Wen, J;
Lu, J;
Spataru, C;
Weng, Y;
Lv, X;
(2024)
Coordinated intelligent control strategy and power management for marine gas turbine under pulsed load using optimized neural network.
Energy
, 313
, Article 133719. 10.1016/j.energy.2024.133719.
![[thumbnail of Manuscript.pdf]](https://discovery-pp.ucl.ac.uk/style/images/fileicons/text.png) |
Text
Manuscript.pdf - Accepted Version Access restricted to UCL open access staff until 6 December 2025. Download (4MB) |
Abstract
Faced with the challenges of potentially irreversible damage due to instantaneous large pulsed power loads onboard, considering all-electric ship's small power grid capacity and complex gas turbine system, this study proposes a coordinated intelligent control strategy. It utilizes multi-objective optimization neural network to control a validated 20 MW ship three-shaft gas turbine, ensuring global optimization tailored to pulsed load variations. Research findings reveal that the gas turbine efficiency at the design point is 32.3 %, with a steady-state error within 2.8 % and a maximum steady time error of 3.4 s, demonstrating acceptable accuracy. Under slope-type pulsed load of 20 MW change in 5s, the proposed control strategy effectively manages speed while utilizing the energy storage system to smooth load inputs. It is worth noting that, increasing the fuel flow rate to 1.28 kg/s and inlet guide vane to 5.8° reduces overshoot by 0.7 % and rotor steady time by 15.4 s. Under transient-type pulsed load of 10 MW back and forth change in 5s, the coordinated inlet guide vane control strategy mitigates surge margin and exhaust temperature issues. Adjusting the fuel flow to 1.17 kg/s and inlet guide vane to 9.96° enables the gas turbine to operate within safety boundaries, reducing rotor speed overshoot by 0.4 % and steady time by 6.1 s. Importantly, the optimized control strategy without energy storage system provides faster and safer dynamic regulation and power transmission for ship acceleration processes under radar load. However, considering only the coordinated strategy of the fuel sub-control loop, it is advisable to avoid compressor surge and over-temperature protection for electromagnetic ejection load scenarios. The research results will lay a valuable technical foundation for the intelligent control and smooth power transfer of the all-electric ship gas turbine power system with pulse loads in the extreme missions of ships in the future.
| Type: | Article |
|---|---|
| Title: | Coordinated intelligent control strategy and power management for marine gas turbine under pulsed load using optimized neural network |
| DOI: | 10.1016/j.energy.2024.133719 |
| Publisher version: | https://doi.org/10.1016/j.energy.2024.133719 |
| Language: | English |
| Additional information: | This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions. |
| Keywords: | All-electric ship gas turbine, Multi-object optimal method, Neural network control, Coordinate security controls, Large pulsed power load |
| UCL classification: | UCL UCL > Provost and Vice Provost Offices > UCL BEAMS UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of the Built Environment UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of the Built Environment > Bartlett School Env, Energy and Resources |
| URI: | https://discovery-pp.ucl.ac.uk/id/eprint/10202838 |
Archive Staff Only
 |
View Item |
