{ "cells": [ { "cell_type": "markdown", "metadata": { "toc": true }, "source": [ "

Table of Contents

\n", "
" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Introduction\n", "The objective is to simulate a nonlinear electro-mechanical system with thermal model and static Coulomb friction. We use three ODE solvers, the embedded Matlab solver `ode45`, and two external solvers, the 4th and 5th order Runge-Kutta algorithm `ode45m`, and the basic Euler algorithm `eufix1`.\n", "\n", "# Pre-requisites\n", "- Previous post of [DC motor modelling](/blog/DCmotor/).\n", "- Jupyter Notebook with Matlab-kernel.\n", "- Matlab.\n", "\n", "# Source code\n", "\n", "Version [PDF](https://raw.githubusercontent.com/paulomarconi/Nonlinear_DCmotor/master/asst02_2017/asst02_2017.pdf)/[HTML](../../files/Nonlinear_DCmotor/asst02_2017.html). Matlab and LaTex source code on [GitHub](https://github.com/paulomarconi/Nonlinear_DCmotor). " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Nonlinear model\n", "The nonlinear dynamic system is\n", "\\begin{align}\n", "\tR_A i_A + L_A \\dot{i}_A+\\alpha \\omega_1 & = e_{i}(t) \\\\\n", "\tJ_1 \\dot{\\omega}_1 + B_1 \\omega_1 - r_1 f_c & = \\alpha i_A \\\\\n", "\tJ_2 \\dot{\\omega}_2 + B_2 \\omega_2 + B_{2C}~sign(\\omega_2) + r_2 f_c & = -\\tau_{L} \\\\\n", "\tC_{TM} \\dot{\\theta}_M + \\frac{(\\theta_M-\\theta_A)}{R_{TM}} & = i^2_A R_A\n", "\\end{align}\n", "where $R_A$ is the stationary resistance, $L_A$ is the stationary inductance, $i_A$ is the input stationary current, $\\alpha$ is the internal parameters, $\\omega$ is the angular speed, $e_i(t)$ is the applied armature voltage, $B$ is the rotational viscous-damping coefficient, $J$ is the moment of inertia, $f_c$ is the contact force between two gears, $r$ is the gear radius, and $B_{2C}$ is the static friction. The thermal model is similar to an electrical capacitor-resistor model with thermal capacity $C_{TM}$, $R_{TM}$ is the resistive losses to ambient temperature, $\\theta_M$ is the motor temperature, and $\\theta_A$ is the ambient temperature. \n", "\n", "Now let us define the sate-vector differential equations: state vector $x = [ i_A~ \\omega_2~ \\theta_M ]^T $, and input vector $u = [ e_{i}~ \\tau_{L}~ \\theta_A ]^T$. \n", "\n", "For $\\omega_1=N \\omega_2$ and $N=\\frac{r_2}{r_1}$, eliminating $f_c$ we have\n", "\n", "\\begin{align}\n", "\t\\dot{i}_A & = -\\frac{R_A}{L_A}\\ i_A - \\frac{N \\alpha}{L_A} \\omega_2 + \\frac{1}{L_A} e_{i} \\\\\n", "\t\\dot{\\omega}_2 & = \\frac{N \\alpha}{J_{eq}} i_A - \\frac{B_{eq}}{J_{eq}} \\omega_2 - \\frac{B_{2C}}{J_{eq}} sign(\\omega_2) - \\frac{1}{J_eq}\\tau_{L} \\\\\n", "\t\\dot{\\theta}_M & = \\frac{R_A}{C_{TM}} i^2_A - \\frac{1}{C_{TM} R_{TM}} \\theta_M + \\frac{1}{C_{TM} R_{TM}} \\theta_A\n", "\\end{align}\n", "where $J_{eq}=J_2+N^2 J_1$ and $B_{eq}=B_2+N^2 B_1$. \n", "\n", "For simulation purpose only we can simplify as \n", "\\begin{align}\n", "\t\\dot{i}_A & = -a~ i_A - b~ \\omega_2 + \\frac{1}{L_A}e_{i} \\\\\n", "\t\\dot{\\omega}_2 & = c~ i_A - d~ \\omega_2 - e~ sign(\\omega_2) - \\frac{1}{J_eq}\\tau_{L} \\\\\n", "\t\\dot{\\theta}_M & = f~ i^2_A - g~ \\theta_M + g~ \\theta_A\n", "\\end{align}\n", "where $a=\\frac{R_A}{L_A}$, $b=\\frac{N \\alpha}{L_A}$, $c=\\frac{N \\alpha}{J_{eq}}$, $d=\\frac{B_{eq}}{J_{eq}}$, $e=\\frac{B_{2C}}{J_{eq}}$, $f=\\frac{R_A}{C_{TM}}$, and $g=\\frac{1}{C_{TM} R_{TM}}$." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Matlab scripts\n", "## ODE solver `ode45m`" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Created file 'C:\\Users\\zurit\\OneDrive\\blog\\files\\Nonlinear_DCmotor\\ode45m.m'.\n" ] } ], "source": [ "%%file 'C:/Users/zurit/OneDrive/blog/files/Nonlinear_DCmotor/ode45m.m'\n", "\n", "function [tout, yout] = ode45m(ypfun, t0, tfinal, y0, tol, trace)\n", "%ODE45\tSolve differential equations, higher order method.\n", "%\tODE45 integrates a system of ordinary differential equations using\n", "%\t4th and 5th order Runge-Kutta formulas.\n", "%\t[T,Y] = ODE45('yprime', T0, Tfinal, Y0) integrates the system of\n", "%\tordinary differential equations described by the M-file YPRIME.M,\n", "%\tover the interval T0 to Tfinal, with initial conditions Y0.\n", "%\t[T, Y] = ODE45(F, T0, Tfinal, Y0, TOL, 1) uses tolerance TOL\n", "%\tand displays status while the integration proceeds.\n", "%\n", "%\tINPUT:\n", "%\tF - String containing name of user-supplied problem description.\n", "%\t Call: yprime = fun(t,y) where F = 'fun'.\n", "%\t t - Time (scalar).\n", "%\t y - Solution column-vector.\n", "%\t yprime - Returned derivative column-vector; yprime(i) = dy(i)/dt.\n", "%\tt0 - Initial value of t.\n", "%\ttfinal- Final value of t.\n", "%\ty0 - Initial value column-vector.\n", "%\ttol - The desired accuracy. (Default: tol = 1.e-6).\n", "%\ttrace - If nonzero, each step is printed. (Default: trace = 0).\n", "%\n", "%\tOUTPUT:\n", "%\tT - Returned integration time points (column-vector).\n", "%\tY - Returned solution, one solution column-vector per tout-value.\n", "%\n", "%\tThe result can be displayed by: plot(tout, yout).\n", "%\n", "%\tSee also ODE23, ODEDEMO.\n", "\n", "%\tC.B. Moler, 3-25-87, 8-26-91, 9-08-92.\n", "%\tCopyright (c) 1984-94 by The MathWorks, Inc.\n", "\n", "% The Fehlberg coefficients:\n", "alpha = [1/4 3/8 12/13 1 1/2]';\n", "beta = [ [ 1 0 0 0 0 0]/4\n", " [ 3 9 0 0 0 0]/32\n", " [ 1932 -7200 7296 0 0 0]/2197\n", " [ 8341 -32832 29440 -845 0 0]/4104\n", " [-6080 41040 -28352 9295 -5643 0]/20520 ]';\n", "gamma = [ [902880 0 3953664 3855735 -1371249 277020]/7618050\n", " [ -2090 0 22528 21970 -15048 -27360]/752400 ]';\n", "pow = 1/5;\n", "if nargin < 5, tol = 1.e-6; end\n", "if nargin < 6, trace = 0; end\n", "\n", "% Initialization\n", "hmax = (tfinal - t0)/16;\n", "h = hmax/8;\n", "t = t0;\n", "y = y0(:);\n", "f = zeros(length(y),6);\n", "chunk = 128;\n", "tout = zeros(chunk,1);\n", "yout = zeros(chunk,length(y));\n", "k = 1;\n", "tout(k) = t;\n", "yout(k,:) = y.';\n", "\n", "if trace\n", " clc, t, h, y\n", "end\n", "\n", "% The main loop\n", "\n", "while (t < tfinal) & (t + h > t)\n", " if t + h > tfinal, h = tfinal - t; end\n", "\n", " % Compute the slopes\n", " temp = feval(ypfun,t,y);\n", " f(:,1) = temp(:);\n", " for j = 1:5\n", " temp = feval(ypfun, t+alpha(j)*h, y+h*f*beta(:,j));\n", " f(:,j+1) = temp(:);\n", " end\n", "\n", " % Estimate the error and the acceptable error\n", " delta = norm(h*f*gamma(:,2),'inf');\n", " tau = tol*max(norm(y,'inf'),1.0);\n", "\n", " % Update the solution only if the error is acceptable\n", " if delta <= tau\n", " t = t + h;\n", " y = y + h*f*gamma(:,1);\n", " k = k+1;\n", " if k > length(tout)\n", " tout = [tout; zeros(chunk,1)];\n", " yout = [yout; zeros(chunk,length(y))];\n", " end\n", " tout(k) = t;\n", " yout(k,:) = y.';\n", " end\n", " if trace\n", " home, t, h, y\n", " end\n", "\n", " % Update the step size\n", " if delta ~= 0.0\n", " h = min(hmax, 0.8*h*(tau/delta)^pow);\n", " end\n", "end\n", "\n", "if (t < tfinal)\n", " disp('Singularity likely.')\n", " t\n", "end\n", "\n", "tout = tout(1:k);\n", "yout = yout(1:k,:);\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## ODE solver `eufix1`" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Created file 'C:\\Users\\zurit\\OneDrive\\blog\\files\\Nonlinear_DCmotor\\eufix1.m'.\n" ] } ], "source": [ "%%file 'C:/Users/zurit/OneDrive/blog/files/Nonlinear_DCmotor/eufix1.m'\n", "\n", "function [tout, xout] = eufix1(dxfun, tspan, x0, stp, trace)\n", "%EUFIX1\tSolve ordinary state-vector differential equations, low order method.\n", "%\tEUFIX1 integrates a set of ODEs xdot = f(x,t) using the most\n", "%\telementary Euler algorithm, without step-size control.\n", "%\n", "%\tCALL:\n", "% [t, x] = eufix1('dxfun', tspan, x0, stp, trace)\n", "%\n", "%\tINPUT:\n", "%\tdxfun - String containing name of user-supplied problem description.\n", "%\t Call: xdot = model(t,x) coded in fname.m => dxfun = 'fname'.\n", "%\t t - Time (scalar).\n", "%\t x - Solution column-vector at time t.\n", "%\t xdot - Returned derivative column-vector; xdot = dx/dt.\n", "%\ttspan - Range of t for the desired solution; tspan = [t0 tf].\n", "%\ttf - Final value of t.\n", "%\tx0 - Initial value column-vector.\n", "% \tstp - The specified integration step (default: stp = 1.e-2).\n", "%\ttrace - If nonzero, each step is printed (default: trace = 0).\n", "%\n", "%\tOUTPUT:\n", "%\tt - Returned integration time points (row-vector).\n", "%\tx - Returned solution, one column-vector per tout-value.\n", "%\n", "%\tDisplay result by: plot(t, x) or plot(t, x(:,2)) or plot(t, x(:,2), x(:,5)).\n", "\n", "% Initialization\n", "if nargin < 4, stp = 1.e-2; disp('H = 0.02 by default'); end\n", "if nargin < 5, trace = 0; end %% disable trace if not requested\n", "t0 = tspan(1); tf = tspan(2);\n", "if tf < t0, error('tf < t0!'); return; end %% check for glaring error\n", "t = t0;\n", "h = stp;\n", "x = x0(:);\n", "k = 1;\n", "tout(k) = t; % initialize output arrays\n", "xout(k,:) = x.';\n", "if trace\n", " clc, t, h, x\n", "end\n", "\n", "% The main loop\n", "\n", "while (t < tf) \n", " if t + h > tf, h = tf - t; end\n", " % Compute the derivative\n", " dx = feval(dxfun, t, x); dx = dx(:);\n", " % Update the solution (with no check on error)\n", " t = t + h;\n", " x = x + h*dx;\n", " k = k+1;\n", " tout(k) = t;\n", " xout(k,:) = x.';\n", " if trace\n", " home, t, h, x, dx\n", " end\n", "end\n", "if (t < tf) % if true, something bad happened!\n", " disp('Singularity or modeling error likely.')\n", " t\n", "end\n", "% ... here is the output (tout in row vector form)\n", "tout = tout(1:k);\n", "xout = xout(1:k,:);\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Nonlinear model\n", "In line 39 and 40, the input $e_i$ can be changed from constant input to sinusoidal input." ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Created file 'C:\\Users\\zurit\\OneDrive\\blog\\files\\Nonlinear_DCmotor\\asst02_2017.m'.\n" ] } ], "source": [ "%%file 'C:/Users/zurit/OneDrive/blog/files/Nonlinear_DCmotor/asst02_2017.m'\n", "\n", "function xdot = asst02_2017(t,x)\n", "global E_0 Tau_L0 T_Amb B_2C \n", "\n", "% motor parameters, Nachtigal, Table 16.5 p. 663\n", "\n", "J_1 = 0.0035; % in*oz*s^2/rad\n", "B_1 = 0.064; % in*oz*s/rad\n", "\n", "% electrical/mechanical relations\n", "K_E = 0.1785; % back emf coefficient, e_m = K_E*omega_m (K_E=alpha*omega)\n", "K_T = 141.6*K_E; % torque coeffic., in English units K_T is not = K_E! (K_T=alpha*iA) \n", "R_A = 8.4; % Ohms\n", "L_A = 0.0084; % H\n", "\n", "% gear-train and load parameters\n", "J_2 = 0.035; % in*oz*s^2/rad % 10x motor J\n", "B_2 = 2.64; % in*oz*s/rad (viscous)\n", "N = 8; % motor/load gear ratio; omega_1 = N omega_2\n", "\n", "% Thermal model parameters\n", "R_TM = 2.2; % Kelvin/Watt\n", "C_TM = 9/R_TM; % Watt-sec/Kelvin (-> 9 sec time constant - fast!)\n", "\n", "Jeq = J_2+N*2*J_1;\n", "Beq = B_2+N^2*B_1;\n", "a = R_A/L_A;\n", "b = K_E*N/L_A;\n", "c = N*K_T/Jeq;\n", "d = Beq/Jeq;\n", "e = B_2C/Jeq;\n", "f = R_A/C_TM;\n", "g = 1/(C_TM*R_TM);\n", "\n", "if t < 0.05\n", " e_i = 0;\n", "else \n", " e_i = E_0; \n", "% e_i = E_0*sin(5*(2*pi)*(t - 0.05)); \n", "end\n", "if t < 0.2\n", " Tau_L = 0;\n", "else\n", " Tau_L = Tau_L0;\n", "end\n", "\n", "xdot(1) = -a*x(1)-b*x(2)+e_i/L_A;\n", "xdot(2) = c*x(1)-d*x(2)-e*sign(x(2))-Tau_L/Jeq;\n", "xdot(3) = f*x(1)^2-g*x(3)+g*T_Amb;\n", "xdot = xdot(:); % force column vector\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Main\n", "\n", "Change the input values in line 5-8, the `input_type` $E_0$ to constant or sinusoidal, and the step size in line 10. \n", "\n", "**Note**\n", "This script is an example only. In the **Simulation Results** section we will analyze different scenarios." ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjAAAAGkCAIAAACgjIjwAAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAB3RJTUUH5AwZFigQ8aFnLAAAACR0RVh0U29mdHdhcmUATUFUTEFCLCBUaGUgTWF0aFdvcmtzLCBJbmMuPFjdGAAAACJ0RVh0Q3JlYXRpb24gVGltZQAyNS1EZWMtMjAyMCAxODo0MDoxNoTkihIAACAASURBVHic7N1pXBPX9zDwQ0ggCcgWUIGAcanUtaKVtSpapGpbtah13+3mr64tttWnLtVqW7RqF6qtWq1arMqirf4VUbGIgqi4Y1QkkmDUhEWWLCRMnhdXpzGEPSQBzvdFP5OZm7lnhmmOc+fOvTY6nQ4QQgghS2NYOgCEEEIIABMSQgghK4EJCSGEkFXAhIQQQsgqYEJCCCFkFTAhIYQQsgqYkBBCCFkFTEgIIYSsAiYkhBBCVgETEkIIIauACQkhhJBVwISEEELIKmBCQgghZBUwISFrUVFRkZ+fT5YVCoVlg0EImR8mpGbjwIEDY8eO9fb2vnLlClmzZcuW3r17v/vuuzdu3Kj7fv7666+XX345MTGRfJw7d+69e/dMH+5zBw4cGD16dLt27RYuXLh48eIlS5YsXLiQTjzEtWvXxo4dGxMTk5mZGRMTk5iYuGTJkqYLyVSqO3V79+7VP8N10YCvNJjBBWAqWq125syZnp6ejaz0xo0bs2fP9vT0XPzc7NmzxWKxaaMFgKKiot27d+/evVs/qoKCgpiYmNjY2KNHj5q8RlQ7HWo+jhw58tVXX3Xr1k2tVpM1O3fubMB+5syZk5CQQJZnzZp18+ZNk4VozIkTJ9q0aUN/FIlEPXr0yMvLIx+vXr3q5+f35MkTusCaNWumTZtmqtrz8vKOHDliqr3p0z91BrXMmjWLPsN1DKwuXzEV/QvAtNzd3Rtf6aFDh8LDw+mP0dHRo0aNMkFwL1q+fDlZOHLkyJ9//kmWBw4c+PDhQ51ON2/evBMnTpi8UlQzvENqZubOnevt7b18+XLykcFo7F9w+/bt3bt3b3Rc9dChQ4cPP/xw0aJF5ON777338ccfe3h40AWioqKYTKapqjty5IipdmVA/9QZ1FKXv0sDvtJK/PvvvwMHDqQ/lpaWcjgc01ahUqmuXr1KlsPDw8+fPw8Aly5dYjKZ5CYvPDx8y5Ytpq0U1cpk/9sjs9mxY0ePHj1GjRoVHBysv/7SpUsZGRkeHh4ymWzu3LkAsHfv3n/++Wf27NllZWUikah79+4RERH6X4mPj4+Li5s/f35gYGB1hc+cOSMUCu3t7du0aRMZGUm+xWAwRCKRv7//oEGDdu3adeTIkXHjxonF4nbt2k2ePLnWQ3j55ZcXLlyo1WqlUumFCxd++OEH/a12dnYTJ040+ErV8FxdXUUi0Y0bN+h8VvUMxMbGrlq1KjAwMDk5ee3atWw226BMzcFv2bLl/PnzPB6vX79+SUlJPB7vu+++++CDDwAgIiLi8OHD5NRVrQUAioqK4uPj8/Lyqp52o4EZ/YrByTeIFgBqPScGf6wa/igNO8MA8ODBg4SEBIFA4OrqSu+t6mVTd2lpaWvXriXLp0+fTklJ2bdvX732UCs2m33//v3IyMiff/5527Zt8+fPB4Dbt2+3bduWFODxeMnJyaatFNXO0rdoqB6OHDkil8t1Ot2OHTteeuklpVL5xx9/kE05OTl0s8bZs2cXLVpElmfMmPG///1Pp9NpNJr27duTlQZNdvRy1cI3b94MCwsjW8PDw3NyctRqNQCQNsMePXqQhUmTJi1fvvzixYtbt26tGrZBk51Op3v06BEAXLhwgdwllJaW1uXw9cNzdXVNT08np4K0vVR3BiZMmEC3jBktU3Pwb7zxxtmzZ3U63ZQpU0gbzv79++/evWtw6vRrIWe46mk3UOtXqp78qtHWfE6M/rF01beeNeAMy+XygIAApVKp0+mEQiFpsjMaeR2b7NRqtb29fVxcXEJCwsaNG4cPHy4SiWoof/PmzYSEBLlcrt/qq9PpJBLJz9Ugf75Hjx7x+XxbW9s1a9aQr2zbto1uK5bL5fb29rVGi0wL75CapZkzZ8bFxS1ZsqR///5kzZYtW/h8PlkODg4ePHjw+vXrGQwGk8kcMGAAADCZzKKioqq70m8pqlp427Zt9vb25Kmvra3tpUuXOnXqpFQq79y5c+fOHTabXVpayuPxmExmv+fqEv/9+/cBoFu3bs7OzuRj79699QtIpdKqj8f1w1MoFIGBgQDA5XIfPnxYwxnQ34PRMjUHP3ny5L/++is0NJTNZicmJoaHh1dUVHTp0gVqa2Sr+bTX5StGT75BtDWfEzs7u6p/rBoCaMAZ3r1798svv0zu8AQCAdlqNPI6noRjx46FhobSN1X9+vULCQkRi8X5+fmXLl3Ky8tjMpnk5qykpGTt2rWTJk0aPXp0YmLiqVOn9G+1vb296Xu4qiiKWr169eXLl5OTkxcuXAgAy5YtYzAYFEWRAlKpFBtRzQ8TUnO1ffv2bt26lZaWTp06FQBkMpmDgwO9VaPRaLVaOzu7RtZSVFTUpUuX0aNHAwD5b1FR0fTp02fMmDF69OgdO3bQJev11OfevXshISGOjo5du3bt0aPHpUuXDBLSsWPHZs6cWa9Qaz4DFEVJpVKjZWoOfvz48UuWLJk/f/60adMWLVqkUqlqiIHU4u3tXfewa/hK1ZNP1P1UV/fHahijZ08ikVS9zKqLvC4MHiCRdEhR1M6dOwcOHDh69OjevXuTTDNlypRffvmFnLoRI0aUlpbq76ekpOTs2bNGqwgMDDx16tSgQYM8PDwmTpwYHh4eGhq6bNkyV1fX8vJyUubhw4d1T6LIVPCfAM1MRUUFWWjXrt3WrVt37txJPo4bN66goIAs375929/fv/HZCAAmT55M92xWqVQZGRmHDx/m8XiRkZEMBoP8Q76+HYiLioo2bNhA/2P2559/3rBhQ1lZGV0gMzPT39+/vqFWdwbIv3MrKiqysrIacJbs7OwiIiKWLl06YMCA4cOHv/fee2PGjKlaTL+WOgZc61eqnvw67pnW+D+WPqNn75133pFKpWQl/Wtea+SpqakGXf9paWlp+glp3bp1s2bNYjKZX3zxxYABA7RaLUmKSUlJvr6+dCJnMplvvfWW/n4qKyurOxCtVktRFH0z5OHhER4eDgChoaF0qGKxOCwsrKbTgZqA7cqVKy0dA6qT1NTUVatWJSUldejQwdfXFwB69Ohx69atcePGAcBLL72Uk5MjFAqfPHkSFxe3adMmZ2fnM2fO/Pzzz0+ePOnfv/+vv/6anJzM5XIrKytjYmKkUmlgYOD169djYmLkcnlwcHBWVlbVwlOmTMnPz7948WJxcXFKSsrIkSNtbW337Nnj5eWVlZVVVlZ29epVBoOxb98+iUTi4+NDAjMIe/PmzdnZ2a6urtevX09ISNi7d+8333xDt48JBIKuXbt+/vnntra2crk8LS2NoijScKTP6LG0b99+w4YNV65c6d27N3lWYXAGAMDe3v7w4cMymWzYsGG9e/c2KEPefKoueMLBwaG4uDg8PNzLy+v06dMTJkwgx0WfOicnJ/1aLly4UDXUkJAQg93W+pWqJ//kyZP60dZ6TgQCgcEfKygoKCcnh74AXFxcGnmGfX19Hz58ePHixadPn547dy4hIYHH440fP94g8rNnz+pXOmbMmLy8vOHDh+ufkDt37mzZsmXv3r39+/e/du3auXPnNm/eHBwcvG7dOgBgMBg2NjabNm367LPPeDze1q1bQ0JCevToQb6rVCqdnJz098bhcF6qhqOjY/fu3X/66ScWi8VisXbs2PHmm2/6+Pg4ODjIZLJ79+517Nhx9erVGzdudHR0rO3/S2RKNjqdztIxIJOhKEqlUnG5XJPvtqKigjwnIBQKBamFoihTNbVnZmYWFBQMGTKkMfd2Rs9ARUUFeZxWQ5maabVa8nV6oSqDWuqiLl+pevLry7R/LKNnj9xz2NnZqVQqOtSaI4+Pj69v77t//vmnX79+BQUFPXv2/PXXX52dncePH082HTt2bNiwYfU9llu3buXl5YWHh+v/CW7fvi0SiRp5HaKGwYSEEDI3sVgskUgM3luo2eHDh7/99lvS2T0+Pl6lUi1evHjmzJkcDicvLy8iIsKE764hS8GEhBAytytXrvTp06fx+7l3756zs7P+W9WoWcOEhBBCyCpgLzuEEEJWARMSQgghq4CPAaulkaeXnjMcUc0lPJXB5RcnD6AUkurW0Dh+C+x9xjK4fHOEixBCzVxLfoY0derUCxcuNOy7b75K/b/x2qrrI9eypEU28Us1nq666tbQjlxkbE+ylRbZNCwGhBAyoYCAgN27d1s6ipq05ITk5+cnFAob9l2NPL1CHMd0D7T3GWuRAEzC4gFYQwwWD8AaYsAArCEGDKBW2GRnHMs9iOUe1MideLrqlMJNAMDxW2iKoBBCqCXDTg3GKYWb1OKDjdyJp5tOKdysFseZJCSEEGrZMCEZpxRuLs+KauROugdPY/KCKIVEI083SVT1NWXKFIvUa1UxWDwAa4gBA7CGGCwegPXDhGRE1f5yDTN16lR73zEAUGGhmyQyM4VlWTwGiwdgDTFgANYQg8UDsH74DMkITUE6ADSmOwONxQty8I82ya4QQqhlwzskI1i8IHufsUz3wMbvisHlk2xkqVY7hBBqLvAOyQgGl+/gH22qvVEKSXHyAAaX7xKeaqp9IoSIxrxu2PJY/5tGNcOEZATpX8fiBZlkkAUGl8/kBWkL0jXy9MZ3JUcI6btw4YKVv1tjTn5+fpYOoVGwyc4Ik3Sx08dyDyS7NeE+EUKohcGEZATpZWfCMejIYyRKKTFV/z2EEGp5MCEZIr0PTDsiKmm1oxSSxr9sixBCLRUmJEOUUgIALJ6JH/Zw/BYAAI7agFBr9vHHHx89erSGAqtXr/7888/J8rp169567vz582YJ0MKwU4Oh5+113qbdrS12bUCo1SstLa2oqKhu6/nz5zdu3Dhp0iTycffu3SdOnGCz2QDQpk0bM4VoUZiQDFGKfDB1kx3Zob3vGJZ7IGYjhJpISk5Ryr1i89cb1sUlrLMr/fGvv/6KjY2lKGrUqFGzZ88GgO3btyckJPB4PJlMRsocPXp0y5YtAPDOO+/MnDkTAMrKyj799NPly5ffu3cPALRabXl5OZvNvn//fv/+/c1/UBaBCckQGabB5E12YKKhHxBC1Um5V7wqKdcSNXekE1JaWtq6detOnDhhb2//5ptvOjk5tW/f/tdffz158uSjR4+Cg4Pff//9K1eufPrpp6dPn3ZwcHj77bd9fHzCw8PnzZv35ZdfymQykpAyMzNLS0unTp2q0WgeP3588uRJDw8PSxyaWWFCMkQe9jQRtfigUrjZ3mcMTkiBkMmFdXEB6Gihep/Zu3fvvHnzSPJYtGhRXFwcj8f78MMPHR0du3TpMmzYMADYt29fz549k5KSAKBbt26xsbElJSUODg7Dhg2jX2v19fU9dOjQgAEDAGDJkiUbNmz45ptvzH9oZoYJyVCT3scwOHxKIVGL4zAhIWRyYZ1d9ZvOLEKpVHK5XLLMYDAAoLS0lDwHAgA7OzsAKCgocHd3J8uDBg3y9fWdNGlSp06d3nrrLYlE8vTp065du86dO7ddu3bkW/369Tt8+LAFDsbssJfdCzTy9PKsqKbrnG37vP83Dm2HUIsUHByckJBAlpOTk3v27DlkyJC4uGfda7OysgAgIiJCJBKNHz9+/Pjx+fn5Eonk+vXrhw4d2rdv39y5cyMiIt5///0dO3ZERkaSbx07diw0NNQih2NmeIf0Akr57FWhJrpPYnD5LPdAbUF6hTgOezcg1PLMmTPn9OnTAwYMsLOzc3Z2Xr9+PZvNTkpKGjRoEABotVoAGDdu3NGjRwMDA3k8Xnl5eWJioqOjI/k6h8NhsVhsNnvWrFlHjx59/fXXKyoqfH19586da8mjMhdMSC/QyjOgCfp867P3GWvvM9bkvfgQQtaAwWDExsaSvt2kUQ4Adu/eXVFRwWQySSMeAPz+++8GZYipU6eSaZOYTGZiYqLRMi0YJiQjmjRbkJ1TCkmlQoI3SQi1SFVTSF3W1GU/LRsmpBfY+Yxhugc2RZ9vfTghBUIIVYUJ6QVmu2XBURsQQsgA9rJ7QUnaRNNOPGEU6doAOCEFQgjpwYT0H7X4oLbATL2x6QkpzFMdQghZP0xIlkFPSKEUbrJ0LAghM8HRvmuGz5D+Q/p8M90DzVMdx29B6bl0HLUBodYDR/uuGSak/1SSiSc4ZnpDCCekQMi0FDfPKW8auZPgvfuJ0fIF+zcYXd/I8k002velS5eOHTsmEonEYvG4ceOYTGZsbKy3t/fPP/9MD03U3GFC+g95omNrrldWyYQU9r5jMBshZBLKm+eN5gxzJqSmG+378ePH33777eXLl9u0adOhQ4cVK1YcPXp05MiRf//997hx42o+M80FJqT/PJ+az3xjKOCEFAiZEKdHcHW5xKh6Fa5j+aYb7XvgwIHh4eFdunQBAC6Xu2jRIgaDERYWVlJSUq+jsGaYkJ4ho50ym/iV2KrU4oPlWVEcvwX4JAmhRuL2COH2CKl7+aZISE062re9vT1ZY2trS3ZO19UyYC+7Z2y5/DYhsU06GZJR5JGVWhxn5noRQk0BR/tuDLxDeobB5VtkwFNbLp/B5ZMJKfBhEkLNXdON9l1zf/GWARPSM0rhJo08g+O3wMxZgcHl2/uMUYvj8CVZhFqAphvte8SIESNGjCDF6N56LWxaCkxIz6jFcZRCYsuNNn/V9j5j8QESQi0JjvbdMPgM6RnSxc4i6AkpcBpZhFBrhgnpBZaaN08jTy9OHlB+pcnHdUUIIavVzJrsLly4AAABAQHkY1lZ2R9//CEWi0eMGEE67DeYZacmwq4NCCHUnO6Qbt68uXDhwry8PHrNnDlz1Gr10KFD169f//fffzdm55bqZUfXbu8zBgAqsP83Qqi1ajYJ6c8//5w3b56Pjw+9Jisrq7y8fNGiRUOGDPnqq6+2bdvW4J0rhZuKkwdY9hEOGbVBY675LxBCFjRx4kQXF5dVq1ZNnz69ujJarXbw4ME1DMba8jSbJrsuXbocOXLk66+/ptfk5eW9/PLLZLlXr153796lKIruVVkvlCKfUkga2fFaVKgSFSkFrhzyUeBWv+EOGc/HWlWLD+KQQgi1bIcOHVIoFAUFBfn5+UYL5OfnT5s2LSUlxbxxWVizSUj0cyOaQqGgB9JgMBg2NjaVlZUGCcnPz48sTJkyhfTuN8rx8VkGwGMlnxKJ6hWVpEQbl10qKdUezC7lOzElJVoACFBlP2zbCwB6uVDTXlUGeXPquDdb5/EOBemlNzdIK1+tVxjVhiex/LtNFo/B4gFYQwwYgDkZjOR96dKlM2fOLF68GAB++OGHwMDAmJgYrVYbGRm5bNmy5ORkoVAoFosXL14sk8k+/fTTrVu3stns6OjoTz755Nq1a/WtXfTij9ju3bv37NljoiNrcs0mIVXFYDAoiqI/6nS6qrdHQqGwLrsqvPYIAHy71qM3gahQtTNTuirpvwdaTCYzrLOjqEi1O+vr/AL3eIeBP5VE/l+eVODGDuvsOr1/+7DOrjXvk1IwSx7xQSHxdnxkqq4NAoHAJPtp1jFYPABriKE1BKCRpxud9Jm856c/GWbVNY0vT1QdybuioiIzM5NsvXTpkq+v765duxISEuLj448ePZqRkTFv3rw+ffoMGjRo2bJlU6dOJQPfbdrUwKk7Dc7zl19++eWXX5Jl+h/oVqsZJyRnZ+fy8nKyXFxczGKxbG1tG7CfBgzynZJTNDgmizTKCdzY01/1XPlGR7KpYP+Ggizw1srnPY2PLP/3ArtbvGrgzsJuOzOlAjf27xO61ZCWGFy+Q59oW4t2r0Co+dIWpCuFm6uuf55ONtewpvHliaojeY8ZM6bmsLlc7t69e19//fVRo0ZNnjy55sItWzNOSAEBAUuXLpXL5e7u7gkJCWRc9wYg/QhYdR7ne+Xx3FVJuQAgcGXnLjMcWpj37idOYe8qb55/mvKX983z75SlvlOWms90n9Vri6hQNXNftn72qorcGFEKCeYkhOqLyQviVH8PUHXo5JoHU65veaLqSN5FRUX0VjKWXVUODg4sFku/yad1asYJyc3NbcWKFePGjevUqZNUKv3jjz8atp+6j9EgKlTN3HcrJacYABbZZX54ajPMfVi1GKutD6utj9Pgd3Mvn3e6d67k9H5vmfj0R313ZkpXJeWuSsrddVFaQ1pSCjcphZsd/KOxawNC9cJyD6qhrbvqAF01D9lV3/JERETE77//HhMTAwDff/89g8FwdnZ+8uQJ2Xr9+vWqk+lVVFRMmDAhMTFx5cqVBw4caDGz7TVAM0tIa9as0f84atSot99+W61Wczh17ThQFaXIBwCme2CtJWfuuyUqUgnc2Ovkv/a9mwQABfs31DBFio2bJ33D5OTGXvlGxxn9PUlKI2np9Ed9q3bGY/KCADYrhZsxISHU7FQdybtNmzarV68eNGgQSU5Vv7J06dLw8PDQ0NAdO3aEhISEhIR4e3ubP3Jr0MwSUlUMBqMx2QgAHPyjOX4Lam0iGxxzOSWnWODG/rdig1J8HgD4qw7WZTYwcsNElgVu7NNzn90qiQpVg3+5HNOrePjIEfrlcdQGhJq1qiN5p6amKhQKNptNd7wi07zqD+ANAD4+PmKxWH9X9KjerUSzT0gmUWs2Wnk8NyWn2FsrW539q1KVzfLwaffxxnrNTalvRn/PsM6uOzOlk34LhSwoUH2if5tFT0jRsJ0jhCyu6ijdLWxq1ybSbEZqaDqFhzsWJ9c0Dl5KTtGqpNwAVfap/EUBqmyWh08d741qIHBjr3yjI8lDBfs3bIj6VFSoordy/Ba6hKfi7RFCqFVp7QmJDBdE5hE3SlSoGhyTBQCkYY3TI7jjLxl0E1wj8d79pP3/NgHA27l/pq/+cGemlN5EKSRK4Sa1+KBJKkIIIevX2pvsyHBBttU32c3cdwsAwjq7rHyjI7xhpE9dIzkNfpfTIzh3bmBfcRJrR/bKwn2k912lQqIUbmZw+di1AaEaBAQEWP/7nmZTdUSb5qW1JySCwTXep2VnppR0ZDg9t2/T1c5q69MxJkOyYqxGJh70+6iVcGjlGx1tnw9th10bEKrB7t2761hSJBJZdrgKiwdg/Vp7k529z1iX8NTqXi+YuS8bAH6f0K2pw2C19eGvOsjy8PHWylcl5a48nsvg8u19x0D1L4cjhFALY4E7JLVaXZdi9MCpTaqGMRFINgrr7FLrGHQmQXLS3gdM2Je9Kin3TE7RyRlBAEApsf83QqhVsEBCmj179iefVPsyKZGfn9+vXz9PT8+mDqY4eQCDy686XSwZOxUAVlQ/zI/Jsdr6zGgLYZ1dB/9yOSWnuPNG1T/Bb3qWHakQx2FCQgi1eBZISAKBwN/fv+Yy7dq1M08wUM3QQYe3bxU++Hbd2JPmuT3SJ3Bjn/6o7+BfLosKVSNODswKPIKz9iGEWgMLPEMyGP7HKC8vLzPcHpFO1VW7saXkFPW6tB0Afu5VZORrTY/kJIEbO0/lnlbsB8+7pyOEUAtmXZ0avvjiC0uHAABwLf4Pb60cABr59mtjkJwU1tnlY+Gs7ccD8xl9LBUJQgiZh1UkJIVC8eeffw4aNOjQoUPmrFcrz4Aqw6qKClXKm+cAwCnsXXMGU5XAjf37hO4nhAvfzv0z8ccv7+ffs2w8CCHUpCyZkNRq9T///DNkyBB/f//ffvtt/vz5o0aNsmA8xKqk3HfKUgGghmG8zUbgxm7/v032nW2n+R90y56ZkmOZJkSEEDIDy7wYe+bMmdWrV4vFYnd39zlz5vTq1atXr1729va1Tq1oWkan5nt6+i8A4PQINtX4QI3kNPhdx8CQgsTXACRd9gxfGRRbw/x+CCHUfFnmDumnn37y9vY+fvx4WlrazJkzjc4RYgZkCDv995B2Zkojy1MBwDlsvEVCMorB5TsNWA8ALE/p9MQJ245esHRECCFkepZJSAcOHNi1a5dWq50+ffqePXtKS0vJejPP4OsUGus2Mld/DRnVGwCcBlv4AZIBFi+I47dAdddTIxO//ve8lcdza/8OQgg1K5Z8htSlS5ddu3ZNmjTJxsbm/fffP3bsmP5cVU2NzICnv2ZnplRUqPLrsKfrQdMPotpIDC6f47fQK+qAY2BIJ+46MryQpYNCCCFTsnwvOwaD4e/vv2vXrtdff12r1ZqtXrX4YOm5ieVZUfSaXZlSMMvIdQ2mLYrj+j9eEdERADAnIYRaGAskpPv37xtdz2Kxli5darYwKEU+vNjnOyWnGABm9G/yF3IbTCPPoBSSzwSHMCchhFoeCySk5cuXq6sRGhpKFu7fvy+VSmvfVyOQLnb01Hxk5DprzkYAwPFbAABqcdzKNzqSOznMSQihFsMC3b4/+eSTW7du1VqsU6dOTRoGixekgXR6aj7SXjeos0uTVtpILPcgepKkGf2DAGDmvuxVSbkAgH3BEULNnQUSUq0jq5qHg3+0/kfrb68j7H3HaAvSy69EuYSnkmhn7sue9Fvoo7vvtv94k6WjQwihhrN8pwZL0e9i1yza6wjyGi/dRXBGf8/cZSEAUJKyP/ejQM0TsYXjQwihhmqlCcmgi92qpFzhgylW3l5HMLh88iSJnklW4MbuGJORz3TXyMSSFWMxJyGEmqlWmpDIsKoMrjf5+Nb9PwFgWMa3loypzux9xtITnBOstj6v/XCE5eFDcpLi5jkLhocQQg3TShMSQQYN2pkpjSz/F6xvdIbqkCluDaZxIjOg/91xkkYmfvzTooL9GywVHkIINUwrTUj6w6paw+xHDaAUbipJm6i/htXWZ37UApKTCvZvEBWqLBUbQgg1gDUmpJKSkqaugkxbTu6Q2kkvg3VMNlEvGnmGtiBdKXyhZx2rrc+YL9b83XGSX4c9g3+5LCkx38gXCCHUSFaRkB4+fHj//v3r169/8cUXkZGR777btE1nJBsROzOlZPYji0/HV1/0S7IG6wVu7DFfrAnr7CIqVE1MeIhTKCGEmgurSEgffPDB7du37ezsRowYsXXr1nXr1jVpxnUqdwAAIABJREFUdaS9jjyDKUnZDwBOYe9ayexHdUdekqUUEv38SpCpZldEdJSUaGfuy8ahHBBCzYJVJKS///7bxcXlwYMHPj4+Hh4eTf3mLIPDZz6flM/t6mFobk+PaBy/BW4jc/Xnc6IJ3Ngz+nsuDHAVFap2XZRiTkIIWT/LzBhbVUhICADs2bPHzs7Oy8urSetiuQex3IMAYGemdGa7ZTP6e/4+2HpH+K4BOQr952H6BG7sBYGuzi4uq5JyyfBC857GN7tHZQih1sMq7pBoU6ZMcXJymjBhQpPWopGnk2EOzuQUg9WPX1czpXBTcfKAMr1JNAysfKMjGRp80m+hJaf3Y3dwhBosNTX1wIEDycnJ+itjYmIsFU/LYy13SDRHR8dvvvmmSasovxJFKSQu4ankgX9YZ9cmra5J2fuMVQo3U0qJRp5ObpiqWvlGxxn9PTdERc6TxZOE1Nrukw4cOLB3797z589PnDiRwWC4urryeLy5c+dWV/6vv/5asWLFN998M3r06Lrsf+7cuYsXL+7SpYvpQm6xbt26lZmZOX369MbvatOmTQsXLiTLDx48SElJUSqVXl5eI0eOrFo4MzPzyJEjHTp0eOuttzw8POiV58+fb9OmDVk5e/bsgQMHstlssnXgwIGenv8NJ6bVamNiYmJjY+Pj47ds2dK+fXuKos6dO9ejR4/GHwsirOsOiRAIBE26f7qNi7ypI3BjN2l1TYrB5dv7jKUUkooq3e30CdzYn0Sv/7vjJAAo2L/h0U8LzRWgVRg3btzHH3+sVCo3bdr0/ffff/nll0lJSfRvWVXjx48fMGBADTsUi8VHjx6lP6rV6oqKClNG3PQMDsFsHj9+7Opqgn8CHj58ODr62fjIFEVNmTLlnXfe+fDDD6Ojo3///XeDwkePHo2Li5sxY4ZGo6H7TJGV8+fPJyspirpx48ahQ4cSExMTExOjo6PbtGmjv5O0tDQulwsAI0aM8PLyGj169FtvvUVR1MyZMxt/OIiwzB3Sxo0b/fz8yITlhYWFbm5uABAVFfXxxx936NChSasmjXUMLr8ZDahaM47fArX4IOk6WAPSHXzKjwN3XP+wJGW/8ub5jr9kmCdCKzRs2LClS5du2tTA8dGPHDni6+tLf9y+fbuJ4jIfg0Mwm8GDBzd+JzKZrKjov/cZKIpycHB49OiRk5NTu3btsrOz9QtTFPXBBx/k5OQ8fPgwJCSEPK6mVwIAWZmVlbV9+/aePXsCQFJSkrOzs6Ojo/5+nJ2dyZTWpaWlJDNFR0d/9tlnjT8cRLNMQnrvvffu3r0LALNnz87NzVWr1R999NFXX3118+ZNLy8vFovVdFXbcvn2PmMZXO8zWc3+ARLB4PLJJElK4SaOX023PgI39p55EcnpiZ3/mOMtE+d+FNhqc1JiYmJU1H8P3s6cOSMUCu3t7du0aRMZGalfMj4+nsFgiEQif3//QYMGAUBsbOyqVasCAwOTk5PXrl1L/0M7MDBwy5Yt58+fd3Z2/uGHH2JjY48dOxYRETF58uSMjIykpCSj+69a+65du44cOTJu3DixWNyuXTutVqv/cfLkyZcuXcrIyPDw8JDJZKTh0eArkydPNqji2LFj7du3p4/C4BDoRipi7969//zzz+zZs8vKykQiUffu3V1dXUUi0Y0bNz7++GPS3lX1tOgf+/z580tLS+fMmRMaGqq/59jYWJFI9NlnnzEYjWqbiYuLmzJlCj3BNJPJPHbsGACUlZXduHHDoM0/KSnJy8vr+vXrqamp/fr1I/e+9Mrbt2/7+vrq3xAXFBTcvn17/vz5BpX26dOHJKRTp06NGzcuIyNDIBC0a9euMQeCDOks6rvvvtPpdOXl5Tt27AgODn7zzTdNuPOuXbvWsBUWn4TFJ3MLlCas0UBubm7T7Vxfhex8wSFB0YnX6hjAb0cyTo3vBYtPhv18qaljM9tJqDmAEydOcDichISEuLi4N954Y9++fXSBmzdvhoWFkeXw8PCcnBydTjdnzpyEhAS1Wg0AarVap9P16NGDLOh0ugkTJhw5coTew6xZsxISEsjy+++/v3PnTp1OV1lZuXbtWrL/oKAgg/3XXPukSZOWL19+8eLFrVu3GnzMyckZNWoUKX/27NlFixaRZYOv6DN6FAaHYGDGjBn/+9//dDqdRqNxdXVNT0/X6XQ7duxYvnx5dTvU6XQJCQkTJkzQ6XRLly59+PChwV8hISHh6dOnQ4YMMTgDOp1OJBIVFhaSZYlE8nM17t69q9Pp/v77b4lEUlpa6uXlpb+Tf//9d8GCBX/++afBznfu3PnSSy9dv349Nzd36dKl5E9PryTR6l8PCxYsUCqN/yw8efJk//79QqFQqVSSM5+VlbVnz54nT55UdyYNTkJdijWdmn8SrYGFOzVERUVduHChTZs2M2fONE9TrFp8EADuZiuEDz7367CnWT9AounPJFtd1wZ9c0YEiIIuwNfnUnKKO3597vRHfVvGeagZk8kknRS6du06atSoUaNGkTuDbdu22dvbJyYmAoCtre2lS5fo2Yrt7OyUSuWdO3fu3LnDZrNLS0t5PF7VPev/e/+jjz5atGjR9OnTjx07NmfOHLJ/Ozs7o/uvrnYmk9nvORI5/XHJkiV8/rMu/sHBwYMHD16/fj2DwTD4ij47O7vbt2/XehQG54rcNDCZTIVCERgYCABcLvfhw4c1nJbRo0f/+++/77333hdffKHfHYDo169feXl5QUFB1cmgP/roo4ULF0ZERACAt7d3Df1NHj9+TMqUlZUZbBowYMCAAQPeeuut/Pz8Tz/9lF7PYDDs7Ox69uwpEokCAwOXL18+fvx4eiUA0CsBQCaT5eXlGdwy0jw8PMaNGwcAq1at+uyzz8Ri8Z49e9avX799+/bZs2fXfEpRXVi+l11AQIA5q1MKN1MKibbkPTsAacfDAEPMWXvT4fgtsOVGG31J1iiBGzt3WcjgXy6LClWDf7k8/VXP1jMJes+ePR8/fpycnPzWW28BQFFRUZcuXUiuMuhWV1RUNH369BkzZowePXrHjh0G+6EoSiqVent766/s06fP06dP79y5k5eXR56SFhUVCQQCo/uvoXYm84X/N+mPMpnMwcGBXq/RaLRarZ2dXdWv6Ffxv//9b+7cuVWPwugh1KqG0zJt2rQpU6boP+Ch+fj4fPvtt9OmTTNYn5qaGhwcrFAoyMeSkpKzZ88arZc0ij59+vTUqVPl5eVFRUWLFy9eu3ZtUVHRqVOnSEPliBEjVq9erZ+QXF1d6adlTCZTJBJVtxIAEhISau1zkZaW1rVr13bt2v3666/9+/cHAB6PR1FUI9shEVhnL7u6u3r16sHnqruIDZAudrdST0GzHaDBKJZ7EIPLNzqSUHUEbuzTH/VdEdFRVKhalZTbqkZzYDAY9+7dA4A7d+5MnjyZLAOASqXKyPjvudrhw4d5PF5kZCSDwSA/suRWhvz0VFRUZGVlVd35+++/v2LFCj8/P/Jx8uTJ9O+dwf7J1upqN2rcuHEFBQVk+fbt2/7+/iQb1eDw4cOurq4GR1HzIdS6Q6Onpaio6MKFC8ePH//ggw+M5qS9e/dOnTr1wIED9JqKigq1Wt2lS5fS0lKyprKysrp6tVrtihUrvv/++++//37cuHGurq7ff/89m81OTk6mM1B+fj7JNGlpaVKpFAAiIiIuXbpEtpaWloaFhVW3EgDOnDnD4XBqOHaFQhEXFzdx4kSyTD/wpiiqplOG6sZ25cqVlo6h4b777rvbt2+r1WqZTGZra9urVy/9rT/99NO8efMMvkImWi2/qHWiFB4zV9k6ODddeMXFxS4u5us0oZGnl5x5U/s02953bB0DcOEww7q4AsCZnOIzOcW7Lkrf54lNO6yfmU+C0QCuX7++efPm7OxsrVbbpUsXZ2fnsrKy1NTUwYMHC4XCiIiI/Pz8ixcvFhcXp6SkjBw58uzZszExMVKpdNCgQYmJiV5eXllZWWVlZVevXg0KCvLy8rK3tz98+LBMJhs2bNjFixdjYmLkcnlwcLCTkxMAdO/effXq1evXrycBdOrUSSgU3rhxg96/jY0NHV6nTp0Maj958mRMTIxEIvHx8fH19U1OTtb/+NJLL+Xk5AiFwidPnsTFxW3atMnZ2dmgjMEZYDAYf/zxh6+vr/5RCAQC+hAMfoLPnDnz888/P3nypH///r/++mtycjKXy23fvv2GDRuuXLnSu3dvgUCwZ88eg9Ny8ODBuXPnBgQEDBkyZPv27bt27RIIBJ07d6b/Ci4uLjt37vTy8nrppZfat29P1v/yyy9sNvvKlStyufz1118HAA6H81I16G5vd+7c+f77769cucJgMIKDgzt37lxRUcHlcktKSjZu3BgTE+Pl5RUZGenm5ubv729ra9umTZt///3Xycnp999/j46OdnV1pVd6eXlt3ryZrASAvXv3enp6hoeHV3c5ffPNNwsWLCCRcLnckydPDhky5NKlS3369KnLpWjZ/xeM/iRaF0s/xGqU4cOH1/CcsOoTPFXegYJDgsO73hCO8RSO8Wza4Mz+DLOyXEy6NlTIztc3gNwCpWBNGiw+KRzjKf1xQcXjPFNFZfEHudUFcPfu3XPnztEfKysrq3uUXV5eTpehV6rVao1GU12lBptyc3Nr2H/NtVdXno6qLnJzc6seRc2HUCujp6WGAEhJugeETqfLyckh3QrS09NnzJjR4EiICxcunDp1qrpg7t69u3PnToPjvXv37v/93//pr8zJyaG7V1RVWlp66NAh/TVxcXH79++/evVqXSK0+P8L2KmhCVVWVorF4pMnT968ebNTp07vvfeevb29QRm62WTKlClTp05lFck5AO2fygDAxtWTbkhpIhJJXVvPTIXjOoxVdKwwe5fSp319A9j9dlvP9F2VD6AkZX/p1VRG/xG2ESZ4Tmv+k1DHAJhMpqdnk18DNcdgNlYYwJUrV3bu3Ll06VKRSHT+/PmrV6+ePn26Y8eGP8j08PDw8PDIy8szupXJZHbu3NkgDCaT+fLLL+uvZDAYT58+ffr0aXW19O7dW/+a6du3L1moy4Vkkb/C7t279+zZY/56G8ZGp9NZOoYGysrKmj179meffda2bdt9+/ZRFPXbb7/pF/Dz8xMKhfpryrOi1OKDd2+6ul565BT2bvuPG/heZB2JRKKmHnXCAKWQlJybyPFbQCbXaEAAd4V3f/hu87yn8QDA8vBxGvxuI8cZMv9JsLYArCEGDMAaYrB4AFV/Eq1NM75D8vf3v3z5MlkOCgrq16+fTCajR6mqQftimbpl9WigMbh8l/DUxuzhJb+XPole///2jBx24dsAWXbB/g0lp/e32vdnEULm1Ix72UmlUnqMEA6HY2trW15eXvNXmO6BUsc3BzI3fDDkH06P4KaP0TKUwk0laRMb/HWBG3vNlIH3pm4b4r3xArvbEP7GVtUBDyFkKc04IeXk5CxYsIC8NJ6UlOTp6Vnr7bC9z9jMNsvyVO4CN06zmyK27jTyDPKSbIP3IHBjr3yj49kVo1ST1pFO4R2/PkfGokUIoSbSjJvsXnvttaFDhw4fPrxjx45isfjHH3+s9SvlWVEV4g4A3VvAEHY1sPcdoy1IVwo3g3ejJoMXuLHnjAhgekhXJeW2wvdnEUJm1owTEgBERUUtXry4oqKi5nfZCLX4oFp8UCMPBejerOdAqhWLFwQA2oJ0W+crAIJG7m1Gf8+wzq47M6Vk5tldF6UkLRXs39Da5lVCCDWpZtxkR9ja2tYlGxlo2UO3Mbh8jt8CALB/vMskOyQteLnLQgRubNKCd2esV8H+DbkfBeIUtAghU2n2CanutPIMAEh76tcC5kCqFen2zdA8asyTJANkqKHfJ3QTuLFJfweNTEzSkuLmOVPVghBqtVpRQmK6B/5VMj6t2K9lP0AiyEyyjIpHNc8kW18CN/aM/p6nP+o7Z0TA1HbLPue9n89018jEkhVjH/20UPNEbMK6EEKtTStKSPY+Y797MDJP5W7pQMyEtNoxuPUbyLku6Ba83pHTp7Zb9qNzJACUpOyXrBhr8roQQq1HK0pI5VlRC6ifhA+mDMv41tKxmAODyy97OZbMIVv3IcDrju4aznv3kyHeG390juzEXTc45nJKjpFhnhFCqFatJSFp5Olq8cGJ7dKghY7RYBRl1x4AStImlpyb2BQ5CV5MSwCQklM8OCar49fndmZKm6I6hFAL1loSEqWUAEBluQ4AWvAYDVVRCgnLPZCMcddEOQn0GvFWRHQkPfFm7svu+PW5lcdz0/OVTVQpQqiFaTUJSSEBAKpMBwAteIyGqkjvBiYviFJIyrKimrQukpZIT7ywzi6kg/jEeOnZeW+WnN6PPfEQQjVr3i/G1h2lyAeAyjJgebSibEQwuHxH/2ilcLODf7QZqhO4sWe4ec7o70kS0hcHXweARz9nAQDLw4fZlk8GEW89DacIoTpqLQlJU5AOAJrHlOOrrai9jsbg8kk20sjTK8RxZstMv0/olts1jvdUfOvsqXvCuwGybI1MrITzJSn7z312qTW8EIYQqrvWkpDoxyet/B/mpecmAgCD601635mBjZunU9/goMHvti9UJadfU9w8f/f23UB19sx92auScsM6uw7q7KKfmRQ3z7XyvxFCrVZrSUgEVaZrVT0aqnLwjy7PilIKNwOA2XISQYZqhREBokLVzkxpWE5RSk7xzkLpzkzpzH3ZAjc2SU4h344lzaqcHsHcHiHMtnzMTwi1Eq0iIVEKyUHtJ3/ctJ3z2eiubVt1MxEZUqg8K0otjmPygljuQeaPgfR9AOgoKlSl5BTtypSm5BSLClUkOe1mdwuQZQOAJkVckrIfAFgePpwewU09vS9CyOJaRUJicPnrsvuKilV7WvQg33Vk7zOWUkiUws3lV6Ic+kRbJCcR+t0fREVKUaFqV6Z0Kizz1sq8tXJvrSxQnf1OWapGJtakiI8FfkbuoiwVLUKoqbWKhKQUbvrU42IsFSpwG2LpWKwCaawjOckpJJbB5Vs2HoEbW+DGhs5AkhMAkDunLUURnxd+4K2VBaiyE/Zlw/Nh2klaGtTZReDGfvn/Vlm2ZY9+PElOo1p8kF7zrG+nQsJVqUCQAAAlaRNtn59t0rWkXK8vvv44T8//Rpvovw65u6VHy6X3Y/E/H0Km0ioSEqXIn9g+LZ/xiqUDsSL2PmM18gxKKalUSKzqF42kHHLnBADPb54GOucUiwqVKTnFALCzUAoAZDAI4YP9dMse6VNOMtNl36ECV07VeUbICLB1fxdNLT4IAJRCQmcXAKCUEpfwVEohKU4eQIoxuHyyptzoy14OfQBAI0/XFqRrCwAAmLwgeD6ASNXiTF4Qxw808nTytA+ev09GKSSkW4oBUjsA/BcPh+8UGgvPEx6nrKy8yBGe5zyyN/roSHlyr0wSnq3eJWFVlwdq2VpFQiJ9vpmWa5uyQuTlJPJbQ1lZTtKnf/MEAPT9EwCcySnWPBFvsVvQTnrZWyvX71MOAIM77CF74DsxmcyHAlc2AAjcOOS9KH2kD4XN12cAwJctf56B8h38o6tLMIwqv9cMzvP7FXagfYcgg03SQgAAWy6/TUgsGTSElLd93h2foG+tKsuelX/s8QEpz+QFFReqtAX33ChvANA8ETMcbUhhW0cbBoefklOklWf0pW/XOPydmVJtQXpkxUEAYAGo9YcYpLzX3erH112JrHh2dPmMPnGs77UF6Ysdlxr5M1Den6h+86au0FvzGX3m533pwy74wfcrI8XLdGMe/uDDLvjJbwdZU/DUYfDeWT72BT+/vKNq+coy3Ts5S/S35ql4Hwtn+dgXJHYxMvJkZZku4EF0fct72j6O6TSOrBGreCOvLvFlyw+9YuQVCKpM1+/md/pbG1Pely0nK50ACq8BAIjkvH43vwt1ER5+5TuyKa3Yb+TVJfpr9Jms/BSz9mNqgFaRkMj/5528u1g6EOtCfi418vTScxPtfcaa5+WkRqLvnwBIiuoGEEGy1F3h3UBVtkYm1jwRi4pUYTwXcjslKdECaEkZyCmexnQHAG+tHAAYjja2Djas9g8BwOfrc75seVbgZ6SiPJW7/55Tvmz5hQ6U5jFFVlaWPRvsQ/M4xz+DDDyxl6Q6AICrlwFg66l/Af6lAybZTqvVdhT8+HydFwAAUADnAOB41nyjR+rXoTcAAAQABDxfdw4AhA/uGyt+ZnCHLF92CcB/P8d5qmxftmaoVqtfztaRHMiDVdrcUJdH6nahzwvzvnuQG+pSPK8dRc6M3ldsNE/EO4XSUJfixc9bGcjdaqjLPcrdyHhUmicU2ap1f9bA+LDYT1So8naReFNXjJQvoQy2ihTPygMjv2p5StGw8o/o3CBW8QDAh11Ar3khnmLKYKtpy3trZVVX1qCpy1sPG51OZ+kYmoqfn59QKKQUkmk7ToiKVGsmTTDzI3GRSCQQCMxZYwMCoNudOH4LmqIjuMVPwtlr9/h8vqhICQCiQlWFOC5Hcvc1F2Goi1C/WL/rewHg8CvR/z50yFPx0p6+nFbsBwDvlP0LzxMYwa+UAcDnvA+MVncqf6F+YZrf8zs2A8IHU6quzGe6D/HeBMamNj4lWUQKdHhx07R2y8iCwO2FCZTXybcCQFlZmaOjI1lDmivJpCEGBG7st3P/NDqaSbzjAP1iZMGXXdD2YZZ+MVZbPgBUlukusLv5sgt87J+diswHWq+Or/qy5ZXGBlSkynXSNv76W8VqdwaHb9ryDPkVPv+FB2/Vje5YWaZjtfXR39qY8pVlz35jJRIJHUB1jcbVTSpmkvLkJ9FoeSvR8hMSANh8cgoAyAzc5gzA4r/FdQyA3CQBQJuQWJN3urPsSaAUksfXt7m4uGjkGeSZSnHyAINuCPY+Y6D6t7Ke3VpVj6Q6AwLXZ1mB/FtVIpHYCPyNft0812RzuRRbdgwWD8D6E1LLb7JTiw9e6vENT/jY9eoGGPyupcOxRiz3IPK8xIJdwE2LfipWlhVlX5CufAwAoJGns9yDWLwgho83/VS/VrUmjNoK+ACAjaLSzP8YQqg5avkJSSvPELgXlApb16wT9UX/OpdnRdn5jGmOmYlSSCoVEm3Bs55p5G7P3neMSqVqwx9EvwXcLB6VIdQ6tfyERLrYUWW6VjXrRMOoxQfV4oOagnRreDmpjiiFRC0+qJFnaAvS6ZUMLp90TrP3GauofLWtpRuLEEJ10fITEmEvaH7/5Dc/Fi+IyQvSFqSXnJto5TmJ3A+x3IMqFZIXX9YZY+8z1pojRwhVp+UnJPIEm35NBNWAvJxUlhVFchJ519LaKIWbSAYib4PacvlMXhDLPdDMY8UihEyuhc8YS1471zyicMToOqJfmK120AGzoxQSpXBTSdrE58vPshHpHcfg8p1CYzEbIdQCtPA7JPIgobK8tc86US8MLt8pJLY4eYBafNCcMyfpo3so6D8cIt3kHPyjWbwgbJRDqOVp6QmJjDyGPRrqibSGFScPIAO4kdF06OHOmgjdV7s8K0p/hDcGl8/g8FnugaT2OnbXRgg1Oy08Ia3L7rcr89sVER1nWTqSZofB5dPvySqFm+mXSR38o+19xpK2UJPkJzI8tn5fbQbXm05Clpq0CSFkfi08IYlV7nkqDTbvNAydCTh+C7TyDNKGxuIFAUD5lahnvUW4fBYviOkeWPf7J9JR+/meF+o/FqL7auMzIYRaoRaekBYwpkW+wuvVOdHSgTRv9j5jSUMZ3arG4gVVciTagnRKIVErDj5r0+PySWdxtfigQX4y+rYQmWSB/i9dHv8BgVDr1MITki9b7suWu+GoLSbCeHFyOXKTpClIp++f6IdA/xXuspseK4/eCekgR5Icyx0b5RBCAC07Ib35KgUAUsc33SwdSUv1bHBS7lj9jgaUQmLvM5bkJ7LelsunkxC2xSGEqtOSExJBj7uMzIPxfNK5Z/0gnmjp+UwRQqgGLfnF2D9K3vbP+PavEhzh2zIYXD4+DUII1V1LTkirFXF5Knf8TUQIoWahJTfZDZxm+4Q1i9fjrKUDQQghVLuWnJA8XXUANniHhBBCzUKLbbJTZh8AfKMFIYSajxabkBiONgBAhhWwlN27d1uwdmsIwBpisHgA1hADBmANMVg8AOvXvBNSWVlZTEzMF198kZpq2KtYLUo3+hVz2rNnTysPwBpisHgA1hADBmANMVg8AOvXvBPSnDlz1Gr10KFD169f//fff+tvsnW0AQCme6CFQkMIIVQ/zbhTQ1ZWVnl5+aJFiwCAx+MtX7787bffprdqCtIBJ4pFCKHmw0an01k6hgY6dOjQ2bNno6OjAYCiqJ49e964cYPBeHbPV5w8gFJIIteypEU2Fg0TIYSsQkBAgJU/x2rGd0gKhcLe3p4sMxgMGxubyspKOiG5hKfez7+Xkt7FcgEihBCqh2b8DInBYFAURX/U6XR0NiI6eWM2QgihZqMZJyRnZ+fy8nKyXFxczGKxbG1tLRsSQgihBmvGCSkgICA1NVUulwNAQkLCsGHDLB0RQgihhmvGnRoA4NChQ5s2berUqZNUKv3jjz/c3d0tHRFCCKEGat4JCQAoilKr1RwOTnqEEELNW7NPSAghhFqGZvwMCSGEUEtiu3LlSkvH0EBlZWXbtm2Lj4+3t7fv0KFDXbZevXr17Nmzt27dunXrVmFhoa+vrxnqNVrp48ePd+zYkZiYKJPJevToYWNTv7d3TRIDUVxcvH79+oEDB5o/gMrKyj///HPfvn35+fm9evWyyEkoLi7esWNHfHy8Vqvt0qWB7wk04FJs5LE3rF6jlZr5UqzhwM1zKRoNwMyXotHqLHIpPnjwIDk5+ZYePz8/g/dnzKkZJ6Rp06a1b98+JCRk48aNjo6Ofn5+tW797rvvbt++rVarZTKZra1tr169zFBv1UrLysrefvvt4ODgfv36HT9+/N/puV9tAAAgAElEQVR//x06dKiZY6ALL1my5Pjx4x988IH5A5g/f35ZWVlERMTJkycvXrw4aNAg88cwbty4jh07hoSEbN++vbS01N/fv14xNCySxh97w+qtWqn5L8UaDtw8l6LRAMx8KRqtziKX4oMHD86ePSuTyWQyWVpa2okTJ6ZPn27BhAS65uny5ctvvfUWWb5y5crIkSPrsnX48OG5ublmrrdqpSdOnFi0aBFZfvr0abdu3cwfAxEfH79kyRJ/f3/zB3Dr1q0333yTLMtksjVr1pg/BplM9sorr5DlEydOvP/++/WKocGRNPLYG1av0UrNfCnWcODmuRSNBmDmS9FodZa6FGnl5eVhYWFXrlxpQL0m1FyHDsrLy3v55ZfJcq9eve7evUtRFJ3YjW7V6XRisfjkyZM3b97s1KnTe++9R4881HT1ajSaqpWGh4eHh4eTYjk5OTwez/wxAIBYLP7jjz+2b99+4sQJ8wdw69atPn36nDt37tixY/7+/suWLTN/DG5ubu7u7klJSa+99tqxY8e6d+9erxgaFglFUY089obVe/PmzaqVmvlSNBoDmPFSNBqAmS9FozFY6lKkt/7444+DBw9+5ZVXGlCvCTXXTg1GB7Kreeu1a9dYLJajo+Pbb799/fr1jz/+2Az1kiEkqqtULpdHRUV98cUXFokhKipqzZo1bDa7XrWbKoDs7OwzZ878/fffAQEB+/bt++qrr8wfA4PBmDZt2ueffz59+vTz589HRkbW91Q0IJLKyspGHnvD6pVKpTVUap5LsboYzHYpGg3AzJei0RgsdSmSjzKZ7M8//5w7d24DKjWt5nqHVPNAdka3+vv7X758mawJCgrq16+fTCbz8PBo0nrd3Nyqq1QkEs2aNWvOnDkjRowwfwwHDhwYOHBgjx49FApFvWo3VQA2Njbe3t7r1q0DgJCQkNdee23ZsmV1H/zJJDHcv39/z549KSkpTk5OZ86cmT59+qlTp+p+HhoWCfktaMyxN6xeR0fH6io126VoNIatW7ea7VI0GoCZL0WjMVy8eNEilyJZPnDgQEREhDUMLNBc75BqHsjO6FapVJqdnU1WcjgcW1tbukzT1fvkyROjlV68eHHq1KlffvnlpEmTLBLD4cOHt23b1rdv39dee628vLxv374ajcacAfTs2bNt27ZkpZubm42NTUVFhZlPwuXLl1999VUnJycAGDRoUEFBQQN+ExtwKTby2BtWr5ubm9FKzXkpGo3BnJei0QDMfCkajcFSlyL5mJycPHz48PpW1yTM+LzKlAoKCvz9/WUymU6n27Fjx5IlS8j6jIyMoqIio1tTU1OHDh2qUql0Ot3x48eHDh1qqnpJpUa3Gq00Pz8/MDDw8uXLJjz2+sZAKy8vr++TZJME8PDhQ39//ydPnuh0utTUVPoxrzljIPWSldeuXQsMDKysrKxXGNVFoqvxUmzksTfsDBit1MyXYs0HboZL0WgAZr4UjVZnqUtRp9Nptdru3btrtdr6VtcUmmtC0ul0iYmJYWFhs2bNGj58ODnLOp3O398/JSWluq3ffffd4MGDZ82aNXTo0Nu3b5uqXrpSo1urVrp27dquLzJ/DLQG/AqYKoAjR46EhYV99NFHr7/++rVr1ywSw/fff09iGDhwYHp6en1jqK4uXW2XYiOPvWFnoGql5r8Uazhw81yKRgMw86VotDpLXYpXrlzp169fw6ozueY9dFDNA9kZ3VpZWVlRUdHIse/qW69JKrW2GEwSQCOHIrSGGBoWiaXqbYqxHy0eg0kCMPOlaD2XhFVp3gkJIYRQi9FcOzUghBBqYTAhIYQQsgqYkBBCCFkFTEgIIYSsAiYkhBBCVgETEkIIIauACQkhhJBVwISEEELIKmBCQgghZBUwISGEELIKmJAQQghZBUxICCGErAImJIQQQlYBExJCCCGrgAkJIYSQVcCEhBBCyCpgQkIIIWQVMCEhhBCyCpiQEEIIWQVMSAghhKwCJiSEEEJWARMSQgghq4AJCSGEkFXAhIQQQsgqYEJCCCFkFTAhIYQQsgqYkBBCCFkFTEgIIYSsAiYkhBBCVgETEkIIIauACQkhhJBVwISEEELIKmBCQgghZBUwISGEELIKmJAQQghZBUxICCGErAImJGScWnywJG2iRp4OAJRCQhYQQqjp2Oh0OkvH0FSmTp164cIFS0fRDHi66gBAWmTz5qvU/xuv3ZZku/2E7eyhlXMiKsny/xuvffNVas1fzCMXGZ6uOk833eUc/KcMQs1MQEDA7t27LR1FjXQtV9euXTGA6jZVlosrZOd1Op0q70DBIcHTsxN0Ol2F7HzBIUHZ5U/1C+h0OsXtjU/PTqgsF+t0uqdnJxQcEqjyDpDydJkGxGAeFg/AGmLAAKwhBgygVkxLJ0RkJpRCUqmQUEqJvc9YjTy99NxEJi+I5R7E4PDpMrZcvtvIXLLM4PIZ3GebOH4LOX7/lQEIYvGCAKD8ShSlkLiEpzK4fLX4IIPDZ7kHmfWoEEItCCakFk4p3MTg8u19xgJA6bmJZNmWy2dw+Sz3QKiShGrdoYN/NL3M4gWpFQfJt8qzogCATk6kRoQQqjtMSE1oypQplg0gav4UpXAznZAYXL69zxiy4BKeSsrUJQlVx8E/muQnSiGhqwCA8qyo8qwokudmvNOr0cfRKBb/K1hDDBiANcRg8QCsX0vu1ODn5ycUCi0YgEgkEggEFqmaUkgYXL5IJGqnTgQAjt9Cc1ZdlhVly+U7+EerxQfLs6KYvCCn0FizBWDAgn8F64kBA7CGGCwegMV/EmuFd0gtU1lWFKWUMHyjzZmKCAaXr59+tA597Ll8AFCLDyqFmzl+C7A1DyFkFCakFohSSCilBAAou/aWjcTeZ6yi8tW2AsGzqBQSsl4p3EQp8u18xmAnCIQQDRNSS0Ma65xCYhlcfrFIZOlw/sPxW2jvM5bx7G4pjlJI7HzGwIvdLhBCrRm+3tiiaOTpJecmkp94S8diBB2VU0isg380uT1Si+PKs6LIzZNafJC+i0IItTZ4h9SiaAvSyR2SpQOpBYPLt+eOBQBKIeH4LdDKM/T7jpPuec3iQBBCJoQJqYUgP98cv4XNq/mLZCYSMN13HADIq7v63dMRQi0eJqQWgu5p3YyykQEGl0+/dUs6ZZDxIEj3PBYvyME/mow3gV0hEGqRMCG1BBp5urYgnVK2nAYue5+xJBsBwH3JvXYKSZ4Dj1OoUgr3tJNtlTq+KeuwWivPGOBVllb8cp6KR38xsszwjkpcDErx/Qx2N4P1Ajc2AHhr5eRjPtMdAASunKplEELmgQmpJWC5B7mEp1Y2t+4AokIVAIiKlKJCFVl+UKQSFSoBQFT0bM07ZdsAIJgZCHCFKp8+wPuOWwfq6YO0waqsn/z29H2ctk04K/ZR6MT2aaHOwtjHoSFXVxvU0g4AAGZ22GM0BuGDZy/PewEAQMXz9X7Gygvc2Lsffa2/poMbO5/pAQBf8N4XuHEMyndwZc97Gg8AlcXFBS4uLA8felO844Cq2U7gymkrvWywknyL1dYHEGrpMCE1bxp5evmVKNLJ25q7AIgKVaIiZcq94kB19l3hXQC4e/suAPArZQDgrZWvarfM6Be/Kfj1hc9CeCoEd3gcNsSly0th9i6eXaiwGT7uY5kPgu3SWO5BTq7v2jAvACNfpw0AyptSSsrlSjabPaObZ5WQlADAUjz7odfIxPpbSbYgSVH/KNpKs/TXKKXgBgAAKR3ehZziqvFPerCBLBS8uL7WBGnAIEHSyex41lgAYDjaUGU6sgAAthz+rF6/hLoIASCt2A8A3nE5xb7Mjn0U8sfjr7mvMAFAcVULAPTy8m5rlnQ4BACxj0PFKvclHQ75sgu+fTByriSO+wrT1tFGcVVbWaajl3/2nP+Z4BClyP/2wUgGh/+Z4JBGnrFeNndBxQX2S48qFRL13fYMLp9e3syZtqbo50qFZL1sLilPvrug4gL3FVtKka+4qmVw+fRy1f3XpTwAfCsaBQD08ryn8dxXbAFAcbWysrhYOYhHln90jqy1PADQyyYpv/nKoP93Ia7p9l9r+dlDK41eXdYDE1LzphRuphQStfig+UdkqAGdfgDg+E1pev59epPwwZzOAAAw7MWvrIjoCAACNzb5qRW4csjCo5/ehef3B+RegdmWDwCne/QF6AsAX/kDAGjkcyjlsDm8IAaXX5w8gFIwXMI3kmU7rdZt4IHfuXz9t3HV4oMAwOIFMbgZpKO5QTrPrea42koPGqzUPpEAwGlff4PsRcrznn4CAMXFxcUU21srozfNcPfMkdyjlBIWLxAAPMuPAMD5itefVPgL2l4DANGT3gAgaHuN4Whz96YrAPzkt8OXLf9YOCtP5f6Dz1ehLkL/jG8BwDXSztbRpjC+girTOUewbB1tCuMlKTnFGwO/92XLd9/8Nk/lvjHwL1+m/Ov8tkrhebafna2jjfpBBVWmo5d3lkkXMOJ82fJdmZ3yVJoFjDg1W74rs/8U4X5bNztbR5unKRVUmY5eXsV7e3TgXl+2fFdm/zyV5vnyO1OEG0g8pDy9vIn39nSbFFJGr3z/KcINlPa/8vRyNfuvvfzvGV55Knd6edLz8kVHK6gyXZnjs+VVPP+6lKeXTVI+/prXR025/1rLz4mw9oSEY9k1ITMMXUUpJJqC9Oo6Mphz7CxRoWpnphQA7gnvul49/JNLpP5WgRtb4Moe1Nl19rlPDdqgWB4+zLZ8bo8QU0Wi3/GhOHmAVqt1G3jgeaJ6NllGdcslaRMppYTccZIExvFbQJbh+ZCABuvp5fKsqEqFxNE/2mC5JG1iRamo7jHUfbm43wkGh+9yeSgAFHb7ncHhu2XPpJSSx7yvpE59+pesrVRILjotY3C9/eSbnezKM5jTgxgltk4PKYUk54GbLYffSVBIli+6Dp3YPo1SSP4qGQ8AE9unAUDso9DIslSpWwUAqHMqAcC+sy1ZjncYSJch5cUq97PFfvOexrPaMQBA85gCAHp5jW7Im4LHAHC22E9///OextP71N//j86RBvuvS3lftjz2UWieyp1eHvX4X/vOtraOoMqhCiVF7frxyPJm23dqLU+V6ehlk5T/5XaP6eprTbf/Wssf+m3lkp/umep/tP/f3p3HNXFtDwA/BBACCrIoLoBIfSJuNWIBWQQsoqUCVYuiQm2V4ns86lq0LoDbcylubV9d+lOoQMWqgLggiAsqm1ZBRVSKSDQsKiCoCRAgye+PW9O8EJAtMwmc7x9+JpPJ3DNxzHHu3DlXHjAhyZFc8wEvN5ihNbj1CyN5JySShJ5W1//6R/ngpooZ3OszedfIMIGprJMLJgwEgBG9633spQcUUOnZn9mmw20BoLEym0wHBf+bVHi5wY1V2X/VtujShPHe5AcAJGlJxkOeDiZVLciVHImZzCJPEm17H9Kivawn7QEoQgy0B0D7T+J7YZedUmqszBb/VFF/64jkoatF1WlFNSQPFbyOF7+r3s9Ex2V2/ijQGjUUANh0ly8SF/STHCwumcglZ3jSsYuFd913TIulksviEhJS68XL2uPCZS7r2MWWlJToaxkDgORjVZIlaCXjkbzelVyWjF+R7xci1GGYkJSSuqGtNiucwaR0IAP7VX1aUfVXxx6K15jpa6bkLv8rpH4mOi6zmaMmdmHnG/Ukv8+WEkNbEoZU8hD2auryUBHqfjAhKRky25COfSxlD8CS66Ejt8rFN+3N9DUXTBjoPKyv8wd6AGVVx3fpOM/GcckIoU7ChKRk6gp+aKrK5uUGS3Y0yQlJRYeSbs7gXt/MfxhiuVkiD/3NYPZKeUeCEOoJMCEpGXLTQt7ZaENK8cXsPNazCzb8h5fr/+qjK16nxH1xCCHFhwlJafA5J9UNbCULvsnDhpTiquO7ZvKuzXtXU6fLh2UjhJBMmJCUg7C2hJcbLJ55r8v3T3rnNl4oBgAyZI4MUiB/dnlzCCHUHCYkpUEeUpFHNtqQUkxSEQCY6WuWBeZb1z/ESyKEEMUwISk6cS0GeRQH+vWPcvEwbjN9zUgfy3cDFjAbIYSopjQJSSAQxMbGPnjwYMSIEb6+vgwGAwC4XG5UVBSHw3F3d3d0dKQ7Rrng5gaTeWC7NiGxX9Wvj7n221M1kE5FCCFEDwbdAbTV0qVLHz16NHXq1Fu3bm3ZsoWs9Pf35/P5U6ZM2blz55kzZ+iNUE5IT13XZqMNKcVzQg6HXvvSTF8zzG1o8To7zEYIIdopxxXSw4cP2Wz22bNnAWDUqFEHDx4EgNzcXB6Pt3z5cgAwMDAIDQ318PCgOdAuRYbVkbmOumqfaUXVLvtyg2rio1/HA8C1hl0mU+O6aucIIdQZypGQHjx4MG7cuMzMzOTkZBaLtW7dOgB49uzZiBEjyAZjxowpLCwUCoWkK0/MwsKCLPj6+vr5+VEcdklJp2bM07kXLOw1gDsi9v2btiGAkjdNwRdfcp5yoqt+sa5/CACqbosEbovkXWuuk19CNwhAEWLAABQhBloCiI6OjomRPfmWAlKOhPTw4cOrV68KBAJ7e/vo6Oi8vLzQ0NDa2loNDQ2yAYPBUFFREQgEUgmJ9tK2HS7uK6wt4RrYqmoZG3auPDAJIK2o2uVI7gzutSNVvwCAej8To6A9lI2jo73MM+0BKEIMGIAixEB9ACEhISEhIWRZ/B90haUcCUlFRWXw4MHbtm0DADs7OwcHh3Xr1jEYDKFQKN5GJBJJZSOlxtAylqwG3RlkVHf0i/+QCyMd59kDgvZ2yZ4RQqgLKccv+OjRo/v370+W9fX1VVRUGhoadHV1eTweWVlTU6Ourq6qqkpfjF2psTK7rmCveL6DDit50+SyL4c8Y/TY7xAAGG88idkIIaSYlCMhWVtbp6enV1RUAEB6evrQoUOZTKa1tfX169crKysBICEhYdq0ae/bjdJo4MTVFfxQV/BDZ3aSVlTteORZWlGNmb7mlUDWhqlDh58sw8ddEUIKSzm67AYOHLhly5bZs2dbWlr++eefe/bsAQB9ff2wsDBvb29zc/Py8vKoqCi6w+wyvUxmwbs6qh0jLr7g/EHfK4HjuywyhBCSG+VISADg7u4+bdo0Pp/PZDLFK728vDw8PKRWdgPqhraSM7y1l8u+nLSiGgBYZq23Zw6r6+JCCCE5Uo4uO4LBYDRPPDJXKrU3GXN5ucEd+yz7Vb3LvpzRtyMA4Eoga6kNPu6KEFIaypSQeoLGyuymqmw+52QHPkuqAS3KDP7mdXzBU18svoAQUi5K02XXQ3S4LkNaUfXq3UejX/wH3j1m1NWhIYSQfGFCUizC2pIOTDCRVlR9Y/M/o7nXAYA5aqLJRqwGhBBSPthlp0B4ucFvMue2t78uNStPtM5pBvc6ABjMXonZCCGkpPAKSYEwtAYLOSUMZjuukH79o/yrkxUFTZUUVwNCCKEuR0NC4vP5bdlMXKeu52BaLNMw+bztXXZpRdVker3M1be//GigPENDCCG5oyEhLVq0aOXKla1vU1paamVlNXBgD/qR5XNOMpjGbX/8iEwkAQCRPpaYjRBC3QANCcnMzIzFes/TmkZGRtQEoyAaK7PJs0f6nsVt2R6zEUKo+6FhUIN4vlcAKCwsLCwsBIDIyEgPD4+HDx+S9YMGDepRl0eqWsZMi6UaJp+3ZWPMRgihbonmUXb//e9/jY2Nnzx5EhERkZiYeO/ePXrjoQuZpFybFf7eLTEbIYS6K5oT0tixY5lM5ubNm1esWMFgMJydnblcLr0hUY/017Vlsom0omq1vb4FT30xGyGEuh+aExKHw+Fyubm5uY6OjgBw8ODB3r170xsS9eoKfuBzTjZWZbe+GftVfdzWkP7luQAwW6tNt5oQQkiJ0Pwc0oYNG8LDwyMjIxkMxrx58wCgtrZWS0uL3qgo1psVzuecbP0GEvtV/ZyQw9Gv4wHAeONJfN4IIdT90JCQVq9e7eTkNHXqVDLBa3DwX5Wtjx49Sn0wtCO1gpgWy1rfbH3MNVKnzmD2SsxGCKFuiYYuux07djg4OPzyyy/z5s0rKCigPgCFUnPR8U3G3Na3cdmXsyjzWwBgjppoMPs9j3AhhJCSouceko6Ozr/+9a+jR4/27dt3/fr1S5cuLSsroyUSejVWZgOAaqulGb469nBRZvDgpkqsmooQ6t5ovodkZGREHkt6+vTp4sWL//GPf/j7+/ft25feqCjz3skmNqQU//pH+Zr6h+r9TDAbIYS6N0UprjpkyJCDBw8CwK1bt/bu3Ttr1qwZM2bQHZR8kcujVmoFpRVVb7xQDABlO/Jxtj2EULenKAlJbMKECTExMXV1dXQHInd1BT80VWVrs8Jljq9jv6oXPwCL2Qgh1BMoXEIimEwm3SHInbqhjbCuRN1A9hXSV8ceAIDzB33xAViEUA+hWBP0RUZG0h0CdZgWy/q6Xpc52YTLvpy0ohrnD/peCRxPfWAIIUQLeq6Q7OzshEJh8/XV1dU+Pj494fKIlxusZmgjs7NuQ0pxWlENAIRNHUp5XAghRBt6EpK5ufmRI0cAICkpycnJSTwX3+HDh3tCNuJzTvI5JwW1Jc0TUlpRNTP2OzBYfCWQhbeOEEI9Cj1ddkeOHFFVVVVVVb1y5YqOjo7GOzNmzOgJwxnUDWy1WeFMi6VS69mv6lfvPjqDe73gqS9mI4SUxfXr10+cOHHx4kXJlfv27aMrHuVFzxUSKRoEAAKBQHL9tWvX5syZQ0dE1CG1gjS0ZHTW7f81jtQHMt54kvK4urMTJ0789ttvWVlZc+fOZTAYenp6BgYGgYGBLW3/+++/h4WFbd++/bPPPmvL/gMDA1esWDFs2LCuCxm9B5fLPXDgwLfffvvelcTTp0/j4uK0tbUHDRrk6elJVkZGRurp6WlpabHZ7MmTJ5O/werq6rNnzwJAnz59yAnw9OnTxMTEkpISV1dXNzc3qT03NTXt27cvNjY2Pj7+wIEDAwYMEAqFmZmZo0aNkseBd280j7LbvXv3+PHjzczM1NXVX79+7e3tTW88FKgr+KGxKrv5dHyHkm4uygwGrFYnB97e3np6epcvX967dy9Z89lnn/3555/il1LmzJkj9b9dKRwOJy8vz93dnbzk8/kNDQ1dG7O8SR2Cctm8eXNVVdWJEyckc4/MlYRQKPT19d2/f//o0aMdHR2rqqq++uorAMjOzr5w4YKpqWlQUJD4/xN79+7duHEjACQlJcXGxk6bNi0uLm7FihVNTU329vYcDmfRokWSO8/IyCD1oN3d3S9cuODp6dnQ0JCenk6aQO1C8yg7VVXVnJycgIAAOzu7AwcOfPLJJ61vX1NTIznhLJfL3bdv35o1a65fb63egUJprMoW1kqP9k4rqtY8ugawWh1Vpk2bFhUV1eGPnzt3TvLl4cOHR44c2emgKCV1CMolJCRk1qxZbVlJCIVCbW3tiooKADAyMhLPTG1tbV1cXHz16lXxf4Xr6+vv3r1Lll1dXbOysq5evbpt2zYAUFNTW7Ro0aFDh6R2rqur29TUBABv374lmSk8PHz16tVdcqQ9Df3PIfH5fGtra2trawCIi4tbtGgRg9FimgwNDU1PT1+/fj156e/vb2NjM2XKlJ07d9bU1Hh4eFAUdCfo2MU2VmVLjfZW2+trjfWBKHTq1ClxmXkAuHr1akFBgYaGRp8+fWbOnCm5ZXx8PIPBYLPZLBbLyckJAGJjYzdu3GhjY3Px4sWtW7cmJSXFxcUtWbLExsbmwIEDWVlZurq6P/74Y2xsbHJyspub2/z582/cuHHhwgWZ+2/e+pEjR86dO+ft7c3hcIyMjJqamiRfzp8///bt2zdu3OjXr19FRQXpeJT6yPz586WaSE5OHjBggPgopA5BU1NTcuPffvvt7NmzixYt4nK5bDZ75MiRenp6bDb7/v37QUFB/fr1k/m1SB77kiVL3r596+/vb29vL7nnrKysiRMnil/ev38/Li4uLCyso3+NbaWmppacnMxms7lc7v3797dv3y5+KyMjo6yszMrKytzcHAA0NTWfPHkyc+bMn3/++dChQ0uWLBkwYID4cZT6+npy+JLGjRtHEtLly5e9vb1v3LhhZmZmZGQk74PqnkS0un//vo2NjYuLi4uLi4ODw5YtW1rZOD4+ftWqVSwWi7zMycmZPn06Wb5z546np6fU9sOHD5dHzG1XXFwstab+2Ynmmzn/fLtg1sCCWQN59zPkHQD1aI+BBJCamspkMhMSEuLi4qZOnXrs2DHxBvn5+c7OzmTZ1dW1qKhIJBL5+/snJCTw+XwA4PP5IpFo1KhRZEEkEvn4+Jw7d068h4ULFyYkJJDlgICAX3/9VSQSCQSCrVu3kv3b2tpK7b/11ufNmxcaGnrr1q2DBw9KvSwqKvLy8iLbp6enL1++nCxLfUSSzKOQOgQpX3755b///W+RSNTY2Kinp5ednS0SiSIiIkJDQ1vaoUgkSkhI8PHxEYlEa9euLSsrk/pbKCkpSU1NlVz5/PlzBweH5q2XlJT83ILCwkKyzbVr1wYNGiT1QZkrxY4fP7506dKjR4+K12zatOn58+cikWjWrFm3bt0SR2VsbKyqqir1cyQQCCwtLdPT05vv+eXLl8ePHy8oKKirqyN/I7m5uTExMS9fvpT6ElqKjRq0/yS+F/33kLKzswsKCiwsLADgzJkzLW3J4XCioqIOHz6cmppK1jx79mzEiBFkecyYMYWFhUKhUOrqiuwWAHx9ff38/ORyDC0rKfmfWclVuXe0nwS/zd/FHRErXpldWpdWVGMxJKbIveal9iBgs+UXAC1oj4EE8OLFCwaDMW7cOABYsWLF119/PW7cOPK8we7du4VC4S+//AIADQ0NycnJ7u7ub9++ffnyZVlZ2aNHj9LS0p48ecJgMPLz8/X09ACAx+O9ePGC/e4vi8fjvXz5krz08vLavHmzk5PTlStXpv0sDcQAACAASURBVE6dymazZe5fHJ7Md+vr601MTAwMDNzc3NhstuTLbdu26erqkrYGDhz4008/BQUFMRgMqY9IfQmXLl2SOgqpQ5BSV1dnZWVF3uXxeEZGRmw2m8fjFRQUkJUyv5Zx48adPXvWx8cnMDCQz+dL7rykpCQnJ8fT05PNZt+7d4/P53/00UcA8Nlnn7HZ7PLy8ry8vLKyMlVVVfLvtJWbW2S3z58/FwgEUvHLXCk2cODAZcuWLVq06P79+19//TUA+Pn51dXVsdlsBweH0NDQn3/+WSgUbty4MTExMT09fdOmTdXV1UFBQeTj27ZtW7Zs2eDBg2XunxzOmjVr5s+fn5WVFRkZuXbt2oiICMlRWrT8W4iOjo6JiaG+3Y6hOSFpa2tzuVwDA4PMzEw7OzsrKysulytzFvPg4OAtW7ZI9i3U1taKH2BiMBgqKioCgUAqIdE+35KZmZl4mc+5xTewVdUyNny3kv2qfu5PmQBwJZBlLp9x3pIB0IX2GMzMzB4/fsxgMEgkZmZmVVVVhYWF06dPBwCBQDBmzJiAgAAAIH8CQJ8+ffr376+rq7tgwYIvv/wyICDg7NmzpqamBgYGAKCtrW1kZGRqalpeXj548GCysXjn69evb2ho4PP548ePJ/sfMWKE1P7FZLbeu3fvQYMGib83yZcNDQ19+vQhy0KhsKmpydjYuFevXlIfkVRdXb1169bAwEDJo5A6BKmP9OnTp1+/fmRvKioqZKFfv36k6erqaplfCwAEBQX5+vpqa2s3j4TD4ZCVe/fuDQwMJPsxNjY2MzOLjo6eNGlSQEDA2LFjQ0JC3rx5k56eLvOv0sbGhrTF4XBUVVWlWpG5EgDKy8svX75sb29vZmY2a9aszZs3/+c//7lz586ePXvIA5EDBw4sLS01MzM7ceLE9OnTx48fP378+Llz59rb2+/cuRMAIiMjfX19HR0dORyOiYmJzNgyMjKsra0nTJjwyy+/fPzxx2ZmZhYWFqamppI/StT/WwgJCQkJCSHL4v+gKyyaBzUMHz7cysqqV69eW7dunTFjhpeXl8xstG/fvkmTJkkNo2QwGJLlHkQiUSs3nxSBhsnnOvax2qxw8RpxwTp86ohiDAbj8ePHAPDnn3/Onz+fLANAfX39jRs3xJudPn3awMBg5syZDAajuroaAE6dOkU+DgANDQ25ubnNdx4QEBAWFib+xz9//nzx/6ml9k/ebal1mby9vauqqsjyo0ePWCxWr169Wv/I6dOn9fT0pI6i9UN47w5lfi3V1dU3b95MSUlZvHgxWS+JDBZ48eLF5cuXzczM3rx5ExISQkYTrFmzxtHRsampSVtbG5o9DSKJ3K1pu4yMjPLy8osXL4qH3pWWlpqamgLA69evxXn0wYMHDg4OACAUCsW/Kv369XN1dQWACxcujBw5kkTY0mCQ2trauLi4uXPnkmV1dXWyXmZJGtQiuvsMRTwejyzk5ORUV1fL3Gbq1Kmsd4YPH85isRoaGs6fP79kyRKyQXV19dixY6U+RXuHqWSXcUNFVu2jPZLvhiU/gRWXzLZ08X2jlgKgC+0xFBcXX7t2zcvLS11dPTQ09NmzZyKRaO3atVOnTi0pKTl//rxIJNq+ffuePXvOnz+/f/9+gUBw7do1S0vLWbNmJScnOzg4nD9//vjx4wEBAWvXriV3Gs6cORMQEHDw4MFXr16RjX18fEpKSkiLPB7P0tJSMobVq1dL7l8qQqnWU1NTzc3Nvby8yO0KqZdk+0OHDp0/f37lypXkcJpvI+nu3bsTJkyQOgrJQ5DaPi0tzcLCYsaMGUVFRdu3b1dVVQ0PDyf3riwtLa9du3b37t3mX8sPP/xgaWm5f/9+kUhkbW1tZWWVkpIi+bewdetWXV3dDz/8kCxYWFjk5eVJtrtr166CgoK2/J2mpKR4eXkxmcz169eLPyJzpZWVVURExOvXr0NDQxMSEvLz862trcW3i3bt2pWbm5ubm+vl5VVZWSkSiQQCQUBAwPnz59ls9q5duzIzM3Nzc8XZBQDEN+2kiG9HkS98/fr1IpEoJiZGchva/y3Q/pP4XvQnpHbh8XjiQQ1VVVUsFquiokIkEkVERKxatUpqY9q/fcnzrzrVoSrRrKEii7y88vgVrLgEKy5deSz9cyCnAOhCewwtBVBYWJiZmSl+KRAI6urqZG4p/j+TZC7h8/mNjY0tNSr1VnFxcSv7b731lrYXR9UWxcXFzY+i9UN4L5lfSysBSH6kefBnzpwpKyuTSlFdKzEx8fLly1LR3rx5kwzZkJSfn3/+/Pm2fzlv375NTEyUXBMXF3f8+PG7d+9KrqT93wLtP4nvRfM9pMjIyKqqKplPVr+Xvr5+WFiYt7e3ubl5eXl5Zx4rkTdhbQnTYin/WZx4Or6NKcUAEOY2FDvr6DJs2DDJ2goMBkNq9LMYebgE3vXUEa13lKmpSf/LamX/731X5vbiqNqo+VG8t6+vvTts+0ekgj99+vSOHTvIUOn4+PjORNWKsWPHNr+FQwYjSBk5cmS7Hizr3bu3uPoD0XxwP2oLmhPSnTt3JJ8gu3v37ocfftjK9lpaWjk5OeKXXl5eHh4efD5fwUuyklpB4tIMVcd3Hby8ayrr5Aas540QgKenp9QPOuqZaB4FsHTp0p9++onD4ZSVlZWVle3evbu9e2AwGAqejQCg5qJjXcFfVWoKCwqrju8CgPw2lUlDCKGeguYrpMDAQHV1dfFI/8LCQnrjkQc+56SwtkRYW0peah79rg5Ax3k2FqxDCCFJNCSkJ0+ekCodAODi4iLZZXf16lXq45E3DZPPGcy/CgXdiz+imZ8FAAOCZJf1RAihHouGLrvQ0FD+O8uWLeNLsLW1JQtPnjwpLy+nPrYuJ6wtAQB1Q1t1Q1txBVWcXQIhhJqj4Qpp5cqVDx48eO9m4qsopcbNDRbWlWiPC1c3tP0gyr8OgDlqInbWIYRQczQkJBaLRX2jNBLWlqgb2p4/nfQB6az7N3bWIYSQDPRPP9G96djHAgD7Vb37VU0YEvN6eqV6f9mFsBBCqIdT6OJvyk7jxa/kHhKpWfflRwN1XGbTHRRCCCkovEKSFz7npMaLI3W9eCeaVqQV1QBApI8l3UEhhJDiwiskeVE3sG3Um6ZmaPPVsYcAcCWwZ905Qwih9sKEJC8MLeM6k9XTzpgDTjCBEEJtgF12ckGqM9wsG5dWJASASJ92FGpECKGeSSGukPbv329vb29nZ/fmzZvy8vJuUCi3ruCHuoIffkx/CABhbkPN9NtRyBkhhHomhUhI/fr1y8jIyMzM3L9/f//+/blcLt0RdZaOXWwJuEfc3Vzw1BdLeiOEUFsoRJedhoZGXl7emDFjVq9eHRMTQ3c4XaCIU6cZlQBYJQghOfPz87t58ybdUSgKa2vr6OhouqPoOIVISNOmTfv+++/HjBkDAL6+voMGDaI7ok6puejY97mwGkDlAxZWCUJIrm7evFlQUEB3FIrCwsKC7hA6heaEtHjx4vv3748cOfKrr74Sr5w8eTKNIXVSY2W2sLZEWCsEALU56+kOByGElAbNCWnTpk2GhoaqqqpCofDrr7/eunVrv3796A2pk9QNbfNO6wxuqhzw772v9AfSHQ5CCCkNmgc1GBkZqaqqAgCDwfi///s/V1dXeuPpvLLwzwfWVGgMscUqQQgh1C4KMcouPz//1atXAKCqqlpbW0t3OB3H55zUtLjd217NYPZKumNBCCmcoKCgpKSkVjbYvHnzd999R5a3bds2/Z2srCxKAqSZQgxq0NPT27p168WLF7dv366lpUV3OB0XcfXWHB14O3nrIBzLgBBq5u3btw0NDS29m5WVtWfPnnnz5pGX0dHRqampmpqaANCnTx+KQqSVQiSkQYMG7dy5UyAQ2NraTps2je5wOujXP8oDc91iB9pf8/SgOxaEeqK0ouq0xzXUt+s87H9qg/3++++xsbFCodDLy2vRokUAcPjw4YSEBAMDg4qKCrJNUlLSgQMHAGDGjBlkSBeXy/32229DQ0MfP34MAE1NTTweT1NT88mTJx999BH1B0ULmhNSYmLigAEDbGxsAEBVVVUgEHC53N69e9MbVcdcSDtk39dw0wwfugNBqIdKe1yz8UIxHS0PFSekjIyMbdu2paamamhofPrppzo6OgMGDPjll18uXbr0/PnziRMnBgQE3Llz59tvv71y5Yq2traHh4eJiYmrq+s333wTEhJSUVFBEtIff/zx9u1bPz+/xsbGFy9eXLp0SdkHfLUFzQnJy8urrq4uJiYmMzPz+fPnfn5+SpqNXPbl7Bly2lSzsu9ADwCso4oQDZyH9QWgoTCK87C+4uXffvvtm2++Iclj+fLlcXFxBgYG//znP3v37j1s2DDSA3Ts2LHRo0dfuHABACwtLWNjY9+8eaOtrT1t2jTxY62mpqaJiYmOjo4AsGrVql27dm3fvp36Q6MY/V12TCbT19fX19eX7kA6Lq2o+knp49gmu1DHXgwtY7rDQaiHcv5Aj/ay+nV1deIb4QwGAwDevn1L7gMBQK9evQCgqqrK0NCQLDs5OZmams6bN8/c3Hz69OklJSWvX78ePnx4YGCgkZER+ZSVldXp06dpOBjKKcQoO2W3MaX4Wb0h02KZNiuc7lgQQnSaOHFiQkICWb548eLo0aMnT54cFxdH1uTm5gKAm5sbm82eM2fOnDlzSktLS0pK8vLyEhMTjx07FhgY6ObmFhAQEBERIS4znZycbG9vT8vhUIz+KyRl9/y/yw6mHU91m+w/8jNaugsQQorD39//ypUrjo6OvXr10tXV3blzp6am5oULF5ycnACgqakJALy9vZOSkmxsbAwMDHg83qlTp8S3KphMprq6uqam5sKFC5OSkj7++OOGhgZTU9PAwEA6j4oqSpOQXrx48fvvv5eXl48dO3bOnDnkWpjL5UZFRXE4HHd3d9LZSrHa/Mw3acc1PlCdOyCjqXKghsnn1MeAEFIcDAYjNjaWjO0mnXIAEB0d3dDQoKamRn64ACAyMlJqG8LPz8/Pzw8A1NTUTp06JXObbkw5uuy4XO7MmTP79Onj5uaWkZGxZs0ast7f35/P50+ZMmXnzp1nzpyhPrCq47sA4OWA6dqs8F4ms6gPACGkgHr16iWVRXr16iXORi1t05b9dG/KcYWUnZ1tY2NDRutbWVnZ2tru2LEjNzeXx+MtX74cAAwMDEJDQz08KH0AqOr4rrr8rFI1w6EL9mjgFHwIIdQ5ypGQXF1dxWXuioqKDAwMAODZs2cjRowgK8eMGVNYWCgUCqX+DyIuxu7r60suhLuK6FV54/FdANDbXr33TZdSowWNetKP9JaUlHRhix1AewCKEAPtAShCDBhAz8FmsyVfRkdHK9Ekc8qRkMQqKyuDg4NJl11tba2GhgZZz2AwVFRUBAKBVEKS30QpnMiVjQA5Jm7TLIuEtSX9TcapG5o138zMTMZKKtEegCLEQHsAihADBtBDSH3PISEhISEhZFnxZ0tSpoTEZrMXLlzo7+/v7u4OAAwGQygUit8ViURS2Uh+avMz6/KzAMDqn+t0TJiNVdnqhrbUNI0QUl5BQUHu7u7kF0ymzZs383g88gzstm3bMjIyyPp169ZNnDiRoijpozQJ6datW8uXL9+0aZOLiwtZo6ury+PxyHJNTY26ujqZyYICn17VTBsSk2hZON2EydAy1tDCwXUIoffD4qqtU46EVFZWFhQUtH//fhaLJV5pbW29du3ayspKQ0PDhIQEyqqyphVVpxXVAMDYmQtqLloytIx17GKxQANCiJBTcdXbt28nJyez2WwOh+Pt7a2mphYbGzt48OCff/5ZXAlC2SlHQjpy5Eh1dbWPz991SwsKCvT19cPCwry9vc3NzcvLy6OioqgJ5qtjDwEg0sdysPBOnYGtqpYxZiOEFIG4L11KS/OTkcc2unZ7+RVXffHixY4dO3Jycvr06TNkyJCwsLCkpCRPT88zZ854e3u3/s0oC+VISGvWrBE/eyTJy8vLw8ODz+czmUxqItmQUsx+Ve/8Qd8vPxoIMBBvHSGkOOrys2TmDCoTkvyKq06aNMnV1XXYsGEAoKWltXz5cgaD4ezs/ObNG9lfhxJSjoTUCgaDQVk2Yr+qJ8Xtw6YOBQA+5ySWZkBIcTBHTWzXZM3tndm5LdvLtbiqeFyxqqoq2blSz2janNInJCp9dewBAHz50UDnD/T4nJO83OCmyhtYUBUhBaE1yk6rPZM1yyMhkeKqc+fOhXfFVY2NjePi4sia3NxcDw8PNze3yMjIffv2AcDu3bsZDEZeXh75+NGjR2/fvk2Kq549e5bkISyuiqT9+kc5GcsQ6WNJ1qgZ2GK5IISQJPkVV01KSqLzwCihIhKJ6I5BXiwsLLrqwdja/MySsM8BoGxHftsnXGGz2fQ+DEh7AIoQA+0BKEIM3TiALvxn3oWaF0WVKq4qc5u27Kd1rX8bivldScIrpDYhNzPzrBbO+kAPAOoK9gKAhsnnOL4OIdRc8xTSljVt2U/3phzVvuklLqJqtXg9WcPnxNUV/EBvVAgh1M3gFdJ7NL7kkMuj+nnbzd6V9NYeF95UlY2XRwgh1IUwIb3H85+XAcCrDz0/8fy7/JS6oS0+gYQQQl0Lu+xa8+bKcfLgd/28beKVNRcdebnB9AWFEELdEyak1pDOukfTwsQj6xors4W1OLMLQqhT5s6d27dv340bNy5YsKClbZqamlxcXFopxtr9YJddizakFG/U2mbG0iz2//tRO3VD276u12mMCiHUDSQmJtbW1lZVVZWWlsrcoLS09IsvvkhLS6M2LprhFZJsaUXVpEqQ+DFYABDWlghrSxhYTRUh1DJS89TT0zMyMhIAbt++vXv3bvLWjz/+eOPGjQULFjQ1Nc2cOZPNZp8/f/7EiRNkg4qKigULFtTX1wNAeHj4ypUrDQ0NaTwQ6uEVkmwbU4oBIMxtqORjsHUFPzRWZTMtlmIJO4QUUGNldlNVNtNiGbx7WJCa5cbKv6fobF7Ju6Gh4Y8//iDv3r5929TU9MiRIwkJCfHx8UlJSTdu3Pjmm2/GjRvn5OS0bt06Pz8/Uvhu79691HxpCgWvkGTYkFKcVlRjpq+5YepQyfUMrcHC2hIGEy+PEFJETVXZ4gcE6wp+oGy5qSpbHIO4kndCQgKp5P3esLW0tH777bePP/7YyMho/vz5HTny7gKvkKTJ7KwjmBbLsDoDQgpLzcCWafHXMtNiqXi9vJfVDP5+CKR5Je/q6mrxu6SWXXPa2trq6upCobC1w+sBMCFJc9mXC8066wCgsTIbAPDxI4QUluQDgqQzjZplyZ+F5pW8dXV1X758Sd7Ny8trPpleQ0ODj4/PqVOnNmzYcOLEiW4z214HYJfd32rzM//8fBAAOH/QV6qzDgDeZs59mzkXx3wjhFrh7e1tZGRkY2Pj7u6emJjo6uo6efLkhoYGJycnFxcXXV3d5h9Zu3atq6urvb19RETEihUrWhp31xPgFdJfGl9ySD3voJr4lT47pd4V1pYwLZYKa0uxvw4h1LrIyEipKt3Xr1+vra3V1NQUV/sm07y6u7u7u/9dAsbExITD4UjuqqKigqKgFQMmpL+QEkE3NS1nrd0srlknxtAylrxCRwihVjSv0t3NpnaVE+yyAwDghM0i9bwf+x2SOd0RLzeY3ENCCCEkJ3iFBLX5maRgXbL16h3Nbh0BAJ9zks85CTiiASGE5KmnJyTxVLAhlpt/XyH7CQB1A1umxVLJkZ0IIQVhbW1tYWHx/u16Bmtra7pD6JSenpD+nae3BuCmpuW/vpzZ0jZ4AwkhhRUdHd3GLbvxPO7dRo++h/TVsYe//lFuMSTG4cdzMm8dAQAvN7jmoiPeQEIIIXnruVdIJBsBwJVAVvNhdWLk7pEqjvZGCCE566EJSTIbtXRtRPR1vd6Is5UjhJD89awuOzLhXtuzUWNlNkPLuMO1vdveuy0ntAegCDHQHoAixIABKEIMtAeg+FREIhHdMXQcl8uNioricDju7u6Ojo5S71pYWBQUFJDl2vzMF/9d3ljBWTz5bFpRDbQhGwHAq9NDGVrGOnaxHbtCkgyAFrQHoAgx0B6AIsSAAShCDBjAeyn3FZK/vz+fz58yZcrOnTvPnDkjc5vGlxxO2KySsM8bKzilaoa1+Vlm+pptyUbk8kjdwBb76xBCiAJKfA8pNzeXx+MtX74cAAwMDEJDQz08PCQ36N9LWHV8F+mmK1UzjNeedNZ8nvMHejeazSshE5mtHKupIoQQNZS4yy4xMTE9PT08PBwAhELh6NGj79+/L65dCACvTg8FgMoo/k+6Mzd6nQOAD7eP1qr6Myu8AQAmBvcCgFaWc4pU/n1AnfrjQgghebC2tlbw+1hKfIVUW1uroaFBlhkMhoqKikAgkExIxKZJvw6z+AfwzwEAJ/MMvEtUpC+1leUJYwbfzwjHckEIIUQNJU5IDAZDcoJFkUgklY30PYsBIOavV8VS69u+jBBCiAJKPKhBV1eXx+OR5ZqaGnV1dVVVVXpDQggh1GFKnJCsra2vX79eWVkJAAkJCdOmTaM7IoQQQh2nxIMaACAxMXHv3r3m5ubl5eVRUVGGhoZ0R4QQQqiDlDshAYBQKOTz+Uwmk+5AEEIIdYrSJySEEELdgxLfQ0IIIdSdqG7YsIHuGDqIy+UeOnQoPj5eQ0NjyJAhbXn37t276enpDx48ePDgwatXr0xNTSloV2ajL168iIiIOHXqVEVFxahRo1RUVKiPgaipqdm5c+ekSZOoD0AgEBw9evTYsWOlpaVjxoyh5UuoqamJiIiIj49vamoaNmxYuwLocCSdP/aOtSuzUYpPxVYOnJpTUWYAFJ+KMpuj5VR8+vTpxYsXH0iwsLBo/jQnZZQ4IX3xxRcDBgyws7Pbs2dP7969paYxlvnu999//+jRIz6fX1FRoaqqOmbMGArabd4ol8v18PCYOHGilZVVSkrKtWvXpkyZQnEM4o1XrVqVkpKyePFi6gNYsmQJl8t1c3O7dOnSrVu3nJycqI/B29t76NChdnZ2hw8ffvv2LYvFalcMHYuk88fesXabN0r9qdjKgVNzKsoMgOJTUWZztJyKT58+TU9Pr6ioqKioyMjISE1NXbBgAY0JCUTKKScnZ/r06WT5zp07np6ebXn3k08+KS4uprjd5o2mpqYuX76cLL9+/drS0pL6GIj4+PhVq1axWCzqA3jw4MGnn35KlisqKrZs2UJ9DBUVFR9++CFZTk1NDQgIaFcMHY6kk8fesXZlNkrxqdjKgVNzKsoMgOJTUWZzdJ2KYjwez9nZ+c6dOx1otwspa6WGZ8+ejRgxgiyPGTOmsLBQKBSKE7vMd0UiEYfDuXTpUn5+vrm5+ddffy2uPCS/dhsbG5s36urq6urqSjYrKioyMDCgPgYA4HA4UVFRhw8fTk1NpT6ABw8ejBs3LjMzMzk5mcVirVu3jvoY9PX1DQ0NL1y44ODgkJycPHLkyHbF0LFIhEJhJ4+9Y+3m5+c3b5TiU1FmDEDhqSgzAIpPRZkx0HUqit/96aefXFxcPvzwww6024WUdVCDzEJ2rb977949dXX13r17e3h45OXlBQUFUdAuKSHRUqOVlZXBwcFr1qyhJYbg4OAtW7ZoarY4fbtcA3j48OHVq1fPnDljbW197NixTZs2UR8Dg8H44osvvvvuuwULFmRlZc2cObO9X0UHIhEIBJ089o61W15e3kqj1JyKLcVA2akoMwCKT0WZMdB1KpKXFRUVR48eDQwM7ECjXUtZr5BaL2Qn810Wi5WTk0PW2NraWllZVVRU9OvXT67t6uvrt9Qom81euHChv7+/u7s79TGcOHFi0qRJo0aNqq2tbVfrXRWAiorK4MGDt23bBgB2dnYODg7r1q1re/GnLonhyZMnMTExaWlpOjo6V69eXbBgweXLl9v+PXQsEvJb0Jlj71i7vXv3bqlRyk5FmTEcPHiQslNRZgAUn4oyY7h16xYtpyJZPnHihJubmyIUFlDWK6TWC9nJfLe8vPzhw4dkJZPJVFVVFW8jv3Zfvnwps9Fbt275+fmFhITMmzePlhhOnz596NCh8ePHOzg48Hi88ePHNzY2UhnA6NGj+/fvT1bq6+urqKg0NDRQ/CXk5ORMmDBBR0cHAJycnKqqqjrwm9iBU7GTx96xdvX19WU2SuWpKDMGKk9FmQFQfCrKjIGuU5G8vHjx4ieffNLe5uSCwvtVXamqqorFYlVUVIhEooiIiFWrVpH1N27cqK6ulvnu9evXp0yZUl9fLxKJUlJSpkyZ0lXtkkZlviuz0dLSUhsbm5ycnC489vbGIMbj8dp7J7lLAigrK2OxWC9fvhSJRNevXxff5qUyBtIuWXnv3j0bGxuBQNCuMFqKRNTqqdjJY+/YNyCzUYpPxdYPnIJTUWYAFJ+KMpuj61QUiURNTU0jR45sampqb3PyoKwJSSQSnTp1ytnZeeHChZ988gn5lkUiEYvFSktLa+nd77//3sXFZeHChVOmTHn06FFXtStuVOa7zRvdunXr8P9FfQxiHfgV6KoAzp075+zs/K9//evjjz++d+8eLTHs3r2bxDBp0qTs7Oz2xtBSW6L3nYqdPPaOfQPNG6X+VGzlwKk5FWUGQPGpKLM5uk7FO3fuWFlZday5LqfcpYNaL2Qn812BQNDQ0NDJ2nftbbdLGlW0GLokgE6WIlSEGDoWCV3tyqP2I+0xdEkAFJ+KinNKKBTlTkgIIYS6DWUd1IAQQqibwYSEEEJIIWBCQgghpBAwISGEEFIImJAQQggpBExICCGEFAImJIQQQgoBExJCCCGFgAkJIYSQQsCEhBBCSCFgQkIIIaQQMCEhhBBSCJiQEEIIKQRMSAghhBQCJiSEEEIKARMSQgghhYAJCSGEkELAhIQQQkghYEJCCCGkRhFDfwAAAD5JREFUEDAhIYQQUgiYkBBCCCkETEgIIYQUAiYkhBBCCgETEkIIIYWACQkhhJBCwISEEEJIIWBCQgghpBD+H+e0x1OYT8bhAAAAAElFTkSuQmCC\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "clear variables; close all; clc;\n", "cd C:/Users/zurit/OneDrive/blog/files/Nonlinear_DCmotor\n", "global E_0 Tau_L0 T_Amb B_2C;\n", "\n", "E_0 = 120; % [V] 120\n", "Tau_L0 = 80; % [N.m] 80\n", "T_Amb = 18; % [deg] 18\n", "B_2C = 80; % [N] 80/300\n", "\n", "t0 = 0; tfinal = 0.3; step = 1e-3;\n", "x0 = [0; 0; 0]; % initial conditions\n", "\n", "input_type = 0; % 0=constant, 1=sinusoidal\n", "%% ode45 vs ode45m vs eufix1\n", "\n", "timer = clock; \n", "[t1,x1] = ode45('asst02_2017',[t0, tfinal],x0);\n", "Tsim1 = etime(clock,timer); % integration time \n", "Len1 = length(t1); % number of time-steps \n", "\n", "timer = clock;\n", "[t2,x2] = ode45m('asst02_2017',t0,tfinal,x0,step);\n", "Tsim2 = etime(clock,timer); % integration time \n", "Len2 = length(t2); % number of time-steps\n", "\n", "timer = clock;\n", "[t3,x3] = eufix1('asst02_2017',[t0 tfinal],x0,step);\n", "Tsim3 = etime(clock,timer); % integration time \n", "Len3 = length(t3); % number of time-steps\n", "\n", "%% Relative error\n", "\n", "% relative error at max current: ode45 vs eufix1\n", "max_iA_ode45 = max(x1(:,1)); \n", "max_iA_eufix1 = max(x3(:,1)); \n", "max_iA_error = 100*abs( (max_iA_ode45-max_iA_eufix1)/max_iA_ode45 ) ; \n", "\n", "% relative error at max angular velocity: ode45 vs eufix1\n", "max_omega2_ode45 = max(x1(:,2)); \n", "max_omega2_eufix1 = max(x3(:,2)); \n", "max_omega2_error = 100*abs( (max_omega2_ode45-max_omega2_eufix1)/max_omega2_ode45 ); \n", "\n", "%% Plotting\n", "if input_type == 0\n", " %% Constant input e_i=E0\n", " figure;\n", " subplot(3,1,1); \n", " plot(t1,x1(:,1),t2,x2(:,1),'--',t3,x3(:,1),'-.','LineWidth',1.5);\n", " title(['Nonlinear DC motor with thermal model, $B_{2C}=$',num2str(B_2C)],'Interpreter','Latex');\n", " ylabel('$i_A$ [A]','Interpreter','Latex');\n", " legend(['ode45: ',num2str(Tsim1),' [s]'],['ode45m: ',num2str(Tsim2),' [s]'],['eufix1: ',num2str(Tsim3),' [s]']);\n", " grid on;\n", "\n", " subplot(3,1,2); \n", " plot(t1,x1(:,2),t2,x2(:,2),'--',t3,x3(:,2),':','LineWidth',1.5);\n", " ylabel('$\\omega_2$ [rad/s]','Interpreter','Latex');\n", " legend('ode45','ode45m','eufix1','Location','southeast');\n", " grid on;\n", "\n", " subplot(3,1,3); \n", " plot(t1,x1(:,3),t2,x2(:,3),'--',t3,x3(:,3),':','LineWidth',1.5);\n", " xlabel('Time [s]','Interpreter', 'Latex');\n", " ylabel('$\\theta_M$ [deg]','Interpreter','Latex');\n", " legend('ode45','ode45m','eufix1','Location','southeast');\n", " grid on;\n", " \n", " % print('../asst02_2017/sinE0_ode45-ode45m-eufix1_1e-4.png', '-dpng', '-r300'); % Save as PNG with 300 DPI \n", "\n", " figure;\n", " subplot(2,1,1);\n", " plot(t1,x1(:,1),t2,x2(:,1),'--',t3,x3(:,1),'-.','LineWidth',1.5);\n", " title(['Nonlinear DC motor with thermal model, $B_{2C}=$',num2str(B_2C)],'Interpreter','Latex');\n", " ylabel('$i_A$ [A]','Interpreter','Latex');\n", " legend('ode45','ode45m','eufix1');\n", " axis([0.05 0.07 -inf inf]);\n", " text(0.058,5.5,['Relative error at max $i_{A}$=',num2str(max_iA_error),' $\\%$'],'Interpreter','Latex');\n", " grid on;\n", "\n", " subplot(2,1,2); \n", " plot(t1,x1(:,2),t2,x2(:,2),'--',t3,x3(:,2),':','LineWidth',1.5);\n", " ylabel('$\\omega_2$ [rad/s]','Interpreter','Latex');\n", " legend('ode45','ode45m','eufix1','Location','southeast');\n", " axis([0.05 0.07 -inf inf]);\n", " text(0.058,40,['Relative error at max $\\omega_{2}$=',num2str(max_omega2_error),' $\\%$'],'Interpreter','Latex');\n", " grid on;\n", "\n", " figure;\n", " plot(t1,x1(:,3),t2,x2(:,3),'--',t3,x3(:,3),':','LineWidth',1.5);\n", " title('Motor temperature $\\theta_M$ over $80[s]$','Interpreter','Latex');\n", " xlabel('Time [s]','Interpreter', 'Latex');\n", " ylabel('$\\theta_M$ [deg]','Interpreter','Latex');\n", " legend('ode45','ode45m','eufix1','Location','southeast');\n", " grid on;\n", "\n", "elseif input_type == 1\n", " %% Sinusoidal input e_i\n", "\n", " figure;\n", " subplot(3,1,1); \n", " plot(t1,x1(:,1),t2,x2(:,1),'--',t3,x3(:,1),'-.','LineWidth',1.5);\n", " title(['Nonlinear DC motor with thermal model, $B_{2C}=$',num2str(B_2C)],'Interpreter','Latex');\n", " ylabel('$i_A$ [A]','Interpreter','Latex');\n", " legend('ode45','ode45m','eufix1','Location','southeast');\n", " grid on;\n", "\n", " subplot(3,1,2); \n", " plot(t1,x1(:,2),t2,x2(:,2),'--',t3,x3(:,2),':','LineWidth',1.5);\n", " ylabel('$\\omega_2$ [rad/s]','Interpreter','Latex');\n", " legend('ode45','ode45m','eufix1','Location','southeast');\n", " grid on;\n", "\n", " subplot(3,1,3); \n", " plot(t1,x1(:,3),t2,x2(:,3),'--',t3,x3(:,3),':','LineWidth',1.5);\n", " xlabel('Time [s]','Interpreter', 'Latex');\n", " ylabel('$\\theta_M$ [deg]','Interpreter','Latex');\n", " legend('ode45','ode45m','eufix1','Location','southeast');\n", " grid on;\n", "\n", " figure;\n", " subplot(3,1,1);\n", " plot(t1,x1(:,2),t2,x2(:,2),'--',t3,x3(:,2),'-.','LineWidth',1.5);\n", " title(['Stiction behaviour on $\\omega_2$, $B_{2C}=$',num2str(B_2C)],'Interpreter','Latex');\n", " ylabel('$\\omega_2$ [rad/s]','Interpreter','Latex');\n", " legend('ode45','ode45m','eufix1','Location','southeast');\n", " axis([0.148 0.157 -0.6 0.4]);\n", " grid on;\n", "\n", " subplot(3,1,2); \n", " plot(t1,x1(:,2),t2,x2(:,2),'--',t3,x3(:,2),':','LineWidth',1.5);\n", " ylabel('$\\omega_2$ [rad/s]','Interpreter','Latex');\n", " legend('ode45','ode45m','eufix1','Location','southeast');\n", " axis([0.148 0.157 -0.015 0.010]);\n", " grid on;\n", "\n", " subplot(3,1,3);\n", " plot(t1,x1(:,2),t2,x2(:,2),'--',t3,x3(:,2),'-.','LineWidth',1.5);\n", " xlabel('Time [s]','Interpreter', 'Latex');\n", " ylabel('$\\omega_2$ [rad/s]','Interpreter','Latex');\n", " legend('ode45','ode45m','eufix1','Location','southeast');\n", " axis([0.148 0.157 -11e-5 5e-5]);\n", " grid on;\n", "\n", " figure;\n", " plot(t1,x1(:,3),t2,x2(:,3),'--',t3,x3(:,3),':','LineWidth',1.5);\n", " title('Motor temperature $\\theta_M$ over $80[s]$','Interpreter','Latex');\n", " xlabel('Time [s]','Interpreter', 'Latex');\n", " ylabel('$\\theta_M$ [deg]','Interpreter','Latex');\n", " legend('ode45','ode45m','eufix1','Location','southeast');\n", " grid on;\n", "\n", "end\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\n", "# Simulation Scenarios\n", "Two types of scenarios are simulated. The first one is submitted to a constant input, low stiction, and two step sizes. The second scenario is more interesting because we study the behaviour due to a sinusoidal input which emulates the reversing mode of the motor at $5[Hz]$ with higher stiction. Both scenarios have load torque at $t=0.2[s]$.\n", "\n", "
Table 1: Scenario 1 and 2
\n", "\n", "| \t\t| Scenario 1 \t| Scenario 2 |\n", "|:-----:|:-------------|:----------|\n", "|`ode45m` step size | $1\\times 10^{-3}$ and $1\\times 10^{-4}$ | $1\\times 10^{-4}$ |\n", "|`eufix1` step size | $1\\times 10^{-3}$ and $1\\times 10^{-4}$ | $1\\times 10^{-4}$ |\n", "|`ode45` step size | auto | auto |\n", "| $e_i$ | $E_0$ | $E_0 \\sin[5(2\\pi)(t-0.05)]$ |\n", "| $E_0$ | $120~[V]$ | $120~[V]$ |\n", "| $\\tau_{L}$ | $80~[N m]$ at $t=0.2[s]$} | $80~[N m]$ at $t=0.2[s]$ |\n", "| $\\theta_A$ | $18~[°C]$ | $18~[°C]$ |\n", "| $B_{2C}$ | $80~[N]$ | $300~[N]$ |\n", "\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Simulation Results\n", "## Scenario 1\n", "The result in Fig. 1 shows the output states due to constant input $E_0=120$, and $B_{2C}=80$. The current overshoot at $0.05[s]$ is due to the inertia that the motor has to overcome. After the inertia is broken, the current $i_A$ drops down to a constant value. The load torque $\\tau_{L}$ is applied at $0.2[s]$ which produces the increment in the current and the decrement in the angular velocity. Also, `eufix1` solves the system with noticeable error, this result is analyzed later.\n", "\n", "
\n", " \"Fig1\"\n", "
Fig. 1 - Scenario 1: step size $1\\times 10^{-3}$
\n", "
\n", "\n", "The time simulation of each solver indicates that `ode45` is 10 times slower than `ode45m` and 30 times slower than `eufix1`. \n", "\n", "
Table 2: Scenario 1: simulation time and number of steps.
\n", "\n", "| \t\t | `ode45` \t | `ode45m` | `eufix1` |\n", "|:---------------------:|:----------:|:--------:|:--------:|\n", "| simulation time [s] | $0.267$ | $0.025$ | $0.009$ |\t\t\t\n", "| number of time steps | $71769$ | $15133$ | $80001$ |\n", "\n", "\n", "The temperature of the motor $\\theta_M$ increases linearly and reaches steady-state at $45[s]$ approximately, which indicates that the motor won't reach unsafe temperatures.\n", "\n", "
\n", " \"Fig2\"\n", "
Fig. 2 - Scenario 1: $\\theta_M$ over $80[s]$
\n", "
\n", "\n", "Although, `eufix1` is the fastest solver, with step size of $1\\times 10^{-3}$, `eufix1` outputs the worst performance. The result can be improved if the steps size is decreased to $1\\times 10^{-4}$. Fig. 3 and Fig. 4 show the relative error at max current and max angular velocity between `ode45` and `eufix1`.\n", "\n", "
\n", " \"Fig3\"\n", "
Fig. 3 - Scenario 1: step size $1\\times 10^{-3}$
\n", "
\n", "\n", "
\n", " \"Fig4\"\n", "
Fig. 4 - Scenario 1: step size $1\\times 10^{-4}$
\n", "
" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Scenario 2\n", "In this scenario we submitted the DC motor to high stiction $B_{2C}=300$ and sinsusoidal input at $5[Hz]$ which simulates the reversing mode. Table 3 shows the output for the three solvers showing that `ode45` is the slowest by far.\n", "\n", "
Table 3: Scenario 2: simulation time and number of steps.
\n", "\n", "| \t\t | `ode45` \t | `ode45m` | `eufix1` |\n", "|:---------------------:|:----------:|:--------:|:--------:|\n", "| simulation time [s] | $517.798$ | $3.463$ | $0.093$ |\t\t\t\n", "| number of time steps | $13559913$ | $9647$ | $3002$ |\n", "\n", "Fig. 5 shows the simulation output. The relevant result is the behavior of the system around the (nonlinear) stiction. $\\omega_2$ sticks at $0.15[s]$ and $0.25[s]$ due to $B_{2C}$. \n", "\n", "
\n", " \"Fig5\"\n", "
Fig. 5 - Scenario 2: reversing mode.
\n", "
\n", "\n", "Even though the solvers were able to solve the dynamics with stiction, `ode45` took too much time to overcome this nonlinearity. Fig. 6 shows the stiction with three different zoom levels for each solver. The fastest but with more integration step error is `eufix1`. `ode45m` has less error $\\pm 0.01$, and finally `ode45` solves with the minimum error, around $\\pm 10\\times 10^{-5}$. In conclusion, `ode45m` is the best choice against the rest because it can obtain the solution with low error and with decent speed.\n", "\n", "
\n", " \"Fig6\"\n", "
Fig. 6 - Scenario 2: stiction behavior at $0.148\\leq t \\leq 0.157$.
\n", "
\n", "\n", "The motor temperature $\\theta_M$ in reversing mode reaches the steady-state at $45[s]$ approximately which indicates the motor won't suffer overheat due to stiction. \n", "\n", "
\n", " \"Fig7\"\n", "
Fig. 7 - Scenario 2: motor temperature.
\n", "
" ] } ], "metadata": { "kernelspec": { "display_name": "Matlab", "language": "matlab", "name": "matlab" }, "language_info": { "codemirror_mode": "octave", "file_extension": ".m", "help_links": [ { "text": "MetaKernel Magics", "url": "https://metakernel.readthedocs.io/en/latest/source/README.html" } ], "mimetype": "text/x-octave", "name": "matlab", "version": "0.16.9", "pygments_lexer": "octave" }, "latex_envs": { "LaTeX_envs_menu_present": true, "autoclose": false, "autocomplete": true, "bibliofile": "biblio.bib", "cite_by": "apalike", "current_citInitial": 1, "eqLabelWithNumbers": true, "eqNumInitial": 1, "hotkeys": { "equation": "Ctrl-E", "itemize": "Ctrl-I" }, "labels_anchors": false, "latex_user_defs": false, "report_style_numbering": false, "user_envs_cfg": false }, "nikola": { "category": "", "date": "2020-11-20 17:48:34 UTC-04:00", "description": "", "link": "", "slug": "Nonlinear_DCmotor", "tags": "nonlinear, DC motor, dynamics model, stiction, Coulomb friction, thermal model, differential equations, Matlab, Jupyter", "title": "Nonlinear DC motor with thermal model and Coulomb friction (stiction)", "type": "text" }, "toc": { "base_numbering": 1, "nav_menu": {}, "number_sections": true, "sideBar": true, "skip_h1_title": false, "title_cell": "Table of Contents", "title_sidebar": "Contents", "toc_cell": true, "toc_position": {}, "toc_section_display": true, "toc_window_display": true } }, "nbformat": 4, "nbformat_minor": 4 }