如何在MATLAB中绘制随时间变色的相图并添加色阶？

Got it, let's walk through exactly how to create those phase plots with color-coded time and a colorbar reference—this is a common need but the documentation doesn't spell it out super clearly, so I'll break it down step by step.

Step 1: Read and Prepare Your Data #

First, load your text file with readtable and extract the relevant columns. I'll assume your test.txt has no headers (if it does, just adjust the variable names):

% Load the data
data = readtable('test.txt');

% Extract columns: time, and the two phase plot pairs
time = data{:, 1};
x1 = data{:, 2};
y1 = data{:, 3};
x2 = data{:, 4};
y2 = data{:, 5};

Step 2: Create Phase Plots with Time-Varying Color #

The trick here is using MATLAB's patch function, which lets you interpolate edge colors based on a third variable (your time sequence). This is way more efficient than looping to draw each line segment individually.

Here's the code to create two subplots (one for each phase plot pair) with color-coded lines and colorbars:

% Set up the figure with two subplots
figure('Position', [100 100 1000 400]);

% First phase plot: Columns 2 vs 3
subplot(1, 2, 1);
% Use patch to draw a line with interpolated color based on time
patch(x1, y1, time, 'EdgeColor', 'interp', 'FaceColor', 'none', 'LineWidth', 1.5);
colorbar; % Add the color scale reference
xlabel('Column 2 Values');
ylabel('Column 3 Values');
title('Phase Plot: Col 2 vs Col 3 (Color = Time)');
caxis([min(time), max(time)]); % Set colorbar to match actual time range
colormap(viridis); % Use a perceptually uniform colormap (swap for jet/plasma if preferred)

% Second phase plot: Columns 4 vs 5
subplot(1, 2, 2);
patch(x2, y2, time, 'EdgeColor', 'interp', 'FaceColor', 'none', 'LineWidth', 1.5);
colorbar;
xlabel('Column 4 Values');
ylabel('Column 5 Values');
title('Phase Plot: Col 4 vs Col 5 (Color = Time)');
caxis([min(time), max(time)]);
colormap(viridis);

% Optional: Sync colorbars if time ranges are identical
linkprop(findobj(gcf, 'Type', 'ColorBar'), 'Limits');

Key Explanations #

• patch with EdgeColor='interp': This tells MATLAB to interpolate the edge color of the patch using the third input (time), creating a smooth color gradient along the line that matches your time sequence. Setting FaceColor='none' ensures we only see the line, not a filled shape.
• caxis: This sets the color axis limits to match your actual time values, so the colorbar shows real time instead of a normalized 0-1 range.
• Colormap: viridis is a great default because it's colorblind-friendly and perceptually uniform, but you can replace it with any MATLAB colormap (e.g., jet, plasma, cool).

Troubleshooting Tips #

• If your text file has headers, use readtable('test.txt', 'VariableNames', {'Time', 'X1', 'Y1', 'X2', 'Y2'}) to name columns explicitly.
• If your line looks choppy, increase LineWidth or check that your time/data sequences are sampled densely enough.
• If you want discrete color steps instead of a smooth gradient, replace time with a discretized version (e.g., discrete_time = discretize(time, 10); to split into 10 time bins).

内容的提问来源于stack exchange，提问作者Marine Galantin