coupling coefficient, and determining the compact state of the three types of signals during the transition centrifugation phase.

[0070] The transitional centrifugation stage is treated as an independent centrifugation stage and its stabilization coupling coefficient is also compared with the qualified coefficient range.

[0071] When the re-judgment result of the transition centrifugation stage is normal, it indicates that the concrete has completed compaction at the connecting speed and enters the next centrifugation stage. When the re-judgment result of the transition centrifugation stage is insufficient compaction, it indicates that insufficient compaction cannot be eliminated by continuing centrifugation at the connecting speed. The cause may be improper mix ratio or uneven material distribution. Continuing to adjust the speed and duration cannot remedy the situation. At this time, the current member is judged to be unqualified for centrifugation, and the machine is stopped and an alarm is triggered to avoid ineffective adjustments that would cause energy waste and equipment damage.

[0072] If the stabilization coupling coefficient is greater than the upper limit of the acceptable coefficient range, it indicates that the first interval is too large compared to the second interval. That is, after the acoustic emission signal stabilizes, the vibration signal fails to stabilize for a long time, and the transmission time from the compaction event inside the concrete to the mechanical response of the mold is too long, reflecting that the concrete compaction speed is too slow and the degree of compaction is insufficient. Increasing the rotation speed of the current centrifugal stage can increase the centrifugal force on the concrete, and extending the duration can increase the compaction time. Both can accelerate the compaction process and shorten the first interval, so that the stabilization coupling coefficient falls back to the acceptable coefficient range. Therefore, for the case where the stabilization coupling coefficient is greater than the upper limit, the adjustment method of increasing the rotation speed and extending the duration is adopted.

[0073] As an optional embodiment, the purpose of the re-judgment after adjustment is to verify whether the adjustment is effective. If the adjustment is effective, the stabilizing coupling coefficient should move closer to the qualified coefficient range. If the deviation between the stabilizing coupling coefficient and the qualified coefficient range does not decrease or even increases after adjustment, it indicates that insufficient compaction cannot be remedied by speed and duration. Continuing to adjust will only waste energy and increase equipment wear. To solve the problem that invalid adjustments cannot be automatically terminated, this embodiment verifies the effectiveness of the adjustment during the re-judgment: Specifically, when re-executing steps S2 to S4 after adjustment, the monitoring system calculates the deviation distance, which is used to characterize the degree to which the stabilizing coupling coefficient deviates from the qualified coefficient range. Its calculation method corresponds to the direction of deviation: when the stabilizing coupling coefficient is greater than the upper limit of the qualified coefficient range, the deviation distance is the difference between the stabilizing coupling coefficient and the upper limit; when the stabilizing coupling coefficient is less than the lower limit of the qualified coefficient range, the deviation distance is the difference between the lower limit and the stabilizing coupling coefficient.

[0074] Let the stabilizing coupling coefficient be r, and the upper limit of the qualified coefficient interval be... The lower limit is If the deviation distance is d, then the deviation direction is when the upper limit is exceeded. When the deviation direction is below the lower limit Among them, r, , All are dimensionless numbers, therefore d is also dimensionless. The deviation distance is always non-negative; the smaller the value, the closer the stabilization coupling coefficient is to the acceptable range, meaning the closer the compaction rhythm of the member is to that of a qualified member. After calculating the deviation distance of this reassessment, it is compared with the deviation distance of the previous assessment. The reason for using the comparison of two adjacent deviation distances as the verification basis is that the change of rotation speed and duration directly changes the centrifugal force and centrifugal time of the concrete, which will inevitably be reflected in the change of compaction speed, and thus cause the stabilization coupling coefficient to change accordingly. Therefore, when the adjustment is effective, the deviation distance will necessarily decrease. If the deviation distance does not decrease, it means that the adjustment has failed to change the compaction state. The cause may be improper mix proportion, uneven material distribution or mold failure. Such defects cannot be eliminated by adjusting the rotation speed and duration.

[0075] When the current deviation distance is less than the previous deviation distance, and the stabilizing coupling coefficient is still outside the qualified coefficient range, it indicates that the adjustment direction is correct but the adjustment amount is not enough to restore the compaction rhythm to the qualified level. The monitoring system continues to adjust and re-judge according to the adjustment method in step S4 until the stabilizing coupling coefficient falls back to the qualified coefficient range, and it is determined that the current centrifugal stage compaction is normal.

[0076] It should be noted that the above "continue to adjust and re-judge" only applies to the case where the adjustment method is to increase the speed of the current centrifugation stage by the first speed increment and extend the duration by the first delay duration. When the adjustment method is to insert a transition centrifugation stage, the transition centrifugation stage is a one-time connection stage. If the re-judgment result is still insufficient compaction, the transition centrifugation stage will not be inserted again. Instead, the current rod centrifugation is directly judged to be unqualified and the machine will be stopped and an alarm will be issued.

[0077] When the current deviation distance is greater than or equal to the previous deviation distance, it indicates that the adjustment is invalid. The monitoring system determines that the current centrifugal forming of the rod is unqualified and stops the machine with an alarm. It should be noted that the judgment is made when the comparison results of two adjacent deviation distances do not converge. There is no need to manually set the allowed number of re-judgments: as long as the adjustment is effective, the re-judgment can continue. Once the adjustment is invalid, it is immediately rejected and the machine is stopped. Invalid adjustment is terminated when it occurs for the first time.

[0078] In this way, the effectiveness of the adjustment is automatically verified by comparing the deviation distances of two adjacent re-judgments, without the need to manually set an upper limit on the number of re-judgments. When the adjustment is effective, iterative adjustments are made until the compacted molding is normal. When the adjustment is ineffective, it is judged as scrapped and the machine is stopped on the first occurrence, avoiding repeated execution of ineffective adjustments, saving energy and reducing equipment wear and tear. At the same time, the calculation of the deviation distance corresponds to the deviation direction, so that the same verification logic is used for both the case of exceeding the upper limit and the case of falling below the lower limit, making the judgment process simple and the standard consistent.

[0079] During centrifugal compaction, the compaction state of concrete is transmitted along the path of the material interior, mold, and drive system. The stabilization order of the three types of signals reflects the sequential response of the transmission chain. When the stabilization order is inconsistent, the transmission chain breaks at a certain link. Different disordered order patterns correspond to faults at different locations on the chain. If only the compaction and molding abnormality is judged and the machine is stopped and alarmed without distinguishing the fault type, maintenance personnel cannot know the location of the fault and need to check item by item, resulting in long downtime.

[0080] Therefore, this embodiment identifies specific situations where the stabilization order is inconsistent and outputs corresponding fault prompt information accordingly.

[0081] Specifically, the misalignment of the stabilization order includes the following four situations: The first scenario: The stabilization time of the operating current signal is earlier than that of the vibration signal. It should be noted that during normal centrifugation, the change in the compaction state of the concrete is transmitted to the drive system through the mold. The stabilization of the operating current signal will necessarily be later than that of the vibration signal. When the operating current signal stabilizes before the vibration signal, it indicates that the load change of the drive system has not been transmitted through the mold, that is, the force transmission between the mold and the concrete is interrupted. The cause is that the mold is loose or the mold is separated from the concrete. At this time, it is determined that the force transmission between the mold and the concrete is interrupted, and a mold loosening fault prompt is output.

[0082] The second scenario: The stabilization time of the vibration signal precedes that of the acoustic emission signal. It should be noted that the acoustic emission signal reflects the microscopic compaction events within the concrete and is the starting point of the transmission chain; therefore, its stabilization must occur earliest. When the vibration signal stabilizes before the acoustic emission signal, it indicates that the acoustic emission signal has not captured the actual compaction process within the concrete, meaning the acoustic emission signal acquisition has failed. This is caused by poor coupling between the acoustic emission sensor and the outer wall of the mold, or by sensor damage. In this case, the acoustic emission signal acquisition is deemed a failure, and an acoustic emission sensor fault warning is output.

[0083] The third scenario: The acoustic emission signal does not stabilize during the duration, while the vibration signal and the operating current signal have stabilized. "Not stabilizing" means that the signal amplitude still does not meet the criteria for stabilization by the end of the centrifugation phase. It should be noted that the vibration signal and the operating current signal have both stabilized, indicating that the response chain of the mold and the drive system is complete and the fault is not in the subsequent links of the transmission path; however, the acoustic emission signal continues to be unstable, indicating that the compaction process inside the concrete has not been completed. Under the premise that the equipment is operating normally, the cause is that the amount of concrete fed is insufficient, and the amount of material cannot form a stable compacted structure at the current speed. At this time, it is determined that the amount of concrete fed is insufficient, and an abnormal feeding prompt is output.

[0084] The fourth scenario: When the stabilization order of the three types of signals does not conform to the above three scenarios and does not conform to the normal stabilization order, it indicates that the transmission chain has experienced disorder outside of the above three typical modes. It cannot be classified into any of the following categories: mold loosening, acoustic emission sensor failure, or insufficient feeding. The monitoring system outputs a transmission chain abnormality prompt and stops the machine with an alarm.

[0085] In some implementations, when the stabilization sequence is determined to be inconsistent, the monitoring system first identifies the specific circumstances of the sequence disorder, outputs the corresponding fault prompt information, and then executes a shutdown alarm. It should be noted that the execution order of identification before shutdown ensures that the fault prompt information is output and recorded before shutdown, allowing maintenance personnel to directly locate the fault location based on the fault prompt information without having to troubleshoot item by item after shutdown.

[0086] In this way, the disordered pattern of the stabilization sequence corresponds one-to-one with the fault location on the transmission chain, upgrading the abnormal handling from a single shutdown alarm to a parallel fault location and shutdown protection, thus shortening the fault investigation time. At the same time, the judgment of the above situations is based solely on the order of the stabilization time and whether the system has stabilized, without the need to add detection equipment or set additional parameters. The judgment criteria are clear and easy to implement.

[0087] As an optional embodiment, each centrifugation stage corresponds to its own pass coefficient range. The pass coefficient range of each centrifugation stage is determined by the statistical analysis of the stabilizing coupling coefficient of the same centrifugation stage in the qualified centrifugation process of the same specification rod. The "qualified centrifugation process" is determined according to the process criterion in step S4. However, the process criterion may miss some cases: some rods are judged to be in normal compaction at each stage of the centrifugation process, but the measured strength after demolding is unqualified. If the stabilizing coupling coefficient of such rods is left in the statistical sample, it will cause the sample mean and standard deviation to gradually deviate from the true qualified level, resulting in the pass coefficient range being slowly pulled off course, and this deviation is difficult to be detected by the process criterion itself.

[0088] Therefore, in this embodiment, after centrifugal molding is completed, the strength test results are introduced to reverse verify the process criteria, and the pass coefficient range is updated accordingly.

[0089] Specifically, the monitoring system acquires the strength test results of the members after demolding and links these results with the centrifugal compaction monitoring files of the members. The strength test results refer to the conclusion of whether the strength value measured by standard test methods after demolding meets the design requirements. The results are divided into qualified and unqualified. Strength is the final criterion for judging the density of concrete. Members with qualified demolding strength must have reached a qualified level in the centrifugal compaction process, while members with unqualified demolding strength must have defects in the centrifugal compaction process.

[0090] Furthermore, the correlation refers to writing the strength test results into the centrifugal density monitoring file of the corresponding rod, so that the process monitoring data and the result test data of the same rod correspond one-to-one and can be mutually verified.

[0091] Reverse verification is only performed on rods whose judgment results at each centrifugation stage are normal compaction. It should be noted that the stability coupling coefficient of rods judged as insufficient or abnormal compaction is not included in the statistical sample and therefore does not need to be verified. The verification object is rods that have passed the process criteria, and the purpose is to find the missed samples.

[0092] When the judgment results of each centrifugation stage are all normal compaction and the strength test results are qualified, it indicates that the rod has passed the dual confirmation of process criteria and result criteria, and its stabilizing coupling coefficient is reliable. The monitoring system adds the stabilizing coupling coefficient of each centrifugation stage of the rod to the sample set of the corresponding centrifugation stage.

[0093] When the judgment results of each centrifugation stage are all normal compaction and the strength test results are unqualified, it indicates that the rod is a missed sample of the process criterion. The monitoring system removes the stabilization coupling coefficient of each centrifugation stage of the rod from the sample set of the corresponding centrifugation stage and outputs a monitoring criterion review prompt.

[0094] The monitoring criterion review prompt refers to the information that prompts maintenance personnel to review the qualified coefficient range and each stabilization judgment parameter; the occurrence of missed samples indicates that there is a gap in the current criterion's discrimination capability, and the review content includes whether the upper and lower limits of the qualified coefficient range are too wide and whether the determination of the stabilization time is accurate.

[0095] Each time a sample is added or removed, the monitoring system re-determines the pass coefficient range based on the added or removed samples. The statistical method is the same as the initial establishment method of the pass coefficient range. It should be noted that the pass coefficient range is continuously updated with production, and its samples are always maintained by the bars confirmed by strength testing. The range gradually converges to the true pass level as the number of samples increases.

[0096] Through the above methods, the judgment results of the process criteria are reverse-verified by the final result criterion of strength testing. Missed samples are promptly removed, and the pass coefficient range is not skewed by missed samples. At the same time, the addition and removal of samples are automatically executed with the strength testing of each member, and the pass coefficient range remains consistent throughout the production process without the need for manual periodic calibration. In addition, the verification prompts output when missed samples occur enable maintenance personnel to promptly identify and correct the judgment gaps of the criteria, and the monitoring reliability continues to improve with production.

[0097] In summary, this invention uses the dense transmission chain of the material interior, mold, and drive system as a physical model. It quantifies the stabilization time sequence relationship of three types of signals—motor current, vibration, and acoustic emission—as the stabilization coupling coefficient and stabilization order criterion. This enables online quantitative determination of the centrifugal compaction of cement poles, automatic location of fault links, and closed-loop self-adjustment of process parameters. Furthermore, the determination benchmark is continuously updated and preserved through historical archive statistics and reverse verification of demolding strength. The entire process requires no human experience intervention, which avoids waste loss and excessive vibration segregation risks caused by insufficient compaction, shortens fault diagnosis time, reduces energy and equipment losses from ineffective adjustments, and significantly improves the quality consistency and production automation level of centrifugal molding of cement poles.

[0098] 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 method for online monitoring of the centrifugal density of cement poles based on multi-parameter fusion, characterized in that: include: S1. Before centrifugal molding, establish a monitoring task for the current rod, and associate the rod's specification information, concrete mix proportion information, and the rotation speed and duration of each centrifugation stage. During the centrifugal molding process, the operating current signal of the centrifuge drive motor, the vibration signal of the centrifugal mold, and the acoustic emission signal attached to the outer wall of the mold are collected simultaneously to obtain three types of signals, and the three types of signals are aligned and stored. S2. Divide the three types of signals into segments according to the centrifugal stage, and perform exponential decay fitting on the signal in each segment to obtain the decay time constant and the stabilization time. S3. The time difference obtained by subtracting the stabilization time of the acoustic emission signal from the stabilization time of the vibration signal is used as the first interval, and the time difference obtained by subtracting the stabilization time of the vibration signal from the stabilization time of the operating current signal is used as the second interval; when the first interval or the second interval is zero, it is determined that the current centrifugation stage is abnormally compacted and an alarm is triggered; when both the first interval and the second interval are not zero, the stabilization coupling coefficient is used as the ratio of the first interval to the second interval. S4. When the acoustic emission signal first stabilizes, the vibration signal second stabilizes, and the operating current signal finally stabilizes, and the stabilization coupling coefficient is within the qualified coefficient range, it is determined that the current centrifugation stage is in normal compaction. The qualified coefficient range is determined by the statistical analysis of the stabilization coupling coefficient of the same centrifugation stage during the qualified centrifugation process of rods of the same specification, and each centrifugation stage corresponds to its own qualified coefficient range. When the stabilization order is inconsistent and both the first interval and the second interval are not zero, the fault link on the dense transmission chain is identified according to the order of the stabilization times of the three types of signals, the corresponding fault prompt information is output, and a shutdown alarm is triggered. When the stabilization sequence is consistent but the stabilization coupling coefficient exceeds the qualified coefficient range, it is determined that the compaction is insufficient. When the stabilization coupling coefficient is greater than the upper limit of the qualified coefficient range, the rotation speed of the current centrifugation stage is increased by the first rotation speed increment, and the duration is extended by the first delay duration. When the stabilizing coupling coefficient is less than the lower limit of the qualified coefficient range, a transition centrifugation stage is inserted after the current centrifugation stage. After adjustment, repeat steps S2 to S4 for re-judgment, and compare the deviation distance of the current stabilizing coupling coefficient from the qualified coefficient range with the deviation distance of the previous judgment: if the current deviation distance is less than the previous deviation distance and the stabilizing coupling coefficient is still outside the qualified coefficient range, continue to adjust and re-judge; if the current deviation distance is greater than or equal to the previous deviation distance, determine that the adjustment is invalid, determine that the current centrifugal forming of the rod is unqualified, and stop the machine with an alarm. S5. After centrifugal molding is completed, the decay time constant, stabilization time, stabilization coupling coefficient, judgment result and adjustment record of each centrifugation stage are associated and stored with the identification information of the rod to form a centrifugal density monitoring file.

2. The method for online monitoring of centrifugal density of cement poles based on multi-parameter fusion according to claim 1, characterized in that: In step S1, the determination of the rotation speed and duration of each centrifugation stage includes: Based on the specifications of the rod, query the historical centrifugal density monitoring records of rods of the same specifications; when a qualified historical centrifugal density monitoring record exists, retrieve the last adjusted rotation speed and duration recorded therein, and use them as the rotation speed and duration for each centrifugation stage of the current rod. When there is no qualified historical centrifugation density monitoring file, the preset initial process parameters are used as the rotation speed and duration of each centrifugation stage of the current rod, and the current rod is marked as the first rod. The stabilizing coupling coefficient obtained in this centrifugation process is used as the initial sample to establish the qualified coefficient range of the rod of this specification.

3. The method for online monitoring of centrifugal density of cement poles based on multi-parameter fusion according to claim 1, characterized in that: In step S1, the alignment storage includes: The rotation pulse signal of the centrifuge mold is collected; whenever the rotation pulse signal indicates that the centrifuge mold has rotated through a preset angle interval, the sampling of the operating current signal, the vibration signal and the acoustic emission signal is triggered simultaneously. The sampled values ​​obtained from each sampling are associated with and stored with the corresponding angular positions, so that the sampling points of the three types of signals are aligned with the angular positions of the centrifugal mold. The average of consecutive sampled values ​​at the same diagonal position is superimposed, and the average value is used as the sampled value at that diagonal position.

4. The method for online monitoring of centrifugal density of cement poles based on multi-parameter fusion according to claim 3, characterized in that: In step S2, the signal within each segment is processed as follows: S21. Obtain the actual rotational speed of the centrifuge in each segment; determine the moment when the actual rotational speed in the segment first reaches the rotational speed of the centrifugation stage as the transition end moment; when the actual rotational speed at the end of the segment still has not reached the rotational speed of the centrifugation stage, determine that the centrifuge speed is abnormal and stop the machine with an alarm; discard the sampled data between the start moment of the segment and the transition end moment, and take the remaining sampled data as the valid sampling interval. S22. Perform exponential decay fitting on the signal within the effective sampling interval to obtain the goodness of fit; S23. Shift the transition end time back by one sampling interval, redetermine the effective sampling interval, and re-perform exponential decay fitting; when the goodness of fit of the current exponential decay fitting is greater than the goodness of fit of the previous exponential decay fitting, repeat step S23. When the goodness of fit of the current exponential decay fitting is less than or equal to the goodness of fit of the previous exponential decay fitting, or when there is no sampled data in the effective sampling interval, step S23 is stopped. S24. When the goodness of fit of each exponential decay fitting is lower than the preset goodness of fit threshold, the current segment signal is determined to be abnormal, an abnormal signal quality prompt is output, and a shutdown alarm is triggered. The fit with the highest goodness among all exponential decay fits is taken as the target fit; the decay time constant obtained by solving the target fit is determined as the decay time constant of this segment, and the stabilization time is determined based on the fitting curve obtained by the target fit.

5. The method for online monitoring of centrifugal density of cement poles based on multi-parameter fusion according to claim 1, characterized in that: In step S2, the stabilization time is determined as follows: Obtain the initial and stable amplitudes obtained from the exponential decay fitting; The absolute value of the difference between the signal amplitude and the stable amplitude of the fitted curve obtained by the target fitting at each time moment is calculated as the deviation value at each time moment; The product of the preset ratio and the initial amplitude is used as the stabilization threshold. The moment when the deviation value first becomes less than or equal to the stabilization threshold is determined as the stabilization moment.

6. The method for online monitoring of centrifugal density of cement poles based on multi-parameter fusion according to claim 1, characterized in that: In step S4, when the stabilizing coupling coefficient is greater than the upper limit of the qualified coefficient range, the first speed increment and the first delay duration are determined as follows: Obtain adjustment records from the historical centrifugal compaction monitoring archives of rods of the same specification; the adjustment records include the first speed increment, the first delay duration, and the stabilizing coupling coefficient obtained by re-judgment after each adjustment. Select qualified adjustment records from the adjustment records after re-judgment, whose stabilizing coupling coefficient falls back to the qualified coefficient range; determine the average value of the first speed increment in each qualified adjustment record as the first speed increment of this adjustment, and determine the average value of the first delay duration in each qualified adjustment record as the first delay duration of this adjustment; When no qualified adjustment record exists, the preset initial speed increment is determined as the first speed increment of this adjustment, and the preset initial delay duration is determined as the first delay duration of this adjustment.

7. The method for online monitoring of centrifugal density of cement poles based on multi-parameter fusion according to claim 1, characterized in that: In step S4: When the stabilizing coupling coefficient is less than the lower limit of the qualified coefficient range and the current centrifugation stage is not the last centrifugation stage, a transition centrifugation stage is inserted after the current centrifugation stage; the rotation speed of the transition centrifugation stage is the average of the rotation speed of the current centrifugation stage and the rotation speed of the next centrifugation stage. When the current centrifugation stage is the last centrifugation stage, the current rod is determined to be unqualified for centrifugation forming, and the machine is stopped and an alarm is issued; The duration of the transition centrifugation phase is half the duration of the current centrifugation phase; during the transition centrifugation phase, steps S2 to S4 are repeated. When the re-judgment result of the transition centrifugation stage is that the compaction is normal, the next centrifugation stage is entered; when the re-judgment result of the transition centrifugation stage is that the compaction is insufficient, the transition centrifugation stage is not continued, the current rod centrifugation is determined to be unqualified, and the machine is stopped and an alarm is issued.

8. The method for online monitoring of centrifugal density of cement poles based on multi-parameter fusion according to claim 7, characterized in that: In step S4, when re-executing steps S2 to S4 after adjustment, the validity of the adjustment is verified as follows: When the stabilizing coupling coefficient is greater than the upper limit of the qualified coefficient range, the deviation distance is the difference between the stabilizing coupling coefficient and the upper limit; when the stabilizing coupling coefficient is less than the lower limit of the qualified coefficient range, the deviation distance is the difference between the lower limit and the stabilizing coupling coefficient. Compare the deviation distance of this re-judgment with the deviation distance of the previous judgment; if the deviation distance of this re-judgment is less than the deviation distance of the previous judgment, and the stabilizing coupling coefficient is still outside the qualified coefficient range, continue to adjust and re-judge according to the adjustment method in step S4; If the current deviation distance is greater than or equal to the previous deviation distance, the adjustment is deemed invalid, the current centrifugal forming of the rod is deemed unqualified, and the machine is stopped with an alarm.

9. The method for online monitoring of centrifugal density of cement poles based on multi-parameter fusion according to claim 8, characterized in that: In step S4, identifying the faulty link in the dense transmission chain based on the chronological relationship of the stabilization times of the three types of signals includes: When the stabilization time of the operating current signal is earlier than the stabilization time of the vibration signal, it is determined that the force transmission between the mold and the concrete is interrupted, and a mold loosening fault prompt is output. When the stabilization time of the vibration signal is earlier than the stabilization time of the acoustic emission signal, it is determined that the acquisition of the acoustic emission signal has failed, and an acoustic emission sensor fault prompt is output. If the acoustic emission signal does not stabilize within the specified duration, while the vibration signal and operating current signal have stabilized, it is determined that the concrete feeding amount is insufficient, and an abnormal feeding prompt is output. When the stabilization order of the three types of signals does not conform to the above three situations and does not conform to the normal stabilization order, an abnormal output transmission chain prompt will be issued, and a shutdown alarm will be triggered.

10. The method for online monitoring of centrifugal density of cement poles based on multi-parameter fusion according to claim 1, characterized in that: Step S5 includes: Obtain the strength test results of the rod after demolding, and associate the strength test results with the centrifugal density monitoring file of the rod; When the judgment results of each centrifugation stage are all normal, the stabilizing coupling coefficients of each centrifugation stage of the rod are first added to the sample set of the corresponding centrifugation stage; when the strength test result is qualified, it is retained in the sample set of the corresponding centrifugation stage; when the strength test result is unqualified, the stabilizing coupling coefficients of each centrifugation stage of the rod are removed from the sample set of the corresponding centrifugation stage, and a monitoring criterion review prompt is output. Based on the sample set after addition or removal, the pass coefficient range for the corresponding centrifugation stage is re-statistically determined.

Images