System of Laser Monitoring of Water Pollution with Application of Relative Description of Signal Shape

As a rule, the wastewater treatment system is not designed to filter substances formed, as a result of beyond design basis accident. The nature of the beyond design basis accident is associated with the shortterm appearance of a clot of these substances in wastewater, determined by the volume of the substance storage tank. Therefore, a rational approach is to divert this portion of the formed substances into a separate branch of the sewage system or sedimentation tanks. The aim of the work is to implement this approach by creating a laser monitoring system for water pollution.The article proposes a system for automatic detection of a clot of emergency discharge of pollutants into the wastewater of an industrial enterprise. The structural diagram of the system and the purpose of its main elements are given. The system should provide clot detection in real time. To ensure this function, a preliminary study is made of the spectral characteristics of all substances that may appear in wastewater in the event of an emergency.Based on these data, the wavelengths of laser radiation in the system are selected. The obtained measurement data from several probes are presented in the form of a lattice function, which is translated into a relative description representing the order relationship matrix on the set of lattice function components. The relative description is invariant to linear changes in the lattice function. The decision to detect any substance from emergency discharges is made based on a comparison of the relative description of the measurements with the standards prepared at the stage of system setup.The article provides an example of the formation of standards for emergency clots from glycerin and allyl alcohol. The graphs of the lattice functions obtained from the IR spectra of emergency discharges of these substances are given; algorithms for constructing a lattice function and comparison of lattice functions. Thus, using the developed mathematical description of the shape of digital signals based on the relative description, the signal of the monitoring curve can be described in the form of a curve of the optical density change of an aqueous medium.

Learning Ratio Mask with Cascaded Deep Neural Networks for Echo Cancellation in Laser Monitoring Signals

Laser monitoring has received more and more attention in many application fields thanks to its essential advantages. The analysis shows that the target speech in the laser monitoring signals is often interfered by the echoes, resulting in a decline in speech intelligibility and quality, which in turn affects the identification of useful information. The cancellation of echoes in laser monitoring signals is not a trivial task. In this article, we formulate it as a simple but effective additive echo noise model and propose a cascade deep neural networks (C-DNNs) as the mapping function from the acoustic feature of noisy speech to the ratio mask of clean signal. To validate the feasibility and effectiveness of the proposed method, we investigated the effect of echo intensity, echo delay, and training target on the performance. We also compared the proposed C-DNNs to some traditional and newly emerging DNN-based supervised learning methods. Extensive experiments demonstrated the proposed method can greatly improve the speech intelligibility and speech quality of the echo-cancelled signals and outperform the comparison methods.