Grey relation analysis of natural frequency and semi-active robust optimal control of a multi-dimensional isolator based on parallel mechanism and magneto-rheological damper

For isolating multi-dimensional vibrations experienced by precise facilities carried on a vehicle, a novel isolator is proposed based on 2-RPC/2-SPC parallel mechanism with magneto-rheological dampers. Kinematics and dynamics of the isolator are analyzed by geometrics and the Lagrange method. Grey relation analysis approach is conducted to determine the contributions of geometric parameters on natural frequency conveniently. Through analysis, the first order natural frequency of the isolator is affected by the length of the fixed platform most significantly. Due to manufacturing and assembling errors which could not be avoided in the isolator, robust optimal control algorithm is conducted to ensure control effect and robustness of the isolator at the same time. The gain of robust optimal control algorithm is obtained by deducing and solving linear matrix inequality. Compared to passive control, velocity root mean square values of robust optimal semi-active control decreased obviously in horizontal, longitudinal, vertical, and roll directions.

A comprehensive investigation on machining of composites by EDM for microfeatures and surface integrity

In electro-discharge machining (EDM), the material removal takes place by precisely controlled sparks that occur between tool and workpiece separated with a spark gap in the presence of a dielectric. Generally, the non-contacting type and less material removal rates are attributed to attain a good surface finish and close dimensional tolerances during an EDM of monolithic metals and alloys. But the dimensional accuracy and surface integrity parameters would considerably affect during EDM of composites due to the existence of more than one material phase constituents. Therefore, the present work aims to study and optimize the performance characteristics under various EDM conditions employed in making rectangular channels on AA6061-B4C composite material. Initially, AA6061-4wt.%B4C composites were fabricated by ultrasonically assisted stir-casting, and the improved properties were obtained from various mechanical characterizations. The EDM experiments were conducted according to the full factorial experimental design. The three levels of input conditions such as discharge Current (I), discharge duration (T On), and discharge idle time (T Off) were considered. The considered output responses are material removal rate (MRR),taper (θ) of the machined channel, tool wear rate (TWR), average surface roughness (R) of the machined surface, and average recast layer thickness (ARLT) of the machined zone. These responses are co-related with multi-objective types in the sense that the MRR has to be maximized with all other responses minimized. Hence, principal component analysis (PCA) coupled with grey relation analysis (GRA) was used for optimization in which the results were normalized, and all the responses were converted into a single response named weighted grey relation grade (WGRG) for each trial. The experimental trial, which had the highest WGRG, was considered as a local optimum. The global optimum parameters were obtained by performing the Taguchi method (TM) (higher-the-better) for the maximization of WGRG. The analysis of variance (ANOVA) was performed to know the contribution of each EDM parameter toward the WGRG. The optimum levels of Current, T On, and T Off were identified as 8 A, 25 µs, and 36 µs, respectively. Results showed that all three input parameters significantly affected the WGRG, and a higher contribution of Current (52.11%) followed by the T On (26.72%) was observed. The interaction between the Current and T Off was found to be greater than other interactions. Taper values were observed to be reduced at the combination of 8 A discharge Current and 25 µs T On. None of the input parameters significantly affected the Ra, except for Current, which showed a slight effect. ARLT values showed an increasing trend of T On from 25 µs to 45 µs but decreased slightly at 65 µs for all Current levels. The moderate Current level 6 A was observed to be favorable in reducing ARLT when compared to low (4 A) and high (8 A) for all Ton values.

Research on short-term wind power Prediction of GRU based on similar days

As the proportion of wind power generation continues to increase, accurate forecasting of wind power output is of great significance to the smooth operation of the entire power grid. However, due to the greater impact of environmental factors, wind power generation has strong randomness, and it becomes difficult to accurately predict the power generation. Thus, a new hybrid model for wind power generation prediction combining GRU neural networks and similar days’ characters analysis is proposed to address solve this problem. The prediction method employs grey relation analysis to screen similar days, which not only reduces the amount of data required to train the model, reduces the computational complexity, and improves the training speed, but also improves the prediction accuracy based on the selected datasets. In addition, this method also filters and processes the data through box-plot analysis and linear smoothing, which further improves the prediction accuracy of the model. The results show that compared with a single GRU network, the MAE of this method has dropped by 1.89, RMSE has dropped by 1.9, and MAPE has dropped by 11.07%. Obviously, the prediction model based on similar days extraction has obvious advantages.

Short-term load prediction of integrated energy system based on neural network

Considering the correlation and nonlinear characteristics of multiple types of loads in the integrated energy system, grey relation analysis (GRA) and long short term Memory (LSTM) neural network are selected to establish the short-term load prediction model of the integrated energy system. The model uses GRA method to analyze the coupling between multiple types of loads and the meteorological factors, and then obtains the load forecast results through the LSTM prediction model. Finally, a practical example is given to verify the validity of the model.

Reliability Analysis of Intelligent Manufacturing Systems Based on Improved FMEA Combined With Machine Learning

Along with the booming of intelligent manufacturing, the reliability management of intelligent manufacturing systems appears increasingly important. Failure mode and effects analysis (FMEA) is a prospective reliability management instrument extensively utilized to manage failure modes of systems, products, processes, and services in various industries. However, the conventional FMEA method has been criticized for its inherent limitations. Therefore, this paper devises a method based on improved FMEA model combined with machine learning for complex systems and applies it to the reliability management of intelligent manufacturing systems. The structured network of failure modes is constructed based on the knowledge graph for the intelligent manufacturing systems. The grey relation analysis (GRA) is applied to determine the risk prioritization of failure modes, hereafter the clustering analysis is employed to extract the features of failure modes. The results show that the proposed method can more accurately reflect the coupling relationship between the failure modes compared with the conventional FMEA method. This research provides significant support for the reliability and risk management of complex systems such as intelligent manufacturing systems.

Vibration Analysis of Partially Cracked Symmetric Laminated Composite Plates Using Grey-Taguchi

Nowadays, the demand for carbon-boron boron-based composites has tremendously increased in the aerospace and automotive industries due to their lightweight and high strength-weight ratio. Hence, it is necessary to study the vibrational behavior of the carbon-boron matrix. The stacking sequences, fiber orientation, crack size, and length-width ratio has all been studied at three levels. A mathematical model has been developed using classical laminated plate theory to evaluate the natural frequency for two boundary conditions; first, all sides were supported (SSSS), second, two adjacent sides clamped, and the other two were simply supported (CCSS). Grey relation analysis has also been applied to optimize the parameters.

The Mathematical Analysis and Classification Research of an Iris Data Set Using Binary Tree and Grey Relation Grade

The data from an Iris flower database is studied. The Iris database is the most commonly used database for machine learning algorithms. The Iris database was developed by Ronald Aylmer Fisher in 1936. The Iris database has 150 records in three categories: Iris Sentosa, Iris Versicolor and Iris Virginic. The database has four attributes: sepal length, sepal width, petal length and petal width. For the machine learning algorithm, 150 Iris flower databases are used. Of the 150 Iris in the Iris database, 80% are used as the training set and the remaining 20% Iris as the test set. In machine learning, to perform classification and discrimination is a complicated and difficult thing. In this study, a grey relation grade is used to extract the main features of the Iris flower and a Binary Tree [1] is used to classify the Irises. The results show that for the same specific attributes, grey relation grade extracts the main attributes and can be used in combination with a binary for classification.