Search

Finite-control-set model predictive control (FCS-MPC) of multiphase (n-phase, n is assumed to be an odd number for simplicity) drives is challenging because of the large number of actual/virtual voltage vectors and the need for current control in (n-1)/2 sub-spaces (or planes; multi-plane current control). Any sub-optimal design (poor or no current control in some of the (n-1)/2 planes) may result in high individual plane current ripples, due to the low reactance. This work therefore investigates continuouscontrol-set (CCS) MPC for constant switching frequency multiphase motor drives as another alternative. The highbandwidth CCS-MPC is designed to accurately account for system non-idealities, namely digital control and pulse width modulation delays, inverter dead time, and measurement noise. It will be shown that the CCS-MPC has the advantages of full voltage vector space access, regular switching characteristic, and improved cycle-by-cycle tracking control, while maintaining some of the known advantages of the FCS-MPC, e.g., intuitive cost function design, model-based control, and fast dynamics. The proposed control scheme is benchmarked experimentally against the classical, proportional-integral-based, fieldoriented control in conjunction with an asymmetrical sixphase induction motor drive.

Recently, inverters with a common ac and dc ground are gaining significant interest due to their zero common-mode voltage that made them particularly attractive for the solar photovoltaic application. However, the dc source current of the existing topologies is discontinuous, and their voltage gains are limited. This article proposes a novel common-ground-type boost inverter that achieves continuous dc source current with significantly enhanced dynamic voltage gain. It consists of one boost inductor, four capacitors, nine power MOSFETs, and two power diodes. Voltage boosting and generation of nine-level ac voltage are achieved concurrently within a single-stage operation. The operation of the proposed CGT-9L-BI inverter is analyzed. Simulation and experimental results are presented for validation.

The emerging switched-capacitor-based multilevel inverters offer interesting merits such as self-balancing of capacitor voltages and boosting of voltage gain. While the switched capacitors (SCs) in these topologies are charged in parallel with the dc source, severe current spikes issue is inevitable, rendering them impractical at high power. This article proposes a novel switched-capacitor-based T-type inverter that mitigates the current spikes by enabling soft charging for its integrated SCs, where both SC in the topological structure charges through a dedicated circuit comprises of an inductor and two switches. The proposed topology is capable of five-level ac voltage generation and when compared to a classical T-type/ANPC (active neutral-point-clamped) inverter, it achieves higher dc-link voltage utilization since its maximum attainable voltage gain is doubled. Theoretical findings of the proposed topology are validated by both the simulation and experimental results.

Multilevel inverters with a common-ground structure and voltage step-up feature are beneficial for transformerless grid-connected photovoltaic applications. In this article, a novel topology of such converters with a five-level output voltage generation is presented. The proposed topology comprises of an integrated switched-boost (SB) module and a dual T-type (D2T) cell constructed with 10 power switches (six unidirectional and two bidirectional). An inductor in the integrated SB module is used to boost the voltage across two involved capacitors of the D2T cell. Therefore, a desirable ac voltage magnitude for the grid-connected application can be achieved over a wide range of input voltage changes. Current stress is also kept within a permissible input current range by the soft-charging operation of the involved capacitors through the inductor. A corresponding dead-beat continuous-current-control-set controller with a sinusoidal PWM modulator is implemented for controlling the real and reactive powers. The controller has an added benefit of achieving constant switching frequency for power switches. Circuit description, theoretical analyses, and comparative study are discussed. Simulation and experimental results are also presented to confirm the feasibility and correctness of the proposed topology.

Recent research on common-ground switched-capacitor transformerless (CGSC-TL) inverters shows some intriguing features, such as integrated voltage boosting ability, possible multilevel output voltage generation, and nullification of the leakage current issue. However, the number of output voltage levels and also the overall voltage boosting ratio of most of the existing CGSC-TL inverters are limited to five and two, respectively. This article presents a generalized circuit configuration of such converters capable of higher voltage gain and output voltage levels generation. A basic five-level (5L) CGSC-TL inverter is first proposed using eight power switches and two self-balanced dc-link capacitors. A generalized extension of the circuit for any output voltage levels and voltage gain is then presented while keeping all the traits of the proposed basic 5L-CGSC-TL inverter. The circuit descriptions, control strategy, design guidelines, comparative study, and the relevant simulation and experimental results for the proposed 5L-CGSC-TL inverters and its seven-level derived topology are presented to validate the effectiveness and feasibility of this proposal.

Switched-capacitor-based multilevel inverters have been gaining increasing attention in recent years in view of their voltage-boosting capability and capacitor self-balancing properties. However, the existing topologies, which comprise frontend switched-capacitor cells and a backend H-bridge, result in longer discharging duration in some of their switched-capacitors. In addition, the adoption of backend H-bridge introduces common-mode voltage and necessitates multiple dc sources for voltage level extension. To resolve these problems, this article proposes a novel topology comprises a T-type inverter and n cascaded switched-capacitor cells. It requires only a single dc source and it manages to mitigate the common-mode voltage by connecting the neutral of ac output to the mid-point of dc link. All switched-capacitors in the proposed topology are ensured to be charged for at least half of the fundamental period to alleviate the capacitor voltage ripples and current spikes. The operation of the proposed topology is analyzed and experimental results are presented for validation.

This article proposed and investigated a near-optimal finite-control-set model-predictive grid current control (NOP-MPGCC) for voltage source inverters (VSIs) with actively damped LCL filters. It carries three advantages: (i) the constant-frequency pulsewidth modulator produces regular switching spectrum—which can ease the LCL filter design process in grid current control applications with medium-to-high carrier ratio; (ii) it avoids the use of the usual first-order assumption in most PI-/PR-based control and the common simplifications adopted in some existing FCS-MPC schemes. This is achieved by considering the third-order behavior using long prediction horizon; and (iii) it can operate near and across the critical frequency, defined using the established classical control definition. Moreover, it retains the intuitive enumeration structure and maximizes the dc-bus utilization through the consideration of entire hexagonal vector space at every control cycle. The theoretical derivation, simulation, and experiment results verify that NOP-MPGCC is potentially a viable direct grid current control scheme for PWM-VSIs with actively damped LCL filters.

The emerging active-neutral-point-clamped (ANPC) inverters with voltage-boosting capability are attractive for their low dc-link voltage requirement. These low voltage requirements enable a single-stage dc-ac power conversion, which improves the overall efficiency, reliability, and power density of the system. A high voltage gain of 1.5 was demonstrated in recent boost type ANPC topology; however, it was achieved at the expense of high voltage stress on some of its switching devices. This brief proposes an improved topology with reduced voltage stress and a lower number of components while retaining the merits of high voltage gain. The proposed topology is a hybrid of a T-type inverter and an H-bridge, which require only one floating capacitor and one less power switch than the aforementioned topology. One floating capacitor with self-voltage balancing capability is integrated to generate 7 output voltage levels. The proposed topology is analyzed and compared with recent boost ANPC topologies. Experimental results are presented for validation.

This paper presents a common-ground five-level transformerless inverter for solar photovoltaic (PV) applications. In the proposed inverter, the grid neutral line is directly connected to the negative pole of the dc bus eliminating the common-mode leakage current which is a critical problem in grid-connected PV systems. Moreover, the proposed inverter has only four power switches, making the first five-level common-ground topology with less than six switches found in the recent literature. In addition to low switch count, only two switches are operating in high-frequency. The operation modes, the design guidelines as well as a modulation strategy are presented. Simulation and experimental results are obtained to demonstrate the performance of the proposed inverter.

The circuit performance of conventional active neutral point-clamped (ANPC) inverter is widely accepted in many renewable energy-based applications like photovoltaic (PV) or electric vehicle grid-connected systems. This is mainly because of its excellent characteristics in terms of voltage/current stress profile of the switches, bidirectional power flow capability, and efficient operation. Nonetheless, due to its half-dc link voltage utilization in the ac output voltage, another power processing stage with additional active and passive elements is required to make its output voltage compatible with the grid when low and wide varying input dc source is available. In this paper, a novel ANPC-based five-level (ANPC5L) inverter with a single-stage boost-integrated circuit design is presented. The proposed topology is able to make the peak output voltage of the conventional ANPC5L inverter followed by a front-end bidirectional boost converter double using the same number of power switches but with less total standing voltage across semiconductors. The working principles of the proposed topology is discussed. Experimental results obtained from 1.3 kW laboratory-built prototype under the grid-connected condition are also given to support the discussion.

The switched-capacitor-based cascaded multilevel inverters (CMI) have been emerging due to their voltage boosting capability. Unfortunately, they suffer from impulse charging current and nonuniform operation. This article presents a topology termed as dual-T-type five-level CMI to resolve these problems without compromising the desirable voltage boosting characteristic. The main idea is to integrate a half bridge and an inductor to soft charge a capacitor that is connected in series with the dc source. The capacitor enables the voltage gain boosted to two, while the control of eight power switches that constitutes a dual-T structure enables five voltage levels generation. In addition, uniform operation is achieved for cascaded extensions. The operating principle of the proposed topology is analyzed and elaborated. For validation, simulation, and experimental results of a prototype are presented.

Conventional active-neutral-point-clamped (ANPC) inverters exhibit low voltage gain that inherently leads to a high dc-link voltage requirement. An improved ANPC inverter that is capable of generating five voltage levels has recently reduced the dc-link voltage twofold to achieve unity gain. This led to the development of a single-stage dc–ac power converter with no frontend boost dc–dc converter. This article proposes novel ANPC inverters capable of unity or boosted voltage gain while generating higher voltage levels. The first topology can provide a voltage gain of 1.5 and extends the number of levels to seven by incorporating only one additional switch. The topology can also be extended by adding three switches and one floating capacitor to generate nine levels with unity voltage gain, or 11 levels with a voltage gain of 2.5. The proposed ANPC inverters and their operations are comprehensively discussed. Experimental results are provided to validate the feasibility of the proposed ANPC inverters.

Direct torque control (DTC) suffers from flux droop owing to the long zero-voltage vectors at low motor speeds. Previous studies were able to attain flux regulation by imposing the continuous switching of forward and reverse active-voltage vectors by causing torque overshoots in the torque hysteresis bands (THBs). This led to an excessive increase in the switching frequency and larger torque and current ripples, thus reducing the drive efficiency. In this paper, a modified flux regulation method is proposed for the classical DTC using a single THB when flux droop occurs in a low-rotor-speed range. The proposed method can protect the DTC from flux droops at low speeds by reducing the duration of zero-voltage vectors and minimizing the number of reverse-voltage vectors. In addition, the proposed strategy is efficient in reducing torque and current ripples while operating at low motor speeds. Furthermore, a significant reduction in the switching frequency is obtained. The effectiveness of the proposed strategy is confirmed by simulation and experimental results.

Recently multilevel inverters (MLIs) have received wide attention from industry and academia, as they are changing into a viable technology for diverse applications. To produce high-quality output using less switch count, development of novel reduced switch MLI (RS MLI) topologies has been a focus of current research theme. This paper presents design and control of a switched-diode dual source single switch MLI (SDDS MLI). The generalized SDDS MLI is first designed using an asymmetric basic unit. Proposed SDDS MLI requires less switch count and driver circuit count compared with the few recently developed RS MLI topologies. To improve the voltage quality by eliminating targeted low-order harmonics, a modified version of fish swarm optimization algorithm is examined for computing optimum switching angles required to control the SDDS MLI. Moreover, suitability and superiority of the derived algorithm are established by comparing with traditional selective harmonic elimination techniques. The developed topology is investigated through several MATLAB simulations as well as experimental tests in the laboratory applying the modified control approach.

This paper proposes a new family of multilevel inverter topology that is able to generate seven voltage levels by utilizing one or two floating capacitors and 10 power switches. This novel boost switched-capacitor seven-level inverter possesses voltage boosting capability with an achievable maximum voltage level 1.5 times the input direct current (dc) voltage. The generation of higher output voltage does not incur high-voltage stress on any power switch in this topology, as the peak inverse voltages of all power switches do not exceed the input source voltage. In addition, capacitor voltage balancing is not essential since the floating capacitors are effectively balanced during the charging and discharging processes. Furthermore, the proposed topology eliminates the need for multiple isolated dc sources, and a single dc source is sufficient in both its single-phase and three-phase topologies. The operating principle and steady-state analysis of the proposed topology are elaborated. Experimental results from a single-phase prototype are then presented to verify the validity of the proposed topology.

Many experiments have been carried out on the (16.8 kJ, 17 kV) dense plasma focus PF-24 device operated with pure deuterium and deuterium-argon admixture. The main purpose of the work was to investigate the influence of the doping argon on the average total neutron yields experimentally and numerically. The 103 discharges (34 for D2, 69 for D2- $x$ %Ar, where $x = 3, 5, 10, 15, 22, 25, 30, 45, 60$ ) were performed in four experimental sessions under constant total initial optimum pressure of about 2.2 torr. The total current and its time derivative traces, as well as the total neutron yield values, were recorded, and the electro-kinetic plasma parameters were estimated for each shot. The Lee model has also been adapted for gas mixture plasma focus modeling, and then the enhanced code was used for simulation of each of the PF-24 discharge with D2 and D2- $x$ %Ar. The fitting parameters for each shot are found and discussed. The measured and computed values of the neutron yield and the electro-kinetic plasma parameters are in good agreement. The obtained results showed that there is no neutron yield enhancement with argon seeding. The scaling law of the neutron yield $Y_{n}$ against argon fraction has been deduced. Additionally, our results illustrate that the PF-24 plasma focus device has an optimized configuration for the DD fusion neutron emission.

The conventional three-level active-neutral-point-clamped (ANPC) inverter requires high dc-link voltage at least twice the peak of ac output. To reduce the dc-link voltage, a recent topology that enhanced the voltage gain from half to unity has been presented. In this letter, an alternative ANPC topology is established by using two T-type inverters. Two floating capacitors with self-voltage balancing capability are integrated to achieve a voltage-boosting gain of 1.5. In addition, the proposed topology is capable of generating seven voltage levels. Its operation is validated through circuit analysis followed by experimental results of a prototype.

This program was designed to introduce a plasma platform to examine the field programmable gate array (FPGA) of the link boards typically used at the CERN’s Large Hadron Collider (LHC). Pulsed plasma systems with accelerating gradient of 1 kV/ $\mu \text{m}$ generate high-intensity, high-energy radiation beams. Single-event upset (SEU) is caused by radiation deposition in the FPGA. In FPGA, the SEU probability for 1-MeV protons and 10-keV X-rays are 0.1 and $2 \times 10^{-9}$ particle $^{-1}$ . The number of SEU induced in the Si by 1-MeV proton irradiation at 0.8-V bias computed from simulation in COMSOL Multiphysics was $1.8 \times 10^{5}$ . Although more experimental research is needed to identify the underlying mechanisms, pulsed plasma is perceived as being a smart alternative to investigate the error structure in FPGA.

This paper shows the correlation of the characteristic Ne soft X-ray (SXR) signal pulse with computed dynamics of the plasma focus using doped neon in Ne (97.5%)/Kr (2.5%) admixtures. The Lee model code is coupled with a specifically designed correlation excel template. The results show that typically, the characteristic Ne SXR pulse signal starts after the start of the reflected shock (RS) and before the beginning of the pinch. Specifically, in the illustrated shot, the RS phase starts at $3.375~\mu \text{s}$ while the pinch phase starts at $3.384~\mu \text{s}$ . The characteristic Ne SXR signal pulse starts at $3.377~\mu \text{s}$ , which is 2 ns after the start of the RS and 7 ns before the pinch starts. There is a consistent trend of Ne SXR correlation with the computed dynamics in the Ne doped with Kr admixtures up to a doping level of 2.5%.

The conventional active neutral-point-clamped (ANPC) inverter inherits excellent high-frequency common mode voltage (CMV) mitigation ability but with a restricted voltage gain. This letter proposes a novel five-level ANPC inverter that is able to accomplish voltage-boosting within a single-stage dc–ac power conversion. A common ground in the proposed topology not only mitigates the high-frequency CMV but also enhances the dc-link voltage utilization. The proposed topology exhibits lower voltage stresses and higher compactness when compared with a conventional two-stage five-level ANPC inverter. The proposed inverter saves three power switches and one capacitor while enhancing the overall efficiency. Simulations and experimental tests have validated the overall operation, efficacy, and practicality of the proposed ANPC inverter.

The plasma focus (PF) emerged as a potential prospect to produce radionuclide, but the experimental data are not significant in favor of practical medical radioactivity. Empirical scaling grants a set of magnetic lens to enhance the nuclear activity. The experimental radioactivity of low-energy devices is in the range of kilobecquerel and rises to megabecquerel by optimizing and operating at a few hertz. Magnetically driven ion beam improves the radioactivity to the prescribed value of gigabecquerel for nuclear imaging. Magnetic radioactivity competence scheme reflects the vision that small PF, as a smart alternative to cyclotron, generates medical grade radioactivity.

Cascaded H-bridge (CHB) multilevel inverters (MLIs) have been widely used for power electronics systems. While high-voltage blocking across power switches is not a constraint for low voltage applications, the research trend has been oriented to the design of more compact module topologies as an alternative for CHB. Despite the generation of more voltage levels with reduced switch count, the existing module topologies in recent literature take no account of the freewheeling current path during dead-time, thus, inducing multistep jumps in voltage levels and giving rise to undesirable voltage spikes. Addressing this concern, this paper proposes two symmetrical compact-module topologies for cascaded MLI, where freewheeling current path during dead-time is provided for smooth transition between voltage levels to prevent voltage spikes. The proposed 7-level and 13-level compact-modules demonstrated low number of conducting switches for all voltage levels. Comprehensive analysis and comparison with the latest module topologies are conducted. To validate the operation of the proposed compact-module topologies, simulation and experimental results are presented.

This letter proposes an improved symmetrical 4-level submodule as a basic cell for generating multiple dc voltage levels. A hybrid cascaded multilevel inverter (HCMLI) topology is formed by the combination of n submodules and a full-bridge. A comparative analysis against the recent multilevel inverters reveals that the proposed topology requires less number of switches and dc sources. In addition, the proposed submodule reduces the number of conducting switch and gate driver requirements compared to the widely used half-bridge submodule. To validate the operation of the proposed HCMLI topology, experimental results of a 9-level single-phase inverter controlled by selective harmonic elimination pulse-width-modulation is presented.

The cascaded H-bridge multilevel inverter (MLI) requires separate isolated dc sources to generate more than three voltage levels and to generate higher output voltage. This paper proposes a new MLI topology that requires only one dc source and is capable of generating seven voltage levels with triple voltage boosting gain. Three H-bridges are interconnected through two bidirectional voltage blocking switches to enable the integration of two switched-capacitors. Unlike the existing two-stage structure switched-capacitor-based MLI, the proposed MLI is a single-stage topology. It alleviates the voltage stress across switches such that low voltage stress of not more than the dc source voltage is ensured on all switches. In addition, capacitors voltage balancing is achieved automatically during operation. The operation of the proposed MLI is analyzed followed by verification through simulation and experimental test of a low power/voltage prototype.

The measured current waveforms of PF1000 plasma focus operating in neon are first fitted to the computed currents using the Lee model for the pressures of 0.5 and 0.8 torr. The fitting fixes the model parameters. The code is then run to study and to optimize the soft X-ray yield from (~352 kJ) PF1000 device. The maximum characteristic soft X-ray (H-like and He-like lines) yield of 4.2 kJ is found to occur at the pressure of 0.85 torr, with the pinch duration of 207 ns and with an all-line yield of 4.8 kJ. Maximum compression (corresponding to smallest pinch radius) of 0.585 cm with the duration of 224 ns is obtained at 1 torr with the greatest all-line yield of 7.1 kJ but a lower characteristic soft X-ray yield of 3.2 kJ. Detailed computation results indicate that the maximum compression (minimum pinch radius) at 1 torr is attributed to two mechanisms: thermodynamics specific heat ratio effects and radiative losses.

The Lee model is extended to study and characterize the electron beams emitted from plasma-focus devices. It is then first applied to characterize the electron beams emitted from low and high-energy plasma focus operated with deuterium gas. The numerical experiments on NX2 device at 15 torr of deuterium give the following results: electron fluence = $5.7 \times 10^{{\mathbf {22}}}~\text{m}^{{\mathbf {-2}}}$ , electron flux = $16 \times 10^{{\mathbf {29}}}~\text{m}^{{\mathbf {-2}}}~\text{s}^{{\mathbf {-1}}}$ , the relativistic kinetic mean energy of the electron = 56 keV, electron number = $2 \times 10^{{\mathbf {16}}}$ , electron current = 91 kA, and damage factor = $2.7 \times 10^{{\mathbf {12}}}~\text {W}\cdot \text {m}^{{\mathbf {-2}}}\text {s}^{{\mathbf {0.5}}}$ . Then, the effect of pressure on the beam characteristics is studied. The energy of the beam at pinch exit changes from around 57 J (1.9% of the stored energy $E_{{\mathbf {0}}}$ ) to 176 J (6% $E_{{\mathbf {0}}}$ ) with maximum value of 180 J (6.1% $E_{{\mathbf {0}}}$ ) at 15 torr, and these results are compared with the measured values 3.2% $E_{{\mathbf {0}}}$ on NX2. Scaling trends are suggested for electron beam characteristics. The energy fluence and the power flow density (energy flux) have a variation $4.3- 265 \times 10^{{\mathbf {7}}}~\text {J}\cdot \text {m}^{{\mathbf {-2}}}$ and $2.2 - 19 \times 10^{{\mathbf {15}}}~\text {W}\cdot \text {m}^{{\mathbf {-2}}}$ , respectively. The electron beam current ranges from 12 to 700 kA being 9%–35% of $I_{{\mathbf {peak}}}$ . The energy beam $Y_{{\mathbf {EB}}}$ scales on average as $Y_{\mathrm {EB}} =6.55\times E_{0}^{1.45} $ at energies in the 1 to 500 kJ regions ( $Y_{{\mathbf {EB}}}$ in J and $E_{{\mathbf {0}}}$ in kJ), and $Y_{\mathrm {EB}} =1\times 10^{-6}I_{\mathrm {pinch}}^{3.53} $ and $Y_{\mathrm {EB}} =2\times 10^{-6}I_{\mathrm {peak}}^{3.14} $ ( $I_{{\mathbf {peak}}}$ and $I_{{\mathbf {pinch}}}$ in kA and $Y_{{\mathbf {EB}}}$ in J). These results provide much needed benchmark reference values and scaling trends for electron beams of a plasma focus operated in deuterium gas.

Measured current waveforms at different neon pressures from two different plasma focus (PF) devices, namely, NX2 and NX1 are used for analysis using the Lee code. The model parameters of mass ( $f_{m}$ and $f_{{\text {mr}}}$ ) and current ( $f_{\text {c}}$ and $f_{\mathrm {cr}}$ ) in axial and radial phase, respectively, are obtained by fitting the computed current waveform to the measured current waveform for each pressure over a range of pressures in neon. The results show that the value of $f_{m}$ and $f_{\mathrm {mr}}$ ranges from 0.07 to 0.12 and 0.09 to 0.20, respectively, over the pressure range 1.0–5.0 Torr for NX2, whereas for NX1 it ranges from 0.020 to 0.024 and 0.048 to 0.08, respectively, over the pressure range 7.5–12 Torr. In all cases, the values of $f_{c}$ and $f_{\mathrm {cr}}$ are fixed at 0.7. Comparison of computed and measured soft X-ray (SXR) yield using fit model parameters for both devices shows good agreement. An important finding is that despite the noticeable variation of mass factors over the whole pressure range, a representative set of model parameters is found for each machine which may be taken as fixed over the range of pressures producing a good yield. In particular, for a high-performance PF using conventional electrode structure the representative set is: $f_{m} = 0.075$ , $f_{c} \,\, = \,\, 0.7$ , $f_{\mathrm {mr}} =0.1$ , and $f_{\mathrm {cr}}= \,\, 0.7$ . This is valuable information for neon SXR design purposes.

A single-stage topology of a three-phase boost inverter known as split-source inverter (SSI) has recently been introduced in the literature. This topology suffers from high frequency current commutations across two diodes and complicated analysis since the inductor is charged with variable duty cycle. This paper presents a single-phase version of SSI with improvements in inverter topology as well as the pulse width modulation (PWM) technique. An inductor is connected to two MOSFETs operating at fundamental frequency to boost the voltage from input source to dc-link voltage. In the proposed hybrid quasi-sinusoidal and constant PWM, one of the full-bridge legs undergoes constant duty cycle switching while the other one undergoes sinusoidally varying duty cycle switching, with the former is accountable for charging and discharging of inductor while the latter is accountable for producing ac output. Therefore, the proposed topology with hybrid quasi-sinusoidal and constant PWM exhibits the merit of simplicity since the control of dc-link voltage and ac output is detached within the single-stage topology. It is not liable to the undesired high frequency current commutation. In addition, a wide range of ac output voltage is achievable in either buck or boost operation. Theoretical analysis is presented and verified through simulation and experimental results.

Object triangulation, 3-D object tracking, feature correspondence, and camera calibration are key problems for estimation from camera networks. This paper addresses these problems within a unified Bayesian framework for joint multi-object tracking and sensor registration. Given that using standard filtering approaches for state estimation from cameras is problematic, an alternative parametrisation is exploited, called disparity space. The disparity space-based approach for triangulation and object tracking is shown to be more effective than non-linear versions of the Kalman filter and particle filtering for non-rectified cameras. The approach for feature correspondence is based on the Probability Hypothesis Density (PHD) filter, and hence inherits the ability to update without explicit measurement association, to initiate new targets, and to discriminate between target and clutter. The PHD filtering approach then forms the basis of a camera calibration method from static or moving objects. Results are shown on simulated data.

Reduced Pease–Braginskii currents are estimated for a linear pinch in a range of gases, namely, D, He, Ne, Ar, Kr, and Xe. A characteristic depletion time is defined as the time it takes for the plasma focus (PF) pinch energy to be radiated away. This quantity is used as an indicator for expectation of radiative collapse. The depletion times in various gases are estimated in units of pinch duration. The values indicate that in D and He, the radiation powers are small, resulting in such long depletion times that no radiative collapse may be expected in the lifetime of the focus pinch. In Ne, low tens of percent are radiated and significant cooling and reduction in radius ratio may be anticipated. In Ar, Kr, and Xe, the depletion time is only a fraction of the estimated pinch duration, so radiative collapse may be expected. Numerical experiments are then carried out with a circuit-coupled code, which incorporates radiation-coupled dynamics with PF pinch elongation and plasma self-absorption. The latter eventually limits the radiated power and stops the radiative collapse. These results show the detailed dynamics and confirm the expectations arising from depletion times discussed above.

In plasma focus (PF), the thermodynamic parameters such as ion fraction $\alpha $ , effective ionic charge number $Z_{\rm eff}$ , and effective specific heat ratio $\gamma $ at different temperatures may be calculated by corona model (CM). In the Lee model code, the neon $Z_{\rm eff}$ and $\gamma $ are stored in subroutines using convenient tables and polynomials derived from the CM (we call this approach approximated CM). In this paper, the thermodynamic parameters of the CM are close fitted to the data, thus replacing the approximate CM data with a more accurate close-fitting CM (CFCM). The comparisons of the Lee model code using the approximated CM and CFCM subroutines are conducted, with the main emphasis on optimum neon soft X-ray (SXR) emission and their properties. The suitable focus pinch temperature window of 200–500 eV is applied to generate the optimum neon SXR yield ( $Y_{\rm sxr}$ ). The optimum neon $Y_{\rm sxr}$ is found to be 3.19 and 3.49 J at the optimum pressure $P_{0}=3.1$ torr with approximated CM and CFCM subroutines, respectively. A high optimum value of SXR yield is obtained using CFCM subroutines in the Lee model, which is nearer to the experimental value compared with the approximated CM subroutines. The use of CFCM in the Lee model contributes to better estimation for further numerical experiment studies and gives confidence that the model is sufficiently realistic in describing the PF dynamics and SXR emission.

Ion beams emitted from the low-inductance kJ NX2 plasma focus have been numerically characterized and trends of these beams in various gases have been reported. In this paper, the ion beams from high-inductance kJ plasma focus machines AECS PF2 and INTI operated with various gases are studied. The obtained results show that the beam ions mean energy depends on the gas type and increases with increasing ion mass of the gas. It also indicates a higher ion fluence for H 2 (4.4 × 10 20 ions m −2 ), and a very small fluence value for Xe (0.085 × 10 20 ions m −2 ). The ion number reduces from H 2 (24 × 10 14 ) to Xe (0.074 × 10 14 ). The ion current decreases from H 2 (20.5% of the discharge current) to 5.4% for Xe. The beam energy drops from H 2 (0.63% of stored energy E 0 ) to 0.4% for O 2 ; however, then increases from Ne to the radiatively-collapsed gases Ar, Kr, and Xe. Argon has the highest damage factor (77 × 10 10 Wm −2 s 0.5 ), while the lightest gases have the lowest (2–6 × 10 10 Wm −2 s 0.5 ). The magnetic field compressing the pinch is higher for the heavier gases. This paper confirms that the trends of ion beam production with various gases in high- inductance machines are the same as the trends in low-inductance machines.

It is known that the radius ratio of a plasma pinch depends on the specific heat ratio γ of the pinch plasma, where radius pinch ratio is defined as: radius of plasma pinch column/radius of anode. The lower the specific heat ratio the lower would be the pinch radius ratio with corresponding increased compression and pinch density. The deuterium plasma focus pinch is invariably fully ionized and has a specific heat ratio of practically the highest possible value of 5/3. If the deuterium plasma focus could have its specific heat ratio reduced below 5/3 we might expect its radius ratio to be correspondingly reduced, increasing the pinch density, thus improving the D-D fusion neutron yield. To demonstrate this effect we run the Lee model code in deuterium but hypothetically fix the specific heat ratio, reducing it at each run. The results show that indeed the radius ratio is reduced, increasing the compression, and the neutron yield is substantially increased. The effect is used to explain the observed neutron yield enhancement when a deuterium plasma focus doped with a small amount of krypton.

The six-phase Lee model code was developed to compute the anomalous resistance phase (RAN) following the pinch phase in a plasma focus (PF) discharge. One important method to check such modeling is to look at the soft X-ray (SXR) emission time profile and to correlate this to the PF dynamics. A two-channel filtered SXR spectrometer coupled with an Excel-based analytical template was recently developed to speed up the correlation process. Using this system, we have determined that the neon PF typically operates in a normal (N) mode in which it emits characteristic He-like H-like neon line SXR (in a photon energy window of 900-1550 eV) reproducibly and efficiently. The characteristic neon line SXR pulse straddles the pinch duration starting strongly 10 ns before the start of the pinch, then diminishes through the 10-ns pinch and tails off into the subsequent RAN1 phase. We present the correlated time profiles of shots operating in the efficient N mode as well as, for comparison, poor shots, which are distinctly different in SXR time profiles. The profiles indicate the difference in dynamics of normal and poor shots. Statistics are presented as well as comparison of the yields from the numerical experiments and measurements. In the series that were studied the proportion of N-mode operation ranges from 70% in one series to 80% in another series over pressure range 1-4 torr. At 2 torr, it was found that 90% recorded the normal N profile. The results reinforce the view that while the Lee Model code incorporates the correct physics in its sequence of phases, refinement is needed to extend the radiative phase to the period before the pinch.

This paper presents the first algorithm for simultaneous localization and mapping (SLAM) that can estimate the locations of both dynamic and static features in addition to the vehicle trajectory. We model the feature-based SLAM problem as a single-cluster process, where the vehicle motion defines the parent, and the map features define the daughter. Based on this assumption, we obtain tractable formulae that define a Bayesian filter recursion. The novelty in this filter is that it provides a robust multi-object likelihood which is easily understood in the context of our starting assumptions. We present a particle/Gaussian mixture implementation of the filter, taking into consideration the challenges that SLAM presents over target tracking with stationary sensors, such as changing fields of view and a mixture of static and dynamic map features. Monte Carlo simulation results are given which demonstrate the filter's effectiveness with high measurement clutter and non-linear vehicle motion Manuscript received September 09, 2012; revised December 09, 2012; accepted February 19, 2013. Date of publication March 06, 2013; date of current version May 13, 2013. This work was supported by an EPSRC grant EP/J012432/1, EU grant FP7-ICT-2011-7 project PANDORA Ref 288273, Spanish Ministry of Science and Innovation project RAIMON ref. CTM2011-29691-C02-02) and the Catalan Government (FI and BE-DGR grants).. The work of C. S. Lee was supported by a Ph.D. FI Scholarship of the Catalan Government. The work of D. E. Clark was supported by an RAEng/EPSRC Fellowship. The associate editor coordinating the review of this manuscript and approving it for publication was Prof. Ba-Ngu Vo

We adapted the Lee Model code as a branch version RADPF5.15K to gases of special interest to us, namely, nitrogen and oxygen and applied numerical experiments specifically to our AECS PF-1 and AECS PF-2. We also generalized the numerical experiments to other machines and other gases to look at scaling laws and to explore recently uncovered insights and concepts. The required thermodynamic data of nitrogen, oxygen, neon, and argon gases (ion fraction, the effective ionic charge number, the effective specific heat ratio) were calculated, the X-ray emission properties of plasmas were studied, and suitable temperature range (window) for generating H- and He-like ions (therefore soft X-ray emissions) of different species of plasmas were found. The code is applied to characterize the AECS-PF-1 and AECS-PF-2, and for optimizing the nitrogen, oxygen, neon, and argon SXR yields. In numerical experiments we show that it is useful to reduce static inductance L0 to a range of 15-25 nH; but not any smaller. These yields at diverse wavelength ranges are large enough to be of interest for applications. Scaling laws for argon and nitrogen SXR were found. Model parameters are determined by fitting computed with measured current waveforms in neon for INTI PF and in argon for the AECS PF-2. Radiative cooling effects are studied indicating that radiative collapse may be observed for heavy noble gases (Ar, Kr, Xe) for pinch currents even below 100 kA. The creation of the consequential extreme conditions of density and pulsed power is of interest for research and applications.

Kansas State University (KSU) dense plasma focus (DPF) is a 2.5-kJ DPF machine newly commissioned at the Plasma Radiation Physics Laboratory at KSU. The machine was designed to be used as a multiradiation source for applications in nuclear science and engineering. Neutrons are emitted from deuterium-deuterium (D-D) fusion reactions during high-power electric discharges at 17 kV, 140 kA, and 5 mbar. The machine circuit parameters were calculated using the short-circuit test. The emitted neutrons were measured using several radiation detection techniques. The 2.45-MeV characteristic D-D neutron energy was confirmed using the time-of-flight technique using a BC-418 plastic scintillator. The maximum neutron yield was roughly measured to be 2.8 × 108 neutrons per shot using a set of BTI BD-PND bubble detectors. Moreover, the neutron yield variation with pressure was measured and compared with the computed neutron yield using Lee model. Finally, the measured current showed good agreement with Lee six-phase model.

Conventional pulse-frequency-modulated (PFM) zero-current switching series resonant (SR) inverter-fed voltage-multiplier-based high-voltage dc power supplies have nearly zero switching loss. However, they have limitations of poor controllability at light load and large output voltage ripple at low switching frequencies. To address these problems, this letter proposes a new control scheme that is based on digitally tuning tank capacitance and slightly varying pulse frequency of SR inverter. For the realization of the proposed control approach, the tank circuit of the resonant inverter is made up of several tank capacitors that are switched into or out of the tank circuit by electromechanical switches. By digitally modulating the tank capacitance, the output voltage changes in steps. The regulation of output voltage between two adjacent steps is achieved by slightly varying the pulse frequency. The proposed control scheme has several features, namely, a wide range of output voltage controllability even at light loads, less output voltage ripple, and less current stress on the inverter's power switches at light loads. Therefore, the proposed control approach alleviates most of the problems associated with conventional PFM. Experimental results obtained from a scaled-down laboratory prototype are presented to verify the effectiveness of the proposed system.

The six-phase Lee model code is used to fit the computed current waveform to the measured current waveform of INTI plasma focus (PF; 2.2 kJ at 12 kV), a T2 PF device, operated as a source of neon soft X-ray (SXR) with optimum yield around 2 torr of neon. The characteristic He-like and H-like neon line SXR pulse is measured using a pair of SXR detectors with selected filters that, by subtraction, has a photon energy window of 900 to 1550 eV covering the region of the characteristic neon SXR lines. From the analysis of the fitted current and the measured SXR pulses, the characteristic neon SXR pulses are correlated to the pinch dynamics, and the subsequent slightly harder SXR pulses are correlated to the anomalous resistance phase. The characteristic neon SXR yield is measured; the pulse has a duration of 25 ns. The characteristic neon SXR typically starts 10 ns before the pinch phase and continues through the end of the 10-ns pinch phase, tailing into the anomalous resistance phase. Harder SXR pulses, probably Bremsstrahlung, are correlated to the anomalous resistance phase, with the main pulse occurring nearly 200 ns after the characteristic neon SXR pulse.

The United Nations University/International Centre for Theoretical Physics Plasma Focus Facility (UNU/ICTP PFF), a 3.3-kJ plasma focus, was designed for operation in deuterium with a speed factor S such that the axial run-down time matches the current rise time at an end axial speed of nearly 10 cm/mus. For operation in neon, we first consider that a focus pinch temperature between 200 and 500 eV may be suitable for a good yield of neon soft X-rays, which corresponds to an end axial speed of 6-7 cm/mus. On this basis, for operation in neon, the standard UNU/ICTP PFF needs to have its anode length z0 reduced by some 30%-40% to maintain the time matching. Numerical experiments using the Lee model code are carried out to determine the optimum configuration of the electrodes for the UNU/ICTP PFF capacitor system. The results show that an even more drastic shortening of anode length z0 is required, from the original 16 to 7 cm, at the same time, increasing the anode radius ldquoardquo from 0.95 to 1.2 cm, to obtain an optimum yield of Y sxr = 9.5 J. This represents a two- to threefold increase in the Y sxr from that computed for the standard UNU/ICTP PFF.

In this work, we propose a broadcast algorithm called Most Request Served (MRS) and its variants with caching strategies for on-time delivery of data in Real-Time Information Dispatch System. This family of algorithms consider request deadline, data object size and data popularity in making scheduling decisions. Although previous scheduling algorithms also base on some or all of these attributes to choose the most beneficial data to be broadcast, they did not consider the loss brought by their scheduling decisions. However, MRS considers both gain and loss in making a scheduling decision. We have performed a series of simulation experiments to compare the performance of various algorithms. Simulation results show that our proposed broadcast algorithm not only succeeds in providing good on-time delivery of data but at the same time provides 20% of improvement in response time over traditional scheduling algorithms like First-In-First-Out (FIFO) and Earliest-Deadline-First (EDF). Simulation results also show that our proposed caching strategy provides further improvement in terms of percentage of requests finished in time over traditional caching strategy like Least Recently Used (LRU).

A concept for a 3-D waveguide system, which can be used for optical phased arrays, is presented in this letter. We implemented static phased arrays with rod waveguides of different widths, which can be used for holographic applications. They are fabricated with direct laser writing, a technique based on two photon polymerization. As the element size of our phased array can be in the order of the visible wavelength, relatively large diffraction angles of up to 40° can be reached.

Dense plasma focus (DPF) can be a powerful source of X-rays at the wavelengths useful for microlithography and micromachining depending on its working gas (Ne or Ar correspondingly) and operating parameters of the device. Experimental investigations were carried out with /spl sim/0.4-nm wavelength radiation from a specially designed medium-power soft X-ray tube with a water-cooled silver anode and highly sensitive chemically amplified resist SU-8. They have shown us that it is possible to produce 10-/spl mu/m structures replicated in 100-/spl mu/m resist layer and 4-/spl mu/m structures produced in a 35-/spl mu/m resist layer for 30 min. To decrease the time, a DPF device must be implemented for the task. Using pure argon and mixtures of argon with deuterium or krypton, we have found regimes ("hot spots," plasma pinching, and plasma compression by a "heavy shell") with appreciable soft X-ray yield. Influence on the results of chamber-circuit matching, pinch dynamics, and electron runaway processes are also discussed.