CN-121433047-B - Pressure-sensitive self-repairing asphalt cold-patch mixing intelligent control system and method

CN121433047BCN 121433047 BCN121433047 BCN 121433047BCN-121433047-B

Abstract

The invention belongs to the technical field of intelligent manufacturing control, and particularly relates to a pressure-sensitive self-repairing asphalt cold-patch mixing intelligent control system and method, wherein the method comprises the steps of collecting torque time sequence data of a mixing motor; the method comprises the steps of constructing dynamic feature vectors of torque time sequence data at each time point based on the torque time sequence data, performing unsupervised analysis on feature spaces formed by the dynamic feature vectors by utilizing a local anomaly factor algorithm, calculating local anomaly factors corresponding to each time point, and executing a closed-loop control strategy according to the features of the local anomaly factors and the features of the dynamic feature vectors, wherein the features are used for representing asymmetry, and therefore intelligent control of a mixing process is achieved. The invention can accurately distinguish different mixing working conditions, can stop in time to protect the activity of the product, and can reasonably judge the mixing uniformity through the asymmetry characteristic at the same time, thereby improving the production efficiency.

Inventors

LI LEI
SUN LIYUAN
Xin Chengyang

Assignees

山东润昇工程材料有限公司

Dates

Publication Date: 20260508
Application Date: 20251105

Claims (8)

1. The intelligent control method for the pressure-sensitive self-repairing asphalt cold-patch mixing is characterized by comprising the following steps: collecting torque time sequence data of a mixing motor; constructing a dynamic feature vector of the torque time sequence data at each time point based on the torque time sequence data, wherein the dynamic feature vector is used for representing the asymmetry and transient kurtosis of the torque time sequence data fluctuation; The feature space formed by the dynamic feature vectors is subjected to unsupervised analysis by utilizing a local anomaly factor algorithm, and the feature space comprises two high-density clusters which respectively represent normal mixing and component agglomeration working conditions, wherein sparse anomaly points which represent gel coat cracking working conditions are identified in the feature space, the asymmetry characteristic of the sparse anomaly points is approximately-1, and the transient spike degree characteristic is far greater than 1; calculating local anomaly factors corresponding to each time point; The closed loop control is carried out on the basis of the local abnormal factors and the dynamic characteristic vectors so as to realize intelligent control of the mixing process, and the closed loop control comprises the steps of diagnosing gel coat breakage and immediately stopping mixing when the local abnormal factors at the moment t are larger than a gel coat breakage alarm threshold, diagnosing component aggregation when the local abnormal factors are not larger than the gel coat breakage alarm threshold and the asymmetry characteristic is larger than a component aggregation diagnostic threshold, prolonging mixing time, and diagnosing mixing completion when the mixing time is larger than standard mixing time and the local abnormal factors are not larger than the gel coat breakage alarm threshold and the corresponding asymmetry characteristic is not larger than a mixing uniformity judgment threshold.
2. The intelligent control method for the pressure-sensitive self-repairing asphalt cold-patch mixing according to claim 1, wherein the dynamic feature vector is a two-dimensional feature vector, and the two-dimensional feature vector comprises an asymmetric fluctuation index for evaluating the directionality of the fluctuation of the torque time sequence data and a transient spike index for evaluating the persistence of the fluctuation of the torque time sequence data.
3. The intelligent control method for mixing the pressure-sensitive self-repairing asphalt cold-patch according to claim 2, wherein the asymmetric fluctuation index satisfies the relation: Wherein, the For the asymmetric fluctuation index at time t, As a point of time in time it is, For the sum of squares of all fluctuations above the local average torque within the sliding window, To sum the squares of all fluctuations below the local average torque within the sliding window, To prevent the denominator from being a very small positive number of zero.
4. The intelligent control method for the pressure-sensitive self-repairing asphalt cold-patch mixing according to claim 2, wherein the transient spike index satisfies the relation: Wherein, the For the transient spike index at time t, As a point of time in time it is, For the current transient fluctuations to occur, For the length of the background window, For the time index of the time index, To prevent the denominator from being a very small positive number of zero.
5. The intelligent control method for the pressure-sensitive self-repairing asphalt cold-patch mixing according to claim 4, wherein the transient fluctuation satisfies the relation: Wherein, the Is that The torque value at the moment in time, And Torque values at time t+1 and time t-1, respectively.
6. The intelligent control method for cold-patch mixing of pressure-sensitive self-repairing asphalt according to claim 1, wherein the feature space formed by the dynamic feature vectors is subjected to unsupervised analysis by using a local anomaly factor algorithm, and the method comprises the following steps: taking all dynamic feature vectors as an input data set; Performing standardization processing on the input data set; And calculating the local abnormality factor corresponding to each time point based on the standardized data set.
7. The intelligent control method for the pressure-sensitive self-repairing asphalt cold-patch mixing is characterized in that the specific process for collecting the torque time sequence data of the mixing motor is that the torque time sequence data is collected in real time in a fixed sampling period through a torque sensor arranged on a driving motor of the mixing machine or through feedback data of a frequency converter of the driving motor.
8. The intelligent control system for the pressure-sensitive self-repairing asphalt cold-patch mixing is characterized by comprising a processor and a memory, wherein the memory stores computer program instructions, and the intelligent control method for the pressure-sensitive self-repairing asphalt cold-patch mixing is realized according to any one of claims 1-7 when the computer program instructions are executed by the processor.

Description

Pressure-sensitive self-repairing asphalt cold-patch mixing intelligent control system and method Technical Field The invention relates to the technical field of intelligent manufacturing control. More particularly, the invention relates to a pressure-sensitive self-repairing asphalt cold-patch mixing intelligent control system and method. Background The pressure-sensitive self-repairing asphalt cold repairing material is a novel high-performance road maintenance material, and the core components of the material mainly comprise a pressure-sensitive adhesive, an aggregate wrapped by a self-repairing gel coat and an active filler, unlike the traditional cold repairing material. The self-repairing gel coat wraps the repairing agent, and when the road surface is in actual use, the pressure of the vehicle load can cause the gel coat to crack, and the repairing agent in the gel coat is released, so that the dynamic self-repairing and pressure-strengthening bonding of the road surface damage are realized. At present, in the mixing production phase of asphalt cold feed, a critical process control point is that the integrity of these gel coat microcapsules must be protected from premature rupture under the stirring shear of the stirrer. Once the gel coat is broken in advance, the restorative agent is released and reacts in advance, which results in the loss of self-healing function of the final product. In the prior art, the uniformity degree of the mixture is generally judged indirectly by adopting a mode of monitoring the torque or the power of the stirring motor, and the stirring is considered to be completed when the mixture reaches a certain viscosity and the torque value tends to be stable. However, the mixing process of the pressure-sensitive cold-patch material is complex and changeable, three distinct working conditions possibly comprising normal mixing are included in a motor torque curve, the torque value is symmetrically and slightly fluctuated near the mean value, component agglomeration refers to periodical increase of mixing resistance caused by agglomeration of aggregate or active filler, torque is continuously expressed as high-amplitude oscillation, gel coat rupture refers to internal fluid released by breaking of the gel coat microcapsules under shearing force, the viscosity of the mixture is instantaneously reduced, and the torque is suddenly and sharply reduced. The prior art produces high intensity anomaly signals for both component agglomeration and gel coat rupture, because both exhibit high volatility, which results in the control system being unable to distinguish between the two anomalies. Such confusion can lead to erroneous control decisions, possibly leading to rejection of the entire batch of products due to premature failure of the gel coat, resulting in significant economic losses. Disclosure of Invention In order to solve the technical problem that the component agglomeration and gel coat breakage are difficult to distinguish in the prior art, the invention provides schemes in various aspects as follows. In a first aspect, the invention provides an intelligent control method for cold repair asphalt material mixing of a pressure-sensitive self-repairing asphalt, which comprises the steps of collecting torque time sequence data of a mixing motor, constructing dynamic feature vectors of the torque time sequence data at each time point based on the torque time sequence data, wherein the dynamic feature vectors are used for representing asymmetry and transient kurtosis of fluctuation of the torque time sequence data, performing unsupervised analysis on feature spaces formed by the dynamic feature vectors by utilizing a local anomaly factor algorithm, calculating local anomaly factors corresponding to each time point, and performing closed-loop control based on the local anomaly factors and the dynamic feature vectors so as to realize intelligent control on a mixing process. According to the invention, by constructing asymmetric fluctuation and transient peak dynamic characteristic vectors and combining an LOF algorithm, component agglomeration and gel coat breakage can be accurately distinguished, so that gel coat breakage can be detected instantaneously and stopped immediately, the self-repairing activity of a product core is protected, and meanwhile, a mixing end point is reasonably judged, so that intelligent closed-loop control is realized. Preferably, the dynamic feature vector is a two-dimensional feature vector, and the two-dimensional feature vector comprises an asymmetric fluctuation index for evaluating the directionality of the torque time series data fluctuation and a transient spike index for evaluating the persistence of the torque time series data fluctuation. Preferably, the asymmetric fluctuation index satisfies the relation: wherein, the method comprises the steps of, For the asymmetric fluctuation index at time t,As a point of time in time it is,For the sum of squares