Biomedical subjects

K M Poo

Publications and source records attributed to K M Poo.

6 recordsLinked to original sources

PubMed2006

Practical approach to parameter estimation for ASM3+ bio-P module applied to five-stage step-feed EBPR process.

Various parameter optimization approaches to a five-stage step-feed EBPR process modeled using the ASM3+bio-P module were examined. Five stoichiometric (Y(STO,NO), Y(H,O2, Y(H,NO,) Y(PAO,O2), Y(PO4)) and seven kinetic parameters (k(STO), eta(NO), b(H), mu(max),PAO, q(PHA), q(PP), mu(max),A) were estimated. The optimization approaches could be classified based on the data sources (batch experiments or CSTR operation data) and the number of target variables used in calculating the objective function. Optimized parameter values obtained by each approach were validated with CSTR operation data that were not used for parameter optimization. The results showed that the parameter optimization only with batch experimental results could not be directly applied to CSTR operation data. ASM3 + bio-P module parameters could be finely optimized only with CSTR operation data when sufficient target variables for objective function calculation were applied. When the number of target variables was increased, prediction performance was significantly improved. Once optimized, the model was able to predict the characteristic features of the five-stage step-feed process; namely, a high PAO yield, fast PAO growth, fast X(pp) storage, slow X(STO) and X(PHA) storage.

Bioreactors↗

PubMed2006

Rule-based fuzzy inference system for estimating the influent COD/N ratio and ammonia load to a sequencing batch reactor.

A fuzzy inference system using sensor measurements was developed to estimate the influent COD/N ratio and ammonia load. The sensors measured ORP, DO and pH. The sensor profiles had a close relationship with the influent COD/N ratio and ammonia load. To confirm this operational knowledge for constructing a rule set, a correlation analysis was conducted. The results showed that a rule generation method based only on operational knowledge did not generate a sufficiently accurate relationship between sensor measurements and target variables. To compensate for this defect, a decision tree algorithm was used as a standardized method for rule generation. Given a set of inputs, this algorithm was used to determine the output variables. However, the generated rules could not estimate the continuous influent COD/N ratio and ammonia load. Fuzzified rules and the fuzzy inference system were developed to overcome this problem. The fuzzy inference system estimated the influent COD/N ratio and ammonia load quite well. When these results were compared to the results from a predictive polynomial neural network model, the fuzzy inference system was more stable.

Algorithms↗

PubMed2006

Full-cyclic control strategy of SBR for nitrogen removal in strong wastewater using common sensors.

A full-cyclic automatic control strategy for sequencing batch reactors (SBR) was proposed using only common sensors such as ORP, DO and pH. The main objective was to develop a generally applicable and robust control strategy. To accomplish this, various control schemes found in the literature or suggested by authors were examined at diverse ammonia loads and SCOD/NH4(+)-N ratios. Advantages and constraints of each scheme were discussed and compared. Ammonia load was estimated with DO lag time during the aerobic stage, and then the influent pump was manipulated to meet the desired load at the next anoxic stage. A partial denitrification scheme was chosen for the anoxic stage period control, to save anoxic time and external carbon. For external carbon dosage control, intermittent feeding at each anoxic stage was concluded to be a suitable scheme. The anoxic stage period could be successfully controlled by the combination of pH increase and DO increase. Every suggested control scheme was incorporated into a full-cyclic control strategy and tested at 0.02, 0.035, 0.08 kg NH4(+)-N/m3/sub-cycle. From the results, it is expected to perform unmanned automatic SBR operation with this strategy.

Ammonia↗

PubMed2006

Equipment fault diagnosis system of sequencing batch reactors using rule-based fuzzy inference and on-line sensing data.

The importance of a detection technique to prevent process deterioration is increasing. For the fast detection of this disturbance, a diagnostic algorithm was developed to determine types of equipment faults by using on-line ORP and DO profile in sequencing batch reactors (SBRs). To develop the rule base for fault diagnosis, the sensor profiles were obtained from a pilot-scale SBR when blower, influent pump and mixer were broken. The rules were generated based on the calculated error between an abnormal profile and a normal profile, e(ORP)(t) and e(DO)(t). To provide intermediate diagnostic results between "normal" and "fault", a fuzzy inference algorithm was incorporated to the rules. Fuzzified rules could present the diagnosis result "need to be checked". The diagnosis showed good performance in detecting and diagnosing various faults. The developed algorithm showed its applicability to detect faults and make possible fast action to correct them.

Algorithms↗

PubMed2004

Treatment of strong nitrogen swine wastewater in a full-scale sequencing batch reactor.

Treatment of swine wastewater containing strong nitrogen was attempted in a full-scale SBR. The strongest swine wastewater was discharged from a slurry-type barn and called swine-slurry wastewater (SSW). Slightly weaker wastewater was produced from a scraper-type barn and called swine-urine wastewater (SUW). TCOD, NH4+-N and TSS in raw SSW were 23,000-72,000 mg/L, 3,500-6,000 mg/L and 17,000-50,000 mg/L, respectively. A whole cycle of SBR consists of 4 sub-cycles with anoxic period of 1 hr and aerobic period of 3 hr. The maximum loading rates of both digested-SSW and SUW were similar to 0.22 kg NH4+-N/m3/day whereas the maximum loading rates of raw SSW was up to 0.35 TN/m3/day on keeping the effluent quality of 60 TN mg/l. The VFAs portion of SCOD in raw SSW was about more than 60%. The VFAs in SUW and digested-SSW were about 22% and 15%, respectively. NH4+-N and PO4(3-)-P in SSW were removed efficiently compared to those in digested-SSW and DUW because SSW had high a C/N ratio and readily biodegradable organic. High concentration of organic was useful to enhance denitrification and P uptake. Also the amount of external carbon for denitrification was reduced to 5% and 10% of those for digested-SSW and SUW.

Agriculture↗

PubMed2004

External carbon feeding strategy for enhancing nitrogen removal in SBR.

Effective method for feeding an external carbon source (ECS) in SBR was investigated to enhance denitrification based on modifying the anoxic/aerobic sub-cycle for swine wastewater treatment. The wastewater discharged from the scraper-type barns contains relatively low readily biodegradable organic. Therefore NOx-N was accumulated during repeating sub-cycle in SBR operation. When acetic acid was fed as ECS during the final sub-cycle, the maximum nitrogen removal rate was 0.22 kg N/m3/d. This was due to both less denitrification rate during the sub-cycle period and inhibition of denitrification by pH drop during the final cycle. The pH drop was caused by a large amount of ECS feeding to remove high concentration of NOx-N in the final period. To overcome these limitations and achieve higher nitrogen removal rate, the intermittent ECS feeding method with raw wastewater at every anoxic period was developed. Using the modifying ECS feeding method, the removal rate was increased to 0.45 kg N/m3/d without NOx-N accumulation.

Agriculture↗