CN-122020020-A - Fault prediction maintenance method and system for silica gel production equipment

CN122020020ACN 122020020 ACN122020020 ACN 122020020ACN-122020020-A

Abstract

The invention relates to the technical field of equipment fault prediction, and discloses a fault prediction maintenance method and a fault prediction maintenance system for silica gel production equipment, wherein the method comprises the steps of collecting a glue injection pressure sequence, a motor rotating speed sequence and a glue temperature sequence in real time in the operation process of the silica gel production equipment, and constructing a sliding window for data analysis; the method comprises the steps of evaluating a fluid elastic damping index, adaptively adjusting a punishment factor of a variation mode decomposition algorithm based on the fluid elastic damping index, decomposing a glue injection pressure sequence by utilizing the adjusted variation mode decomposition algorithm, extracting high-frequency characteristic modes, calculating energy of the high-frequency characteristic modes, reversely compensating the energy of the high-frequency characteristic modes, calculating a fault early warning index, comparing the fault early warning index with a preset reference threshold, and judging whether the equipment has faults according to comparison results. The invention can prevent the fault signal of the non-Newtonian fluid from being absorbed and disturbed due to high viscoelasticity, can identify the initial fault and avoid false alarm and missing judgment during working condition switching.

Inventors

MA XINCHUAN
YANG HUITAO
SU YONGHUI
DONG JUNLIANG
MA YONGKANG

Assignees

尉氏县百茂硅胶科技有限公司

Dates

Publication Date: 20260512
Application Date: 20251229

Claims (10)

1. The fault prediction maintenance method of the silica gel production equipment is characterized by comprising the following steps of collecting a glue injection pressure sequence, a motor rotating speed sequence and a glue temperature sequence in real time in the operation process of the silica gel production equipment, and constructing a sliding window for data analysis; According to the glue injection pressure sequence, the motor rotating speed sequence and the glue temperature sequence in the sliding window, evaluating a fluid elastic damping index, wherein the fluid elastic damping index is used for representing the covering capacity of the current fluid on fault signals; the punishment factors of the variation modal decomposition algorithm are adaptively adjusted based on the fluid elastic damping index, the glue injection pressure sequence is decomposed by utilizing the adjusted variation modal decomposition algorithm, the high-frequency characteristic mode is extracted, and the energy of the high-frequency characteristic mode is calculated; And reversely compensating the energy of the high-frequency characteristic mode by using the fluid elastic damping index, calculating a fault early warning index, comparing the fault early warning index with a preset reference threshold value, and judging whether the equipment has faults or not according to a comparison result.
2. The method of claim 1, wherein the fluid elastic damping index is calculated by the following formula: ; In the formula, Indicating the fluid-elastic damping index (f-i), Indicating a preset basic viscosity coefficient of the composition, Indicating that the preset reference temperature is to be reached, Representing the average value of the glue temperature sequence in the current sliding window, Represents the average value of the motor rotation speed sequence in the current sliding window, Indicating a preset logarithmic correction constant, Represents the standard deviation of the sequence of injection pressures within the current sliding window, Representing the average value of the sequence of injection pressures within the current sliding window, Indicating a preset pressure reference constant.
3. The silica gel production facility failure prediction maintenance method of claim 2, wherein the penalty factor is calculated by the following formula: ; In the formula, A penalty factor is indicated and is indicated, Representing a preset minimum value of the penalty factor, Representing a preset maximum value of the penalty factor, Indicating a preset decay slope coefficient, Indicating the fluid elastic damping index.
4. A silica gel production facility failure prediction maintenance method according to claim 3, wherein the energy of the high frequency characteristic mode is calculated by the following formula: ; In the formula, Represents the energy of the high-frequency characteristic mode, Indicating the length of the sliding window and, Indicating that the high frequency characteristic mode is within the sliding window Data values for the sampling points.
5. The method for predicting and maintaining a fault in a silica gel production facility according to claim 4, wherein the fault pre-warning index is calculated by the following formula: ; In the formula, The fault early warning index is indicated, Represents the energy of the high-frequency characteristic mode, Indicating the preset compensation gain factor(s), Indicating the fluid elastic damping index.
6. The method according to claim 1, wherein the glue injection pressure sequence is obtained by a piezoelectric pressure sensor installed at the outlet of the mixing pump, the motor rotation speed sequence is obtained by a feedback encoder of a servo driver, and the glue temperature sequence is obtained by a thermocouple or an infrared sensor attached to the wall of the conveying pipe.
7. The method for predicting and maintaining the failure of the silica gel production equipment according to claim 3, wherein the decomposing the injection pressure sequence by using the modified variation mode decomposition algorithm to extract the high-frequency characteristic mode comprises dynamically adjusting the search bandwidth of the variation mode decomposition algorithm by using the penalty factor obtained by calculation, decomposing the injection pressure sequence of the current sliding window, and selecting the component with the highest central frequency as the high-frequency characteristic mode.
8. The method for predicting and maintaining a failure of a silica gel production facility according to claim 1, wherein the step of determining whether the failure of the facility exists based on the comparison result comprises the steps of, when continuously When the fault early warning indexes calculated by the sliding windows exceed a preset reference threshold value, judging that the equipment has blockage or mechanical abrasion faults, and triggering maintenance early warning, wherein the detection method comprises the following steps of Is an integer greater than 1.
9. The method according to claim 1, wherein the fluid elastic damping index is positively correlated with the viscosity and compressibility of the current fluid, negatively correlated with the shear rate of the fluid, and wherein the penalty factor is adaptively decreased when the fluid elastic damping index is increased to expand the search bandwidth of the variation modal decomposition algorithm.
10. The system for predicting and maintaining the faults of the silica gel production equipment is characterized by comprising a memory and a processor, wherein the memory stores computer instructions, and the processor realizes the method for predicting and maintaining the faults of the silica gel production equipment according to any one of claims 1 to 9 when executing the computer instructions.

Description

Fault prediction maintenance method and system for silica gel production equipment Technical Field The invention relates to the technical field of equipment fault prediction, in particular to a fault prediction maintenance method and system for silica gel production equipment. Background In the precise manufacturing process of high-end silica gel products, a feeding system generally adopts A/B double-component liquid silica gel as a basic raw material. The technological process requires that the raw materials are pressurized and conveyed by a high-precision metering pump, and after uniform mixing of molecular level is completed in a static or dynamic mixer, the raw materials are finally injected into a die cavity in a high-pressure jet flow mode through a terminal nozzle or are precisely coated on the surface of a product. However, since liquid silica gel exhibits remarkable non-newtonian fluid properties, and has complex rheological characteristics of high viscosity, high compressibility and high resilience, it presents a great technical challenge for real-time accurate monitoring of the operating state of production equipment. In the prior art, the health state of the glue injection system is mainly monitored by a fixed threshold method, namely, a pressure sensor is arranged on a pipeline, and an alarm is given when the monitored pressure value exceeds a set upper limit or is lower than a lower limit. However, this method suffers from the disadvantage that, firstly, the fluid viscoelasticity has a significant low-pass filtering effect, and when early minor clogging or mechanical wear of the nozzle or mixing tube at the end of the device occurs, the resulting high-frequency pressure pulse signal is rapidly absorbed and smoothed by the viscous fluid, resulting in failure of the control system to identify the incipient fault. In addition, the change of working conditions can also cause interference to monitoring, the viscosity of silica gel is extremely sensitive to temperature and shear rate, the viscosity of the equipment is high during cold start, and the viscosity of the equipment is reduced due to the shear thinning effect during high-speed operation. The filter with fixed bandwidth cannot adapt to the dynamically changing physical environment, and misjudgment or missed judgment is easy to cause. Disclosure of Invention The invention provides a fault prediction maintenance method and a fault prediction maintenance system for silica gel production equipment, which are used for solving the problems that initial faults cannot be identified and misjudgment or missed judgment is easy to occur when working conditions are switched by a fixed threshold method in the prior art. In a first aspect, the fault prediction maintenance method for silica gel production equipment of the present invention includes the following steps: In the operation process of the silica gel production equipment, a glue injection pressure sequence, a motor rotating speed sequence and a glue temperature sequence are collected in real time, and a sliding window for data analysis is constructed; According to the glue injection pressure sequence, the motor rotating speed sequence and the glue temperature sequence in the sliding window, evaluating a fluid elastic damping index, wherein the fluid elastic damping index is used for representing the covering capacity of the current fluid on fault signals; the punishment factors of the variation modal decomposition algorithm are adaptively adjusted based on the fluid elastic damping index, the glue injection pressure sequence is decomposed by utilizing the adjusted variation modal decomposition algorithm, the high-frequency characteristic mode is extracted, and the energy of the high-frequency characteristic mode is calculated; And reversely compensating the energy of the high-frequency characteristic mode by using the fluid elastic damping index, calculating a fault early warning index, comparing the fault early warning index with a preset reference threshold value, and judging whether the equipment has faults or not according to a comparison result. The method has the beneficial effects that continuous and synchronous physical state data base can be provided for subsequent analysis by collecting multidimensional data in real time and constructing a sliding window. By evaluating the fluid elastic damping index, the physical damping degree of the current fluid environment on the signal can be accurately identified. Through self-adaptive adjustment of penalty factors, accurate feature extraction under different viscosity states can be ensured. The real fault intensity absorbed by the fluid phagocytosis can be restored by reversely compensating and calculating the fault early warning index, and the detection rate of early faults is improved. Preferably, the fluid elastic damping index is calculated by the following formula: ; In the formula, Indicating the fluid-elastic damping index (f-i),Indicati