Erping Zhou

Maximum Power Point Tracking technologies are being used in traditional PV system charge controllers to enhance the power conversion efficiency. An MPPT controller will ensure power extracted from the PV panels during varying climatic conditions is always maximum. This will ensure that maximum power is flowing between the panel and load. As both Temperature and Irradiation levels vary during the day, maximum power point trackers are an inevitable component in a PV system. As solar energy holds a major share in renewable energy in the world market, an improvement in MPPT technique makes the efficiency of the PV system increase and in turn cost reduction possible. However, the efficiency of conventional MPPT Techniques suffers from failing in tracking MPPT at fast varying climatic conditions and falling in local maxima of maximum power point than global maxima. The issues of stability, tracking speed, and accuracy can be solved using intelligent MPPT techniques methods based on soft computing tools: Therefore, this paper aims to study and provide a comparative analysis of two AI-based MPPT techniques such as ANN and ANFIS. The MPPT techniques considered in this study are ANN and ANFIS. Performance evaluation is carried out using MATLAB Simulation. Experimental results indicate that the two methods ANN and ANFIS are more efficient than conventional MPPT techniques due to its capability to avoid local MPP and partially shaded conditions.
Maximum Power Point Tracking technologies are being used in traditional PV system charge controllers to enhance the power conversion efficiency. An MPPT controller will ensure power extracted from the PV panels during varying climatic conditions is always maximum. This will ensure that maximum power is flowing between the panel and load. As both Temperature and Irradiation levels vary during the day, maximum power point trackers are an inevitable component in a PV system. The renewable energy source has a vital role in supplying sustainable power to meet the rising electricity demands. However, the PV system performance heavily depends on environmental conditions. This, in turn, causes efficiency to be less and in turn higher cost. For maximum power to be transferred from the PV panels under varying climatic conditions PV systems should operate at Maximum Power Point.

This paper aims to use machine learning methods to predict the service life of rotary lip seals to aid manufacturers and users improving the current maintenance procedures. Seals are widely used in most engineering applications. The knowledge of condition of seals throughout their working life is important due to the fact that they are often used on high value engineering products. As the current material properties of the seal and the working environment various, it is difficult to predict useful life of the rotary lip seal. In this paper, the factors relating to life of rotary lip seals are investigated and discussed. The application of machine learning methods using actual testing data in order to estimate the useful life of the seals has been presented. The early results show good agreement between actual and predicted values.

This paper investigates design and implementation of Maximum Power Point Tracking (MPPT) algorithm for Photo Voltaic (PV) applications on Field-Programmable Gate Array (FPGA) platform. The algorithm is developed by means of Hardware Description Language (Verilog). Algorithms were simulated on MODELSIM software and synthesized using ALTERA Quartus II software. Furthermore, hardware implementation is done using ALTERA Cyclone II FPGA starter board to verify the performance of the designed algorithms. The paper will investigate whether FPGA is a suitable hardware platform for real time MPPT controller.

This paper focuses on the mathematical modelling and simulation of Maximum Power Point algorithms to investigate tracking efficiency at different atmospheric conditions. This paper will review existing Maximum Power Point Tracking approaches. A 60W PV panel is modelled in MATLAB since panel current is taken as the input for maximum power point tracking. This paper presents a simulation based comparative study between two most popular techniques perturb and observe (P&O) and incremental conductance (InCond) to optimize the energy conversion efficiency of PV system.

Direct flux control (DFC) is one of the sensorless methods, which is used to acquire the electrical rotor position of permanent magnet synchronous machines (PMSMs) for all speeds. The DFC is implemented on a TriCore PXROS platform, which is connected to the real PMSM. Flux linkage signals are measured to calculate the electrical rotor position by using an approximation function. Due to the weak intensity of the measured flux linkage signals of the buried magnet PMSM, a fuzzy logic is selected as a tool to overcome the mentioned problem. The fuzzy logic is designed based on the flux intensity characteristics, which are recognized by driving the motors. Thus, a flux intensity management (FIM) strategy by fuzzy logic is proposed in this paper to solve the obstacle. Consequently, the buried magnet PMSM electrical rotor positions can be acquired for all speeds by employing the DFC and the FIM methods.

A Direct Flux Control (DFC) method is implemented in this paper in order to acquire the rotor position of a permanent magnet synchronous machine (PMSM), which can be obtained for all speeds and also at standstill. Flux linkage signals are utilized to calculate the rotor position based on trigonometric approximations. Electrical rotor positions are results of this method, which are validated by both software and hardware executions. The measured flux linkage signals are taken into account to analyze their behaviors, especially in frequency domain, which provide a new approximation function of the flux linkage signals. The estimated flux linkage signal function is analyzed and shows that a magnetization curve (BH curve) is one of the most important factors to reach the actual performances of the PMSM. Thus, the exact nonlinear BH curve is necessary to be considered, which has been achieved in the mutual software simulation with Simplorer and Maxwell. A TriCore PXROS platform is connected to the real PMSM to employ the DFC method as the hardware implementation on a real time system.

Eddy current techniques are favoured for measuring the properties of conductive coatings on magnetic substrates because the skin effect ensures that only the coating or region at or just below the surface of a magnetic substrate is inspected. The data acquired from eddy current sensors, however, is affected by a large number of variables, which include sample conductivity, permeability, geometry, and temperature, as well as sensor lift-off. The multivariable properties of sample coatings add an even greater level of complexity. This research project is therefore motivated by the need for a measuring instrument, which can intelligently adapt to the large number of variables that affect eddy current measurements on steel. Sensor model optimisation against experimentally generated data is undertaken, leading to the development of accurate and fast inversion models based on artificial neural networks. Neural network architecture, operation and training are discussed, which includes an algorithm for neural network normalisation and calibration. System tests are fmally undertaken on a wide range of plated samples. This research demonstrates that an intelligent measuring system incorporating a ferrite-cored sensor can provide high accuracy while operating over a wide frequency range.