Integrator error: Integrator in Simulink 2021a, I input a negative nymber but the output is a positive integral value

Fig. 1 the -2.6 is input and the 1.53e+6 is the output of the integrator

Fig. 1 the -2.7 is input and the 1.58e+6 is the output of the integrator

Compare Fig.1 and Fig.2, the input of the integrator is both negative numbers, while the integral value become larger, which is wrong. Who can solve my problem? Many thanks!

Answer 1

[Sam Chak]

The integrator could be integrating from an initial positive value.

[壮 王]

Thanks for your answer! The integrator in my simulation is shown as Fig.1, and the setting is shown in Fig.2. But, I don not kown where the wrong is ?

Fig.1 The integrator in my simulation

Fig.2 The parameter of the integrator

Also, I find that if the input number is all positive as shown in Fig.3, the integrator will sometimes give a decline trend as shown in Fig. 4, marked in red circle shape. Can you help me an advise to find the reason?

Fig.3 The input of the positive number

Fig.3 The wrong output marked in red circle shape

[Sam Chak]

Is the input signal to the Integrator (as shown in Fig. 3) some kind of a bounded random signal?

Also, test if your Integrator behaves correctly by integrating . The output on the scope should show a sine signal.

[壮 王]

Thank you, the input signal is not a bounded random number, It is just the battery current, and the time step in my Simulation is 600s. Besides, I have tested the Cos function and the result was same as yours. Maybe, I should analysis how the Ingegrator works furtherly. I test two different blocks, namely the “Ingegrator” and the "Discrete-Time Integrator1" , the result showed that the the "Discrete-Time Integrator1" gives a correct result.