Search

Considering the load uncertainty and unmodeled dynamics in multicylinder hydraulic systems, this paper proposes a balance control algorithm based on safe reinforcement learning to release the restrictions of classical model-based control methods that depend on fixed gain. In this paper, the hydraulic press is controlled by a trained agent that directly maps the system states to control commands in an end-to-end manner. By introducing an action modifier into the algorithm, the system states are kept within security constraints from the beginning of training, making safe exploration possible. Furthermore, a normalized exponential reward function has been proposed. Compared with a quadratic reward function, the precision is greatly improved under the same training steps. The experiment shows that our algorithm can achieve high precision and fast balance for multicylinder hydraulic presses while being highly robust. To the best of our knowledge, this research is the first to attempt the application of a reinforcement learning algorithm to multi-execution units of hydraulic systems. [ABSTRACT FROM AUTHOR]

Digital hydraulics is a novel alternative to proportional or servovalve-controlled systems in fluid power engineering, providing hydraulic systems with high-energy efficiency, good controllability, and insensitivity to contamination. Switched inertance hydraulic converters (SIHCs) are new digital hydraulic devices that can adjust flow and pressure by digital switching instead of throttling the flow. In this paper, an efficient closed-loop control system is proposed for SIHCs subject to time-varying loading conditions in which the load pressure and/or flow varies with time. The control system is designed to operate SIHCs at optimized switching frequencies and ratios that maximize system efficiency when the load varies. With the proposed controller, the SIHC can effectively adapt to the time-varying load and has achieved up to 10% efficiency improvement and up to 65% pressure ripple reduction without affecting the system's dynamic responses, compared with using a nonoptimized controller. The work shows the feasibility and advantages of simultaneously controlling the switching ratio and switching frequency of SIHCs with a time-varying load. As time-varying loading conditions are commonly found in hydraulic applications, the research outcomes constitute an important aspect in the design and development of highly efficient SIHCs and their practical use in hydraulic machinery. [ABSTRACT FROM AUTHOR]

This paper presents the formulation of a variable load sense control strategy suitable to achieve power savings in hydraulic systems using postcompensated load sensing (LS) hydraulic control architectures. Such architecture is typical in off-road construction machinery. The paper also describes the application of the proposed control strategy referred to as variable load sensing margin (VLM) on a full-size wheel loader. The paper first presents the rationale for the proposed strategy, showing how the state-of-the-art LS architecture present in commercial machines has a margin for lowering the throttling losses present at the control valves. A feedforward controller, derived from an empirical study on a reference vehicle, is used to control the flow to the front-end loader functions. Test results show improvements of the hydraulic power consumption up to 45%, based on the commanded speed of each front-end loader actuator. The paper also describes a gain scheduling pressure feedback control strategy, which is used to allow controlling also functions that include priority. For the case of off-road vehicles, this is typically the steering function. The experimental results show good performances with an error in controlled velocity below 5%, which is achieved when the front-end loader functions are used concurrently with the steering. [ABSTRACT FROM AUTHOR]

Counterbalance valves (CBVs) are widely used in hydraulic machines handling gravitational loads. They are usually adopted in hydraulic circuits that set the flow to the actuator with either metering control strategies or primary control strategies. In these circuits, the CBV does not determine the actuator velocity, but it establishes a proper counterpressure that balances the actuator during overrunning loads. A well-known problem of CBVs is the excessive power consumption introduced into the circuit, which is due to an excessive pressurization of the flow supply. This paper presents a solution for using CBVs to control the actuator velocity while also reducing energy consumption. The method consists in controlling the pilot port of the CBV through an external pressure source (adjustable pilot). The solution is first studied analytically, considering all the possible loading conditions at the actuator. Two different control strategies are presented: the "Smart CBV," where the CBV does not control the actuator velocity but it minimizes the system pressure; and the "Smart System" that uses the CBV to efficiently control the actuator velocity during overrunning load conditions. An experimental validation of the proposed solutions is also provided, taking as reference a truck-mounted hydraulic crane. The results of the proposed solutions are compared to those achieved by the traditional circuit of the reference hydraulic crane which uses standard CBVs. For both cases of smart CBV and smart system, a remarkable power saving, respectively, up to 75% and up to 90% is observed. [ABSTRACT FROM AUTHOR]

Hydraulic pressure ripple in a pump, as a result of converting rotational power to fluid power, continues to be a problem faced when developing hydraulic systems due to the resulting noise generated. In this paper, we present simulation results from leveraging an actor-critic reinforcement learning method as the control method for active noise control in a hydraulic system. The results demonstrate greater than 96%, 81%, and 61% pressure ripple reduction for the first, second, and third harmonics, respectively, in a single operating point test, along with the advantage of feed forward like control for high bandwidth response during dynamic changes in the operating point. It also demonstrates the disadvantage of long convergence times while the controller is effectively learning the optimal control policy. Additionally, this work demonstrates the ancillary benefit of the elimination of the injection of white noise for the purpose of system identification in the current state of the art. [ABSTRACT FROM AUTHOR]

Artificial muscles (AMs) traditionally rely on pneumatic sources of fluid power. The use of hydraulics can increase the power and force to weight and volume ratios of AM actuators. This paper develops a control-centric third-order single-input single-output (SISO) lumped-parameter dynamic model and sliding mode position controller based on Filippov's principle of equivalent dynamics for a braided hydraulic artificial muscle (HAM) actuator. The model predicts the nonlinear behavior of the HAM free contraction and captures the fluid and actuator nonlinear dynamic interactions in addition to the braid deformation. Model simulations are compared to experimental results for quasi-static pressurization, isometric pressurization, and open-loop square wave commands at 0.25, 0.5, and 1 Hz. Experiments of sine wave tracking at 0.25, 0.5, and 1 Hz and continuous square wave tracking at 0.067 Hz are conducted using a sliding mode controller (SMC) derived from the model. The SMC achieves a steady-state error of 6 μm at multiple setpoints within the actuator's 17 mm stroke. Compared to a proportional-integral-derivative (PID) controller, the SMC root-mean-square (RMS) error, mean error, and absolute maximum error are reduced on average by 53%, 61%, and 44%, respectively, demonstrating the benefit of model-based approaches for controlling HAMs. [ABSTRACT FROM AUTHOR]

This paper proposes an active roll control system for passenger cars. The roll actuator with electric motor is expected to replace the hydraulic roll control actuator with the better performance. In order to meet those expectations, highly precise control methods are needed for adjusting the roll motion of the vehicle. In this study, roll dynamics of vehicle and the properties of active roll actuators are investigated first including the latency of the actuator. The identification method is designed to estimate the key parameters of the suspension system. The reference model method is proposed to determine the target roll states and the model predictive control (MPC) method is adopted to control the rolling motion. The roll moment distribution method is also designed between the front and rear actuators considering the load transfer. The proposed control system is validated via simulations and experiments. [ABSTRACT FROM AUTHOR]

In this paper, a hierarchical control logic for two-channel hydraulic active roll control (ARC) system, which includes vehicle level control and actuator level control is proposed. Vehicle level control consists of antiroll torque controller and antiroll torque distributor. The antiroll torque controller is designed with "PID + feedforward" algorithm to calculate the total antiroll moment. The antiroll torque distributor is devised based on fuzzy control method to implement an antiroll moment allocation between the front and rear stabilizer bar. Actuator level control is designed based on pressure and displacement, respectively. The contrastive analysis of the two proposed actuator control method is presented. The hardware-in-the-loop (HIL) test platform is proposed to evaluate the performance of the devised control algorithm. The HIL simulation result illustrates that actuator displacement control could generate a relatively accurate antiroll moment, and the vehicle roll stability, yaw stability can be enhanced by the proposed ARC control method. [ABSTRACT FROM AUTHOR]

This paper reports on experimental investigations of a switched inertance hydraulic system (SIHS), which is designed to control the flow and pressure of a hydraulic supply. The switched system basically consists of a switching element, an inductance (inertance), and a capacitance. Two basic modes, a flow booster and a pressure booster, can be configured in a three-port SIHS. It is capable of boosting the pressure or flow with a corresponding drop inflow or pressure, respectively. This technique makes use of the inherent reactive behavior of hydraulic components. A high-speed rotary valve is used to provide sufficiently high switching frequency and to minimize the pressure and flow loss at the valve orifice, and a small diameter tube is used to provide an inductive effect. In this paper, a flow booster is introduced as the switched system for investigation. The measured steady-state and dynamic characteristics of the rotary valve are presented, and the dynamics characteristics of the flow booster are investigated in terms of pressure loss, flow loss, and system efficiency. The speed of sound is measured by analysis of the measured dynamic pressures in the inertance tube. A detailed analytical model of an SIHS is applied to analyze the experimental results. Experimental results on a flow booster rig show a very promising performance for the SIHS. [ABSTRACT FROM AUTHOR]

Switch-mode hydraulic control is a compact and theoretically efficient alternative to throttling valve control or variable displacement pump control. However, a significant source of energy loss in switch-mode circuits is due to throttling during valve transitions. Hydraulic soft switching was previously proposed as a method of reducing the throttling energy loss, by absorbing, in a small variable volume chamber, the flow that would normally be throttled across the transitioning high-speed valve. An active locking mechanism was previously proposed that overcomes the main challenge with soft switching, which is a lock mechanism that releases quickly and with precise timing. This prior work demonstrated a reduction in energy losses by 66% compared to a control circuit. In this paper, a numerical model is developed for a switch-mode virtually variable displacement pump (VVDP) circuit, utilizing the proposed soft switch. The model is then used as a means of designing a proof of concept prototype to validate the model. The prototype design includes methods for controlling the soft switch spring preload, travel distance, piston displacement required to unlock the soft switch, valve command duty cycle, switching cycle period, and load pressure. Testing demonstrated that the soft switch circuit performed as expected in a baseline condition. The operating region for this prototype was found to be quite narrow. However, the model does a good job of predicting the displacement of the soft switch. [ABSTRACT FROM AUTHOR]

Flow control valves are extensively used to control the motion of hydraulic actuators in industry. In this paper, a two-stage proportional flow control valve with pilot digital pressure compensator is presented. The proposed valve is composed of two pressure sensors, a proportional pilot valve, a Valvistor poppet valve (main stage), and an electronic controller. The pressure drop across the pilot valve metering orifice is detected by two pressure sensors and fed into the controller which changes the pilot valve input voltage to compensate pressure influence on the outlet flow. The presented valve is investigated theoretically by simplified analytical models as well as numerical methods and also evaluated experimentally. The results show that the proposed scheme is feasible. It has a reasonable static performance and an acceptable dynamic characteristic for low bandwidth disturbances. The flow force on the Valvistor poppet should be estimated and considered in the controller design. [ABSTRACT FROM AUTHOR]

In this paper, reinforcement learning is applied to coordinate, in a decentralized fashion, the motions of a pair of hydraulic actuators whose task is to firmly hold and move an object along a specified trajectory under conventional position control. The learning goal is to reduce the interaction forces acting on the object that arise due to inevitable positioning errors resulting from the imperfect closed-loop actuator dynamics. Each actuator is therefore outfitted with a reinforcement learning neural network that modifies a centrally planned formation constrained position trajectory in response to the locally measured interaction force. It is shown that the actuators, which form a multiagent learning system, can learn decentralized control strategies that reduce the object interaction forces and thus greatly improve their coordination on the manipulation task. However, the problem of credit assignment, a common difficulty in multiagent learning systems, prevents the actuators from learning control strategies where each actuator contributes equally to reducing the interaction force. This problem is resolved in this paper via the periodic communication of limited local state information between the reinforcement learning actuators. Using both simulations and experiments, this paper examines some of the issues pertaining to learning in dynamic multiagent environments and establishes reinforcement learning as a potential technique for coordinating several nonlinear hydraulic manipulators performing a common task. [ABSTRACT FROM AUTHOR]

This paper deals with modeling and simulation of a class of three-way pressure reducing valves. The study aims to point out the peculiarities of function and operation of this class of valves in the steady-state and transient modes of operation. A comprehensive nonlinear mathematical model is deduced in order to predict the performance of the studied valve in both modes. The proposed modeI takes into consideration most nonlinearities of the studied valve. A computer simulation, based on the proposed model, is performed to predict the steady-state and transient performance. During the simulation study, it was found that nonlinearity occurs due to the following factors: the transient change in the valve operating pressures and the change in the throttling areas of the valve restrictions and their discharge coefficients. The transient change in the valve operating pressures causes nonlinear velocity changes of the fluid flow passing through the throttling areas of the valve restrictions. These throttling areas usually have nonlinear mathematical formulas. The discharge coefficients of these throttling areas are assumed constant independent of the flow rates, Reynolds number, and dimensions of these areas. However, these parameters affect the discharge coefficient in a complicated manner. The validity of the proposed model is assessed experimentally in the steady-state and transient modes of operation. The results show good agreement between simulation and experiment in both modes. The study shows that the geometry of the throttling orifice, which connects the upstream port to the downstream port, plays an important role in the studied valve steady-state and transient performance. This result implies the need for further investigation of the effect of the dimensions of the throttling orifices on the steady-state and transient performance of hydraulic control valves. [ABSTRACT FROM AUTHOR]

When modeling flexible robots and structures for control purposes, most often the assumed modes (AMs) method is used to describe the deformation in combination with a floating reference frame formulation. This typically has the benefit of obtaining a low-order, but accurate model of the flexible structure, if the number of modes and AMs are properly chosen. The basis for using this method is, however, that the vibrations (deflections) are time and position independent, i.e., the expression is separable in space and time. This holds for the classic Euler-Bernoulli beam equation, but essentially does not hold for translational links. Hence, special care has to be taken when including flexible translational links. In the current paper, different methods for modeling a hydraulic loader crane with a telescopic arm are investigated and compared using both the finite segment (FS) and AMs method. The translational links are approximated by a single beam, respectively, multiple beam elements, with both one and two modes and using different mode shapes. The models are all validated against experimental data and the comparison is made for different operating scenarios. Based on the results, it is found that in most cases a single beam, low mode order approximation is sufficient to accurately model the mechanical structure and this yields similar results as the FS method. [ABSTRACT FROM AUTHOR]

The air content in hydraulic transmission fluids significantly reduces bulk modulus of the fluid and causes a drop in the stiff-ness and response of the hydraulic system. It is consequently very important to monitor the air content in hydraulic fluid for ensur-ing the hydraulic works in good condition. In this paper, a novel method is presented in which the sampled fluid flows slowly into a vacuum chamber and the pressure of separated air is measured. A model of pressure-time characteristics is established, with mois-ture content taken into account as well, since moisture is volatile in vacuum and its content in tributyl phosphate (TBP) based fluid is usually too high to be neglected. The model can be simplified, which turned out to be a nonlinear least square problem. Compar-ison between the measured and calculated value shows that the model matches well with the experimental data. [ABSTRACT FROM AUTHOR]

This research is aimed at generating an efficiency map for a linear hydraulic-actuator that is controlled using an open-centered 4-way valve. Using the basic configuration of a double rod, double-acting linear actuator, equations for describing the input and output power of the actuator are written and nondimensionalized to produce a nondimensional map for the efficiency of the actuator itself. In conclusion, the actuator is shown to operate below a wide open valve (WOV) line with efficiencies that are typically less than 50%. Note that the WOV line describes the maximum actuator-force that may be achieved at a given actuator-velocity. While the actuator chosen for this paper is common, extensions of the methods used in this study may be applied to double-acting, single-rod designs or single-acting, single-rod designs. Furthermore, it is anticipated that this type of analysis may also be used to study the effects of power regenerating valves and for predicting the efficiency improvements that may be gained for linear hydraulic-actuators that use this type of valve design. [ABSTRACT FROM AUTHOR]