Comparisons of Three-Dimensional Variational Data Assimilation and Model Output Statistics in Improving Atmospheric Chemistry Forecasts

Atmospheric chemistry models usually perform badly in forecasting wintertime air pollution because of their uncertainties. Generally, such uncertainties can be decreased effectively by techniques such as data assimilation（DA） and model output statistics（MOS）. However, the relative importance and combined effects of the two techniques have not been clarified. Here,a one-month air quality forecast with the Weather Research and Forecasting-Chemistry（WRF-Chem） model was carried out in a virtually operational setup focusing on Hebei Province, China. Meanwhile, three-dimensional variational（3 DVar） DA and MOS based on one-dimensional Kalman filtering were implemented separately and simultaneously to investigate their performance in improving the model forecast. Comparison with observations shows that the chemistry forecast with MOS outperforms that with 3 DVar DA, which could be seen in all the species tested over the whole 72 forecast hours. Combined use of both techniques does not guarantee a better forecast than MOS only, with the improvements and degradations being small and appearing rather randomly. Results indicate that the implementation of MOS is more suitable than 3 DVar DA in improving the operational forecasting ability of WRF-Chem.

The Construction of Statistically Recursive Sets

We have constructed a class of random sets by statistical contraction operators in this paper.When the probability space is some special product space and the statistical contraction operators are affine or similar,these statistically recursive sets are investigated by many authors.It will be very convenient for us to study their distributions and dimensions and measures using our model in this paper.

Design of DC Partial Discharge Detecting System for Electrical Equipment

A balancing method was used to build a DC partial discharge （PD） testing circuit for electrical equipment,and a narrow-band detection system was designed using detection resistance and a filter. After signal accessed a high-speed digital acquisition （DAQ） card,the system was triggered to extract a single partial discharge （PD） signal. To eliminate the interference pulses caused by power supply ripple,etc.,the time domain and frequency domain features of pulses were extracted. Based on the features,cluster analysis was used to exclude interference pulses. Two-dimensional and three-dimensional histograms were obtained by use of the Δt method. Then,22 discharge statistical operators were calculated for the two-dimensional charts. Lastly,the defective capacitors were tested to verify the system＇s ability. The results show that the system is capable of PD detection in electrical equipment.

The Classification of Statistically Recursive Sets

The main aim of this paper is to make a classification of random sets K m(ω) constructed in theorem 2.1 and theorem 2.1' in . We provide five criterions for the classification. Many kinds of random sets such as Hawkes constructed in , Graf constructed in and Mauldin constructed in are the special cases of K m(ω) constructed in ,and then these random sets belong to some model respectively according to our classification.

Fractals Generated by Statistical Contraction Operators

In the theory of random fractal, there are two important classes of random sets, one is the class of fractals generated by the paths of stochastic processes and another one is the class of factals generated by statistical contraction operators. Now we will introduce some things about the probability basis and fractal properties of fractals in the last class. The probability basis contains (1) the convergence and measurability of a random recursive setK(ω) as a random element, (2) martingals property. The fractal properties include (3) the character of various similarity, (4) the separability property, (5) the support and zero-one law of distributionP k =P·K ?1, (6) the Hausdorff dimension and Hausdorff exact measure function.

MRF based construction of statistical operator and its application

Based on the Markov random field (MRF) theory, a new nonlinear operator isdefined according to the statistical information in the image, and the corresponding 2Dnonlinear wavelet transform is also provided. It is proved that many detail coefficientsbeing zero (or almost zero) in the smooth gray-level variation areas can be achievedunder the conditional probability density function in MRF model, which shows that thisoperator is suitable for the task of image compression, especially for lossless codingapplications. Experimental results using several test images indicate good performancesof the proposed method with the smaller entropy for the compound and smooth medicalimages with respect to the other nonlinear transform methods based on median andmorphological operator and some well-known linear lifting wavelet transform methods(5/3, 9/7, and S+P).