George N. Nyakoe

A neural fuzzy based maximum power point tracker for a photovoltaic system

The global electrical energy consumption is steadily rising and therefore there is need to increase the power generation capacity. The required capacity increase can be based on renewable energy. Photovoltaic energy remains a largely unexploited renewable energy source due to low conversion efficiency of the photovoltaic modules. To maximize the power derived from the PV systems it is important to operate the panel at its optimal power point by use of a maximum power point tracker (MPPT). MPPTs find and maintain operation at the maximum power point, using an MPPT algorithm. This paper presents a high performance tracking of maximum power delivered from photovoltaic systems using adaptive neural fuzzy inference systems (ANFIS). This method combines the learning abilities of artificial neural networks and the ability of fuzzy logic to handle imprecise data. It is therefore able to handle non linear and time varying problems hence making it suitable for this work. It is expected that this method will be able to accurately track the maximum power point. This will ensure efficient use of PV systems and therefore leading to reduced cost of electricity. The performance of the proposed method was compared to that of a fuzzy logic based MPPT to demonstrate its effectiveness over other previously used MPPT techniques.

Fuzzy logic based controller for micro-electro discharge machining servo systems

In this paper, a tunable fuzzy logic based servo controller for monitoring and control of the micro-electro discharge machining (micro-EDM) process has been developed, which uses the behavior of discharge pulses. The fuzzy logic applies a heuristic approach in dealing with the non-linear and time varying nature of the micro-EDM process. The control parameters affecting the performance of micro-EDM include gap voltage and gap current discharge pulses. It is important to discriminate between different levels of pulses for proper operation of micro-EDM. Discrimination of pulses from an RC-type power source is still an ill-defined problem relying on heuristics. The choice of appropriate values of discharge pulses is generally based on the knowledge and experiences of operators. In this study experiments were carried out to distinguish the discharge pulses, which were then classified into open, sparking, arcing, off and short circuit. The classified pulses were utilized as the input fuzzy sets of the fuzzy logic based controller that drives the servo system to maintain the desired gap width. The simulated results obtained demonstrate that the fuzzy logic controller is able to provide stable machining and improve the performance of the micro-EDM process.

Load flow analysis with a neuro-fuzzy model of an induction motor load

In the conventional load flow study, the active and reactive powers of all load buses are generally specified. Although the constant power model is applicable for approximate studies, it may not be suitable for the motors because the reactive power is very sensitive to the voltage variations while the active power is dependent upon the torque being driven. This paper proposes to solve the load flow equations using a neuro-fuzzy model of an induction motor. Both the active and reactive powers of the motor are estimated at each iteration using neuro fuzzy techniques. The efficiency is estimated using the IEEE 30 bus system. The results indicate that the inclusion of induction motor loads has an effect on the convergence characteristic of the active and reactive power mismatches of the modified load flow.

Novel Modeling of a Fast DC Breaker for a VSC-HVDC Transmission System Protection

The inability to quickly isolate the faulty sections of the direct current (DC) network damages the converters and the transmission system. Further delay in fault interruption can be escalated to the generation system thus shutting down the whole power system. Currently, Voltage Source Converter-High Voltage Direct Current (VSC-HVDC) systems provide the best mode of bulky power transmission over long distance. Though VSC-HVDC have their own internal protection, they still suffer from short-circuit faults especially on the DC side hence requiring a novel fast DC breaker. The novel design of the DC breaker in this paper utilizes a systematically calculated mutual inductance to divert the energy generated during fault to resistive elementson both sides of the DC breaker thus, preventing the transient voltage and backward current generated at the time of clearing the fault. The major challenge for DC breaker is that current is not alternating inferring that it has no natural zero current occurring point. Thus, the current has to be artificially forced to cross a current zero point so that the fault can be easily interrupted.This study aims at reviewing the currently available DC breakers with a fault clearing time of 5msby modelinga novel fast DC breaker therefore, providing the best possible protection against short-circuit faults on VSC systems. The results indicate that the proposed DC breaker model has a fault clearing time of 1ms. This is a major contribution towards the development a fast DC breaker.

Investigating the impact of power system stabilizers in a multi-machine system with an induction motor load

The high reactive power demand by the induction motor load during fault condition due to reduced bus voltages may cause a generator to behave like a voltage source behind the synchronous reactance and its terminal voltage reduces leading to the possibility of a voltage collapse scenario. For reliability of these systems, and in an attempt to reduce system oscillations, Power System Stabilizers (PSS) have used to add damping by controlling the excitation system. Studies on a SMIB and those using static loads have shown that a well-tuned PSS using a Fuzzy Logic Controller (FLPSS) can effectively improve power system dynamic stability. This paper investigates the impact of the FLPSS in maintaining voltage stability in a system with induction motor loads. A large induction motor is introduced as a load in a multi machine system, and the impact of the FLPSS are investigated by introducing a temporary three phase fault. For comparison the FLPSS is compared to other PSS found in literature. Results indicate that the FLPSS may lead the generator to lose its capability to maintain constant voltage and hence lead to the stalling of the induction motor load soon after the fault is cleared.

Optimization of Pulse Pattern for a Multi-robot Sonar System Using Genetic Algorithm

Sonar sensors are widely used in mobile robotics research for local environment perception and mapping. Mobile robot platforms equipped with multiple sonars have been build and used by many researchers. A significant problem with the use of multiple sonars is that, when the sonars are operated concurrently, signal interference occurs, making it difficult to determine which received signal is an echo of the signal transmitted by a given sensor. In this paper, a technique for acquiring suitable modulation pulses for the signals emitted in a multisonar system is presented. We propose a technique to reduce the probability of erroneous operation due to interference by satisfying conditions for minimizing the signal length and the variation in the signal length of each sonar, using the Niched Pareto genetic algorithm (NPGA). The basic technique is illustrated for the case where two or more robots operate in the same environment.

Acquisition of modulation pulses for a multi-robot system using genetic algorithm

Ultrasonic sensors (sonars) are widely used in mobile robotics research for local environment perception and mapping. Mobile robot platforms equipped with multiple sonars have been build and used by many researchers. A significant problem with the use of multiple sonars is that, when the sonars are operated concurrently, signal interference occurs, making it difficult to determine which received signal is an echo of the signal transmitted by a given sonar. One way of solving the problem of signal interference is to use different pulse patterns to modulate the signals emitted by the sonars. Furthermore, in order to reduce the time taken to perform measurements, a concurrently operated sonar system is desirable. In this paper, a technique for acquiring suitable modulation pulses for signals emitted in a multi-sonar system is presented. In order to optimize the length of modulation signal, we propose a technique to reduce the probability of erroneous operation due to interference, using the Niched Pareto genetic algorithm (NPGA). The basic technique is illustrated for the case where two or more robots equipped with multiple sonars operate in the same environment.