| 1. |
- Khan, Adil, et al.
(författare)
-
Predictive modeling for depth of wear of concrete modified with fly ash : A comparative analysis of genetic programming-based algorithms
- 2024
-
Ingår i: Case Studies in Construction Materials. - : Elsevier. - 2214-5095. ; 20
-
Tidskriftsartikel (refereegranskat)abstract
- There has been increasing growth in incorporating fly ash as a supplementary cementitious material in concrete mixtures due to its potential to enhance the durability and strength properties of concrete. However, there is a lack of research on predicting the depth of wear of fly ash-based concrete. The laboratory methods available for estimating the depth of wear often involve destructive and expensive tests. Therefore, to avoid costly and laborious tests, this study utilized two machine learning methods, including multi-expression programming (MEP) and gene expression programming (GEP), to predict the depth of wear of fly ash-modified concrete. A comprehensive dataset of 216 experimental records was compiled from published studies for model training and validation. This extensive dataset encompasses the depth of wear as the target variable, along with nine explanatory parameters, namely fly ash, cement content, fine and coarse aggregate, water content, plasticizer, age of concrete, air-entraining agent, and testing time. The models were trained with 70% of the data, and the remaining 30% of data was used for validating the models. The models were developed by a continuous trial-and-error process and iterative refinement of hyperparameters until optimal results were achieved. The efficacy of the models was assessed via multiple statistical indicators. Furthermore, the SHapley Additive exPlanation (SHAP) was utilized for the interpretability of the model prediction from both global and local perspectives. The GEP model exhibited excellent accuracy with a correlation coefficient (R) of 0.989 (training) and 0.992 (validation). Similarly, the MEP model provided prediction accuracy with R values of 0.965 and 0.968 for training and validation sets, respectively. In addition, the MEP and GEP models outperformed the traditional multi-linear regression model. The SHAP interpretation revealed that testing time and age have a higher contribution in determining the depth of wear. The findings of this study can assist practitioners and designers in avoiding costly and laborious tests for durability assessment and promoting sustainable use of fly ash in the construction sector.
|
|
| 2. |
- Khan, Majid, et al.
(författare)
-
Intelligent prediction modeling for flexural capacity of FRP-strengthened reinforced concrete beams using machine learning algorithms
- 2023
-
Ingår i: Heliyon. - : Cell Press. - 2405-8440. ; 10
-
Tidskriftsartikel (refereegranskat)abstract
- Fiber-reinforced polymers (FRP) are widely utilized to improve the efficiency and durability of concrete structures, either through external bonding or internal reinforcement. However, the response of FRP-strengthened reinforced concrete (RC) members, both in field applications and experimental settings, often deviates from the estimation based on existing code provisions. This discrepancy can be attributed to the limitations of code provisions in fully capturing the nature of FRP-strengthened RC members. Accordingly, machine learning methods, including gene expression programming (GEP) and multi-expression programming (MEP), were utilized in this study to predict the flexural capacity of the FRP-strengthened RC beam. To develop data-driven estimation models, an extensive collection of experimental data on FRP-strengthened RC beams was compiled from the experimental studies. For the assessment of the accuracy of developed models, various statistical indicators were utilized. The machine learning (ML) based models were compared with empirical and conventional linear regression models to substantiate their superiority, providing evidence of enhanced performance. The GEP model demonstrated outstanding predictive performance with a correlation coefficient (R) of 0.98 for both the training and validation phases, accompanied by minimal mean absolute errors (MAE) of 4.08 and 5.39, respectively. In contrast, the MEP model achieved a slightly lower accuracy, with an R of 0.96 in both the training and validation phases. Moreover, the ML-based models exhibited notably superior performances compared to the empirical models. Hence, the ML-based models presented in this study demonstrated promising prospects for practical implementation in engineering applications. Moreover, the SHapley Additive exPlanation (SHAP) method was used to interpret the feature's importance and influence on the flexural capacity. It was observed that beam width, section effective depth, and the tensile longitudinal bars reinforcement ratio significantly contribute to the prediction of the flexural capacity of the FRP-strengthened reinforced concrete beam.
|
|
| 3. |
- Saif-Ul-Allah, Muhammad Waqas, et al.
(författare)
-
Computationally Inexpensive 1D-CNN for the Prediction of Noisy Data of NOx Emissions From 500 MW Coal-Fired Power Plant
- 2022
-
Ingår i: Frontiers in Energy Research. - : FRONTIERS MEDIA SA. - 2296-598X. ; 10
-
Tidskriftsartikel (refereegranskat)abstract
- Coal-fired power plants have been used to meet the energy requirements in countries where coal reserves are abundant and are the key source of NOx emissions. Owing to the serious environmental and health concerns associated with NOx emissions, much work has been carried out to reduce NOx emissions. Sophisticated artificial intelligence (AI) techniques have been employed during the past few decades, such as least-squares support vector machine (LSSVM), artificial neural networks (ANN), long short-term memory (LSTM), and gated recurrent unit (GRU), to develop the NOx prediction model. Several studies have investigated deep neural networks (DNN) models for accurate NOx emission prediction. However, there is a need to investigate a DNN-based NOx prediction model that is accurate and computationally inexpensive. Recently, a new AI technique, convolutional neural network (CNN), has been introduced and proven superior for image class prediction accuracy. According to the best of the author's knowledge, not much work has been done on the utilization of CNN on NOx emissions from coal-fired power plants. Therefore, this study investigated the prediction performance and computational time of one-dimensional CNN (1D-CNN) on NOx emissions data from a 500 MW coal-fired power plant. The variations of hyperparameters of LSTM, GRU, and 1D-CNN were investigated, and the performance metrics such as RMSE and computational time were recorded to obtain optimal hyperparameters. The obtained optimal values of hyperparameters of LSTM, GRU, and 1D-CNN were then employed for models' development, and consequently, the models were tested on test data. The 1D-CNN NOx emission model improved the training efficiency in terms of RMSE by 70.6% and 60.1% compared to LSTM and GRU, respectively. Furthermore, the testing efficiency for 1D-CNN improved by 10.2% and 15.7% compared to LSTM and GRU, respectively. Moreover, 1D-CNN (26 s) reduced the training time by 83.8% and 50% compared to LSTM (160 s) and GRU (52 s), respectively. Results reveal that 1D-CNN is more accurate, more stable, and computationally inexpensive compared to LSTM and GRU on NOx emission data from the 500 MW power plant.
|
|
| 4. |
- Khan, Wali Ullah, et al.
(författare)
-
Integration of NOMA with Reflecting Intelligent Surfaces : A Multi-cell Optimization with SIC Decoding Errors
- 2023
-
Ingår i: IEEE Transactions on Green Communications and Networking. - : Institute of Electrical and Electronics Engineers (IEEE). - 2473-2400. ; 7:3, s. 1554-1565
-
Tidskriftsartikel (refereegranskat)abstract
- Reflecting intelligent surfaces (RIS) has gained significant attention due to its high energy and spectral efficiency in next-generation wireless networks. By using low-cost passive reflecting elements, RIS can smartly reconfigure the signal propagation to extend the wireless communication coverage. On the other hand, non-orthogonal multiple access (NOMA) has been proven as a key air interface technique for supporting massive connections over limited resources. Utilizing the superposition coding and successive interference cancellation (SIC) techniques, NOMA can multiplex multiple users over the same spectrum and time resources by allocating different power levels. This paper proposes a new optimization scheme in a multi-cell RIS-NOMA network to enhance the spectral efficiency under SIC decoding errors. In particular, the power budget of the base station and the transmit power of NOMA users while the passive beamforming of RIS is simultaneously optimized in each cell. Due to objective function and quality of service constraints, the joint problem is formulated as non-convex, which is very complex and challenging to obtain the optimal global solution. To reduce the complexity and make the problem tractable, we first decouple the original problem into two sub-problems for power allocation and passive beamforming. Then, the efficient solution of each sub-problem is obtained in two-steps. In the first-step of For power allocation sub-problem, we transform it to a convex problem by the inner approximation method and then solve it through a standard convex optimization solver in the second-step. Accordingly, in the first-step of passive beamforming, it is transformed into a standard semi-definite programming problem by successive convex approximation and different of convex programming methods. Then, penalty based method is used to achieve a Rank-1 solution for passive beamforming in second-step. Numerical results demonstrate the benefits of the proposed optimization scheme in the multi-cell RIS-NOMA network.
|
|
| 5. |
- Khan, Wali Ullah, et al.
(författare)
-
Rate Splitting Multiple Access for Next Generation Cognitive Radio Enabled LEO Satellite Networks
- 2023
-
Ingår i: IEEE Transactions on Wireless Communications. - : Institute of Electrical and Electronics Engineers (IEEE). - 1536-1276. ; , s. 1-
-
Tidskriftsartikel (refereegranskat)abstract
- Low Earth Orbit (LEO) satellite communication (SatCom) has drawn particular attention recently due to its high data rate services and low round-trip latency. It has low launching and manufacturing costs than Medium Earth Orbit (MEO) and Geostationary Earth Orbit (GEO) satellites. Moreover, LEO SatCom has the potential to provide global coverage with a high-speed data rate and low transmission latency. However, the spectrum scarcity might be one of the challenges in the growth of LEO satellites, impacting severe restrictions on developing ground-space integrated networks. To address this issue, cognitive radio and rate splitting multiple access (RSMA) are the two emerging technologies for high spectral efficiency and massive connectivity. This paper proposes a cognitive radio enabled LEO SatCom using RSMA radio access technique with the coexistence of GEO SatCom network. In particular, this work aims to maximize the sum rate of LEO SatCom by simultaneously optimizing the power budget over different beams, RSMA power allocation for users over each beam, and subcarrier user assignment while restricting the interference temperature to GEO SatCom. The problem of sum rate maximization is formulated as non-convex, where the global optimal solution is challenging to obtain. Thus, an efficient solution can be obtained in three steps: first we employ a successive convex approximation technique to reduce the complexity and make the problem more tractable. Second, for any given resource block user assignment, we adopt KarushKuhnTucker (KKT) conditions to calculate the transmit power over different beams and RSMA power allocation of users over each beam. Third, using the allocated power, we design an efficient algorithm based on the greedy approach for resource block user assignment. For comparison, we propose two suboptimal schemes with fixed power allocation over different beams and random resource block user assignment as the benchmark. Numerical results provided in this work are obtained based on the Monte Carlo simulations, which demonstrate the benefits of the proposed optimization scheme compared to the benchmark schemes.
|
|
| 6. |
- Nawaz, Muhammad Ul Saqlain, et al.
(författare)
-
Efficient resource prediction framework for software-defined heterogeneous radio environmental infrastructures
- 2023
-
Ingår i: Advanced Engineering Informatics. - : Elsevier Ltd.. - 1474-0346. ; 56
-
Tidskriftsartikel (refereegranskat)abstract
- Artificial Intelligence (AI) is defining the future of next-generation infrastructures as proactive and data-driven systems. AI-empowered radio systems are replacing the existing command and control radio networks due to their intelligence and capabilities to adapt to the radio environmental infrastructures that include intelligent networks, smart cities and AV/VR enabled factory premises or localities. An efficient resource prediction framework (ERPF) is proposed to provide proactive knowledge about the availability of radio resources in such software-defined heterogeneous radio environmental infrastructures (SD-HREIs). That prior information enables the coexistence of radio users in SD-HREIs. In a proposed framework, the radio activity is measured in both the unlicensed bands that include 2.4 and 5 GHz, respectively. The clustering algorithms k- means and DBSCAN are implemented to segregate the already measured radioactivity as signal (radio occupancy) and noise (radio opportunity). Machine learning techniques CNN and LRN are then trained and tested using the segregated data to predict the radio occupancy and radio opportunity in SD-HREIs. Finally, the performance of CNN and LRN is validated using the cross-validation metrics.
|
|
