Grey prediction of natural aging mechanical proper

  • Detail

Grey prediction of natural aging mechanical properties of plastics Huang Wei and Qiu Jun (School of mechanical engineering, Guangxi University, Nanning, Guangxi 530004) have no obvious distribution characteristics, so it is difficult to study them with regression analysis method. In this paper, a prediction method of natural aging mechanical properties of plastics based on grey prediction theory is proposed. It is proved by examples that this method has good accuracy of the fatigue test method gb/t2656 (1) 981 for weld metal and welded joints. Plastic is a new material rising with the development of petrochemical industry. Because of its excellent performance and high economic value, its utilization field is expanding day by day. It has become an indispensable important material in many industrial fields, such as machinery, electronics, automobile, chemical architecture, packaging and so on. The per capita consumption of plastics has become an important indicator of a country's industrialization level. Plastics have good mechanical properties, electrical properties, chemical properties and dimensional stability, but their major disadvantage is that they are not resistant to aging. In the process of processing, storage and use, due to the long time and the comprehensive effect of various internal and external factors, its physical and mechanical properties gradually decline, resulting in its loss of service performance. This property of plastics determines that various effects caused by aging must be considered in the research of design, use and improvement, and the change law of mechanical properties during aging must be studied. The aging mechanism of plastics is very complex. The main reasons for aging are light aging, oxygen and ozone aging, biodegradation, water degradation and so on. These factors lead to the destruction of chemical structure, the breaking of macromolecules, the decline of molecular weight, and the loss of original properties. The change of its mechanical properties more often than other properties dominate the design, processing and application of plastic parts. Due to the complexity of aging mechanism, it is difficult to directly study the performance changes caused by various aging causes, but to analyze and study them through test data. At present, the current aging test is divided into two categories. One is the natural aging test method, which is an aging test directly using the natural environmental conditions, and the other is the artificial accelerated aging test method, which is an aging test in the laboratory using the aging box to simulate some aging factors of the natural environmental conditions, so as to speed up the aging process of materials and obtain the test results. Due to the diversity of aging factors and the complexity of aging mechanism, artificial accelerated aging test is still difficult to replace natural aging test. Natural aging test is still an important and reliable aging test method at present. However, this method is time-consuming and expensive, and the experimental data obtained are very limited. In engineering, regression analysis is often used to process the test data and predict after obtaining the regression formula. This method requires that the test data have obvious distribution rules and sufficient data volume, so it is quite limited in the prediction of aging performance. The prediction method based on grey theory grey system theory has made great progress both in theory and in practical application since it came out in the early 1980s. It has become a unique idea and new method for system analysis, modeling, prediction, decision-making and control in China. Grey prediction is a prediction method based on grey system theory. It considers that all random quantities are grey quantities and grey processes that change in a certain range and time. The processing of grey quantity is not to seek its statistical law and probability distribution through large samples, but to process the irregular original data into regular time series data through the generation method, that is, to find the law between the data according to the data processing method, then establish the grey dynamic model according to the differential equation fitting method, and then use the solution of the differential equation to realize the system prediction. Because no matter how complex the objective system is, it is always regular and orderly, with overall functions. Therefore, the data, as the behavior characteristics of the system, always contains some laws. The grey prediction method only needs more than 4 original data to analyze, model and predict the grey system. The modeling method is simple and easy, the calculation is simple, but the prediction can achieve high accuracy. Especially because it has no special requirements for the distribution characteristics of original data and sample size, the study of aging performance is undoubtedly a powerful tool. The system with some known information and some unknown information is called grey system. The plastic aging test system is regarded as a grey system with unknown information. The main mechanical indexes of plastic aging are tensile strength and elongation at break. The original data can be obtained through the test, and then the grey prediction model can be established for prediction and analysis. The basic principle and main steps of establishing the grey GM (1,1) model for the data series are as follows: determine a data series and solve the albino equation coefficient AA by * small two multiplication to list the differential equation factory +ax (1) =u. (3) construct the matrix B and vector y, where (8) carry out error check on the results. 3 plastic natural aging tensile strength and elongation at break predictor the plastic natural climate exposure test is to expose the plastic sample to the natural climate environment, make it subject to the comprehensive effects of solar, temperature, oxygen and other climatic factors, and evaluate the aging performance of the plastic by measuring the change of its performance. The natural climate exposure test method for plastics is to use an exposure frame, The free height shall be tested in accordance with the national standard gb3681 (natural climate exposure test method for plastics). The aging mechanical properties of plastics shall be tested in accordance with the 9 plastic mechanical properties test method for the customer's differentiated requirements. The tensile strength and elongation at break of plastics are important performance parameters of plastic aging * according to gb3681 (the results of LDPE shed mold atmospheric natural aging test are shown in Table 1. Table 1 LDPE shed mold atmospheric aging test results aging time/monthly tensile strength/mpa elongation at break rate grey prediction of tensile strength is as follows: use 6 data to determine the original data sequence. In order to facilitate calculation and ensure calculation accuracy, take x (0) (T) The calculated generated number sequence is obtained by restoring the data sequence. Using the established GM (1,1) for extrapolation, X (0) (7) =1.18808 can be obtained. The above calculation results and error analysis of ax (0) are listed in Table 2. Table 2 tensile strength prediction results aging time/monthly test series/mpa prediction series/mpa relative error can be seen from the table. In this paper, six data are used to establish GM (1,1) model for tensile strength, and its fitting * large relative error is 3 64%, the average relative error is 202%, and then the established model is used to predict the tensile strength with aging time of 18 months and 21 months, the error is 3.40% and 6.75%, and the prediction accuracy is quite good. The same method can be used to model and predict the elongation at break, and the results are shown in Table 3. Table 3 prediction results of elongation at break aging time/monthly test sequence/mpa prediction sequence/mpa relative error can be seen from the table * the large error is 4.89%, and the average error is 3 4 Summary by using the Grey Prediction GM (1,1) model to predict the natural aging mechanical properties of plastics, it can be seen that the grey prediction method is simple in calculation, easy to program the calculation steps, and requires less data. These characteristics are very suitable for the study of aging properties. The prediction results of tensile strength and elongation at break of LDPE shed mold aging in this paper have good accuracy. The method can be extended to similar problems

Copyright © 2011 JIN SHI