| As one of the extremely important resources on earth, water is the source of living beings and indispensable to modern civilization. The safety and health of water resource determines the development and the future of mankind, and its significance is self-evident.In recent years, water pollution events havefrequently occured in China, in which water pollution from heavy metals accounts for a considerable proportion. People in contaminated areas have suffered remarkable economica loss, and at the same time physically, they also suffered the torture of pain for a long time. In our country, the degree of water pollution is no worse than haze, even deteriorating as a leading cause of “cancer villages”. The protection of water resources and water pollution prevention and tacking is urgent and has become an important task in the current environmental problem relief effort. Concerningthe above-mentioned work, the premise is to detect the water body scientifically and precisely. Consequently, it is of great significance to develop suitable quick and accurate detection and intelligent analysis technology research for heavy metals in water, providing scientific and reliable basis for evaluation and governance of pollution, especially for such a country as China, which is extremely devoid of water resource avalbale.Based on deep study of existing detection technology of heavy metals, and coupled withthe characteristics of electrochemical detection technology, such as being economical and practical of the instruments adopted, as well as the easyof implement of on-line detection, the researcher conducts a series of studies on the detecting system and auto analysis methodstudy of in the respect of the prompt and precise detection of heavy metal elements in water, and designs an electrochemicalheavy metals system which bears the virtues of stable performance and simple manipulation. This system can achieve the purpose of accurate and simultaneous monitoring for heavy metal pollution online, especially focusing on the detection of Sb, Pb, Zn, As, and working together with its superior terminal, eventually builds a suitable monitoring platform,which is of low cost, high efficiency and rapidness, for heavy metals in water.This study belongs to the multi-disciplinary research, involving electrochemical, intelligent detection and control, data processing and analysis, and other relevant fields.The specific research content is as follows:(1) Based on the theory of differential pulse stripping voltammetry, the research develops the heavy metal detection system with three-electrode-sensor system. With excellent MSP430F5438 as MCU, adopting the modular design concept, the research completes module design works respectively including the power module, main control module, water treatment module, electrochemical detection module, communication and keyboard display module, among which, the water processing module simplifies the complicated channel control system via the multi-pass reversing valve and quantitative pump system, thereupon improving the stability of the whole system; at the same time, the design focuses on electrochemical detection module, introducing its excitation source circuit, the potentiostat circuit, I/V converting circuit, signal modulation and detection circuit in detail. In addition, the implementation of system health check and variable error automatic compensation function partially further improves the detection efficiency of the whole system and the stability, when working in field.(2) To obtain the useful weak isolated signals in a complex noisy environment, the software filter is adopted to get rid of the gross error from discrete data and to make processing smoother. After filtering out gross error with the difference method, the research respectively analyzes the methods of the five-point-three-times smoothing, wavelet de-noising, and the adaptive LMS smoothing, and proposes one improved adaptive algorithm by updating the error factor. Subsequently, it also analyzes and compares the results of being smoothed by the above-mentioned methods, and the comparison shows that the improved LMS algorithm has higher precision, stronger anti-interference ability and better astringency.(3) For getting more scientific and correct evaluation results, the study performs mathematical approach for the existing discrete data to get the optimal mathematical expressions of source data, and provides more reliable basis for scientific evaluation. According to the comprehensive analysis of test results, it is deduced that choosing 8th fitting mathematical model in practical testing could meet the design requirements. Regarding the mixture of multi-peak dissolution curve fitting, the study puts forward the idea of piecewise fitting, using the maximum method to segment, and independently to perform fitting processing for each piecewise interval data, which can effectively reduce the data redundancy and ensure the precision of fitting. Due to the fact that the background of solution composition and other factors on the detection may cause drifting, the drift correction is also a main research content. Through the calibration experiment and analysis, the results show that the tangent method with high sensitivity identification and algorithm is easy to implement, and is a kind of optimization algorithm which can realize automatic recognition and analysis.(4) With respect to the problem of overlapping peaks in the analytical chemistry, on the basis of existing testing equipment, with the aid of mathematical theory, employs the computer technology to perform mathematical analysis for the overlapped peak data, and to separate the peak form which the instrument has not been able to completely separate via mathematical decomposition. In this method, the requirement of experimental hardware conditions is not too strict relatively. The papers, based on Gaussian model, puts forwards one method which is a kind of overlapping peak resolution algorithm with nonlinear curve fitting for heavy metals stripping voltammetry curve. Considering the recognition task of this system at present, the algorithm is then optimized and adjusted, by simulation and measurement show that the algorithm is effective.(5) As an auto detection system, it is the necessity to analyze the kinds and concentrations of heavy metals in the test sample automatically. In the process of auto analysis, stripping peak position can be used for qualitative analysis, but for quantitative basis, there are two methods: concentration=f(H)(Peak height) and concentration=f(S)(Peak area). Through the experimental analysis, it is concluded that the regression of H and concentration is better than S, and considering the amount of calculation in addition, H is adopted as the quantitative basis for detecting element concentration.(6) To test the overall performance of the system, such four kinds of heavy metal elements as Sb, Pb, Zn and As, are chosen as the research object. Under this detection system, through a series of experiments such as relationship experiment between the enrichment of the time and peak current, relationship experiment between stripping voltage scanning speed and the peak current and so on, the researcher, after comprehensive analysis, gets the respective best parameter settings for the detection system to determinate the concentration of four elements based on stripping voltammetry. Under the condition of adopting the best parameter settings, the paper further researches the linearization of each element’s concentration and the peak current, carries out the linear regression analysis, and then finds out the standard linear equations of each element separately as follows:;the minimum detection limit(PPB) is respectively: 0.5,1,1,5. The researcher tests the whole heavy metal detection system with the known concentrations of Zn and Pb samples. Through multiple-set result analysis of the test concentrations and the actual concentrations, it is concluded that the average error rate of Zn is 4.15%, the average error rate of Pb is 3.88%, and all have met the requirement of actual accuracy of detection and the system stability is also good. |
PDF Full Text Request