radu/Master-Project
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
Master-Project/Notebooks/50_mpc_formulation.ipynb

2376 lines
94 KiB
Text
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
"from pathlib import Path\n",
"from shutil import copyfile\n",
"import pickle"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Load the general math/data manipulation packages"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"import numpy as np\n",
"import pandas as pd"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Load the packages related to the Gaussian Process Regressor:"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
"import gpflow\n",
"import tensorflow as tf"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [],
"source": [
"from gpflow.utilities import print_summary"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [],
"source": [
"gpflow.config.set_default_summary_fmt(\"notebook\")"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [],
"source": [
"tf.config.set_visible_devices([], 'GPU')"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [],
"source": [
"import matplotlib.pyplot as plt"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Load the CasADi package used for optimization:"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [],
"source": [
"import casadi"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Import MATLAB engine and start it in the background since this takes a while:"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [],
"source": [
"import matlab.engine"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [],
"source": [
"eng = matlab.engine.start_matlab()"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"<matlab.engine.futureresult.FutureResult at 0x7f1ca0156fa0>"
]
},
"execution_count": 11,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"eng.load_system(\"../Simulink/polydome\", background = True)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Copy the experimental data set to the CARNOT input location: "
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [],
"source": [
"exp_id = 'Exp5'"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"'../Data/input_WDB.mat'"
]
},
"execution_count": 13,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"copyfile(f\"../Data/Experimental_data_WDB/{exp_id}_WDB.mat\", \"../Data/input_WDB.mat\")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Load the existing GP model"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [],
"source": [
"model_path = Path(Path.cwd(), 'model')"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {},
"outputs": [],
"source": [
"def load_gpr(model_path):\n",
" x_scaler = pickle.load(open(Path(model_path, 'x_scaler.pkl'), 'rb'))\n",
" m_params = pickle.load(open(Path(model_path, 'gp_params.gpf'), 'rb'))\n",
" m_data = pickle.load(open(Path(model_path, 'gp_data.gpf'), 'rb'))\n",
"\n",
" k = gpflow.kernels.SquaredExponential(lengthscales=([1] * m_data[0].shape[1])) + gpflow.kernels.Constant()\n",
"\n",
" m = gpflow.models.GPR(\n",
" data = m_data, \n",
" kernel = k, \n",
" mean_function = None\n",
" )\n",
" \n",
" gpflow.utilities.multiple_assign(m, m_params)\n",
" \n",
" return x_scaler, m"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<table>\n",
"<thead>\n",
"<tr><th>name </th><th>class </th><th>transform </th><th>prior </th><th>trainable </th><th>shape </th><th>dtype </th><th>value </th></tr>\n",
"</thead>\n",
"<tbody>\n",
"<tr><td>GPR.kernel.kernels[0].variance </td><td>Parameter</td><td>Softplus </td><td> </td><td>True </td><td>() </td><td>float64</td><td>284790.175758825 </td></tr>\n",
"<tr><td>GPR.kernel.kernels[0].lengthscales</td><td>Parameter</td><td>Softplus </td><td> </td><td>True </td><td>(7,) </td><td>float64</td><td>[8.23155895e+04, 1.32876975e+05, 2.21167354e+02...</td></tr>\n",
"<tr><td>GPR.kernel.kernels[1].variance </td><td>Parameter</td><td>Softplus </td><td> </td><td>True </td><td>() </td><td>float64</td><td>220815.62255492577 </td></tr>\n",
"<tr><td>GPR.likelihood.variance </td><td>Parameter</td><td>Softplus + Shift</td><td> </td><td>True </td><td>() </td><td>float64</td><td>0.1263230455247145 </td></tr>\n",
"</tbody>\n",
"</table>"
],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"x_scaler, model = load_gpr(model_path)\n",
"print_summary(model)"
]
},
{
"cell_type": "code",
"execution_count": 85,
"metadata": {},
"outputs": [],
"source": [
"def get_gpr_horizon_array(W, x0, u):\n",
" \n",
" \n",
" n_rows = W.shape[0]\n",
" \n",
" n_w = W.shape[1]\n",
" n_inputs = u.shape[1]\n",
" n_cols = n_w + n_inputs + x0.shape[1]\n",
" \n",
" X = np.zeros([n_rows, n_cols])\n",
" X[:, :2] = W\n",
" X[0, (n_w + n_inputs):] = x0\n",
" X[:, 2] = u.reshape((-1,))\n",
" \n",
" # autoregressive inputs\n",
" X[1: , 3] = X[:-1, 2]\n",
" \n",
" for idx in range(1, n_rows):\n",
" x_sc = x_scaler.transform(X[idx-1, :].reshape((1, -1)))\n",
" X[idx, 4], _ = model.predict_y(x_sc)\n",
" X[idx, 5:] = X[idx -1, 4:-1]\n",
" \n",
" return X"
]
},
{
"cell_type": "code",
"execution_count": 86,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"array([[ 1. , 1. , 3. , 2. , 2. ,\n",
" 2. , 2. ],\n",
" [ 1. , 1. , 3. , 3. , 3.16887448,\n",
" 2. , 2. ],\n",
" [ 1. , 1. , 3. , 3. , 6.02687796,\n",
" 3.16887448, 2. ],\n",
" [ 1. , 1. , 3. , 3. , 11.22601695,\n",
" 6.02687796, 3.16887448],\n",
" [ 1. , 1. , 3. , 3. , 19.6304083 ,\n",
" 11.22601695, 6.02687796]])"
]
},
"execution_count": 86,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"test_W = np.ones([5, 2])\n",
"test_x0 = 2 * np.ones([1, 4])\n",
"test_u0 = 3 * np.ones([5, 1])\n",
"\n",
"test_X = get_gpr_horizon_array(test_W, test_x0, test_u0)\n",
"test_X"
]
},
{
"cell_type": "code",
"execution_count": 87,
"metadata": {},
"outputs": [],
"source": [
"n_states = model.data[0].shape[1]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Set up the CasADi optimization problem"
]
},
{
"cell_type": "code",
"execution_count": 88,
"metadata": {},
"outputs": [],
"source": [
"# Package the regression model in a CasADi callback\n",
"class GPR(casadi.Callback):\n",
" def __init__(self, name, opts={}):\n",
" casadi.Callback.__init__(self)\n",
" self.construct(name, opts)\n",
" \n",
" # Number of inputs and outputs\n",
" def get_n_in(self): return 1\n",
" def get_n_out(self): return 1\n",
"\n",
" def get_sparsity_in(self,i):\n",
" return casadi.Sparsity.dense(n_states,1)\n",
"\n",
" def eval(self, arg):\n",
" x_input = x_scaler.transform(np.array(arg[0]).reshape(1, -1))\n",
" #x_input = np.array(arg[0]).reshape(1, -1)\n",
" [mean, _] = model.predict_y(x_input)\n",
" return [mean.numpy()]"
]
},
{
"cell_type": "code",
"execution_count": 89,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"GPR:(i0[7])->(o0) CallbackInternal\n"
]
}
],
"source": [
"# Instantiate the Callback (make sure to keep a reference to it!)\n",
"gpr = GPR('GPR', {\"enable_fd\":True})\n",
"print(gpr)"
]
},
{
"cell_type": "code",
"execution_count": 90,
"metadata": {},
"outputs": [],
"source": [
"T_set = 20\n",
"N_horizon = 5"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Define optimization variables"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"<img src=\"Images/gpr_X.JPG\" alt=\"Gaussian Process Input Array\" width=\"750\"/>"
]
},
{
"cell_type": "code",
"execution_count": 91,
"metadata": {},
"outputs": [],
"source": [
"X = casadi.MX.sym(\"X\", N_horizon, n_states)\n",
"W = casadi.MX.sym(\"W\", N_horizon, 2)\n",
"x0_lags = casadi.MX.sym(\"lags\", 1, n_states - 3)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Impose initial lags:"
]
},
{
"cell_type": "code",
"execution_count": 92,
"metadata": {},
"outputs": [],
"source": [
"g = casadi.vec(X[0,3:] - x0_lags)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Impose disturbances:"
]
},
{
"cell_type": "code",
"execution_count": 93,
"metadata": {},
"outputs": [],
"source": [
"g = casadi.vertcat(\n",
" g,\n",
" casadi.vec(X[:, :2] - W)\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Compute objective:"
]
},
{
"cell_type": "code",
"execution_count": 94,
"metadata": {},
"outputs": [],
"source": [
"Y = gpr(X.T)\n",
"J = casadi.dot(Y - T_set, Y - T_set)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Fix input/output lags between time steps (equality constraints):"
]
},
{
"cell_type": "code",
"execution_count": 95,
"metadata": {},
"outputs": [],
"source": [
"for idx in range(1, N_horizon):\n",
" g = casadi.vertcat(\n",
" g,\n",
" X[idx, 3] - X[idx-1, 2],\n",
" X[idx, 4] - gpr(X[idx-1,:]),\n",
" X[idx, 5] - X[idx-1, 4],\n",
" X[idx, 6] - X[idx-1, 5]\n",
" )"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Impose input inequality constraints:"
]
},
{
"cell_type": "code",
"execution_count": 96,
"metadata": {},
"outputs": [],
"source": [
"g = casadi.vertcat(\n",
" g,\n",
" X[:, 2]\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Compile the optimization problem"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Compile the parameter vector:"
]
},
{
"cell_type": "code",
"execution_count": 97,
"metadata": {},
"outputs": [],
"source": [
"p = casadi.vertcat(\n",
" casadi.vec(W),\n",
" casadi.vec(x0_lags)\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 98,
"metadata": {},
"outputs": [],
"source": [
"prob = {\"x\": casadi.vec(X), \"f\": J, \"p\": p, \"g\": g}\n",
"options = {\"ipopt\": {\"hessian_approximation\": \"limited-memory\", \"max_iter\": 10,\n",
" #\"acceptable_tol\": 1e-6, \"tol\": 1e-6,\n",
" #\"linear_solver\": \"SPRAL\",\n",
" #\"acceptable_obj_change_tol\": 1e-5, \n",
" #\"mu_strategy\": \"adaptive\",\n",
" \"expect_infeasible_problem\": \"yes\"\n",
" }}"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Get the solver object:"
]
},
{
"cell_type": "code",
"execution_count": 99,
"metadata": {},
"outputs": [],
"source": [
"solver = casadi.nlpsol(\"solver\",\"ipopt\",prob, options)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Compute the `lbg` `ubg` since they are always the same:"
]
},
{
"cell_type": "code",
"execution_count": 100,
"metadata": {},
"outputs": [],
"source": [
"Pel_max = 6300\n",
"COP_heat = 5\n",
"COP_cool = 5\n",
"u_min = - COP_cool * Pel_max\n",
"u_max = COP_heat * Pel_max"
]
},
{
"cell_type": "code",
"execution_count": 101,
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"#u_min_sc = x_scaler.transform(np.array([1, 1, u_min, u_min, 1, 1, 1]).reshape((1, -1)))[0, 2]\n",
"#u_max_sc = x_scaler.transform(np.array([1, 1, u_max, u_max, 1, 1, 1]).reshape((1, -1)))[0, 2]"
]
},
{
"cell_type": "code",
"execution_count": 102,
"metadata": {},
"outputs": [],
"source": [
"real_lbg = [0] * (4 + 2 * N_horizon + 4 * (N_horizon - 1)) + [u_min] * (N_horizon)\n",
"real_ubg = [0] * (4 + 2 * N_horizon + 4 * (N_horizon - 1)) + [u_max] * (N_horizon)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Load CARNOT building with MATLAB backend"
]
},
{
"cell_type": "code",
"execution_count": 103,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Power</th>\n",
" <th>Setpoint</th>\n",
" <th>OutsideTemp</th>\n",
" <th>SupplyTemp</th>\n",
" <th>InsideTemp</th>\n",
" <th>SolRad</th>\n",
" <th>Heat</th>\n",
" <th>SimulatedTemp</th>\n",
" </tr>\n",
" <tr>\n",
" <th>timestamp</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>2017-06-30 20:00:00+02:00</th>\n",
" <td>8.620690</td>\n",
" <td>22.5</td>\n",
" <td>19.0</td>\n",
" <td>22.1</td>\n",
" <td>22.283333</td>\n",
" <td>178.618267</td>\n",
" <td>8.620690</td>\n",
" <td>22.283285</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-06-30 20:05:00+02:00</th>\n",
" <td>8.620690</td>\n",
" <td>22.5</td>\n",
" <td>19.0</td>\n",
" <td>22.1</td>\n",
" <td>22.233333</td>\n",
" <td>135.122100</td>\n",
" <td>8.620690</td>\n",
" <td>22.266734</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-06-30 20:10:00+02:00</th>\n",
" <td>7.566667</td>\n",
" <td>22.5</td>\n",
" <td>19.0</td>\n",
" <td>22.1</td>\n",
" <td>22.366667</td>\n",
" <td>126.219967</td>\n",
" <td>7.566667</td>\n",
" <td>22.258667</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-06-30 20:15:00+02:00</th>\n",
" <td>8.172414</td>\n",
" <td>22.5</td>\n",
" <td>19.0</td>\n",
" <td>22.1</td>\n",
" <td>22.366667</td>\n",
" <td>123.772467</td>\n",
" <td>8.172414</td>\n",
" <td>22.255146</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-06-30 20:20:00+02:00</th>\n",
" <td>8.379310</td>\n",
" <td>22.5</td>\n",
" <td>17.5</td>\n",
" <td>22.4</td>\n",
" <td>22.566667</td>\n",
" <td>106.163600</td>\n",
" <td>-25.137931</td>\n",
" <td>22.255204</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Power Setpoint OutsideTemp SupplyTemp \\\n",
"timestamp \n",
"2017-06-30 20:00:00+02:00 8.620690 22.5 19.0 22.1 \n",
"2017-06-30 20:05:00+02:00 8.620690 22.5 19.0 22.1 \n",
"2017-06-30 20:10:00+02:00 7.566667 22.5 19.0 22.1 \n",
"2017-06-30 20:15:00+02:00 8.172414 22.5 19.0 22.1 \n",
"2017-06-30 20:20:00+02:00 8.379310 22.5 17.5 22.4 \n",
"\n",
" InsideTemp SolRad Heat SimulatedTemp \n",
"timestamp \n",
"2017-06-30 20:00:00+02:00 22.283333 178.618267 8.620690 22.283285 \n",
"2017-06-30 20:05:00+02:00 22.233333 135.122100 8.620690 22.266734 \n",
"2017-06-30 20:10:00+02:00 22.366667 126.219967 7.566667 22.258667 \n",
"2017-06-30 20:15:00+02:00 22.366667 123.772467 8.172414 22.255146 \n",
"2017-06-30 20:20:00+02:00 22.566667 106.163600 -25.137931 22.255204 "
]
},
"execution_count": 103,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df = pd.read_pickle(f\"../Data/CARNOT_output/{exp_id}_full.pkl\")\n",
"df.head()"
]
},
{
"cell_type": "code",
"execution_count": 104,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>time</th>\n",
" <th>timestamp</th>\n",
" <th>zenith</th>\n",
" <th>azimuth</th>\n",
" <th>dni</th>\n",
" <th>dhi</th>\n",
" <th>OutsideTemp</th>\n",
" <th>Tsky_rad</th>\n",
" <th>relative_humidity</th>\n",
" <th>precipitation</th>\n",
" <th>cloud_index</th>\n",
" <th>pressure</th>\n",
" <th>wind_speed</th>\n",
" <th>wind_direction</th>\n",
" <th>aoi</th>\n",
" <th>incidence_main</th>\n",
" <th>incidence_second</th>\n",
" <th>poa_direct</th>\n",
" <th>poa_diffuse</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>0</td>\n",
" <td>201706302000</td>\n",
" <td>77.058396</td>\n",
" <td>290.138593</td>\n",
" <td>461.587394</td>\n",
" <td>75.759999</td>\n",
" <td>19.0</td>\n",
" <td>13.0</td>\n",
" <td>50</td>\n",
" <td>-9999</td>\n",
" <td>0.5</td>\n",
" <td>96300</td>\n",
" <td>0</td>\n",
" <td>-9999</td>\n",
" <td>77.058396</td>\n",
" <td>-9999</td>\n",
" <td>-9999</td>\n",
" <td>103.376122</td>\n",
" <td>75.759999</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>300</td>\n",
" <td>201706302005</td>\n",
" <td>77.859576</td>\n",
" <td>290.978550</td>\n",
" <td>206.014761</td>\n",
" <td>92.042087</td>\n",
" <td>19.0</td>\n",
" <td>13.0</td>\n",
" <td>50</td>\n",
" <td>-9999</td>\n",
" <td>0.5</td>\n",
" <td>96300</td>\n",
" <td>0</td>\n",
" <td>-9999</td>\n",
" <td>77.859576</td>\n",
" <td>-9999</td>\n",
" <td>-9999</td>\n",
" <td>43.326628</td>\n",
" <td>92.042087</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>600</td>\n",
" <td>201706302010</td>\n",
" <td>78.655742</td>\n",
" <td>291.820500</td>\n",
" <td>205.632053</td>\n",
" <td>86.034806</td>\n",
" <td>19.0</td>\n",
" <td>13.0</td>\n",
" <td>50</td>\n",
" <td>-9999</td>\n",
" <td>0.5</td>\n",
" <td>96300</td>\n",
" <td>0</td>\n",
" <td>-9999</td>\n",
" <td>78.655742</td>\n",
" <td>-9999</td>\n",
" <td>-9999</td>\n",
" <td>40.448556</td>\n",
" <td>86.034806</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>900</td>\n",
" <td>201706302015</td>\n",
" <td>79.446620</td>\n",
" <td>292.664742</td>\n",
" <td>254.541692</td>\n",
" <td>77.502756</td>\n",
" <td>19.0</td>\n",
" <td>13.0</td>\n",
" <td>50</td>\n",
" <td>-9999</td>\n",
" <td>0.5</td>\n",
" <td>96300</td>\n",
" <td>0</td>\n",
" <td>-9999</td>\n",
" <td>79.446620</td>\n",
" <td>-9999</td>\n",
" <td>-9999</td>\n",
" <td>46.619695</td>\n",
" <td>77.502756</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>1200</td>\n",
" <td>201706302020</td>\n",
" <td>80.231464</td>\n",
" <td>293.511567</td>\n",
" <td>186.928631</td>\n",
" <td>74.726041</td>\n",
" <td>17.5</td>\n",
" <td>11.5</td>\n",
" <td>50</td>\n",
" <td>-9999</td>\n",
" <td>0.5</td>\n",
" <td>96300</td>\n",
" <td>0</td>\n",
" <td>-9999</td>\n",
" <td>80.231464</td>\n",
" <td>-9999</td>\n",
" <td>-9999</td>\n",
" <td>31.715871</td>\n",
" <td>74.726041</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" time timestamp zenith azimuth dni dhi \\\n",
"0 0 201706302000 77.058396 290.138593 461.587394 75.759999 \n",
"1 300 201706302005 77.859576 290.978550 206.014761 92.042087 \n",
"2 600 201706302010 78.655742 291.820500 205.632053 86.034806 \n",
"3 900 201706302015 79.446620 292.664742 254.541692 77.502756 \n",
"4 1200 201706302020 80.231464 293.511567 186.928631 74.726041 \n",
"\n",
" OutsideTemp Tsky_rad relative_humidity precipitation cloud_index \\\n",
"0 19.0 13.0 50 -9999 0.5 \n",
"1 19.0 13.0 50 -9999 0.5 \n",
"2 19.0 13.0 50 -9999 0.5 \n",
"3 19.0 13.0 50 -9999 0.5 \n",
"4 17.5 11.5 50 -9999 0.5 \n",
"\n",
" pressure wind_speed wind_direction aoi incidence_main \\\n",
"0 96300 0 -9999 77.058396 -9999 \n",
"1 96300 0 -9999 77.859576 -9999 \n",
"2 96300 0 -9999 78.655742 -9999 \n",
"3 96300 0 -9999 79.446620 -9999 \n",
"4 96300 0 -9999 80.231464 -9999 \n",
"\n",
" incidence_second poa_direct poa_diffuse \n",
"0 -9999 103.376122 75.759999 \n",
"1 -9999 43.326628 92.042087 \n",
"2 -9999 40.448556 86.034806 \n",
"3 -9999 46.619695 77.502756 \n",
"4 -9999 31.715871 74.726041 "
]
},
"execution_count": 104,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df_wdb = pd.read_pickle(f\"../Data/Experimental_python/{exp_id}_WDB.pkl\")\n",
"df_wdb.head()"
]
},
{
"cell_type": "code",
"execution_count": 105,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Experiment runtime: 208500\n"
]
}
],
"source": [
"runtime = df_wdb['time'].iloc[-1]\n",
"print(f\"Experiment runtime: {runtime}\")"
]
},
{
"cell_type": "code",
"execution_count": 106,
"metadata": {},
"outputs": [],
"source": [
"eng.workspace['t0'] = float(df['InsideTemp'][0])"
]
},
{
"cell_type": "code",
"execution_count": 107,
"metadata": {},
"outputs": [],
"source": [
"day_air_exchange_rate = 2.75\n",
"night_air_exchange_rate = 2.75"
]
},
{
"cell_type": "code",
"execution_count": 108,
"metadata": {},
"outputs": [],
"source": [
"air_exchange_rate = np.zeros((df_wdb.shape[0], 2))\n",
"air_exchange_rate[:, 0] = df_wdb['time']\n",
"air_exchange_rate[:, 1] = np.where(df['Power'] < 100, day_air_exchange_rate, night_air_exchange_rate)\n",
"eng.workspace['air_exchange_rate'] = matlab.double(air_exchange_rate.tolist())"
]
},
{
"cell_type": "code",
"execution_count": 109,
"metadata": {},
"outputs": [],
"source": [
"power = np.array([df_wdb['time'], df['Heat']]).T\n",
"eng.workspace['power'] = matlab.double(power.tolist())"
]
},
{
"cell_type": "code",
"execution_count": 110,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Power</th>\n",
" <th>Setpoint</th>\n",
" <th>OutsideTemp</th>\n",
" <th>SupplyTemp</th>\n",
" <th>InsideTemp</th>\n",
" <th>SolRad</th>\n",
" <th>Heat</th>\n",
" <th>SimulatedTemp</th>\n",
" <th>time</th>\n",
" </tr>\n",
" <tr>\n",
" <th>timestamp</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>2017-06-30 20:00:00+02:00</th>\n",
" <td>8.620690</td>\n",
" <td>22.5</td>\n",
" <td>19.0</td>\n",
" <td>22.1</td>\n",
" <td>22.283333</td>\n",
" <td>178.618267</td>\n",
" <td>8.620690</td>\n",
" <td>22.283285</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-06-30 20:05:00+02:00</th>\n",
" <td>8.620690</td>\n",
" <td>22.5</td>\n",
" <td>19.0</td>\n",
" <td>22.1</td>\n",
" <td>22.233333</td>\n",
" <td>135.122100</td>\n",
" <td>8.620690</td>\n",
" <td>22.266734</td>\n",
" <td>300</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-06-30 20:10:00+02:00</th>\n",
" <td>7.566667</td>\n",
" <td>22.5</td>\n",
" <td>19.0</td>\n",
" <td>22.1</td>\n",
" <td>22.366667</td>\n",
" <td>126.219967</td>\n",
" <td>7.566667</td>\n",
" <td>22.258667</td>\n",
" <td>600</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-06-30 20:15:00+02:00</th>\n",
" <td>8.172414</td>\n",
" <td>22.5</td>\n",
" <td>19.0</td>\n",
" <td>22.1</td>\n",
" <td>22.366667</td>\n",
" <td>123.772467</td>\n",
" <td>8.172414</td>\n",
" <td>22.255146</td>\n",
" <td>900</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-06-30 20:20:00+02:00</th>\n",
" <td>8.379310</td>\n",
" <td>22.5</td>\n",
" <td>17.5</td>\n",
" <td>22.4</td>\n",
" <td>22.566667</td>\n",
" <td>106.163600</td>\n",
" <td>-25.137931</td>\n",
" <td>22.255204</td>\n",
" <td>1200</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Power Setpoint OutsideTemp SupplyTemp \\\n",
"timestamp \n",
"2017-06-30 20:00:00+02:00 8.620690 22.5 19.0 22.1 \n",
"2017-06-30 20:05:00+02:00 8.620690 22.5 19.0 22.1 \n",
"2017-06-30 20:10:00+02:00 7.566667 22.5 19.0 22.1 \n",
"2017-06-30 20:15:00+02:00 8.172414 22.5 19.0 22.1 \n",
"2017-06-30 20:20:00+02:00 8.379310 22.5 17.5 22.4 \n",
"\n",
" InsideTemp SolRad Heat SimulatedTemp \\\n",
"timestamp \n",
"2017-06-30 20:00:00+02:00 22.283333 178.618267 8.620690 22.283285 \n",
"2017-06-30 20:05:00+02:00 22.233333 135.122100 8.620690 22.266734 \n",
"2017-06-30 20:10:00+02:00 22.366667 126.219967 7.566667 22.258667 \n",
"2017-06-30 20:15:00+02:00 22.366667 123.772467 8.172414 22.255146 \n",
"2017-06-30 20:20:00+02:00 22.566667 106.163600 -25.137931 22.255204 \n",
"\n",
" time \n",
"timestamp \n",
"2017-06-30 20:00:00+02:00 0 \n",
"2017-06-30 20:05:00+02:00 300 \n",
"2017-06-30 20:10:00+02:00 600 \n",
"2017-06-30 20:15:00+02:00 900 \n",
"2017-06-30 20:20:00+02:00 1200 "
]
},
"execution_count": 110,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df = pd.DataFrame(df).assign(time = df_wdb['time'].values)\n",
"df.head()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Control loop"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Initiation setup"
]
},
{
"cell_type": "code",
"execution_count": 111,
"metadata": {},
"outputs": [],
"source": [
"current_timestamp = 1500"
]
},
{
"cell_type": "code",
"execution_count": 112,
"metadata": {},
"outputs": [],
"source": [
"df_power = df['Heat']\n",
"df_power = pd.DataFrame(df_power).assign(time = df_wdb['time'].values)\n",
"df_power.loc[df_power['time'] >= current_timestamp, 'Heat'] = np.NaN"
]
},
{
"cell_type": "code",
"execution_count": 113,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Heat</th>\n",
" <th>time</th>\n",
" </tr>\n",
" <tr>\n",
" <th>timestamp</th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>2017-06-30 20:00:00+02:00</th>\n",
" <td>8.620690</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-06-30 20:05:00+02:00</th>\n",
" <td>8.620690</td>\n",
" <td>300</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-06-30 20:10:00+02:00</th>\n",
" <td>7.566667</td>\n",
" <td>600</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-06-30 20:15:00+02:00</th>\n",
" <td>8.172414</td>\n",
" <td>900</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-06-30 20:20:00+02:00</th>\n",
" <td>-25.137931</td>\n",
" <td>1200</td>\n",
" </tr>\n",
" <tr>\n",
" <th>...</th>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-07-03 05:35:00+02:00</th>\n",
" <td>NaN</td>\n",
" <td>207300</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-07-03 05:40:00+02:00</th>\n",
" <td>NaN</td>\n",
" <td>207600</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-07-03 05:45:00+02:00</th>\n",
" <td>NaN</td>\n",
" <td>207900</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-07-03 05:50:00+02:00</th>\n",
" <td>NaN</td>\n",
" <td>208200</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-07-03 05:55:00+02:00</th>\n",
" <td>NaN</td>\n",
" <td>208500</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>696 rows × 2 columns</p>\n",
"</div>"
],
"text/plain": [
" Heat time\n",
"timestamp \n",
"2017-06-30 20:00:00+02:00 8.620690 0\n",
"2017-06-30 20:05:00+02:00 8.620690 300\n",
"2017-06-30 20:10:00+02:00 7.566667 600\n",
"2017-06-30 20:15:00+02:00 8.172414 900\n",
"2017-06-30 20:20:00+02:00 -25.137931 1200\n",
"... ... ...\n",
"2017-07-03 05:35:00+02:00 NaN 207300\n",
"2017-07-03 05:40:00+02:00 NaN 207600\n",
"2017-07-03 05:45:00+02:00 NaN 207900\n",
"2017-07-03 05:50:00+02:00 NaN 208200\n",
"2017-07-03 05:55:00+02:00 NaN 208500\n",
"\n",
"[696 rows x 2 columns]"
]
},
"execution_count": 113,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df_power"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Compute input to apply at current time step"
]
},
{
"cell_type": "code",
"execution_count": 114,
"metadata": {},
"outputs": [],
"source": [
"u_1 = float(df_power.loc[df['time'] == (current_timestamp - 300 * 1), 'Heat'])\n",
"\n",
"y_0 = float(df.loc[df['time'] == (current_timestamp - 300 * 0), 'SimulatedTemp'])\n",
"y_1 = float(df.loc[df['time'] == (current_timestamp - 300 * 1), 'SimulatedTemp'])\n",
"y_2 = float(df.loc[df['time'] == (current_timestamp - 300 * 2), 'SimulatedTemp'])"
]
},
{
"cell_type": "code",
"execution_count": 115,
"metadata": {},
"outputs": [],
"source": [
"real_x0 = np.array([u_1, y_0, y_1, y_2])"
]
},
{
"cell_type": "code",
"execution_count": 116,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"array([-25.13793103, 22.25542067, 22.25520408, 22.25514583])"
]
},
"execution_count": 116,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"real_x0"
]
},
{
"cell_type": "code",
"execution_count": 117,
"metadata": {},
"outputs": [],
"source": [
"#real_x0 = x_scaler.transform(np.hstack([np.zeros((3)), real_x0]).reshape((1, -1)))[0, 3:]"
]
},
{
"cell_type": "code",
"execution_count": 118,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"array([-25.13793103, 22.25542067, 22.25520408, 22.25514583])"
]
},
"execution_count": 118,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"real_x0"
]
},
{
"cell_type": "code",
"execution_count": 119,
"metadata": {},
"outputs": [],
"source": [
"iter_idx = (df['time'] >= current_timestamp)\n",
"real_W = df[iter_idx].iloc[:N_horizon, [5, 2]].to_numpy()"
]
},
{
"cell_type": "code",
"execution_count": 120,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"array([[92.377 , 17.5 ],\n",
" [54.69793333, 16. ],\n",
" [51.83236667, 16. ],\n",
" [40.2653 , 16. ],\n",
" [29.8469 , 16. ]])"
]
},
"execution_count": 120,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"real_W = casadi.DM(real_W)"
]
},
{
"cell_type": "code",
"execution_count": 123,
"metadata": {},
"outputs": [],
"source": [
"real_p = casadi.vertcat(\n",
" casadi.vec(real_W),\n",
" casadi.vec(real_x0)\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 124,
"metadata": {},
"outputs": [],
"source": [
"real_X0 = get_gpr_horizon_array(real_W, real_x0.reshape((1, -1)), np.zeros((N_horizon, 1)))"
]
},
{
"cell_type": "code",
"execution_count": 125,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"array([[ 92.377 , 17.5 , 0. , -25.13793103,\n",
" 22.25542067, 22.25520408, 22.25514583],\n",
" [ 54.69793333, 16. , 0. , 0. ,\n",
" 22.27224362, 22.25542067, 22.25520408],\n",
" [ 51.83236667, 16. , 0. , 0. ,\n",
" 22.31347714, 22.27224362, 22.25542067],\n",
" [ 40.2653 , 16. , 0. , 0. ,\n",
" 22.39125122, 22.31347714, 22.27224362],\n",
" [ 29.8469 , 16. , 0. , 0. ,\n",
" 22.52311219, 22.39125122, 22.31347714]])"
]
},
"execution_count": 125,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"real_X0 = casadi.DM(real_X0)"
]
},
{
"cell_type": "code",
"execution_count": 126,
"metadata": {
"tags": []
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"This is Ipopt version 3.13.4, running with linear solver mumps.\n",
"NOTE: Other linear solvers might be more efficient (see Ipopt documentation).\n",
"\n",
"Number of nonzeros in equality constraint Jacobian...: 70\n",
"Number of nonzeros in inequality constraint Jacobian.: 5\n",
"Number of nonzeros in Lagrangian Hessian.............: 0\n",
"\n",
"Total number of variables............................: 35\n",
" variables with only lower bounds: 0\n",
" variables with lower and upper bounds: 0\n",
" variables with only upper bounds: 0\n",
"Total number of equality constraints.................: 30\n",
"Total number of inequality constraints...............: 5\n",
" inequality constraints with only lower bounds: 0\n",
" inequality constraints with lower and upper bounds: 5\n",
" inequality constraints with only upper bounds: 0\n",
"\n",
"iter objective inf_pr inf_du lg(mu) ||d|| lg(rg) alpha_du alpha_pr ls\n",
" 0 8.5409843e+03 6.54e+01 2.57e+01 0.0 0.00e+00 - 0.00e+00 0.00e+00 0\n",
" 1 5.4574789e+02 7.59e+00 1.87e+02 -1.5 6.54e+01 - 9.89e-01 1.00e+00f 1\n",
" 2 3.4479029e+01 4.23e-01 6.34e+01 -3.5 7.50e+00 - 9.90e-01 1.00e+00f 1\n",
" 3 1.7089734e+01 2.18e-01 7.77e+00 -1.4 1.43e+00 - 1.00e+00 1.00e+00f 1\n",
" 4 1.3407625e+01 1.02e-01 5.52e+00 -3.2 1.47e+00 - 1.00e+00 1.00e+00f 1\n",
" 5 2.9545703e+01 1.38e-01 1.14e+01 -5.1 1.45e+00 - 1.00e+00 1.00e+00h 1\n",
" 6 2.6625453e+01 2.30e-02 1.46e+01 -7.0 3.08e-01 - 1.00e+00 1.00e+00f 1\n",
" 7 1.1836963e+01 4.04e-01 1.68e+01 -8.9 2.76e+00 - 1.00e+00 1.00e+00f 1\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"CasADi - 2021-04-15 12:50:16 WARNING(\"solver:nlp_jac_g failed:Error in Function::operator() for 'nlp_jac_g' [MXFunction] at .../casadi/core/function.cpp:1368:\n",
"Error in Function::operator() for 'fwd7_GPR' [CentralDiff] at .../casadi/core/function.cpp:1368:\n",
"Error in Function::operator() for 'GPR' [CallbackInternal] at .../casadi/core/function.cpp:1368:\n",
".../casadi/core/function_internal.cpp:3366: Failed to evaluate 'eval_dm' for GPR:\n",
".../casadi/core/callback_internal.cpp:122: Error calling \"eval\" for object GPR:\n",
"KeyboardInterrupt\") [.../casadi/core/oracle_function.cpp:223]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n",
"Number of Iterations....: 8\n",
"\n",
"Number of objective function evaluations = 9\n",
"Number of objective gradient evaluations = 9\n",
"Number of equality constraint evaluations = 9\n",
"Number of inequality constraint evaluations = 9\n",
"Number of equality constraint Jacobian evaluations = 8\n",
"Number of inequality constraint Jacobian evaluations = 9\n",
"Number of Lagrangian Hessian evaluations = 0\n",
"Total CPU secs in IPOPT (w/o function evaluations) = 8.005\n",
"Total CPU secs in NLP function evaluations = 71.463\n",
"\n",
"EXIT: Invalid number in NLP function or derivative detected.\n",
" solver : t_proc (avg) t_wall (avg) n_eval\n",
" nlp_f | 1.36 s (150.85ms) 778.94ms ( 86.55ms) 9\n",
" nlp_g | 1.07 s (118.45ms) 616.40ms ( 68.49ms) 9\n",
" nlp_grad | 8.54 s ( 8.54 s) 5.05 s ( 5.05 s) 1\n",
" nlp_grad_f | 45.21 s ( 4.52 s) 26.23 s ( 2.62 s) 10\n",
" nlp_jac_g | 33.52 s ( 3.35 s) 19.31 s ( 1.93 s) 10\n",
" total | 89.70 s ( 89.70 s) 52.01 s ( 52.01 s) 1\n"
]
}
],
"source": [
"res = solver(x0 = real_X0.reshape((-1, )), p = real_p, lbg = real_lbg, ubg = real_ubg)"
]
},
{
"cell_type": "code",
"execution_count": 59,
"metadata": {},
"outputs": [],
"source": [
"x_res = np.array(res['x'].reshape((N_horizon, -1)))\n",
"#x_res_sc = x_scaler.inverse_transform(x_res)"
]
},
{
"cell_type": "code",
"execution_count": 60,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"DM(\n",
"[[92.377, 17.5, 0.415039, -25.1379, 22.2554, 22.2552, 22.2551], \n",
" [54.6979, 16, 144.993, 0.415039, 22.2437, 22.2554, 22.2552], \n",
" [51.8324, 16, -103.495, 144.993, 22.272, 22.2437, 22.2554], \n",
" [40.2653, 16, -48.8636, -103.495, 23.3874, 22.272, 22.2437], \n",
" [29.8469, 16, 6.33578, -48.8636, 24.9003, 23.3874, 22.272]])"
]
},
"execution_count": 60,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"casadi.DM(x_res)"
]
},
{
"cell_type": "code",
"execution_count": 61,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"array([ 0.41503917, 144.99327115, -103.49457825, -48.8635556 ,\n",
" 6.33578304])"
]
},
"execution_count": 61,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"x_res[:, 2]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Apply the first computed input as the next input:"
]
},
{
"cell_type": "code",
"execution_count": 62,
"metadata": {},
"outputs": [],
"source": [
"df_power.loc[df_power['time'] == current_timestamp, 'Heat'] = res['x'].reshape((N_horizon, -1))[0, 2]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Simulate the building including the current input:"
]
},
{
"cell_type": "code",
"execution_count": 63,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Heat</th>\n",
" <th>time</th>\n",
" </tr>\n",
" <tr>\n",
" <th>timestamp</th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>2017-06-30 20:00:00+02:00</th>\n",
" <td>8.620690</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-06-30 20:05:00+02:00</th>\n",
" <td>8.620690</td>\n",
" <td>300</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-06-30 20:10:00+02:00</th>\n",
" <td>7.566667</td>\n",
" <td>600</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-06-30 20:15:00+02:00</th>\n",
" <td>8.172414</td>\n",
" <td>900</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-06-30 20:20:00+02:00</th>\n",
" <td>-25.137931</td>\n",
" <td>1200</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-06-30 20:25:00+02:00</th>\n",
" <td>0.415039</td>\n",
" <td>1500</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Heat time\n",
"timestamp \n",
"2017-06-30 20:00:00+02:00 8.620690 0\n",
"2017-06-30 20:05:00+02:00 8.620690 300\n",
"2017-06-30 20:10:00+02:00 7.566667 600\n",
"2017-06-30 20:15:00+02:00 8.172414 900\n",
"2017-06-30 20:20:00+02:00 -25.137931 1200\n",
"2017-06-30 20:25:00+02:00 0.415039 1500"
]
},
"execution_count": 63,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df_power.dropna()"
]
},
{
"cell_type": "code",
"execution_count": 64,
"metadata": {},
"outputs": [],
"source": [
"power = np.array(df_power[['time', 'Heat']].dropna())"
]
},
{
"cell_type": "code",
"execution_count": 65,
"metadata": {},
"outputs": [],
"source": [
"eng.workspace['power'] = matlab.double(power.tolist())"
]
},
{
"cell_type": "code",
"execution_count": 66,
"metadata": {},
"outputs": [],
"source": [
"eng.set_param('polydome', 'StopTime', str(current_timestamp + 300), nargout = 0)"
]
},
{
"cell_type": "code",
"execution_count": 67,
"metadata": {},
"outputs": [],
"source": [
"eng.workspace['result'] = eng.sim('polydome')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Interpret the MATLAB results as python:"
]
},
{
"cell_type": "code",
"execution_count": 68,
"metadata": {},
"outputs": [],
"source": [
"dict_simulation = {}\n",
"dict_simulation['values'] = np.asarray(eng.eval('result.SimulatedTemp.Data')).reshape(-1)\n",
"dict_simulation['time'] = np.asarray(eng.eval('result.SimulatedTemp.Time')).reshape(-1)"
]
},
{
"cell_type": "code",
"execution_count": 69,
"metadata": {},
"outputs": [],
"source": [
"df_simulation = pd.DataFrame(dict_simulation)\n",
"#df_simulation['time'] = df_simulation['time'].astype(int)\n",
"df_simulation.set_index('time', inplace = True, drop = True)"
]
},
{
"cell_type": "code",
"execution_count": 70,
"metadata": {},
"outputs": [],
"source": [
"df_simulation['timestamp'] = df.index[0] + df_simulation.index.map(lambda x: pd.Timedelta(seconds = x))"
]
},
{
"cell_type": "code",
"execution_count": 71,
"metadata": {},
"outputs": [],
"source": [
"df_simulation = df_simulation.reset_index().set_index('timestamp')"
]
},
{
"cell_type": "code",
"execution_count": 72,
"metadata": {},
"outputs": [],
"source": [
"df_resampled_5 = df_simulation['values'].resample('5min').mean().pad()"
]
},
{
"cell_type": "code",
"execution_count": 73,
"metadata": {},
"outputs": [],
"source": [
"df_simulation = pd.concat([df['time'], df_resampled_5], axis = 1)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Update the simulated temperature (output data) with the new info from this step:"
]
},
{
"cell_type": "code",
"execution_count": 74,
"metadata": {},
"outputs": [],
"source": [
"df.loc[:, 'SimulatedTemp'] = df_simulation['values']"
]
},
{
"cell_type": "code",
"execution_count": 75,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Power</th>\n",
" <th>Setpoint</th>\n",
" <th>OutsideTemp</th>\n",
" <th>SupplyTemp</th>\n",
" <th>InsideTemp</th>\n",
" <th>SolRad</th>\n",
" <th>Heat</th>\n",
" <th>SimulatedTemp</th>\n",
" <th>time</th>\n",
" </tr>\n",
" <tr>\n",
" <th>timestamp</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>2017-06-30 20:00:00+02:00</th>\n",
" <td>8.620690</td>\n",
" <td>22.5</td>\n",
" <td>19.0</td>\n",
" <td>22.1</td>\n",
" <td>22.283333</td>\n",
" <td>178.618267</td>\n",
" <td>8.620690</td>\n",
" <td>22.258776</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-06-30 20:05:00+02:00</th>\n",
" <td>8.620690</td>\n",
" <td>22.5</td>\n",
" <td>19.0</td>\n",
" <td>22.1</td>\n",
" <td>22.233333</td>\n",
" <td>135.122100</td>\n",
" <td>8.620690</td>\n",
" <td>21.924169</td>\n",
" <td>300</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-06-30 20:10:00+02:00</th>\n",
" <td>7.566667</td>\n",
" <td>22.5</td>\n",
" <td>19.0</td>\n",
" <td>22.1</td>\n",
" <td>22.366667</td>\n",
" <td>126.219967</td>\n",
" <td>7.566667</td>\n",
" <td>21.781290</td>\n",
" <td>600</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-06-30 20:15:00+02:00</th>\n",
" <td>8.172414</td>\n",
" <td>22.5</td>\n",
" <td>19.0</td>\n",
" <td>22.1</td>\n",
" <td>22.366667</td>\n",
" <td>123.772467</td>\n",
" <td>8.172414</td>\n",
" <td>21.651448</td>\n",
" <td>900</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-06-30 20:20:00+02:00</th>\n",
" <td>8.379310</td>\n",
" <td>22.5</td>\n",
" <td>17.5</td>\n",
" <td>22.4</td>\n",
" <td>22.566667</td>\n",
" <td>106.163600</td>\n",
" <td>-25.137931</td>\n",
" <td>21.471403</td>\n",
" <td>1200</td>\n",
" </tr>\n",
" <tr>\n",
" <th>...</th>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-07-03 05:35:00+02:00</th>\n",
" <td>-22.266667</td>\n",
" <td>22.5</td>\n",
" <td>16.0</td>\n",
" <td>21.5</td>\n",
" <td>21.016667</td>\n",
" <td>3.911200</td>\n",
" <td>-22.266667</td>\n",
" <td>NaN</td>\n",
" <td>207300</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-07-03 05:40:00+02:00</th>\n",
" <td>-21.310345</td>\n",
" <td>22.5</td>\n",
" <td>16.0</td>\n",
" <td>21.5</td>\n",
" <td>20.850000</td>\n",
" <td>4.535500</td>\n",
" <td>-21.310345</td>\n",
" <td>NaN</td>\n",
" <td>207600</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-07-03 05:45:00+02:00</th>\n",
" <td>-22.000000</td>\n",
" <td>22.5</td>\n",
" <td>16.0</td>\n",
" <td>21.5</td>\n",
" <td>20.850000</td>\n",
" <td>5.259500</td>\n",
" <td>-22.000000</td>\n",
" <td>NaN</td>\n",
" <td>207900</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-07-03 05:50:00+02:00</th>\n",
" <td>-23.034483</td>\n",
" <td>22.5</td>\n",
" <td>16.0</td>\n",
" <td>21.5</td>\n",
" <td>20.866667</td>\n",
" <td>6.644067</td>\n",
" <td>-23.034483</td>\n",
" <td>NaN</td>\n",
" <td>208200</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-07-03 05:55:00+02:00</th>\n",
" <td>-20.666667</td>\n",
" <td>22.5</td>\n",
" <td>16.0</td>\n",
" <td>21.5</td>\n",
" <td>20.850000</td>\n",
" <td>9.509900</td>\n",
" <td>-20.666667</td>\n",
" <td>NaN</td>\n",
" <td>208500</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>696 rows × 9 columns</p>\n",
"</div>"
],
"text/plain": [
" Power Setpoint OutsideTemp SupplyTemp \\\n",
"timestamp \n",
"2017-06-30 20:00:00+02:00 8.620690 22.5 19.0 22.1 \n",
"2017-06-30 20:05:00+02:00 8.620690 22.5 19.0 22.1 \n",
"2017-06-30 20:10:00+02:00 7.566667 22.5 19.0 22.1 \n",
"2017-06-30 20:15:00+02:00 8.172414 22.5 19.0 22.1 \n",
"2017-06-30 20:20:00+02:00 8.379310 22.5 17.5 22.4 \n",
"... ... ... ... ... \n",
"2017-07-03 05:35:00+02:00 -22.266667 22.5 16.0 21.5 \n",
"2017-07-03 05:40:00+02:00 -21.310345 22.5 16.0 21.5 \n",
"2017-07-03 05:45:00+02:00 -22.000000 22.5 16.0 21.5 \n",
"2017-07-03 05:50:00+02:00 -23.034483 22.5 16.0 21.5 \n",
"2017-07-03 05:55:00+02:00 -20.666667 22.5 16.0 21.5 \n",
"\n",
" InsideTemp SolRad Heat SimulatedTemp \\\n",
"timestamp \n",
"2017-06-30 20:00:00+02:00 22.283333 178.618267 8.620690 22.258776 \n",
"2017-06-30 20:05:00+02:00 22.233333 135.122100 8.620690 21.924169 \n",
"2017-06-30 20:10:00+02:00 22.366667 126.219967 7.566667 21.781290 \n",
"2017-06-30 20:15:00+02:00 22.366667 123.772467 8.172414 21.651448 \n",
"2017-06-30 20:20:00+02:00 22.566667 106.163600 -25.137931 21.471403 \n",
"... ... ... ... ... \n",
"2017-07-03 05:35:00+02:00 21.016667 3.911200 -22.266667 NaN \n",
"2017-07-03 05:40:00+02:00 20.850000 4.535500 -21.310345 NaN \n",
"2017-07-03 05:45:00+02:00 20.850000 5.259500 -22.000000 NaN \n",
"2017-07-03 05:50:00+02:00 20.866667 6.644067 -23.034483 NaN \n",
"2017-07-03 05:55:00+02:00 20.850000 9.509900 -20.666667 NaN \n",
"\n",
" time \n",
"timestamp \n",
"2017-06-30 20:00:00+02:00 0 \n",
"2017-06-30 20:05:00+02:00 300 \n",
"2017-06-30 20:10:00+02:00 600 \n",
"2017-06-30 20:15:00+02:00 900 \n",
"2017-06-30 20:20:00+02:00 1200 \n",
"... ... \n",
"2017-07-03 05:35:00+02:00 207300 \n",
"2017-07-03 05:40:00+02:00 207600 \n",
"2017-07-03 05:45:00+02:00 207900 \n",
"2017-07-03 05:50:00+02:00 208200 \n",
"2017-07-03 05:55:00+02:00 208500 \n",
"\n",
"[696 rows x 9 columns]"
]
},
"execution_count": 75,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Keep track of the all the prediction horizon, to add to the graph:"
]
},
{
"cell_type": "code",
"execution_count": 76,
"metadata": {},
"outputs": [],
"source": [
"gpr_horizon = np.array(gpr(res['x'].reshape((N_horizon, -1)).T)).flatten()"
]
},
{
"cell_type": "code",
"execution_count": 77,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"array([22.25542067, 22.2722139 , 22.23139534, 22.34675811, 25.0869712 ,\n",
" 27.19437496])"
]
},
"execution_count": 77,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"T_sim_horizon = np.hstack([np.array(y_0), gpr_horizon])\n",
"T_sim_horizon"
]
},
{
"cell_type": "code",
"execution_count": 78,
"metadata": {},
"outputs": [],
"source": [
"simul_idx = (df_simulation['time'] >= current_timestamp) & (df_simulation['time'] <= (current_timestamp + N_horizon * 300))"
]
},
{
"cell_type": "code",
"execution_count": 79,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>time</th>\n",
" <th>values</th>\n",
" </tr>\n",
" <tr>\n",
" <th>timestamp</th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>2017-06-30 20:25:00+02:00</th>\n",
" <td>1500</td>\n",
" <td>22.255421</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-06-30 20:30:00+02:00</th>\n",
" <td>1800</td>\n",
" <td>22.272214</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-06-30 20:35:00+02:00</th>\n",
" <td>2100</td>\n",
" <td>22.231395</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-06-30 20:40:00+02:00</th>\n",
" <td>2400</td>\n",
" <td>22.346758</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-06-30 20:45:00+02:00</th>\n",
" <td>2700</td>\n",
" <td>25.086971</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-06-30 20:50:00+02:00</th>\n",
" <td>3000</td>\n",
" <td>27.194375</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" time values\n",
"timestamp \n",
"2017-06-30 20:25:00+02:00 1500 22.255421\n",
"2017-06-30 20:30:00+02:00 1800 22.272214\n",
"2017-06-30 20:35:00+02:00 2100 22.231395\n",
"2017-06-30 20:40:00+02:00 2400 22.346758\n",
"2017-06-30 20:45:00+02:00 2700 25.086971\n",
"2017-06-30 20:50:00+02:00 3000 27.194375"
]
},
"execution_count": 79,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df_T_sim_horizon = df_simulation[simul_idx].copy()\n",
"df_T_sim_horizon.loc[:, 'values'] = T_sim_horizon.reshape((-1, ))\n",
"df_T_sim_horizon"
]
},
{
"cell_type": "code",
"execution_count": 80,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 1080x360 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"plt.figure(figsize = (15, 5))\n",
"plt.plot(df_simulation.index, df_simulation['values'], label = 'Simulated Temperature')\n",
"plt.plot(df_T_sim_horizon.index, df_T_sim_horizon['values'], label = 'Prediction Horizon', color = 'red', marker = 'x')\n",
"#plt.plot(df.index, df['InsideTemp'], label = 'Inside Temperature')\n",
"#plt.plot(df.index, df['OutsideTemp'], label = 'Outside Temperature')\n",
"plt.title(f'Temperatures step {current_timestamp}')\n",
"plt.legend()\n",
"plt.ylim((15, 30))\n",
"plt.savefig(f\"sim_{current_timestamp}.png\")\n",
"plt.show()\n"
]
},
{
"cell_type": "code",
"execution_count": 81,
"metadata": {
"scrolled": true,
"tags": []
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Timestamp 1500\n"
]
},
{
"ename": "RuntimeError",
"evalue": ".../casadi/core/function_internal.hpp:1257: Input 1 (p) has mismatching shape. Got 15-by-1. Allowed dimensions, in general, are:\n - The input dimension N-by-M (here 14-by-1)\n - A scalar, i.e. 1-by-1\n - M-by-N if N=1 or M=1 (i.e. a transposed vector)\n - N-by-M1 if K*M1=M for some K (argument repeated horizontally)\n - N-by-P*M, indicating evaluation with multiple arguments (P must be a multiple of 1 for consistency with previous inputs)",
"output_type": "error",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[0;31mRuntimeError\u001b[0m Traceback (most recent call last)",
"\u001b[0;32m<ipython-input-81-92d0385ddb27>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m 19\u001b[0m )\n\u001b[1;32m 20\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 21\u001b[0;31m \u001b[0mres\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0msolver\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mp\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mreal_p\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlbg\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mreal_lbg\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mubg\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mreal_ubg\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 22\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 23\u001b[0m \u001b[0mdf_power\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloc\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mdf_power\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'time'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0mcurrent_timestamp\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'Heat'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mres\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'x'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreshape\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mN_horizon\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m2\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/usr/lib/python3.9/site-packages/casadi/casadi.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m 8504\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 8505\u001b[0m \u001b[0;31m# Named inputs -> return dictionary\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 8506\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcall\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 8507\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 8508\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mbuffer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/usr/lib/python3.9/site-packages/casadi/casadi.py\u001b[0m in \u001b[0;36mcall\u001b[0;34m(self, *args)\u001b[0m\n\u001b[1;32m 7677\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 7678\u001b[0m \"\"\"\n\u001b[0;32m-> 7679\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0m_casadi\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mFunction_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 7680\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 7681\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mmapsum\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m->\u001b[0m \u001b[0;34m\"std::vector< casadi::MX,std::allocator< casadi::MX > >\"\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;31mRuntimeError\u001b[0m: .../casadi/core/function_internal.hpp:1257: Input 1 (p) has mismatching shape. Got 15-by-1. Allowed dimensions, in general, are:\n - The input dimension N-by-M (here 14-by-1)\n - A scalar, i.e. 1-by-1\n - M-by-N if N=1 or M=1 (i.e. a transposed vector)\n - N-by-M1 if K*M1=M for some K (argument repeated horizontally)\n - N-by-P*M, indicating evaluation with multiple arguments (P must be a multiple of 1 for consistency with previous inputs)"
]
}
],
"source": [
"for loop_idx in range(100):\n",
" current_timestamp = 1500 + 300*loop_idx\n",
" print(f\"Timestamp {current_timestamp}\")\n",
" \n",
" u_1 = float(df_power.loc[df['time'] == (current_timestamp - 300 * 1), 'Heat'])\n",
" u_2 = float(df_power.loc[df['time'] == (current_timestamp - 300 * 2), 'Heat'])\n",
"\n",
" y_1 = float(df.loc[df['time'] == (current_timestamp - 300 * 1), 'SimulatedTemp'])\n",
" y_2 = float(df.loc[df['time'] == (current_timestamp - 300 * 2), 'SimulatedTemp'])\n",
" y_3 = float(df.loc[df['time'] == (current_timestamp - 300 * 3), 'SimulatedTemp'])\n",
" \n",
" real_x0 = np.array([u_1, u_2, y_1, y_2, y_3])\n",
" iter_idx = (df['time'] >= current_timestamp)\n",
" real_W = df[iter_idx].iloc[:N_horizon, [5, 2]].to_numpy()\n",
"\n",
" real_p = casadi.vertcat(\n",
" casadi.vec(real_W),\n",
" casadi.vec(real_x0)\n",
" )\n",
"\n",
" res = solver(p = real_p, lbg = real_lbg, ubg = real_ubg)\n",
" \n",
" df_power.loc[df_power['time'] == current_timestamp, 'Heat'] = res['x'].reshape((N_horizon, -1))[1, 2]\n",
" \n",
" power = np.array(df_power[['time', 'Heat']].dropna())\n",
" eng.workspace['power'] = matlab.double(power.tolist())\n",
" eng.set_param('polydome', 'StopTime', str(current_timestamp + 300), nargout = 0)\n",
" eng.workspace['result'] = eng.sim('polydome')\n",
" \n",
" \n",
" dict_simulation = {}\n",
" dict_simulation['values'] = np.asarray(eng.eval('result.SimulatedTemp.Data')).reshape(-1)\n",
" dict_simulation['time'] = np.asarray(eng.eval('result.SimulatedTemp.Time')).reshape(-1)\n",
" \n",
" df_simulation = pd.DataFrame(dict_simulation)\n",
" #df_simulation['time'] = df_simulation['time'].astype(int)\n",
" df_simulation.set_index('time', inplace = True, drop = True)\n",
" \n",
" df_simulation['timestamp'] = df.index[0] + df_simulation.index.map(lambda x: pd.Timedelta(seconds = x))\n",
" df_simulation = df_simulation.reset_index().set_index('timestamp')\n",
" df_resampled_5 = df_simulation['values'].resample('5min').mean().pad()\n",
" df_simulation = pd.concat([df['time'], df_resampled_5], axis = 1)\n",
" \n",
" df.loc[:, 'SimulatedTemp'] = df_simulation['values']\n",
" T_sim_horizon = np.array(gpr(res['x'].reshape((N_horizon, -1)).T))\n",
" simul_idx = (df_simulation['time'] >= current_timestamp) & (df_simulation['time'] < (current_timestamp + N_horizon * 300))\n",
" \n",
" \n",
" df_T_sim_horizon = df_simulation[simul_idx].copy()\n",
" df_T_sim_horizon.loc[:, 'values'] = T_sim_horizon.reshape((-1, ))\n",
" \n",
" plt.figure(figsize = (15, 5))\n",
" plt.plot(df_simulation.index, df_simulation['values'], label = 'Simulated Temperature')\n",
" plt.plot(df_T_sim_horizon.index, df_T_sim_horizon['values'], label = 'Prediction Horizon', color = 'red', marker = 'x')\n",
" #plt.plot(df.index, df['InsideTemp'], label = 'Inside Temperature')\n",
" #plt.plot(df.index, df['OutsideTemp'], label = 'Outside Temperature')\n",
" plt.title(f'Temperatures step {current_timestamp}')\n",
" plt.ylim((15, 30))\n",
" plt.legend()\n",
" plt.savefig(f\"sim_{current_timestamp}.png\")"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.9.3"
}
},
"nbformat": 4,
"nbformat_minor": 4
}
Reference in a new issue View git blame Copy permalink