Process data compression and trending are essential for improving control system performances. Swing Door Trending (SDT) algorithm is well designed to adapt the process trend while retaining the merit of simplicity. B...Process data compression and trending are essential for improving control system performances. Swing Door Trending (SDT) algorithm is well designed to adapt the process trend while retaining the merit of simplicity. But it cannot handle outliers and adapt to the fluctuations of actual data. An Improved SDT (ISDT) algorithm is proposed in this paper. The effectiveness and applicability of the ISDT algorithm are demonstrated by computations on both synthetic and real process data. By applying an adaptive recording limit as well as outliers-detecting rules, a higher compression ratio is achieved and outliers are identified and eliminated. The fidelity of the algorithm is also improved. It can be used both in online and batch mode, and integrated into existing software packages without change.展开更多
Augmented UD identification (AUDI) technique is derived from the traditional recursive least-squares (RLS) algorithm and has been developed rapidly during the last decade. AUDI is a cluster of identification algorithm...Augmented UD identification (AUDI) technique is derived from the traditional recursive least-squares (RLS) algorithm and has been developed rapidly during the last decade. AUDI is a cluster of identification algorithms based on matrix factorization methods (such as QR and LDL) and thus shows its stable performance in system identification applications. An AUDI algorithm with resetting strategy (RAUDI) has much ability in rapid time-varying SISO system identification. In this paper, an endeavor to expand the RAUDI in MIMO system identification is made and a comparative experiement is done to exhibit its good ability in rapidly changing parameter estimate in MIMO system.展开更多
Augmented UD identification (AUDI) technique is derived from the traditional recursive least squares algorithm and had been developed rapidly during last decade. However, as the identification process evolves, AUDI al...Augmented UD identification (AUDI) technique is derived from the traditional recursive least squares algorithm and had been developed rapidly during last decade. However, as the identification process evolves, AUDI algorithm falls easily into identification saturation, which means that AUDI algorithm cannot respond to time varying system parameters unless a set of very strong identification signals is utilized or a long identification period is occupied. To overcome such a difficulty, a novel resetting AUDI (RAUDI) strategy is advanced by resetting the augmented information matrix based on MF (Monitor Function) monitoring the conspicuous change of process parameters. The numeric experiment demonstrates that the RAUDI has a good performance in estimation of rapid parameter changes.展开更多
Many industrial process systems are becoming more and more complex and are characterized by distributed features. To ensure such a system to operate under working order, distributed parameter values are often inspecte...Many industrial process systems are becoming more and more complex and are characterized by distributed features. To ensure such a system to operate under working order, distributed parameter values are often inspected from subsystems or different points in order to judge working conditions of the system and make global decisions. In this paper, a parallel decision model based on Support Vector Machine (PDMSVM) is introduced and applied to the distributed fault diagnosis in industrial process. PDMSVM is convenient for information fusion of distributed system and it performs well in fault diagnosis with distributed features. PDMSVM makes decision based on synthetic information of subsystems and takes the advantage of Support Vector Machine. Therefore decisions made by PDMSVM are highly reliable and accurate.展开更多
基金The authors would like to acknowledge the support from Project“973”of the State Key Fundamental Research under grant G1998030415.
文摘Process data compression and trending are essential for improving control system performances. Swing Door Trending (SDT) algorithm is well designed to adapt the process trend while retaining the merit of simplicity. But it cannot handle outliers and adapt to the fluctuations of actual data. An Improved SDT (ISDT) algorithm is proposed in this paper. The effectiveness and applicability of the ISDT algorithm are demonstrated by computations on both synthetic and real process data. By applying an adaptive recording limit as well as outliers-detecting rules, a higher compression ratio is achieved and outliers are identified and eliminated. The fidelity of the algorithm is also improved. It can be used both in online and batch mode, and integrated into existing software packages without change.
文摘Augmented UD identification (AUDI) technique is derived from the traditional recursive least-squares (RLS) algorithm and has been developed rapidly during the last decade. AUDI is a cluster of identification algorithms based on matrix factorization methods (such as QR and LDL) and thus shows its stable performance in system identification applications. An AUDI algorithm with resetting strategy (RAUDI) has much ability in rapid time-varying SISO system identification. In this paper, an endeavor to expand the RAUDI in MIMO system identification is made and a comparative experiement is done to exhibit its good ability in rapidly changing parameter estimate in MIMO system.
文摘Augmented UD identification (AUDI) technique is derived from the traditional recursive least squares algorithm and had been developed rapidly during last decade. However, as the identification process evolves, AUDI algorithm falls easily into identification saturation, which means that AUDI algorithm cannot respond to time varying system parameters unless a set of very strong identification signals is utilized or a long identification period is occupied. To overcome such a difficulty, a novel resetting AUDI (RAUDI) strategy is advanced by resetting the augmented information matrix based on MF (Monitor Function) monitoring the conspicuous change of process parameters. The numeric experiment demonstrates that the RAUDI has a good performance in estimation of rapid parameter changes.
文摘Many industrial process systems are becoming more and more complex and are characterized by distributed features. To ensure such a system to operate under working order, distributed parameter values are often inspected from subsystems or different points in order to judge working conditions of the system and make global decisions. In this paper, a parallel decision model based on Support Vector Machine (PDMSVM) is introduced and applied to the distributed fault diagnosis in industrial process. PDMSVM is convenient for information fusion of distributed system and it performs well in fault diagnosis with distributed features. PDMSVM makes decision based on synthetic information of subsystems and takes the advantage of Support Vector Machine. Therefore decisions made by PDMSVM are highly reliable and accurate.