Kerbala Journal for Engineering Sciences

The ability to relate information about languages heard through visual and audio data is a crucial
aspect of audio-visual speech recognition (AVSR), which has uses in data manipulation for audio-visual
correspondence, including AVE-Net and SyncNet. The technique described in this research uses feature
disentanglement to simultaneously handle the tasks listed above. By developing cross-modal standard
learning methods, this model can transform visual or aural linguistic characteristics into modality
independent representations. AVE-Net and SyncNet can all be performed with the help of such derived
linguistic expressions. Furthermore, audio and visual data output can be modified based on the required
subject identity and linguistic content information. We do comprehensive trials on various recognition
and synthesis tasks on both tasks separately, and that solution can successfully take on both audio-visual
learning problems. The system gives great results in the enhanced video with 91.5% with 5 frames, while
this will increase with the increase of frames with 99.03% with 15 frames, which is more efficient than
the previous methods.

This study discusses the prediction of secondary structure protein from features of amino acid
images. We define a predication algorithm depending on the spatial information of amino acid
sequences. Algorithm A and algorithm B are applied to extract spatial information for clustering types
of amino acids in three states: helix(H), strand(E), and coil(C). The accuracy of clustering depends on
the improvement of algorithm B and the parameters of algorithms A and B. We apply a novel approach
to spatial information extraction from primary protein structure to predicate three states of secondary
protein structure. Primary protein sequences are divided into sub-images depending on the win window
size. The accuracy of prediction is variable due to the size of the windows. Algorithms A and B are
applied to achieve accuracy in each type of secondary protein structure. Helix (H) is 91.93%, Sheet(E)
is 93.15%, and Coil (C) is 89.0126 %. In algorithms A and B, spatial information provides automated
feature extraction to predict the secondary protein structure ion.

This paper includes a review of several previous studies associated with the charging behavior of
phase-change materials (PCM) in horizontal and vertical tabular (annular) Latent Heat Thermal Energy
Storage (LHTES) units. The effects of various factors on charging, such as heat transfer fluid (HTF) inlet
temperature, the eccentricity of the inner tube, the inclination angle of the storage unit ( −90◦, −60◦, −30◦,
0◦, 30◦, 60◦, 90◦), and the mass flow rate of HTF are investigated. Natural convection controls the charging
process in the upper part of the thermal storage unit, while thermal diffusion dominates the melting in the
lower part. Also, the HTF inlet temperature has a clear and significant effect on reducing the melting time
of PCM. In addition, increasing the inner tube eccentricity results in a reduction in the charging time.

This paper presents a set of investigations achieved through a simulation involving a swarm of
six robots transporting heavy objects of complex shapes that require the cooperative efforts of all six
robots. This study focuses on the complexity of the objects’ shapes, starting from a basic shape (i.e.,
Cuboid) to more complex shapes built from multiple basic shapes (i.e., Star and H-shape). The controller
of the robots is a continuous-time recurrent neural network synthesized using artificial evolutionary
techniques. The results indicate that evolution could find solutions for each object separately. However,
a single generalized neuro-controller could not be obtained. This work unveils an interesting relationship
between design choices and the complexity of object shapes. We analyze the effectiveness of cooperative
transport strategies in terms of two metrics: the time required to sustain the transport and the quality of
transport trajectories (i.e., sinuosity metrics).

This paper presents a set of investigations achieved through a simulation involving a swarm of
six robots transporting heavy objects of complex shapes that require the cooperative efforts of all six
robots. This study focuses on the complexity of the objects’ shapes, starting from a basic shape (i.e.,
Cuboid) to more complex shapes built from multiple basic shapes (i.e., Star and H-shape). The controller
of the robots is a continuous-time recurrent neural network synthesized using artificial evolutionary
techniques. The results indicate that evolution could find solutions for each object separately. However,
a single generalized neuro-controller could not be obtained. This work unveils an interesting relationship
between design choices and the complexity of object shapes. We analyze the effectiveness of cooperative
transport strategies in terms of two metrics: the time required to sustain the transport and the quality of
transport trajectories (i.e., sinuosity metrics).

This paper presents the computational investigation of the charging characteristics of PCM inside a latent
heat thermal energy storage (LHTES) unit that consists of a horizontal heat exchanger type of the inner
tube and outer shell using ANSYS/FLUENT software. The melting process numerically tested for the
heat transfer fluid HTF (water) inlet temperature (60, 70, and 80 °C), and the PCM's initial temperature
is 15 °C. Also, this study examined the effect of the mushy zone constant. Extensive measures show that
charging time decreases with increased water inlet temperature. The melting of PCM completes in a time
of 260 minutes when using a water inlet temperature of 60 °C, while the melting time reaches 182 minutes
when the water inlet temperature is 70 °C, and the melting completes in a time of 145 minutes by using
the hot water entry temperature of 80 °C. When the water inlet temperature increased from 60 °C to 70
°C and 80 °C, respectively, the melting time was saved by about 30% and 44%. The best melting time
reduction is achieved at an HTF inlet temperature of 80 °C. In addition, the melting time increased as
constant values rose in the mushy zone. The melting times are 182, 191, and 200 minutes for mushy zone
constants 105, 106, and 107, respectively. So, the mushy zone constant of 105 leads to the best reduction
in melting time.