Future Operating Conditions Research Articles

Adaptive neurofuzzy control of a robotic gripper with on-line machine learning

Pre-programming complex robotic systems to operate in unstructured environments is extremely difficult because of the programmer’s inability to predict future operating conditions in the face of unforeseen environmental conditions, mechanical wear of parts, etc. The solution to this problem is for the robot controller to learn on-line about its own capabilities and limitations when interacting with its environment. At the present state of technology, this poses a challenge to existing machine learning methods. We study this problem using a simple two-fingered gripper which learns to grasp an object with appropriate force, without slip while minimising chances of damage to the object. Three machine learning methods are used to produce a neurofuzzy controller for the gripper. These are off-line supervised neurofuzzy learning and two on-line methods, namely unsupervised reinforcement learning and an unsupervised/supervised hybrid. With the two on-line methods, we demonstrate that the controller can learn through interaction with its environment to overcome simulated failure of its sensors. Further, the hybrid is shown to out perform reinforcement learning alone in terms of faster adaptation to the changing circumstances of sensor failure. The hybrid learning scheme allows us to make best use of such pre-labeled datasets as might exist and to remember effectively good control actions discovered by reinforcement learning.

LabHouse: System simulation and emulation within boiler development

The LabHouse is an emulation test rig for boilers in which future operating conditions are reproduced dynamically in real time. The environment of the boiler is reproduced by a TRNSYS-simulation. The control of the test rig hardware and the interface to simulation use the software LabVIEW. The hardware interfaces include not only electrical signals, but also water temperatures and hydraulic resistances. This paper sheds light on the structure, the function and some examples for the practical application of this tool inside industrial development: early system testing (fault detection), derivation of a solid base for long-run bench tests and derivation of future requirements.

OPTIMAL SIZING OF IRRIGATION DELIVERY SYSTEMS USING A TWO-STAGE STOCHASTIC PROGRAMMING APPROACH

In arid areas where the available water is limited and randomly changing with time, the surface irrigation might still be the best option due to economic and other considerations. Hence, the irrigation delivery system should be sized in such a way that the allocation of available surface water to different crops at any time yields maximum possible revenues. This can be ultimately achieved with the aid of a two-stage optimal formulation combining the present design stage with the future operational conditions. This work applies two stochastic mathematical techniques to determine the optimal sizes of a hypothetical surface irrigation delivery system subjected to varying deficit levels of available water and considering the uncertainties associated with the available water and the crop revenues. Canal sizes represented the first-stage decisions while the future water allocations were made in the second stage. The Regularized Stochastic Decomposition approach solved the case of stochastic available water present at the right hand side, while the L-Shaped method solved the case of stochastic second-objective coefficients associated with the crop selling prices. The L-Shaped method was also used to consider the problem of uncertain parameters in both the right hand side and the second-objective coefficients. The problem was investigated by considering a system of nine irrigation canals and six farms grown with different crops during a season of two distinctive irrigation periods. Results obtained from the stochastic formulation were compared with the corresponding ones obtained from an equivalent deterministic formulation. The comparison showed that the effect of parameter uncertainties on canal sizes and associated revenues increases with the increasing levels of deficit in water supply.

Effect of long-term oxidation on creep and failure of Si 3N 4 and Si 3N 4/SiC nanocomposites

The high-temperature mechanical behaviour of an Si 3N 4/SiC nanocomposite and its monolithic Si 3N 4 reference material was studied after long-term oxidation treatments intended to simulate future operating conditions in a severe environment. Creep and failure at elevated temperature were significantly affected, in the direction of increased brittleness. The transition stress between the ductile range present at low stresses and the brittle range existing at high stresses was shifted to distinctly lower values. The creep resistance in the low-stress range was increased by the oxidation treatment. The failure time under a given stress was drastically reduced; this was attributed to an increased sensitivity to subcritical crack growth. The failure stress for a given failure time was decreased by about half. The phenomena are explained in terms of a purification of the intergranular phase and by the formation of surface defects and of a uniformly distributed pore population.

A methodology for risk-based inspection of pressurized systems

This paper describes a methodology for risk-based inspection and focuses on the planning of inspection activities. Brunei Shell Petroleum (BSP) operates a large number of ageing production and evacuation facilities in the South China Sea, and is facing the challenge to identify the optimal inspection effort in a period of time when more production is required and operating costs have to be minimized. Since 1989 major investments (annual costs about 15 million US$) were made to perform base line inspections which resulted in corrective work and a large repository of equipment integrity data. These data require analysis and updating, as defined in this methodology. The challenge is for BSP to implement risk-based inspection of static equipment, which so far has been very successful. The methodology is based on the determination of risk by evaluating the consequences and the likelihood of equipment failure. Likelihood of failure is assessed, by means of extrapolation, at the future planned maintenance campaign to identify the necessary corrective work. The extrapolation is based on historical inspection and maintenance data as well as on anticipated future operating conditions. Such an analysis requires in-depth knowledge of the equipment regarding functions, operating conditions, materials, imposed potential failure modes and causes, and their controlling parameters (integrity parameters). The novelty of the method is that equipment integrity is expressed in terms of integrity parameters, which can be measured and trended. Consequently, based on defined limits for the integrity parameters (i.e. anomaly classifications), actions for managing risk to an acceptable low level can be identified. The transparent linkage between monitoring activities (e.g. inspection) and integrity parameters offers an effective framework to manage and to demonstrate integrity against clear quantifiable acceptance limits. By pursuing the above methodology for all equipment, a list of inspection activities, prioritized on equipment condition and the risk of failure, can be derived. Within BSP this has resulted in a demonstrated increased efficiency of the inspection effort, a focused strategy for repairs executed on a campaign basis, and an improved quantification of BSP's facilities integrity. The approach has been proven in the company to be practical and it can be implemented relatively quickly.

SOME MECHANICAL CHARACTERISTICS OF NITROGEN STRUCTURAL STEELS

In this paper, test conditions and microstructural factors have been investigated for newly developed nitrogen steels at load rates and test temperatures suitable for future operating conditions. Static tests have been carried out at e∼ 10−4 s-1 and dynamic tests at e∼ 102 s−1. Different test temperatures enable fracture mechanisms and fracture toughness variations to be studied. Anisotropy of some characteristics in the initial (non-heat-treated) state, together with heat treatment effects and structural changes under two load rates have been analysed. Fracture appearance transition temperature (FATT) values have been obtained from absorbed energy and fracture surface crystallinity values. Using yield strength and fracture toughness data, an estimate has been made of critical crack lengths for unstable fracture.

Operational estimation and prediction of nitrification dynamics in the activated sludge process

This paper examines the feasibility and discusses the potential of applications of on-line real-time state estimation and prediction in operational control of the activated sludge process. In particular, the dynamics of nitrification are considered with reference to the activated sludge unit at the Norwich Sewage Works in eastern England. A recursive estimation algorithm, the extended Kalman filter, is applied both for reconstructing operating information on the variations in nitrifying bacterial population concentrations and for making predictions of process performance under assumed scenarios for the short-term future operating conditions of the plant. Time-series field data from the Norwich Works are used for the former analysis. Considerations of uncertainty and the possibility of rapid major perturbations in performance, for example, due to spillages of toxic substances or the loss of solids over the clarifier weir, are of special importance to the discussion. The paper is introduced and concluded with some more general comments on the roles of operator experience and decision-making and man-machine interaction in wastewater treatment plant control.