Electricity Generation Forecast of Shanghai Municipal Solid Waste Based on Bidirectional Long Short-Term Memory Model

Bingchun Liu, Ningbo Zhang, Lingli Wang, Xinming ZhangInternational Journal of Environmental Research and Public Health 2022 · DOI

TL;DR

This paper introduces a Bidirectional Long Short-Term Memory (BiLSTM) model to accurately forecast Municipal Solid Waste (MSW) generation and subsequent electricity generation potential in Shanghai. By integrating six key influencing factors, the BiLSTM model achieves superior prediction accuracy compared to traditional methods, offering valuable insights for sustainable urban waste management and energy production.

What Problem Does This Paper Address?

The escalating volume of Municipal Solid Waste (MSW) presents a significant environmental and management challenge globally, particularly in rapidly urbanizing areas like Shanghai. Efficiently managing this waste, including its conversion into energy, is crucial for sustainable development. Accurate prediction of MSW generation is a prerequisite for optimizing waste-to-energy (WtE) facilities, bridging the gap between energy demands and environmental protection, and reducing reliance on landfills. Existing prediction models often struggle with the complex, non-linear dynamics of MSW generation, leading to uncertainties in planning and resource allocation for WtE technologies.

Methods

The core of this research lies in the application and development of a Bidirectional Long Short-Term Memory (BiLSTM) neural network model for MSW generation forecasting. BiLSTM is chosen for its ability to capture long-term dependencies and contextual information from sequential data by processing sequences in both forward and backward directions. This approach aims to overcome the limitations of simpler models like standard LSTM, which only consider unidirectional temporal relationships.

The study also incorporates a novel combination of influencing factors for MSW generation, including:

  • Economic Indicators: Gross Domestic Product (GDP), per capita disposable income, and per capita consumption expenditure.
  • Population Indicators: Population density and permanent resident population.
  • Social Indicators: Number of urban public transport vehicles.

These six factors are selected based on a review of existing literature and their established correlation with MSW generation. After predicting MSW generation using the BiLSTM model, the paper employs a specific formula to estimate the potential electricity generation from MSW incineration, considering the lower heating value (LHV) of different waste components and conversion efficiencies.

Datasets

The study utilizes historical data for Shanghai from 1978 to 2018, sourced from the Shanghai Statistical Yearbook. This dataset comprises annual data for the six identified influencing factors:

  • Gross regional product
  • Permanent resident population at year-end
  • Per capita disposable income of urban households
  • Per capita consumption expenditure of urban households
  • Number of urban public transport vehicles operating
  • Population density

This data is split into training (1978-2011) and testing (2012-2019) sets for model evaluation. The paper does not introduce a new benchmark dataset but rather uses publicly available historical data for a specific city.

Key Insights and Ideas

  • BiLSTM's Superiority for MSW Forecasting: The research strongly suggests that BiLSTM models offer a significant improvement in prediction accuracy for MSW generation compared to traditional statistical methods and even simpler deep learning models like LSTM. This is attributed to its ability to capture complex temporal dependencies from multiple influencing factors.
  • Holistic Factor Integration: The selection of a diverse set of six influencing factors (economic, social, and demographic) provides a more comprehensive understanding of MSW generation drivers than models relying on fewer variables. This multi-faceted approach enhances the robustness of the predictions.
  • Scenario-Based Forecasting for Uncertainty Reduction: By developing three distinct scenarios (low-growth, base growth, and high-growth), the study effectively addresses the inherent uncertainty in future MSW generation. This allows for a range of potential outcomes, aiding in more resilient planning.
  • Direct Link to Energy Potential: The paper successfully bridges the gap between MSW generation prediction and its practical application in estimating electricity generation potential. This direct link makes the findings highly relevant for energy policy and urban planning.

Key Results

  • Model Accuracy: The BiLSTM model achieved a Mean Absolute Percentage Error (MAPE) of 7.390%, a Mean Absolute Error (MAE) of 42.31, and a Root Mean Square Error (RMSE) of 63.32. These metrics indicate a high level of accuracy in predicting MSW generation.
  • MSW Generation Forecast (2025): Under the three scenarios, the projected MSW production in Shanghai by 2025 ranges from a minimum of 8.82 million tons (low-growth) to a maximum of 17.35 million tons (high-growth).
  • Electricity Generation Forecast (2025): Correspondingly, the estimated electricity generation from MSW in Shanghai by 2025 is projected to range from 260.668 GWh/y (low-growth) to 512.752 GWh/y (high-growth).
  • Comparative Performance: The BiLSTM model outperformed other tested models, including SVR, GRU, LSTM, Bi-SVR, and Bi-GRU, in terms of prediction accuracy (lowest MAPE).

Limitations and Future Work

  • Data Granularity: The study relies on annual data. Future work could explore higher temporal resolution data (e.g., monthly or daily) to capture more nuanced patterns.
  • Geographic Specificity: The model is trained and evaluated on Shanghai data. While the methodology is transferable, its direct applicability to other cities would require recalibration with local data and influencing factors.
  • Waste Composition Dynamics: While the study considers waste components for electricity generation calculation, a more dynamic modeling of waste composition changes over time could further refine energy generation estimates.
  • Policy Integration: The authors suggest that the findings can serve as a scientific information source for environmental sustainability decision-making. Future research could explore how these predictions can be directly integrated into policy frameworks for waste management and renewable energy development.

Summary updated

Abstract

The accurate prediction of Municipal Solid Waste (MSW) electricity generation is very important for the fine management of a city. This paper selects Shanghai as the research object, through the construction of a Bidirectional Long Short-Term Memory (BiLSTM) model, and chooses six influencing factors of MSW generation as the input indicators, to realize the effective prediction of MSW generation. Then, this study obtains the MSW electricity generation capacity in Shanghai by using the aforementioned prediction results and the calculation formula of theMSW electricity generation. The experimental results show that, firstly, the mean absolute error (MAE), mean absolute percentage error (MAPE), and root mean square error (RMSE) values of the BiLSTM model are 42.31, 7.390, and 63.32. Second, it is estimated that by 2025, the maximum and minimum production of MSW in Shanghai will be 17.35 million tons and 8.82 million tons under the three scenarios. Third, it is predicted that in 2025, the maximum and minimum electricity generation of Shanghai MSW under the three scenarios will be 512.752 GWh/y and 260.668 GWh/y. Finally, this paper can be used as a scientific information source for environmental sustainability decision-making for domestic MSW electricity generation technology.
BibTeX
@article{liu2022electricity,
  author = {Bingchun Liu and Ningbo Zhang and Lingli Wang and Xinming Zhang},
  title = {{Electricity Generation Forecast of Shanghai Municipal Solid Waste Based on Bidirectional Long Short-Term Memory Model}},
  journal = {International Journal of Environmental Research and Public Health},
  year = {2022},
  doi = {10.3390/ijerph19116616},
}
Figure