TY - JOUR
T1 - Autonomous-Jump-ODENet
T2 - Identifying Continuous-Time Jump Systems for Cooling-System Prediction
AU - Yuan, Zhaolin
AU - Wang, Yewan
AU - Ban, Xiaojuan
AU - Ning, Chunyu
AU - Dai, Hong Ning
AU - Wang, Hao
N1 - Funding Information:
This work was supported in part by the National Natural Science Foundation of China under Grant 61873299, 61902022, and 61972028, in part by the Scientific and Technological Innovation Foundation of Shunde Graduate School, USTB under Grand BK21BF002, in part by the Macao Science and Technology Development Fund under Macao Funding Scheme for Key R & D Projects under Grand 0025/2019/AKP, and in part by the Fundamental Research Funds for the Central Universities of China under Grand FRF-TP-20-061A1Z. Paper no. TII-22-2194.
PY - 2023/7
Y1 - 2023/7
N2 - Periodic Jump processes commonly occur in complex industrial systems. As the systems vary dynamically between different stages, learning their dynamics in an unified model, so as to forecast and simulation accurately is challenging. In this study, we propose autonomous jump ordinary differential equation net (AJ-ODENet) to learn the continuous-time periodic jump system. The model consists of several Hierarchical ODENets (H-ODENets) and a stage transition predictor. Each H-ODENet is an advanced version of ordinary differential equations network to individually learn specific dynamics in each stage from irregularly sampled sequence data. The stage transition predictor realizes autonomous stage transition during open-loop simulation. Furthermore, an encoder-decoder framework built on AJ-ODENet is employed on a real cooling system of data center to simulate some variables in runtime. With multivariate data given, such as server power and environmental temperature, the model can simulate the working patterns as in reality, and the relative error of the predicted energy consumption is within 5%. Furthermore, based on the model, we infer the optimal cooling temperature settings under different heat loads. The simulation results indicate that 6%-25% of cooling energy consumption can be optimized.
AB - Periodic Jump processes commonly occur in complex industrial systems. As the systems vary dynamically between different stages, learning their dynamics in an unified model, so as to forecast and simulation accurately is challenging. In this study, we propose autonomous jump ordinary differential equation net (AJ-ODENet) to learn the continuous-time periodic jump system. The model consists of several Hierarchical ODENets (H-ODENets) and a stage transition predictor. Each H-ODENet is an advanced version of ordinary differential equations network to individually learn specific dynamics in each stage from irregularly sampled sequence data. The stage transition predictor realizes autonomous stage transition during open-loop simulation. Furthermore, an encoder-decoder framework built on AJ-ODENet is employed on a real cooling system of data center to simulate some variables in runtime. With multivariate data given, such as server power and environmental temperature, the model can simulate the working patterns as in reality, and the relative error of the predicted energy consumption is within 5%. Furthermore, based on the model, we infer the optimal cooling temperature settings under different heat loads. The simulation results indicate that 6%-25% of cooling energy consumption can be optimized.
KW - Contained cooling system
KW - jump system modeling
KW - multi-input-multi-output system prediction
KW - ordinary differential equations network (ODENet)
UR - http://www.scopus.com/inward/record.url?scp=85139407558&partnerID=8YFLogxK
U2 - 10.1109/TII.2022.3207835
DO - 10.1109/TII.2022.3207835
M3 - Journal article
SN - 1551-3203
VL - 19
SP - 7894
EP - 7904
JO - IEEE Transactions on Industrial Informatics
JF - IEEE Transactions on Industrial Informatics
IS - 7
ER -
