Updated PyomoDAE notebooks (typos) and several contrib notebooks

Author: adowling2

notebooks/3/PyomoDAE_TCLab.ipynb

@@ -6,7 +6,7 @@
    "source": [
     "# Pyomo.DAE Example: Temperature Control Lab\n",
     "\n",
-    "This page is [adapted from our process control class](https://github.com/ndcbe/controls/blob/main/notebooks/06.02-Simulation-and-Open-Loop-Optimal-Control.ipynb) at Notre Dame; it was developed by Prof. Jeff Kantor."
+    "This page is [adapted from our process control class](https://github.com/ndcbe/controls/blob/main/notebooks/6/Simulation-and-Open-Loop-Optimal-Control.ipynb) at Notre Dame; it was developed by Prof. Jeff Kantor."
    ]
   },
   {

notebooks/3/PyomoDAE_car.ipynb

@@ -85,15 +85,15 @@
     "Mathematically, you want to solve the following optimal control problem:\n",
     "\n",
     "$$\\begin{align*}\n",
-    "\\min_{u} \\quad & t_f \\\\\n",
+    "\\min_{u,t_f} \\quad & t_f \\\\\n",
     "\\mathrm{s.t.} \\quad & \\frac{dx}{dt} = v \\\\\n",
     "& \\frac{dv}{dt} = u - R v^2 \\\\\n",
     "& x(t=0) = 0, ~~ x(t=t_f) = L \\\\\n",
     "& v(t=0) = 0, ~~ v(t=t_f) = 0 \\\\\n",
     "& -3 \\leq u \\leq 1\n",
     "\\end{align*}$$\n",
     "\n",
-    "where $a$ is the acceleration/braking (your control variable) and $R$ is the drag coefficient (parameter)."
+    "where $u$ is the acceleration/braking (your control variable) and $R$ is the drag coefficient (parameter)."
    ]
   },
   {
@@ -107,7 +107,7 @@
     "Let $t = \\tau \\cdot t_f$ where $\\tau \\in [0,1]$. Thus $dt = t_f d\\tau$. The optimal control problem becomes:\n",
     "\n",
     "$$\\begin{align*}\n",
-    "\\min_{u} \\quad & t_f \\\\\n",
+    "\\min_{u,t_f} \\quad & t_f \\\\\n",
     "\\mathrm{s.t.} \\quad & \\frac{dx}{d\\tau} = t_f v \\\\\n",
     "& \\frac{dv}{d\\tau} = t_f (u - R v^2) \\\\\n",
     "& x(\\tau = 0) = 0, ~~ x(\\tau = 1) = L \\\\\n",
@@ -960,7 +960,7 @@
    "provenance": []
   },
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "summer2024",
    "language": "python",
    "name": "python3"
   },

notebooks/3/PyomoDAE_example.ipynb

@@ -737,7 +737,7 @@
    "provenance": []
   },
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "summer2024",
    "language": "python",
    "name": "python3"
   },

notebooks/5/Parmest-tutorial.ipynb

@@ -13,7 +13,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 24,
+   "execution_count": null,
    "metadata": {
     "colab": {
      "base_uri": "https://localhost:8080/"
@@ -31,12 +31,10 @@
     "    import helper\n",
     "    helper.install_idaes()\n",
     "    helper.install_ipopt()\n",
-    "    helper.set_plotting_style()\n",
-    "\n",
     "else:\n",
-    "    # Make sure helper is installed locally on your system in the same file path.\n",
+    "    sys.path.insert(0, '../')\n",
     "    import helper\n",
-    "    helper.set_plotting_style()"
+    "helper.set_plotting_style()"
    ]
   },
   {

notebooks/contrib/Deterministic_Global_Optimization.ipynb

@@ -2233,7 +2233,12 @@
     "* Solvers like Couenne or BARON have the capability to find the global optima of certain NLPs.\n",
     "* Methods like Spatial branch and bound can guarantee the optimiality of solutions found for non linear programs with poor initialization given the complexity of the GO problem falls within certain assumptions.\n",
     "*Spatial branch and bound methods guarantee convergence but not in a finite number of steps.\n",
-    "* Spatial Branch and Bound Methods partition the feasible region into subregions and enumerate each subregions optimal solution. These solutions converge to a globally optimal solution."
+    "* Spatial Branch and Bound Methods partition the feasible region into subregions and enumerate each subregions optimal solution. These solutions converge to a globally optimal solution.\n",
+    "\n",
+    "Future work ideas:\n",
+    "* Experiment with different settings in Counene\n",
+    "* Add instructions on running BARON on the [CRC](https://crc.nd.edu)\n",
+    "* Show how solver time (Counene versus Ipopt) changes with the number of circles."
    ]
   },
   {

notebooks/contrib/Disaster_Response_Plan.ipynb

@@ -1502,8 +1502,18 @@
    "source": [
     "Through this example, we were able to go beyond the reference material, by creating a new and unique application of the optimization problem. By applying this to a real life situation, we were able to create a simplified model to represent a real life situation that could be expanded to create real social good.\n",
     "\n",
-    "For example, government planning can be used to understand based on existing data, what regions to distribute to which can expediate the process."
+    "For example, government planning can be used to understand based on existing data, what regions to distribute to which can expediate the process.\n",
+    "\n",
+    "Future Work Ideas:\n",
+    "* Refine the Monte Carlo sensitivity analysis\n",
+    "* Adjust input data such that there will be a supply shortage at the final time"
    ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "5592440c",
+   "metadata": {},
+   "source": []
   }
  ],
  "metadata": {

notebooks/contrib/advanced_stochastic_programming.ipynb

@@ -1452,7 +1452,9 @@
     "* Gain a deeper understanding of two-stage stochastic optimization, focusing on its application to agricultural systems under uncertainty\n",
     "* Interpret and analyze optimization results, drawing meaningful conclusions about resource allocation and decision-making in uncertain environments\n",
     "\n",
-    "A major takeaway from this work is the recognition that alternative numerical methods, such as the half-grid approach, can effectively address the \"curse of dimensionality\" associated with full-grid methods. By accounting for the reality that perfect information is rarely available, we demonstrated how modeling uncertainty using Gaussian distributions leads to more realistic scenarios. Finally, this project highlighted the significance of two-stage stochastic optimization in tackling real-world problems, emphasizing its role in enabling robust and informed decision-making under uncertainty."
+    "A major takeaway from this work is the recognition that alternative numerical methods, such as the half-grid approach, can effectively address the \"curse of dimensionality\" associated with full-grid methods. By accounting for the reality that perfect information is rarely available, we demonstrated how modeling uncertainty using Gaussian distributions leads to more realistic scenarios. Finally, this project highlighted the significance of two-stage stochastic optimization in tackling real-world problems, emphasizing its role in enabling robust and informed decision-making under uncertainty.\n",
+    "\n",
+    "Future work: refactor code to use [blocks](../../notebooks/4/blocks.ipynb)."
    ]
   },
   {

notebooks/contrib/hot_air_balloon.ipynb

@@ -1284,7 +1284,11 @@
     "* Pyomo continues to be a powerful tool in solving complex dynamic optimization problems.\n",
     "* Multi-objective optimization provides control over desired outcomes and products.\n",
     "* Multi-objective optimization identifies solutions that are robust to stakeholders decision.\n",
-    "* Proper modeling of the problem is key to obtaining quality results."
+    "* Proper modeling of the problem is key to obtaining quality results.\n",
+    "\n",
+    "Future work ideas:\n",
+    "* Compare the Pareto sets generated with the two methods\n",
+    "* Add another example with a non-convex Pareto set to highlight the difference between the methods"
    ]
   },
   {

notebooks/contrib/multi_objective.ipynb

@@ -2931,7 +2931,12 @@
     "\n",
     "The results further highlight that sugar and ethanol are particularly sensitive to regulatory constraints, with the regulated market setting a ceiling on potential earnings, especially when production surpluses increase supply. This condition may compel producers to prioritize free markets when aiming for higher profitability, albeit with increased exposure to market risks.\n",
     "\n",
-    "Overall, the model provides a comparative framework, guiding decision-makers to balance between predictable returns and potential profits within an uncertain market environment. This perspective is especially valuable for industries where the regulated and free markets can operate in tandem, allowing for diversified approaches that mitigate risk while maximizing revenue opportunities.\n"
+    "Overall, the model provides a comparative framework, guiding decision-makers to balance between predictable returns and potential profits within an uncertain market environment. This perspective is especially valuable for industries where the regulated and free markets can operate in tandem, allowing for diversified approaches that mitigate risk while maximizing revenue opportunities.\n",
+    "\n",
+    "Future work ideas:\n",
+    "* Add more narrative at the beginning to discuss how this goes beyond the previous example/notebook\n",
+    "* More clearly explain (using mathematical symbols) how the uncertainty model is being constructed\n",
+    "* Explore [advanced stochastic programming](../contrib/advanced_stochastic_programming.ipynb) methods.\n"
    ]
   },
   {