{"title":"The calibration system","authors":"P. Thijssen","doi":"10.1533/9780857099372.70","DOIUrl":"https://doi.org/10.1533/9780857099372.70","url":null,"abstract":"Abstract In order to evaluate unknown samples, it is often necessary to calibrate an instrument by measuring known concentration standards or reference values. A state space model will be derived to estimate online a linear calibration graph involving stochastic drift. The predicted state of the Kalman filter is employed online for the evaluation of unknown samples. After each measurement, a quality control algorithm decides online whether to calibrate again or to evaluate the next unknown. The criterion for quality control is based on the maximal imprecision allowed for the expected evaluated results. The decision to recalibrate is followed by the online optimal selection of the concentration standard next to process. If a time delay is not detrimental, the preset quality can be improved by performing fixed-interval smoothing offline after an entire batch of samples has been measured. The calibration model and the noise variances involved can be selected and verified in practice. In case of full calibration, the system and measurement noise variances can be adapted offline afterward in an iterative way. Finally, nonlinear estimation is applied for a calibration graph with an exponential shape, where the previously described algorithms can be employed again.","PeriodicalId":437309,"journal":{"name":"State Estimation in Chemometrics","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124422075","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"State estimation","authors":"P. Thijssen, S. Hadjiloucas","doi":"10.1016/b978-0-08-102603-8.00003-1","DOIUrl":"https://doi.org/10.1016/b978-0-08-102603-8.00003-1","url":null,"abstract":"","PeriodicalId":437309,"journal":{"name":"State Estimation in Chemometrics","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130058859","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"The multicomponent system","authors":"P. Thijssen","doi":"10.1533/9780857099372.50","DOIUrl":"https://doi.org/10.1533/9780857099372.50","url":null,"abstract":"This chapter presents that in multicomponent analysis, the composition of unknown samples can be estimated from the measurements when the responses of the pure components involved are known a priori. A prerequisite is that the measurement model is linear and additive for the pure components. By recursive least squares or the Kalman filter, the number of components as well as their concentrations in the sample can be evaluated online. Optimization of the experimental design can be performed either offline or online. The performance of the measurements can be checked with respect to the model implemented. The measuring process can be terminated when a preset precision is reached. Concentrations are restricted to nonnegative values and its estimates may be corrected accordingly. The measurement noise variance can be adapted. A valuable extension is the incorporation of stochastic drift in the state space model. A novel application arises for the compensation of unknown impurities or when the pure responses deviate somewhat in the measured sample. The estimated stochastic drift by the Kalman filter can be improved by fixed-interval smoothing afterward. The system noise variances and the measurement noise variance can be adapted offline in an iterative way. The chapter discusses the state space model.","PeriodicalId":437309,"journal":{"name":"State Estimation in Chemometrics","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128865677","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Statistics","authors":"P. Thijssen, S. Hadjiloucas","doi":"10.1016/b978-0-08-102603-8.00004-3","DOIUrl":"https://doi.org/10.1016/b978-0-08-102603-8.00004-3","url":null,"abstract":"","PeriodicalId":437309,"journal":{"name":"State Estimation in Chemometrics","volume":"28 10","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131671874","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"The titration system","authors":"","doi":"10.1533/9780857099372.86","DOIUrl":"https://doi.org/10.1533/9780857099372.86","url":null,"abstract":"This chapter discusses the titration system. In titrimetry, unknown samples are evaluated by adding the titrant of a known concentration. The progress of the chemical reaction is indicated by a change of a suitable property such as a color, absorbance or potential. A state space model is designed to estimate in potentiometry the titration curve and its derivatives. From the estimated state, the set point and the inflection point can be evaluated offline. The discrete titration is investigated, where the titrant is delivered in discrete volume increments and the solution is allowed to equilibrate. The Kalman filter causes a systematic delay in the estimated state and so in the evaluated inflection point or set point, which is completely removed by subsequent fixed-interval smoothing. Implementation of the Kalman filter allows the online control of variable volume additions in order to obtain equidistant measurements. For continuous titrations, the extension of the state space model based upon first-order dynamics is proposed. The Kalman filter allows the online control toward an empirically established endpoint in the titration curve. An attempt to estimate in a nonlinear fashion the exponential factor simultaneously with the titration curve and its derivatives was unsuccessful.","PeriodicalId":437309,"journal":{"name":"State Estimation in Chemometrics","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132061300","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Classical estimation","authors":"P. Thijssen, S. Hadjiloucas","doi":"10.1016/b978-0-08-102603-8.00002-x","DOIUrl":"https://doi.org/10.1016/b978-0-08-102603-8.00002-x","url":null,"abstract":"","PeriodicalId":437309,"journal":{"name":"State Estimation in Chemometrics","volume":"143 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126574693","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Nonlinear estimation","authors":"P. Thijssen, S. Hadjiloucas","doi":"10.1016/b978-0-08-102603-8.00005-5","DOIUrl":"https://doi.org/10.1016/b978-0-08-102603-8.00005-5","url":null,"abstract":"","PeriodicalId":437309,"journal":{"name":"State Estimation in Chemometrics","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121847516","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}