how to find time interval of peak value in a timeseries ?

조회 수: 3 (최근 30일)
i have a load profile data set.. Time vs Load Demand as a timeseries object. If I have a threshold for peak as 800KW How do I find at what time my value goes above 800 kW ? The time series object is for every 30 mins interval over 24 hrs.
I need to find the highlighted time interval in the picture. Any idea?

채택된 답변

Star Strider
Star Strider 2017년 8월 4일
If you want to interpolate to find the ‘exact’ time, this works:
t = 0 : 30 : 1440;
LD = 500 + 400*sin(2*pi*t/2880);
zci = @(v) find(v(:).*circshift(v(:), [-1 0]) <= 0); % Returns Approximate Zero-Crossing Indices Of Argument Vector
X800_idx = zci(LD - 800); % Subtract 800, Find Approximate Zero-Crossing
LD800_Rng = LD(X800_idx-1 : X800_idx+1)'; % Define Range For Interpolation
t800_Rng = t(X800_idx-1 : X800_idx+1)'; % Define Range For Interpolation
t800 = interp1(LD800_Rng, t800_Rng, 800, 'linear', 'extrap'); % Find ‘Exact’ Time (Minutes)
figure(1)
plot(t, LD)
hold on
plot(t800, 800, '*r')
hold off
grid
  댓글 수: 1
Star Strider
Star Strider 2017년 8월 4일
A slightly more robust version that finds the points where ‘LD’ crosses 800:
t = 0 : 30 : 1440; % Create Data
LD = 500 + 400*sin(2*pi*t/2880); % Create Data
zci = @(v) find(v(:).*circshift(v(:), [-1 0]) <= 0); % Returns Approximate Zero-Crossing Indices Of Argument Vector
X800_idx = zci(LD - 800); % Subtract 800, Find Approximate Zero-Crossing
for k1 = 1:numel(X800_idx)
LD800_Rng = LD(X800_idx(k1)-1 : X800_idx(k1)+1)'; % Define Range For Interpolation
t800_Rng = t(X800_idx(k1)-1 : X800_idx(k1)+1)'; % Define Range For Interpolation
t800(k1) = interp1(LD800_Rng, t800_Rng, 800, 'linear', 'extrap'); % Find ‘Exact’ Time (Minutes)
end
figure(1)
plot(t, LD) % Plot Load Curve
hold on
plot(t800, [1 1]*800, '*r') % Plot ‘Load = 800’ Points
hold off
grid

댓글을 달려면 로그인하십시오.

추가 답변 (1개)

fbaillon
fbaillon 2017년 8월 4일
You can use the find function:
T= ... % Time
LD= .... % Load Demand
timeYouWant=T(find(LD>800e3,1,'first))
Something like that...

카테고리

Help CenterFile Exchange에서 Correlation and Convolution에 대해 자세히 알아보기

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by