We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Comparative Analysis of Machine Learning Approaches to Predict Impact Energy of Hydraulic Breakers
Summary
Researchers developed a neural network-based model to predict the impact energy of hydraulic breakers using 1,451 data points covering parameters such as working pressure, flow rate, chisel diameter, nitrogen gas pressure, operating frequency, and power. Comparative analysis with linear regression and correlation methods confirmed the neural network approach provided the most reliable predictions across breaker classes.
Impact energy, the main performance subject of hydraulic breakers, is required to evaluate value from consumers. This study proposes a neural network algorithm-based model to predict the impact energy of a hydraulic breaker without measuring it. The proposed model was developed using 1451 data points for various parameters as an input to predict the impact energy of hydraulic breakers in a small class to a large class. Different machine learning methods have been studied, including correlation analysis, linear regression, and neural networks. The results revealed that the working pressure, working flow rate, chisel diameter, nitrogen gas pressure, operating frequency, and power significantly influenced impact energy formation. The results obtained provide a reliable model for predicting the impact energy of hydraulic circuit breakers of various sizes.
Sign in to start a discussion.
More Papers Like This
Neural network based prediction of the efficacy of ball milling to separate cable waste materials
Researchers developed a neural network — a type of artificial intelligence — to predict how well ball milling separates copper from plastic (PVC) in cable recycling, finding that the weight of cables loaded and the force of impact were the most critical factors. Machine learning tools like this could help scale up plastic and metal recycling to industrial levels.
Fine-tuning optimization of poly lactic acid impact strength with variation of plasticizer using simple supervised machine learning methods
Researchers applied machine learning methods — including K-nearest neighbors, support vector regression, and artificial neural networks — to predict and optimize the impact strength of polylactic acid (PLA) with different plasticizers, finding KNN offered the best balance of accuracy and stability for guiding formulation experiments.
Electrical Power Prediction through a Combination of Multilayer Perceptron with Water Cycle Ant Lion and Satin Bowerbird Searching Optimizers
Researchers developed a water cycle algorithm-optimized neural network to predict electrical power output from combined cycle power plants, demonstrating improved prediction accuracy compared to standard optimization algorithms on a publicly available dataset.
Machine Learning-Based Models for Basic Sediment & Water and Sand-Cut Prediction in Matured Niger Delta Fields
Researchers developed neural network-based models to predict basic sediment and water (BS&W) and sand-cut production from wellhead variables in 43 mature oilfields in the Niger Delta, using 457 datasets in both multiple-inputs single-output and multiple-inputs multiple-output network architectures. Model performance was evaluated against new test datasets using statistical measures including coefficient of determination.
Loading Frequency Classification in Shape Memory Alloys: A Machine Learning Approach
Researchers applied machine learning methods to predict the loading frequency of nickel-titanium shape memory alloys based on experimental data from cyclic tensile tests. They tested multiple algorithms across different loading frequencies and found that machine learning could effectively classify the frequency conditions. The study demonstrates the potential of data-driven approaches for characterizing the behavior of these materials used in engineering applications.