Evgenii Maltsev

Dr. Evgenii Maltsev

orcid

Research Focus & Metrics

Academic Career Length

16Years (2011-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
5 October 2021 in Sensors
Verified
Journal Article
Open access
PDF file
DWDaniel Stephen Wamriew
Daniel Stephen Wamriew
RPRoman Pevzner
Roman Pevzner
EMEvgenii Maltsev
Evgenii Maltsev
DPDimitri Pissarenko
Dimitri Pissarenko
Fiber-optic cables have recently gained popularity for use as Distributed Acoustic Sensing (DAS) arrays for borehole microseismic monitoring due to their physical robustness as well as high spatial and temporal resolutions. As a result, the sensors record large amounts of data, making it very difficult to process in real-/semi-real-time using the conventional processing routines. We present a novel approach, based on deep learning, for handling the large amounts of DAS data in real-/semi-real-time. The proposed neural network was trained on synthetic microseismic data contaminated with real-ambient noise from field data and was validated using field DAS microseismic data obtained from a hydraulic fracturing operation. The results indicate that the trained network is capable of detecting and locating microseismic events from DAS data and simultaneously update the velocity model to a high degree of precision. The mean absolute errors in the event locations and the velocity model parameters are 2.04, 0.72, 2.76, 4.19 and 0.97 percent for distance (x), depth (z), P-wave velocity, S-wave velocity and density, respectively. In addition to automation and computational efficiency, deep learning reduces human expert data handling during processing, thus preserving data integrity leading to more accurate and reproducible results.

14 October 2020 in Applied Sciences
Verified
Journal Article
Open access
PDF file
Alexander SafonovAlexander Safonov
Alexander Safonov
EMEvgenii Maltsev
Evgenii Maltsev
SCSvyatoslav Chugunov
Svyatoslav Chugunov
ATAndrey Tikhonov
Andrey Tikhonov
Stepan KonevStepan Konev
Stepan Konev
Iskander AkhatovIskander Akhatov
Iskander Akhatov
3D printing allows the fabrication of ceramic implants, making a personalized approach to patients’ treatment a reality. In this work, we have tested the applicability of the Function Representation (FRep) method for geometric simulation of implants with complex cellular microstructure. For this study, we have built several parametric 3D models of 4 mm diameter cylindrical bone implant specimens of four different types of cellular structure. The 9.5 mm long implants are designed to fill hole defects in the trabecular bone. Specimens of designed ceramic implants were fabricated at a Ceramaker 900 stereolithographic 3D printer, using a commercial 3D Mix alumina (Al2O3) ceramic paste. Then, a single-axis compression test was performed on fabricated specimens. According to the test results, the maximum load for tested specimens constituted from 93.0 to 817.5 N, depending on the size of the unit cell and the thickness of the ribs. This demonstrates the possibility of fabricating implants for a wide range of loads, making the choice of the right structure for each patient much easier.

17 July 2022 in Remote Sensing
Verified
Journal Article
Open access
PDF file
DWDaniel Stephen Wamriew
Daniel Stephen Wamriew
DDDesmond Batsa Dorhjie
Desmond Batsa Dorhjie
DBDaniil Aleksandrovich Bogoedov
Daniil Aleksandrovich Bogoedov
RPRoman Pevzner
Roman Pevzner
EMEvgenii Maltsev
Evgenii Maltsev
DKDmitry Koroteev
Dmitry Koroteev
Microseismic monitoring is a useful enabler for reservoir characterization without which the information on the effects of reservoir operations such as hydraulic fracturing, enhanced oil recovery, carbon dioxide, or natural gas geological storage would be obscured. This research provides a new breakthrough in the tracking of the reservoir fracture network and characterization by detecting the microseismic events and locating their sources in real-time during reservoir operations. The monitoring was conducted using fiber optic distributed acoustic sensors (DAS) and the data were analyzed by deep learning. The use of DAS for microseismic monitoring is a game changer due to its excellent temporal and spatial resolution as well as cost-effectiveness. The deep learning approach is well-suited to dealing in real-time with the large amounts of data recorded by DAS equipment due to its computational speed. Two convolutional neural network based models were evaluated and the best one was used to detect and locate microseismic events from the DAS recorded field microseismic data from the FORGE project in Milford, United States. The results indicate the capability of deep neural networks to simultaneously detect and locate microseismic events from the raw DAS measurements. The results showed a small percentage error. In addition to the high spatial and temporal resolution, fiber optic cables are durable and can be installed permanently in the field and be used for decades. They are also resistant to high pressure, can withstand considerably high temperature, and therefore can be used even during field operations such as a flooding or hydraulic fracture stimulation. Deep neural networks are very robust; need minimum data pre-processing, can handle large volumes of data, and are able to perform multiple computations in a time- and cost-effective way. Once trained, the network can be easily adopted to new conditions through transfer learning.

23 July 2021 in Applied Sciences
Verified
Journal Article
Open access
PDF file
EMEvgenii Maltsev
Evgenii Maltsev
DPDmitry Popov
Dmitry Popov
SCSvyatoslav Chugunov
Svyatoslav Chugunov
APAlexander Pasko
Alexander Pasko
Iskander AkhatovIskander Akhatov
Iskander Akhatov
Complex 3D objects with microstructures can be modelled using the function representation (FRep) approach and then manufactured. The task of modelling a geometric object with a sophisticated microstructure based on unit cell repetition is often too computationally expensive for CAD systems. FRep provides efficient tools to solve this problem. However, even for FRep the slicing step required for manufacturing can take a significant amount of time. An accelerated slicing algorithm for FRep 3D objects is proposed in this paper. This algorithm allows the preparation of FRep models for 3D printing without surface generation stage. The spatial index is employed to accelerate the slicing process. A novel compound adaptive criterion and a novel acceleration criterion are proposed to speed up the evaluation of the defining function of an FRep object. The use of these criteria is significantly reducing the computational time for contour construction during the slicing process. The K-d tree and R-tree data structures are used as spatial indexes. The performance of the accelerated slicing algorithm was tested. The contouring time was reduced 100-fold due to using the novel compound adaptive criterion with the novel acceleration criterion.

12 August 2021 in Applied Sciences
Verified
Journal Article
Open access
PDF file
DPDmitry Popov
Dmitry Popov
YKYulia Kuzminova
Yulia Kuzminova
EMEvgenii Maltsev
Evgenii Maltsev
Stanislav EvlashinStanislav Evlashin
Stanislav Evlashin
Alexander SafonovAlexander Safonov
Alexander Safonov
APAlexander Pasko
Alexander Pasko
Additive manufacturing erases the distance between design ideas and finished parts. However, designers must use several software tools to use these advantages. Moreover, these tools operate with different representations of geometry. This paper describes the architecture of a new CAD/CAM system that uses only the function representation of the geometry (FRep). It provides all widely used design operations and allows an engineer to employ robust and efficient topology optimization algorithms. The developed CAD/CAM system consists of 3D modeling, simulation, topology optimization, and direct manufacturing modules. We successfully printed designed parts and performed mechanical tests of printed parts. The results of tests show good agreement with simulation data. The system makes it possible to create structures with the desired properties in a fast and flexible way. The proposed approach significantly helps in designing additive manufacturing process and saves time for its users.

11 January 2024 in IEEE Access
Verified
Journal Article
Open access
AKAndrey Kvochko
Andrey Kvochko
EMEvgenii Maltsev
Evgenii Maltsev
ABArtem Balyshev
Artem Balyshev
SMStanislav Malakhov
Stanislav Malakhov
AEAlexander Efimov
Alexander Efimov
Many modern Systems-on-Chip (SoCs) are equipped with specialized Machine Learning (ML) accelerators that use both on-chip and off-chip memory to execute neural networks. While on-chip memory usually has a hard limit, off-chip memory is often considered large enough to hold the network’s inputs, outputs, weights, and any intermediate results that may occur during model execution. This assumption may not hold for edge devices, such as smartphones, which usually have a limit on the amount of memory a process can use. In this study, we propose a novel approach for minimizing a neural network’s off-chip memory usage by introducing a tile-aware allocator capable of reusing memory occupied by parts of a tensor before the entire tensor expires. We describe the necessary conditions for such an off-chip memory allocation approach and provide the results, showing that it can save up to 33% of the peak off-chip memory usage in some common network architectures.