GT China coal model

Overview

GT China coal model

The full version of a China coal transport model with a very high spatial reslution.

What it does

The code works in a few steps:

  1. Take easily understandable and readable xlsx input files on networks, plants, demand etc, and create project build files form this (done in R).
  2. Take the build files and create an LP problem file from it (in python, either locally or on AWS Sagemaker).
  3. Solve the problem from the LP problem file, and write solution to a txt file (either in Cplex interactive or in python).
  4. Process the solution txt file and write to easily understandable and readable xlsx (in R). The packages required to run these scripts are included in the environment.yml (for python) and the renv.lock file (for R; this first requires installation of renv package: https://rstudio.github.io/renv/articles/renv.html; after installation run renv:restore()).

The model

The model optimizes for a minimum cost of production + transport + transmission.
Production meaning coal mining costs, transport meaning rail/truck/riverborne/ocean-going transport and handling costs, and transmission meaning inter-provincial transport of electricity via UHV cables.

Constraints in the optimization

The constraints in the mini testbench are the same as in the full model. These are:

  • Mines (or any other node) cannot supply types of coal they do not produce.
  • The flow of coal of each type out of a node cannot exceed flows of coal of each type into a node plus supply by the node (with supply being non-zero only for mines).
  • The energy content of the supply and the flows of coal of each type into a node have to be at least equal to the demand for electricity, plus other thermal coal demand, plus the energy content of flows of coal of each type out of a node. Note that only mines can supply coal, all demand for electrical power occurs in provincial demand nodes, and demand for other thermal coal is placed at city-level nodes.
  • The amount of hard coking coal (HCC) flowing into a node has to at least be equal to the steel demand multiplied by 0.581. Note that all steel demand is placed in provincial level steel demand nodes, which are connecte with uni-directional links from steel plants to steel demand nodes. This means no coal can flow out of a steel demand node and we do not need further forumalae for mass balances. Also note that we presume a mix of coking coal need to produce a ton of steel of 581 kg Hard coking coal (HCC), 176 kg of soft coking coal (SCC), and 179 kg of pulverized coal for injection (PCI).
  • The amount of soft coking coal (SCC) flowing into a node has to at least be equal to the amount of HCC flowing into that node, mulitplied with 0.581/0.176.
  • The amount of pulverized coal for injection (PCI) flowing into a node has to at least be equal to the amount of HCC flowing into that node, mulitplied with 0.581/0.179.
  • The total volume of all coal types transported along a link cannot exceed the transport capacity of that link. Note that this constraint is applied only to those links with a non-infinite transport capacity. In practive this means rail links are assumed to have a transport capacity, ocean routes, rivers, and road links are assumed to have infinite capacity.
  • The total amount of energy transported along a link cannot exceed the transmission capacity of that link. That is, the amount of each coal type multiplied with the energy content of each coal type cannot exceed the electrical transmission capacity of links. This constraint is applied only links between power plant units and provincial electricity demand nodes, as well as UHV transmission links between provincial electricity demand nodes. These are the only links along which electrical energy is transported. All other links transport physcial quantities of coal. This line simultaneously deals with the production capapcity (MW) and conversion efficiency of power plants: the energy transported over a link cannot exceed the volume of each coal type, multiplied with the energy content of each coal type, multiplied with the energy conversion factor of the link. For links between coal fired power plant units and provincial electricity demand nodes, this is equal to the conversion effincy of the power plant unit. For UHV transmission links between two provincial level electricity demand nodes, this is equal to (1- transmission losses) over that UHV line, with transmission losses calculated based on transmission distance and a benchmark loss for UHV-DC or UHV-AC lines.
  • The handling capacity of ports cannot be exceeded. Specifically, the total amount of coal flowing out of a port cannot exceed its handling capacity.
  • The production capacity of steel plants cannot be exceeded. Specifically, the total amount of hard coking coal, soft coking coal, and pulverized coal for injection flowing out of a steel plant node (and towards a provincial steel demand node) cannot exceed the steel plant's production capacity multiplied by 0.581+0.176+0.179, the mix of different coking coals needed to produce steel.

Technical notes

  • All transport costs are pre-calculated for each link, and include a fixed handling costs and a distance based transport cost, based on the type handling (origin and destaination) and type of transport (separate for rail, truck, riverborne, ocean-going. A small number of coal rail lines has specific handling and transport costs).
  • Some of the capacities are already reported in the input sheet for the edges. The physical transport capacity from this sheet is used. For capacities of ports, steel plants, and electrical transmission capacities, the data from the separate port/steel plant/electrical capacities sheets is used.
  • An exmaple lp file is included to make this reporsitory as self-conatined as possible. This lp file is zipped to stay within github file size limits.

Contributions

This project was developed by:

  • Jorrit Gosens: conceptualization, bulk of the data preparation, software implementation and debugging in R & Python;
  • Alex H. Turnbull: conceptualization, some data preparation, proof-reading and debugging of software.

License

MIT License as separately included.
In short, do what you want with this script, but refer to the original authors when you use or develop this code.

License

MIT License as separately included.
In short, do what you want with this script, but refer to the original authors when you use or develop this code.

GeneDisco is a benchmark suite for evaluating active learning algorithms for experimental design in drug discovery.

GeneDisco is a benchmark suite for evaluating active learning algorithms for experimental design in drug discovery.

22 Dec 12, 2022
This repository contains part of the code used to make the images visible in the article "How does an AI Imagine the Universe?" published on Towards Data Science.

Generative Adversarial Network - Generating Universe This repository contains part of the code used to make the images visible in the article "How doe

Davide Coccomini 9 Dec 18, 2022
Learning from Synthetic Shadows for Shadow Detection and Removal [Inoue+, IEEE TCSVT 2020].

Learning from Synthetic Shadows for Shadow Detection and Removal (IEEE TCSVT 2020) Overview This repo is for the paper "Learning from Synthetic Shadow

Naoto Inoue 67 Dec 28, 2022
Combine Tacotron2 and Hifi GAN to generate speech from text

EndToEndTextToSpeech Combine Tacotron2 and Hifi GAN to generate speech from text Download weights Hifi GAN - hifi_gan/checkpoint/ : pretrain 2.5M ste

Phạm Quốc Huy 1 Dec 18, 2021
Customer-Transaction-Analysis - This analysis is based on a synthesised transaction dataset containing 3 months worth of transactions for 100 hypothetical customers.

Customer-Transaction-Analysis - This analysis is based on a synthesised transaction dataset containing 3 months worth of transactions for 100 hypothetical customers. It contains purchases, recurring

Ayodeji Yekeen 1 Jan 01, 2022
ConformalLayers: A non-linear sequential neural network with associative layers

ConformalLayers: A non-linear sequential neural network with associative layers ConformalLayers is a conformal embedding of sequential layers of Convo

Prograf-UFF 5 Sep 28, 2022
Official codebase for Legged Robots that Keep on Learning: Fine-Tuning Locomotion Policies in the Real World

Legged Robots that Keep on Learning Official codebase for Legged Robots that Keep on Learning: Fine-Tuning Locomotion Policies in the Real World, whic

Laura Smith 70 Dec 07, 2022
Unofficial PyTorch implementation of Fastformer based on paper "Fastformer: Additive Attention Can Be All You Need"."

Fastformer-PyTorch Unofficial PyTorch implementation of Fastformer based on paper Fastformer: Additive Attention Can Be All You Need. Usage : import t

Hong-Jia Chen 126 Dec 06, 2022
Código de um painel de auto atendimento feito em Python.

Painel de Auto-Atendimento O intuito desse projeto era fazer em Python um programa que simulasse um painel de auto atendimento, no maior estilo Mac Do

Calebe Alves Evangelista 2 Nov 09, 2022
A simple rest api serving a deep learning model that classifies human gender based on their faces. (vgg16 transfare learning)

this is a simple rest api serving a deep learning model that classifies human gender based on their faces. (vgg16 transfare learning)

crispengari 5 Dec 09, 2021
Codes for AAAI 2022 paper: Context-aware Health Event Prediction via Transition Functions on Dynamic Disease Graphs

Context-Aware-Healthcare Codes for AAAI 2022 paper: Context-aware Health Event Prediction via Transition Functions on Dynamic Disease Graphs Download

LuChang 9 Dec 26, 2022
A Transformer-Based Siamese Network for Change Detection

ChangeFormer: A Transformer-Based Siamese Network for Change Detection (Under review at IGARSS-2022) Wele Gedara Chaminda Bandara, Vishal M. Patel Her

Wele Gedara Chaminda Bandara 214 Dec 29, 2022
SimplEx - Explaining Latent Representations with a Corpus of Examples

SimplEx - Explaining Latent Representations with a Corpus of Examples Code Author: Jonathan Crabbé ( Jonathan Crabbé 14 Dec 15, 2022

Source code for paper "Document-Level Relation Extraction with Adaptive Thresholding and Localized Context Pooling", AAAI 2021

ATLOP Code for AAAI 2021 paper Document-Level Relation Extraction with Adaptive Thresholding and Localized Context Pooling. If you make use of this co

Wenxuan Zhou 146 Nov 29, 2022
Reinforcement-learning - Repository of the class assignment questions for the course on reinforcement learning

DSE 314/614: Reinforcement Learning This repository containing reinforcement lea

Manav Mishra 4 Apr 15, 2022
Learning multiple gaits of quadruped robot using hierarchical reinforcement learning

Learning multiple gaits of quadruped robot using hierarchical reinforcement learning We propose a method to learn multiple gaits of quadruped robot us

Yunho Kim 17 Dec 11, 2022
Vehicle direction identification consists of three module detection , tracking and direction recognization.

Vehicle-direction-identification Vehicle direction identification consists of three module detection , tracking and direction recognization. Algorithm

5 Nov 15, 2022
Keras like implementation of Deep Learning architectures from scratch using numpy.

Mini-Keras Keras like implementation of Deep Learning architectures from scratch using numpy. How to contribute? The project contains implementations

MANU S PILLAI 5 Oct 10, 2021
Code for TIP 2017 paper --- Illumination Decomposition for Photograph with Multiple Light Sources.

Illumination_Decomposition Code for TIP 2017 paper --- Illumination Decomposition for Photograph with Multiple Light Sources. This code implements the

QAY 7 Nov 15, 2020