Active Research Projects:

Compressed Air Energy Storage Using a Liquid Piston

As utilizing renewable energy resources becomes more crucial in today's age, there arises a need to effectively integrate these often unpredictable and intermittent energy sources onto the electric grid. This is difficult to do since the energy available often does not align with the power demand. For example wind blows most heavily at night when power demand is low. Thus, there arises a need for large scale energy storage so that energy harvested can be supplied to the electric grid when it is needed.

One such method of energy storage is compressed air energy storage (CAES). In order to store compressed air effectively, a highly efficient gas compression is needed. Rapidly compressing the air will produce an increase in temperature. This rise in internal energy is then gradually lost as the air cools off in the storage vessel over time. Compressing the air with the least amount of temperature increase will yield the highest thermal efficiency. To achieve a near-isothermal compression quickly, large amounts of heat transfer are necessary. In this system we use a liquid piston for our gas compression. Since liquid can flow through a tortuous path, we can insert a high surface area heat exchanger into the compression chamber to aide in heat transfer. This method of gas compression results in near-isothermal compression in a relatively short period of time, which is ideal for energy storage/recovery. We plan to use this technology with large offshore (5 MW) windmills and have the liquid piston be driven by a hydraulic water pump.

Crosstalk Reduction in Multi-Actuator Systems

In industrial testing setups, a common practice is to drive multiple hydraulic testing stations with a single Hydraulic Power Unit (HPU). The HPU is usually mounted at a large distance from the load frames and is connected to them by rigid tubing. Each testing station typically contains high and low pressure accumulators and a servo valve. A primary objective of such a testing facility is to permit the usage of multiple test stations simultaneously with precise control of the actuator position or force. However, due to the temporally varying flow rate at test station, pressure waves are generated in the pressure and supply lines. These pressure waves result in a varying pressure supply at each test station, possibly causing challenges with control precision.

The goal of this project is to establish design guidelines to reduce pressure variations (crosstalk) within a multi-actuator hydraulic system. We aim to create a modeling and design tool of pressure wave propegation in hydraulic pipelines with multipe hydraulic stations. The guidelines can then be used to mitigate the effects of crosstalk to improve control accuracy.

Hydraulic Flywheel Accumulator

The flywheel-accumulator is a novel energy storage device that integrates rotating kinetic and pneumatic energy storage into a single device. This approach addresses two primary challenges of conventional hydraulic accumulators: limited energy density and system pressure varying with energy storage. The flywheel-accumulator is a cylindrical pressure vessel containing compressed gas and hydraulic fluid, which are separated by a moving piston. Energy is added to the device by either adding hydraulic fluid, which increases the pressure and the moment of inertia, or by applying a torque to the device. Due to the centripetal acceleration acting on the hydraulic fluid, a pressure gradient is formed, causing the hydraulic fluid pressure at the center port to decrease with increasing angular velocity. Thus by modulating the method of adding energy to the device, the system pressure can be controlled independent of the energy stored. The combination of constant pressure and increasing the energy density by an order of magnitude make this technology revolutionary for hydraulic energy storage. Research in the MEPS group to date has resulted in steady-state system models, concept demonstration through a low-energy bench top prototype, and early development of a mathematical simulation of the transient behavior. Current work involves studying the fluid dynamics related to fluid swirl and addressing machine design challenges of a high-energy system.

Phase Shift Variable Displacement Pump

This research aims to reduce power loss in flow-control hydraulic applications by designing a variable displacement pump that offers high efficiency across a large displacement range and can be compactly stacked in line. In the phase-shift pump, pairs of pistons share a common displacement chamber. When the two pistons move in phase, their displacement is combined, resulting in maximum flow from the displacement chamber. When the piston motion is out of phase, fluid is shuttled between the pistons, resulting in no net flow. Partial displacements are credited by intermediate phase angles between the pistons. In this research, the phase-shift pump is realized with a double bank radial piston pump where the cam lobes of the two banks are shifted relative to each other.

Soft Burrowing Robot Locomotion

Current practices for pipe inspection, pipe repair, and underground utility installation require expensive and disruptive soil and infrastructure excavation. The MEPS Laboratory is conducting research in high-force, hydraulically powered, soft robotic actuators to solve this problem by developing multi-segment, hydraulic, soft robots that are capable of providing the large forces and kinematics to burrow into and through soil. The lab is working to design, model, and control the individual robot segments.

One type of soft robotic, fiber reinforced actuator that is being studied is the McKibben actuator which is able to provide extension and contraction forces, as determined by the initial fiber wrap angles, when internal pressure is applied. By combining the McKibben actuator in series with radially expanding soft, traction segments, "inch worm" type locomotion can be achieved. By utilizing the high power density of hydraulics to power the segments, large forces can be achieved and the soil-actuator interactions can be optimized to accomplish a lasting burrow or hole. This technology may also be utilized in search and rescue robots, exploration for natural resources, and anchoring of heavy, off-shore oil rigs.

Soft Burrowing Robot Head Design

The soft burrowing robot must be able to both locomote through a subsurface burrow and propagate the burrow in an efficient manner. The burrow propagation is perfored by the penetrating head of the robot. The design of the head work is largely inspired by two mechanisms existing in nature, penetration of soil by plant roots and the burrowing action of earthworms. These two natural evolved processes effectively penetrating soil in an efficient manner. The aim of this research is to mimic such motions and design a penetrating head that will perform the task of burrowing as efficiently as possible. In the later stages of the research, the aim is to use sensors to predict the path the robot should take to encounter least resistance from the soil.

Soft Robot Locomotion in Medical Applications

Medical robots are continually becoming more advanced, but are limited due to their size and rigidity. Soft robotics provides an opportunity to overcome those limitations by offering compliant, inherently safe robots for use in environments like the human body. Despite these advantages, soft robots have yet to conquer the ability to navigate small spaces while avoiding higher-than-necessary forces on delicate anatomy.

The medical soft robot project focuses on the design and modeling of a hydraulically-actuated robot that inchworms through tube-like structures. The robot consists of three segments that anchor into tissue walls without blocking blood flow and extend to move through the anatomy. The robot is actuated using one input line of fluid (serially) instead of three fluid lines (in parallel) to conserve space, but this adds to the complexity of controlling three segments with only one input signal. Other branches of exploration include a traction model to better predict the behavior of the robot in contact with various anatomical structures as well as new valve design and actuation methodologies to improve the overall control of the robot.

Soft Robot Inverse Design

The research objective of this work is to establish and validate a mathematical framework for the inverse design of universal soft robots that:

  • Provide exceptional force and power densities.
  • Dictate a new means of robotic, automated manufacturing via 3D printed materials exploiting highly anisotropic elastomers, inextensible fibers, and beam elements.

For the application of a catheter robot, the actuator geometry's are created to fit specific anatomy given by CT scans. Software is used to extract the centerlines out of the CT scan as indicated by the yellow line. Next a centerline of the actuated robot is created, this could be anatomy centerline or a spiral centerline as indicated in red. Convex blending is then implemented to create a centerline between the robot centerline and anatomy centerline via a joint angle q. A surface is then fit to the centerline using Frenet-Serret parameterization. The soft robot design provides the displacements needed for every point of the actuator. We then define an input pressure and apply an inverse finite element scheme solving for material properties at any point on the actuator.

Switch-Mode Hydraulic Circuits

Switch mode power converters act as a DC-DC hydraulic transformer, analogous to buck and boost converters in power electronics. By utilizing a fluid inertial element, these converters can efficiently boost the pressure above a supply level or buck the pressure below it. In application, a converter would be applied to modulate power at each actuator and supplied by a common pressure rail. These circuits have the potential to eliminate throttling losses associated with the present metering valve approach, which can exceed 50% losses. While the models of buck/boost circuits developed in the lab have demonstrated potential for 85%+ efficiency across their range of power, innovative approaches to rapid hydraulic switching and fluid compressibility analysis are needed before they can be physically realized. In the MEPS lab, we are approaching this problem with a multi-pronged approach. We previously demonstrated an active soft switch to absorb flow during valve transition to reduce transitional throttling losses. Current work involves designing a variable duty ratio hydraulic valve that will cycle at 100Hz. Work in the near future will center around deriving models for fluid compressibility that account for gasses going in and out of solution during the rapid transients which are characteristic of switch mode hydraulics. With these pieces solved, the converter models will be updated and optimized for a laboratory scale prototype converter.

Variable Displacement Linkage Pump/Motor

Researchers in the MEPS Lab are designing a novel, highly efficient hydraulic motor for the propulsion of off-highway vehicles. This motor uses an adjustable four-bar linkage that drives a cam to vary the displacement of the piston, resulting in a Variable Displacement Linkage Motor (VDLM). The efficiency of this motor is high across a wide range of displacements, speeds, and pressures. A robust model of the motor has been constructed that captures the mechanism kinematics and kinetics, tribology in the bearings and piston-cylinder, pressure dynamics in the cylinder, valve timing, and other sources of energy loss. This model is being used in a multi-objective optimization of the motor parameters for specific applications and drive cycles. The project team includes individuals from the University of Minnesota, Milwaukee School of Engineers, and several industrial partners.

VDLM Active Valves

Active valve timing is necessary for all hydraulic motors. This valve timing can either be fixed or vary with motor operating conditions like speed and working pressure. Fixed valve timing is simpler and less expensive than variable valve timing. However, variable valve timing yields higher efficiency. A research thrust for the VDLM is determining if fixed valve timing is sufficient or if the added gain from variable valve timing justifies the added cost and complexity. A computer model of the fluid dynamics of the valves, piston, and cylinders in the VDLM has been created to study the effects of fixed and variable valve timing. Three main system variables have been of interest to study the operation of the VDLM: working pressure, displacement, and bottom dead center location of the piston.

To actuate the active valves of the VDLM, two primary options were considered: poppet valves, or spool valves. A single spool valve per cylinder is preferred out of simplicity compared with two poppet valves per cylinder. However, spool valves have the potential for higher leakage, especially at low motor speed. To test this, a model was developed to examine how the leakage of the spool valve behaved under many different operating conditions. The results of this study show that the energy lost to leakage are insignificant across most of the operating range and do not exceed one percent of the energy output of the cylinder during extreme operating conditions. This result strongly supports that leakage past the spool valve is negligible, driving the design decision to use spool valves in the VDLM.

Hydraulic Power Take-Off for Wave Energy Conversion

Power can be harvested from ocean waves through the reaction of a mechanical device against the forces of waves. The oscillatory motion of these devices is slow and involves high forces. The amplitude of the motions is irregular with wave forces varying wave-to-wave; the ratio of the peak to mean power harvested can be as high as ten to one. It is a challenge for power systems to remain cost effective and efficient with this degree of variability.

It is common for fluid power to be applied to systems with a high peak-to-mean power ratios or low duty, take off-highway vehicles and construction equipment as examples. However, fluid power systems, conventionally, are not known for efficiency and rely on converting large portions of energy to waste heat to manage low-duty, high-power applications. The application of fluid power to wave energy, with all its challenges, offers the opportunity to rethink fluid power architectures. Switch-mode hydraulics, which involves switching between discrete and efficient modes of operation, may offer solutions that enable wave energy technologies and lend insight leading to more efficient fluid power systems, in general.

Electric to Hydraulic Energy Conversion

The electrification of vehicles is an area that has seen tremendous growth over the past few years, and that growth has also extended to off-highway vehicles. However, with the high power drivetrain requirements and high force auxiliary loads of off-highway vehicles make hydraulics a logical power transmission choice. As such, there is a need for compact and efficient conversion between electric and hydraulic power. The MEPS lab is conducting research to design, model, and prototype an integrated linear electric to hydraulic piston pump for use as a charge pump for hydrostatic circuits in off-highway vehicles.

Current electric to hydraulic conversion uses modularized components and has multiple different energy conversions. Typically, a system would consist of an electric motor, a mechanical transmission, and a piston pump to convert the shaft power to piston oscillation, and then to hydraulic power output. The integrated machine eliminates these multiple energy conversions by directly using electromagnetic forces to magnets mounted on the piston. This allows the ends of the moving element of the electric motor to act as pistons of a double acting pump. As the pistons reciprocate in the cylinders, check valves control fluid flow into the cylinders from a low-pressure tank and direct the fluid out of the cylinders to the high-pressure outlet.

Hydraulic Motor Design for Seawater Operation

Reverse osmosis systems are used to produce drinkable water using seawater sources. Energy recovery devices used in large scale plants can reduce energy consumption by up to 60%. A major problem with these devices is their reduced efficiency when they are scaled down.

The MEPS Laboratory is currently working on the development of an energy recovery device in the form of a hydraulic motor for smaller applications, using triplex piston pump technology. Modeling and component testing have been carried out to determine the performance characteristics of this type of motors. Other important areas of application of this technology include mud-pumping, oil-drilling, corrosive fluids, etc.

Dormant Research Projects:

Variable Timing Pump/Motor Valves

Valve timing is important to the efficiency, power, and noise of hydraulic pumps and motors. The valves of current pumps and motors are fixed and tuned for a specific operating condition. When the operating conditions or fluid properties change, the performance of the pump motor suffers. The goal of this research is to achieve high volumetric efficiency through dynamically varying the opening and closing of the pressure and tank valves to match changing operating conditions, including pressure, angular velocity, fluid compressibility, and displacement. This is being achieved through three objectives:

  1. A simulation that creates the efficiency map for different operation conditions.
  2. Experimentally validate the efficiency map using valve prototype.
  3. Study of active valve timing on the impact of system volumetric efficiency.

Characterizing Check Valve Dynamics

The check valve is a passive device that allows uni-directional flow. Check valves are used in a variety of hydraulic circuits and components, including applications requiring fast valve response such as pumps, digital hydraulics, and servo hydraulic circuits. The dynamic performance of the check valve is critical for determining system parameters such as pressure pulsations and flow ripple. To design better hydraulic systems, the MEPS lab is developing a mathematical model of the fluid-mechanical dynamics of a check valve.

While the behavior of a valve could be predicted through a multi-dimensional multi-physics computational simulation, this approach is computationally expensive and not appropriate for iterative design and optimization. The need for a lumped parameter dynamic model of a check valve is being addressed through a coupled analytical and experimental approach.

A novel experimental setup has being developed to measure the position the valve moving element using a laser displacement sensor. The experimental results are being used to validate the model and construct appropriate empirical corrections. The validated model will then be used for design and optimize check valves for specific high speed applications.

Switch-Mode Continuously Variable Transmission

Analogous to switch-mode hydraulics, a continuously variable transmission (CVT) with flywheel energy storage is enabled by applying switch-moded control to the rotating mechanical domain. In the switch-mode CVT, the energy stored in a flywheel is transferred to an output shaft through a high-speed clutch and a torsion spring. The primary limiting component of this system is a clutch capable of high frequency pulsation with low energy loss. To address this challenge, the MEPS group has designed and demonstrated a custom variable duty cycle clutch for this application. In addition, a custom torsion spring with high deflection and low inertia has been designed and tested. Current work includes system integration, more in-depth system modeling, and design for high energy and power levels.

Mobile Hydraulic Power Supply / Stirling Engine

There exists great potential to improve the compactness and efficiency of mobile hydraulic power sources. Leveraging the liquid piston gas compression concept from the open accumulator, the MEPS group is studying a liquid piston Stirling engine pump. This novel hydraulic power supply eliminates much of the kinematic linkage in the engine and improves heat transfer in the working chambers with the use of liquid pistons. The research team is currently approaching this problem through computational fluid dynamics studies of liquid piston gas compression, creating a computational model of the Stirling cycle for system optimization, and experimental studies of a liquid ring air compressor.

Hydraulic Applications: Hybrid Vehicles, Rescue Spreader, Etc.

The MEPS group is also involved in a variety of projects involving research challenges in specific applications of hydraulics. For example, the research team is developing a mobile hydraulic power supply for a hand-held hydraulic rescue spreader (Jaws of Life). In addition, the research team has extensive experience in hydraulic hybrid vehicles of parallel, series, and power-split architectures.


Sample image caption

U of M Stock Photos