Deep Learning-Based Prediction of Throttle Value and State for Wheel Loaders

Accurate prediction of the throttle value and state for wheel loaders can help to achieve autonomous operation, thereby reducing the cost and accident rate. However, existing methods based on a physical model cannot accurately reflect the operator’s driving habits and the interaction between wheel loaders and the environment. In this paper, a deep-learning-based prediction model is developed to predict the throttle value and state for wheel loaders by learning from driving data. Multiple long–short-term memory (LSTM) networks are used to extract the temporal features of different stages during the operation of the wheel loader. Two backward-propagation neural networks (BPNNs), which use the temporal feature extracted by LSTM as the input, are designed to output the final prediction results of throttle value and state, respectively. The proposed prediction model is trained and tested using the data from two different conditions. The end-to-end LSTM prediction model and BPNNs are used as benchmark models. The results indicate that the proposed prediction model has good prediction accuracy and adaptability. Furthermore, the relationship between the prediction performance and signal sampling frequency is also studied. The proposed prediction method that combines driving data and deep learning can make the throttle action conform to the decisions of an experienced operator, providing technical support for the autonomous operation of construction machinery.

Probabilistic Analysis of Wheel Loader Failure under Rockfall Conditions Based on Bayesian Network

Rockfall is one of the most serious geological hazards in mountain regions. During the rescue situations after rockfall, the wheel loader, a vital type of modern engineering mechanism, plays an important role in relieving the obstruction of the catastrophic site. Increasing the reliability of the wheel loader during the rescue situation is quite important. This study aims to build a fault diagnosis model based on Bayesian network (BN) to diagnose the probability and path of the fault occurrence in the wheel loader during a rockfall disaster. Meanwhile, to reduce the influence of subjective factors, the fuzzy set theory is introduced into BN. The result showed that the probability of failure of the wheel loader under rockfall disaster is 13.11%. In addition, the key cause of the failure of the wheel loader under the rockfall disaster is the malfunction of mechanical parts. The probability of mechanical component failures in this case is as high as 88%, while the probability of human error is 6%. The research results not only show the ability of the BN to incorporate subjective judgment but also can provide a reference for fault diagnosis and risk assessment of wheel loaders under rockfall disaster conditions.

Design and Efficiency Analysis of Closed Loop Pump Controlled Circuit Hydraulic Lifting System of Wheel Loaders Based on Gravity Self-Balancing Hydraulic Cylinder

In the hydraulic lifting systems of wheel loaders, the valve controlled systems are used to drive the hydraulic cylinder to complete frequent lifting and falling operations. The gravitational potential energy of the lifting system, accumulated in the lifting process, is converted into heat energy through the throttling port of the valve during the falling processes, which results in significant throttling loss and severe system overheating. To solve the problems, a potential energy regeneration and utilization system is proposed, where the closed loop pump controlled circuit based on the gravity self-balancing hydraulic cylinder is adopted to eliminate throttling loss, and the gravity self-balancing chamber of the cylinder is directly connected with accumulator to recycle gravity potential energy. In the research, the structure and working principle of the proposed hydraulic system is analyzed first, then the co-simulation model and the test prototype are established to investigate the working and energy characteristics of the proposed system. Test results indicate that, compared with the traditional valve controlled hydraulic system, 58.9% energy consumption reduction can be expected for the hydraulic pump by adopting the proposed system under the same working condition.

Incorporating Valve Switching Losses Into a Static Optimal Control Algorithm for the Hybrid Hydraulic-Electric Architecture (HHEA)

The Hybrid Hydraulic-Electric Architecture (HHEA) has, in recent years, been proposed as an energy efficient alternative to conventional load-sensing architectures in mobile machines such as excavators and wheel-loaders. HHEA leverages the advantages of hydraulic power and electric power to eliminate throttling valves while also improving the energy and control performance of the system. The architecture utilizes a set of common pressure rails to provide a majority of power and and a small electric motor driven pump to modulate this power to meet the exact demand. Previous work has developed a computationally efficient Lagrange Multiplier approach for determining the optimal pressure rail selections that minimizes the energy losses in the system. The static model used considers only the energy use for each pressure rail selection but not the losses associated with the valves during the transition. This paper presents an approach to include the switching losses in the model and in the optimization procedure. To capture the switching losses, switching events between different rails and at various input and output flow rates were simulated with consideration of valve spool dynamics. A parameterized model that summarizes the losses is then obtained, allowing switching losses to be added to the previous energy analysis. The performance of the switching loss model was compared with reference data obtained from a high-fidelity simulation model. To incorporate the switching losses into optimal control algorithm, an efficient dynamic programming approach that prevents frequent switching is adopted in place of the Lagrange multiplier approach. The overall effect of switching losses on the energy consumption and optimal control decisions is presented. In general, switching losses contribute to about 9–10% of input energy.

Data-Driven Reinforcement-Learning-Based Automatic Bucket-Filling for Wheel Loaders

Automation of bucket-filling is of crucial significance to the fully automated systems for wheel loaders. Most previous works are based on a physical model, which cannot adapt to the changeable and complicated working environment. Thus, in this paper, a data-driven reinforcement-learning (RL)-based approach is proposed to achieve automatic bucket-filling. An automatic bucket-filling algorithm based on Q-learning is developed to enhance the adaptability of the autonomous scooping system. A nonlinear, non-parametric statistical model is also built to approximate the real working environment using the actual data obtained from tests. The statistical model is used for predicting the state of wheel loaders in the bucket-filling process. Then, the proposed algorithm is trained on the prediction model. Finally, the results of the training confirm that the proposed algorithm has good performance in adaptability, convergence, and fuel consumption in the absence of a physical model. The results also demonstrate the transfer learning capability of the proposed approach. The proposed method can be applied to different machine-pile environments.

Machine Maintenance Design using Markov Chain Method to Reduce Maintenance Costs

PT. XYZ is a company that focuses on construction with ready mix concrete product (cast). The problem faced by the company is that the schedule is not suitable for machine maintenance activities so that it still results in high maintenance costs incurred by the company. By using the markov chain method can plan maintenance time in order to reduce downtime so as to minimize maintenance costs. The results obtained by the proposal for the company are for proposal I it takes 49.78 hours = 50 hours at a cost of Rp. 16,984,605, the cost savings of Rp. 73,545,395 (81.24%). Schedule for each machine such as wheel loaders every 14,009 hours, batching plant machines every 16,604 hours, truck mixer machines every 19,168 hours. Scheduling the second proposal will take 26.62 hours = 27 hours at a cost of Rp. 9,080,664, the cost savings of Rp. 81,449,336 (89.97%). Schedule for every machine such as wheel loaders every 7,490 hours, batching plant machines every 8,877 hours, mixer truck machines every 10,248 hours. Judging from the results obtained, the recommendation given is

A Novel Real-Time Center of Gravity Estimation Method for Wheel Loaders with Front/Rear-Axle-Independent Electric Driving

Estimation of the center of gravity (CG) is the basis for intelligent control of the front-and-rear-axis-independent electric driving wheel loaders (FREWLs). This paper presents a novel real-time method for estimating the CG of FREWLs, which is suitable for driving and spading conditions on bumpy roads. A FREWL dynamical model is proposed to set up the state-space model. The CG estimator is used to estimate the longitudinal tire force using the state-space model and the improved square-root unscented Kalman filter (ISR-UKF) algorithm. The simulation and experimental results indicate that this method is suitable for FREWL dynamics and operational characteristics, and the estimated value of CG basically converges to the reference value. Finally, the probable reasons for error occurring in two experiments and the practical challenges of this method are discussed. The research in this paper establishes a partial theoretical basis for intelligent control of construction machinery.

An online estimation algorithm to predict external forces acting on a front-end loader

One of the most significant goals of earthmoving equipment is to maximize productivity during loading cycles. A real-time knowledge of the forces exchanged between the machine implement and the surrounding, that is, while digging, can be used in different ways to increase productivity. It can be used to determine the amount of material moved by the machine; or to find the optimal bucket trajectory; moreover, as input to traction control systems. This article presents an online force estimation algorithm able to predict vertical and horizontal forces exchanged between the front-loader and the surrounding environment, as well as the reaction forces through the implement itself. Taking the case of a 14-ton wheel loader as reference, this article illustrates the development of a simulation model for the analysis of the machine digging system, along with the instrumentation and testing of the proposed estimation algorithm. The model is divided into two sections describing, respectively, system kinematic and system dynamics. The kinematic model of the front-loader is compared against measurements, and results show an average error lower than 1%. The dynamic model predicts both hydraulic and dynamic features of the machine implement, achieving an accuracy on the payload mass within 2%–3%, even during dynamic conditions. The estimated pushing force reflects the expected behavior when tested for various pushing efforts and in different ground conditions. Eventually, the algorithm was tested on a complete loading cycle and the results show good consistency considering the measured front-loader trajectory and vehicle speed. The proposed model overcomes some drawbacks of the currently used technologies. For example, it allows for an online estimation of the bucket forces for any position of the implement. Although the results discussed in this article pertain to a specific reference machine, the proposed method can be extended to most wheel loaders equipped with standard digging equipment.

Justification of design factors for front loader attachments in quarrying and handling of natural stone blocks

After analyzing the technology of natural stone mining in Russia in comparison with European countries, we conclude that one of the ways to increase the efficiency of the mining transport equipment system in natural stone quarries is to use wheel loaders more widely. Wheel loaders are characterized by high maneuverability, productivity, and are able to combine excavation, transportation, shipping, as well as perform auxiliary work. At the same time, the loader is very effective for providing access to any section of the quarry after stripping loose overburden and waste that impede movement. To increase the profitability and economic efficiency of natural stone quarrying, it is advisable to use a wheel loader as the main handling, transport and auxiliary equipment. The main reason why wheel loaders equipped with attachments are not widely used in the development of natural stone deposits in Russia is their high cost, which significantly increases capital and operating costs since the main manufacturers of wheel loaders equipped with quick-mounted attachments are such brands as Comatsu, Caterpillar, Volvo, Hyundai, etc. For these purposes, the mounted auxiliary equipment for the CHETRA PK 120 wheel loader of domestic production was designed. The strength of the coupler metalworks and and its hydraulic exponents are calculated. During the design, hydraulic equipment was selected to control and regulate attachments. In case of using the proposed system, the investment costs will decrease, and the proposed organizational and technical measures will increase the efficiency of mining companies due to reduced operating costs.