Estimating the accurate consumption energy of gradient power supply and gradient coil leads to advantage on an optimized power supply for environment and to achieve an economical system. In addition, providing sufficient power supply permits a flexibility in pulse sequences. If the gradient power supply has enough margins for power to keep stability of MRI system, greater motion probing gradient (MPG) can be achieved. Especially, in the large bore Actively Shielded Gradient Coil, when the distance between main coils and shield coils becomes a half, the consumption energy becomes 3 to 4 times greater to generate any gradient magnetic field. Thus, the gradient power supply provides a large proportion of consumption energy of MRI system in the large bore Actively Shielded Gradient Coil. For the point of area, foot prints of MRI system very much depend on the gradient power supply design and performance. On the other word, flexibility of parameter setting of sequences is limited by gradient design itself. The gradient power energy consumption depends on the gradient waveform. Because the gradient coil resistance is affected by frequency of gradient waveform, a simulation model is proposed to estimate the dynamic consumption energy with equivalent circuit for the gradient coil and evaluated actual and simulated consumption energy of DWI imaging with various conditions to ensure accuracy of the model.

1) Figure 1 shows the gradient power supply and the simulation model.The consumption energy of gradient power supply and gradient coil depends on gradient power supply output current I(t). The gradient power supply output current I(t) was applied in the simulation model. Power supply unit and capacitor bank in gradient power supply provide energy to balance the consumption energy. Especially, the capacitor bank has a larger contribution when applying large gradient pulse. Furthermore, the amount of capacitor bank voltage Vc(t) drop depends on the amount of supply energy from the capacitor bank. Therefore, we selected the capacitor bank voltage Vc(t) drop as simulation output signal. By comparing the simulated Vc(t) with the actual Vc(t), the parameters of equivalent circuit model of the gradient coil were optimized from these results.

2) The value of an echo train spacing (ETS) of DWI-EPI sequences was varied to evaluate the model accuracy on gradient coil frequency response. We compared the simulated Vc(t) with the actual Vc(t) by varying ETS values.

3) DWI b value and TE were varied to evaluate the Vc(t) to extend parameter ranges of the b value and TE. The possibility of expanding the control range of b value and TE was considered with acceptance of Vc(t) drops.

1) Using the optimized model, Vc(t) can be calculated dynamically using a 50us sampling pitch. The error of model was measured within 6% (Fig. 2).

2) We observed Vc(t) changes as a function of ETS. The simulated Vc(t) changes produced the gradient coil frequency response (Fig. 3).

3) We evaluated Vc(t) using variable DWI b values and TEs with matrix 128x128, FOV 25x26 and ETS 0.65ms (Fig. 4). Figure 4 shows the data within 88% Vc(t) drops. The value of Vc(t) drop is equal to a maximum slew rate of the evaluated MRI system. The range of Vc(t) drop is about 88%, and DWI imaging parameters were controlled as the simulated Vc(t) within the allowable range. As a result, the flexibility of DWI sequence permits to shorten about 20 ms for TEs and to double the b value at the same TE with ensuring stability of MRI system.

The dynamic consumption energy simulation model was calculated for the large bore Actively Shield Gradient Coil, and evaluated its accuracy. As the result, the error of model was measured to be within 6% and the model allows producing a reasonable gradient coil frequency response. Furthermore, by controlling DWI imaging parameters with the model, the DWI sequence permits to gaining the flexibility to shorten 20 ms in TE and to double the b value at the same TE with ensuring stability of MRI system.