Gas-liquid flows abound in a great variety of industrial processes. Correct recognition of the regimes of a gasliquid flow is one of the most formidable challenges in multiphase flow measurement. Here we put forward a novel approach to the classification of gas-liquid flow patterns. In this method a flow-pattern map is constructed based on the average energy of intrinsic mode function and the volumetric void fraction of gas-liquid mixture. The intrinsic mode function is extracted from the pressure fluctuation across a bluff body using the empirical mode decomposition technique. Experiments adopting air and water as the working fluids are conducted in the bubble, plug, slug, and annular flow patterns at ambient temperature and atmospheric pressure. Verification tests indicate that the identification rate of the flow-pattern map developed exceeds 90%. This approach is appropriate for the gas-liquid flow pattern identification in practical applications.
The unbalance of the neutral point voltage is an inherent problem of three-level neutral-point-clamped (NPC) inverter, the effect of neutral point voltage balancing which is caused by voltage vector is analyzed, and the relationship of the voltage offset and neutral point voltage is studied in this paper. This paper proposes a novel neutral point balance strategy for three-level NPC inverter based on space vector pulse width modulation (SVPWM). A voltage offset is added to the modulation wave, and a closed-loop neutral point voltage balance control system is designed. In the control system, the dwelling time of synthesis voltage vectors for SVPWM is varied to solve the problem of the unbalance of the neutral point voltage, the sequence of the voltage vectors maintains unchanging. Simulation and experimental results show the neutral point voltage balancing control strategy based on SVPWM is effective.
Fault detection and location are important and front-end tasks in assuring the reliability of power electronic circuits. In essence, both tasks can be considered as the classification problem. This paper presents a fast fault classification method for power electronic circuits by using the support vector machine (SVM) as a classifier and the wavelet transform as a feature extraction technique. Using one-against-rest SVM and one-against-one SVM are two general approaches to fault classification in power electronic circuits. However, these methods have a high computational complexity, therefore in this design we employ a directed acyclic graph (DAG) SVM to implement the fault classification. The DAG SVM is close to the one-against-one SVM regarding its classification performance, but it is much faster. Moreover, in the presented approach, the DAG SVM is improved by introducing the method of Knearest neighbours to reduce some computations, so that the classification time can be further reduced. A rectifier and an inverter are demonstrated to prove effectiveness of the presented design.
The drainage consolidation method has been efficiently used to deal with soft ground improvement. Nowadays, it has been suggested to use a new sand soil which is a composite of sand and recycled glass waste. The permeability performance of glass-sand soil was explored to judge the feasibility of glass-sand soil backfilled in the drainage consolidation of sand-drained ground. For comparison purposes, different mix proportions of recycled glass waste, fineness modulus, and glass particle size were analyzed to certify the impact on the permeability coefficient and the degree of consolidation. The numerical results show that adding a proper amount of recycled glass waste could promote the permeability performance of glass-sand soil, and the glasssand soil drain could be consolidated more quickly than a sand drain. Experiments showed that glass-sand soil with the a 20% mix of recycled glass waste reveals the optimum performance of permeability.