Woon Yang Tan’s research while affiliated with University of Malaya and other places

Flooding remains a persistent and ongoing challenge in Malaysia, posing continuous difficulties and hardships; therefore, combating flood risk has become a main priority for sustainable development in this country. The application of sophisticated engineering in hard structure approaches and flood control systems are often incongruous to the public. Therefore, there is a need to integrate soft engineering approaches and best practices of flood management to mitigate flood risk. There is also a need to integrate the concept of sustainable development into its planning policies towards flood hazard reduction. Mobile flood protection measures are useful as an alternative solution in flood protection and mitigation purposes. Mobile protective systems serve as a temporary solution to enhance the height of permanent flood defence structures during extreme events. They can also be deployed as emergency measures to mitigate flooding in vulnerable low-lying regions. As the available mobile systems differ in the type of material, method of installation and available protection height, a description of their features, and potential application are examined including their respective opportunities and drawbacks. This paper presents a review on different types of lightweight mobile flood wall barrier (MFWB) that were implemented to cope with floods in Malaysia with examples of application in other countries such as United Kingdom, Slovakia, and Netherlands. The MFWB products are reviewed and compared to each other according to the types, characteristics, mechanisms, drawbacks, and how these measures are integrated into spatial planning. Based on the findings, several recommendations are provided regarding enhancing flood risk management in areas prone to flooding and the way forward for Malaysia.

Water security and urban flooding have become major sustainability issues. This paper presents a novel method to introduce rates of change as the state-of-the-art approach in artificial intelligence model development for sustainability agenda. Multi-layer perceptron (MLP) and deep learning long short-term memory (LSTM) models were considered for flood forecasting. Historical rainfall data from 2008 to 2021 at 11 telemetry stations were obtained to predict flow at the confluence between Klang River and Ampang River. The initial results of MLP yielded poor performance beneath normal expectations, which was R = 0.4465, MAE = 3.7135, NSE = 0.1994 and RMSE = 8.8556. Meanwhile, the LSTM model generated a 45% improvement in its R-value up to 0.9055. Detailed investigations found that the redundancy of data input that yielded multiple target values had distorted the model performance. Qt was introduced into input parameters to solve this issue, while Qt+0.5 was the target value. A significant improvement in the results was detected with R = 0.9359, MAE = 0.7722, NSE = 0.8756 and RMSE = 3.4911. When the rates of change were employed, an impressive improvement was seen for the plot of actual vs. forecasted flow. Findings showed that the rates of change could reduce forecast errors and were helpful as an additional layer of early flood detection. HIGHLIGHTS A highly accurate flood forecasting system based on deep learning has been developed.; Multi-lead ahead forecasting for streamflow has been investigated.; The novel architecture model has been successfully applied for controlling streamflow in SMART tunnel.; The proposed new model architecture could be applied to forecast river streamflow in different hydrological areas;

Elevating the accuracy of streamflow forecasting has always been a challenge. This paper proposes a three-step artificial intelligence model improvement for streamflow forecasting. Step 1 uses long short-term memory (LSTM), an improvement on the conventional artificial neural network (ANN). Step 2 performs multi-step ahead forecasting while establishing the rates of change as a new approach. Step 3 further improves the accuracy through three different kinds of optimization algorithms. The Stormwater and Road Tunnel project in Kuala Lumpur is the study area. Historical rainfall data of 14 years at 11 telemetry stations are obtained to forecast the flow at the confluence located next to the control center. Step 1 reveals that LSTM is a better model than ANN with R 0.9055, MSE 17,8532, MAE 1.4365, NSE 0.8190 and RMSE 5.3695. Step 2 unveils the rates of change model that outperforms the rest with R = 0.9545, MSE = 8.9746, MAE = 0.5434, NSE = 0.9090 and RMSE = 2.9958. Finally, Stage 3 is a further improvement with R = 0.9757, MSE = 4.7187, MAE = 0.4672, NSE = 0.9514 and RMSE = 2.1723 for the bat-LSTM hybrid algorithm. This study shows that the δQ model has consistently yielded promising results while the metaheuristic algorithms are able to yield additional improvement to the model’s results.

Climate change have led to extreme weather events such as higher rainfall frequency that can cause severe flooding. In Malaysia, there is an increasing trend on extreme rainfall events and short temporal rainfall, particularly during the inter-monsoon season. In order to protect private properties and public premises from flooding, Mobile Flood Wall Barrier (MFWB) has been found to be more suitable as it is less costly, easier to deploy and does not require large space. Buildings such as factories and commercial shops that have larger entrances, they would require heavy-duty type of MFWB as compared to those for residential buildings. Heavy-duty MFWB has a better ability to withstand higher hydrostatic pressure from floodwater, hence suitable for public premises and buildings in industrial and commercial areas. In this paper, various types of heavy-duty MFWB and their application will be presented and discussed. The standards for testing MFWB products presented in this paper are summarised. Some existing testing requirements are also presented. Based on the review, the mobility characteristic indicated that the heavy MFWB can be installed temporarily to prevent flooding and be removed easily to ensure no interruption to the daily activities after flood events. There are many potential advantages for flood protection, in particular, it serves as the way forward for Malaysia.

Streamflow forecasting is the most well studied hydrological science but still portray numerous unknown knowledge. The conventional physical-based model is unable to yield a high accuracy of forecast due to the embedded noises, non-linear and stochastic nature of hydrological data. This paper is to review the recent development of the artificial intelligence (AI) methods for river streamflow forecasting over the last two decades. The original establishments of forecasting model are derived from the neural network theory, supervised machine learning and fuzzy logic theory. The first wave of development comes with evolutional theory of optimization algorithm and the probabilistic theory. The distinction of AI model is the ability to process complex data that are non-linear, non-stationary and stochastic without prior physical knowledge. However, due to the limitations of the AI model, issues arise in the form of over-fitting, trap in local minima and slow learning process; and subsequently limit the possibility of gaining higher accuracy. This has prompted researchers to try numerous new approaches to increase the efficiency of data processing. The second wave comes with the hybrid models, modular techniques and ensemble models which has revolutionize into contemporary streamflow forecasting. With the latest hybrid models, results have shown retrieval of additional information are made possible that lead to better conclusion. This paper has suggested the research gaps and some future recommendations. It will be useful for new researcher and existing one to identify and be acquainted with the trends of different techniques of artificial intelligence.