Kevin A. Edge

Fault diagnosis of a hydraulic actuator circuit using neural networks—an output vector space classification approach

Abstract This paper presents a neural network approach to fault diagnosis of dynamic engineering systems based on the classification of surfaces in system output vector space. A simple second-order system is used to illustrate graphically the nature of the diagnosis problem and to develop theory. The approach is then applied to the diagnosis of a laboratory-based hydraulic actuator circuit. Results are presented for networks trained on both simulation and experimental data. An important achievement is the diagnosis of experimental faults using a network trained only on simulation data.

Cylinder Pressure Transients in Oil Hydraulic Pumps with Sliding Plate Valves

This paper reports an experimental study of the cylinder pressure within an axial piston pump. This study revealed that existing theoretical models, which are based on the effects of fluid compliance within the cylinder, are highly inaccurate at high speeds or high loads. Fluid momentum at the point of port opening was found to be of considerable importance and an improved digital computer model was developed as an aid to pump design. The inclusion of fluid momentum effects resulted in a significant improvement in the agreement between theory and experiment. Cavitation within the cylinder bore was predicted at both high speed and high load conditions; this was confirmed experimentally. The theoretical approach is applicable to any sliding valve plate unit.

A controlled friction damper for vehicle applications

This paper examines the performance of a servo-driven dry-friction damper in a car suspension application; this device is a potential alternative to a traditional viscous damper. The friction damper is semi-active: damping is controlled without energy introduction into the system and hence the power required is much smaller than fully active systems. Models for the friction damper hydraulic drive and vehicle ride are developed. It is shown through simulation and experimental studies that a VSC-controlled friction damper has potentially superior performance to a conventional damper. Limitations of the current design are identified and suggestions for improvements are outlined.

Modeling Requirements for the Parallel Simulation of Hydraulic Systems

Parallel simulation of systems offers the benefit of increased speed of execution, but requires the system model to be partitioned to enable numerical tasks to be performed concurrently. Hydraulic systems are characterised by a transport delay in the pipelines connecting physical components, which is due to the propagation of waves at the speed of sound through the fluid medium. The transmission delay allows component models to be decoupled for the current time step, enabling a parallel solution; the inputs to each component model are delayed outputs from connected models. This paper describes a simulation environment suitable for the simulation of hydraulic system performance, using the transmission line modelling approach for the pipelines, decoupling the component models in a hydraulic circuit simulation. Computationally efficient models for cavitation and friction are developed and evaluated. In addition, partitioning strategies for parallel operation are outlined, although these have yet to be implemented

Pressure pulsations in reciprocating pump piping systems Part 1: Modelling

Abstract A distributed parameter model of pipeline transmission line behaviour is presented, based on a Galerkin method incorporating frequency-dependent friction. This is readily interfaced to an existing model of the pumping dynamics of a plunger pump to allow time-domain simulations of pipeline pressure pulsations in both suction and delivery lines. A new model for the pump inlet manifold is also proposed.

Steady-state steering of a tilting three-wheeled vehicle

The design of a narrow-track enclosed vehicle for urban transport was the subject of the CLEVER project. Due to its narrow track and requirement for car-like controls, an actively controlled tilting system was integrated into the chassis to allow for high lateral accelerations without rolling over. The cornering behaviour of this unique vehicle concept is investigated and compared with the ideal Ackermann response. The steer kinematics of this 1F1T (one front wheel, one wheel tilting) configuration are assessed through the use of a steady-state steering model, with attention focused on how steer parameters such as tilt axis height and inclination can be tuned to provide the required response. A prototype vehicle was designed and built and the results of experimental testing are presented to illustrate the real balancing performance of the combined steering and tilting approach used for the CLEVER vehicle. The experimental results follow the trends demonstrated in the model.

Automated Fault Analysis for Hydraulic Systems: Part 2: Applications

The paper provides further details of the automated failure modes and effects analysis (FMEA) program outlined in Part 1. Some of the more difficult development problems are discussed, and solutions are presented. The functionality of the program was tested through application to two experimental rigs, namely a closed-loop hydrostatic transmission with a dynamometer and a regenerative pump test rig. Non-destructive faults, such as abnormally low relief valve settings and excessive loads, were manually inserted into these rigs, and the measured effects were compared with the predictions from the program to validate the software. The difference in complexity and configuration evident in the two examples considered serves to highlight the generality of the approach. The ease of reconfigurability of the software reflects the key aim of producing a program capable of analysing a wide range of hydraulic circuits.

Automated Fault Analysis for Hydraulic Systems: Part 1: Fundamentals

Early expert systems for fault analysis tended to be based on shallow, heuristic knowledge. For success in engineering applications, it is argued that the complementary knowledge of the underlying principles (deep knowledge) should also be modelled. An object-oriented software library representing models of components of hydraulic circuits is being built using deep knowledge alone. The software modules within this library are reusable for the construction of model-based expert systems for the performance of failure mode and effects analysis and fault tree analysis on any arbitrary hydraulic circuit.

Damp-by-wire: Magnetorheological vs Friction Dampers

Abstract A comparison is made of a magnetorheological damper (MRD) and a friction damper (FD) in semi-active suspension design; both devices are closed-loop controlled. Experimentally validated models of both dampers have been developed and a hybrid variable structure-fuzzy control algorithm designed. It is shown that the two devices, albeit based on very different physical principles, can be described by a general common mathematical model, with the FD model being regarded as a particular case of the model for a MRD. The control strategy is targeted to improve occupant comfort. Performance has been investigated numerical simulation in a quarter car vehicle model and compared to that obtained with a classical passive damper.

Development of a Steer Tilt Controller for a Three Wheeled Tilting Vehicle

A three wheeled tilting vehicle called CLEVER was developed at the University of Bath. The tilt mechanism of this vehicle consisted of two hydraulic actuators that tilted the cabin in response to the driving conditions. Although this system was reliable, it had high power requirements, so a different method was needed. One way in which such a vehicle could be tilted was using the same principle that a motorcycle rider applies to tilt his bike, namely countersteer. This type of tilt control was expected to reduce the power required to lean. First, the vehicle and a basic steer controller were modelled. The simulations showed that the steer controller balanced the vehicle well, but deviated significantly from the intended path. A controller that could combine both the balance and the path following function was required. A good controller for this task is clearly the driver, so a pilot study was launched where the steering inputs of various drivers were measured. This study was carried out using a three wheeled tilting moped. The results of this study showed that the frequency of the steering inputs depended on the driver’s experience and the more experienced the driver, the lower the frequency of steer input for a given manoeuvre. The steady state manoeuvre showed that all drivers achieved a lean angle depending on the speed and turning radius in compliance with the theory. The countersteer was difficult to determine, because the drivers shifted their weight to aid the tilting. This indicated that countersteer is dependent on the driver and the driving conditions. However, a correlation between the countersteer and the countersteer rate was found, showing that either little countersteer could be applied for a short time, or a lot of countersteer could be applied for a short time. Another correlation was found between the countersteer rate and the maximum tilt acceleration, where the larger the countersteer rate, the larger the tilt acceleration. Since the tilt acceleration was related to the lateral acceleration, these correlations could aid the development of a steer tilt controller.Copyright