{"id":21437789,"url":"https://github.com/ymorsi7/seismiclstm","last_synced_at":"2025-03-16T23:42:30.528Z","repository":{"id":244159845,"uuid":"810451724","full_name":"ymorsi7/SeismicLSTM","owner":"ymorsi7","description":"Mitigating the high computational costs associated with applying Bayesian model updating in inverse problems / Uncertainty Quantification and Efficient Sensitivity Analysis by using Surrogate Models","archived":false,"fork":false,"pushed_at":"2024-06-13T04:04:02.000Z","size":57978,"stargazers_count":2,"open_issues_count":0,"forks_count":0,"subscribers_count":2,"default_branch":"main","last_synced_at":"2025-01-23T09:44:00.142Z","etag":null,"topics":["bayesian","ddml","lstm","pbsd","seismic","seismology","structural-engineering","surrogate-modelling","surrogate-models"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/ymorsi7.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-06-04T18:13:39.000Z","updated_at":"2024-12-28T16:21:33.000Z","dependencies_parsed_at":"2024-06-13T07:12:19.129Z","dependency_job_id":null,"html_url":"https://github.com/ymorsi7/SeismicLSTM","commit_stats":null,"previous_names":["ymorsi7/lstmsfornonlinearstructuralsystems","ymorsi7/seismiclstm"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ymorsi7%2FSeismicLSTM","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ymorsi7%2FSeismicLSTM/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ymorsi7%2FSeismicLSTM/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ymorsi7%2FSeismicLSTM/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/ymorsi7","download_url":"https://codeload.github.com/ymorsi7/SeismicLSTM/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":243950798,"owners_count":20373664,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["bayesian","ddml","lstm","pbsd","seismic","seismology","structural-engineering","surrogate-modelling","surrogate-models"],"created_at":"2024-11-23T00:29:20.828Z","updated_at":"2025-03-16T23:42:30.506Z","avatar_url":"https://github.com/ymorsi7.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Application of Long Short-Term Memory (LSTM) Networks-Based Surrogate Modeling for Nonlinear Structural Systems\n\n#### Abdoul Aziz Sandotin Coulibaly, Enrique Simbort Zeballos, Yusuf Morsi, Ramin Sarange\n\n## Abstract\nPerformance-based seismic design (PBSD) of structural systems relies on computationally expensive high-fidelity finite element (FE) models to predict how structures will respond to seismic excitation. For risk-based assessments, FE response simulations must be run thousands of times with different realizations of the sources of uncertainty. Consequently, data-driven machine learning (DDML) surrogate models have gained prominence as fast emulators for predicting seismic structural responses in probabilistic analyses. This paper leverages deep Long Short-Term Memory (LSTM) networks, known for their powerful and flexible framework for time series prediction tasks. The advantages of using LSTM networks include their ability to model continuous-time processes, adapt to varying temporal resolutions, maintain implicit memory of past information, model complex nonlinear dynamics, perform interpolation and extrapolation, handle noisy data robustly, and scale effectively to high-dimensional datasets. The effectiveness of the proposed method is validated through three proof-of-concept studies: one involving a linear elastic 2D 8-degree-of-freedom (DoF) shear building model, a nonlinear single degree of freedom system (NL-SDoF), and a 2D nonlinear 3DoF shear building model. The findings indicate that the proposed LSTM network is a promising, dependable, and computationally efficient technique for predicting nonlinear structural responses.\n\n![Test-bed structures for surrogate modeling study](files/testbed.png)\n\n![The LSTM network architecture: Full sequence to sequence LSTM network (LSTM-f)](files/lstm.png)\n\n## Instructions to Set Up and Run the LSTM Model\n\n1. **Clone the Repository**\n   \n   Clone this repository to your local machine using the following command:\n   \n   ```bash\n   git clone https://github.com/ymorsi7/LSTMsForNonlinearStructuralSystems.git\n   cd LSTMsForNonlinearStructuralSystems\n   ```\n2. **Install Dependencies**\n   Ensure you have Python 3.x installed on your system. Install the required dependencies using pip:\n   ```bash\n   pip install numpy matplotlib tensorflow keras scikit-learn joblib\n   ```\n3. **Prepare the Data** \u003cbr\u003e\n\n   Place the MATLAB data file `data_2DOF_SB_BWWN.mat` in the `LSTMsForNonlinearStructuralSystems` directory. This file should be available in the repository or provided separately.\n\n4. **Run the Python File**\n   \n   Execute the provided Python script `2DOF_ShearBuild_LSTM_f.py` to train the LSTM model:\n\n   ```bash\n   python 2DOF_ShearBuild_LSTM_f.py\n   ```\n\n  This will:\n  - Load the preprocessed data from the MATLAB file.\n  - Normalize the data using MinMaxScaler.\n  - Set up and train the LSTM model.\n  - Evaluate the model performance.\n  - Save the best-performing model.\n\n5. **A/B Testing**\n   The script also includes a section for A/B testing different LSTM architectures. This can be executed within the same script to compare performance metrics between two models.\n\n## Report\nTo read our report, [click here](files/paper.pdf)\n\n## Results\nThe results are much more interpretable with the context on the full report (linked above). However, here are some of our output images:\n\n### Linear Elastic 8-Story Shear Building\n![elsatic 8 story shear building](files/1.png)\n\n### Nonlinear Inelastic Single Degree Of Freedom (SDOF) Structure\n![sdof](files/2.png)\n\n### Nonlinear Inelastic Multi Degree Of Freedom (MDOF) Structure\n![mdof](files/3.png)\n\n### A/B Testing\n![ab testing](files/ab.png)\n\n\n\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fymorsi7%2Fseismiclstm","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fymorsi7%2Fseismiclstm","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fymorsi7%2Fseismiclstm/lists"}