A real-time wastewater quality monitoring system based on multi-parameter sensors
By adaptively optimizing the secondary penalty factor and combining chemical oxygen demand and ammonia nitrogen data, the zero-point drift caused by biofilm is identified and corrected, solving the problems of false detection and false negative detection caused by fixed penalty factors in traditional methods, and improving the accuracy and robustness of wastewater quality monitoring.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHAANXI ZHENGWEI ENVIRONMENTAL TESTING CO LTD
- Filing Date
- 2026-04-24
- Publication Date
- 2026-05-26
AI Technical Summary
In wastewater quality monitoring, existing variational mode decomposition algorithms, with their fixed-size secondary penalty factors, struggle to adapt to the nonlinear growth process of biofilms, leading to nonlinear zero-point drift of the sensor measurement baseline and resulting in false or missed detections.
By calculating the monotonic trend factor, multi-source synchronization coefficient, and drift confidence index, the secondary penalty factor is adaptively optimized, and combined with chemical oxygen demand and ammonia nitrogen data, zero-point drift caused by biofilm is identified and corrected.
It improves the accuracy and anti-interference ability of wastewater quality monitoring, can promptly identify sensor contamination, enhances the robustness of data correction, and ensures the accuracy and stability of monitoring data.
Smart Images
Figure CN122084853A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of wastewater treatment technology. More specifically, this invention relates to a real-time wastewater quality monitoring system based on multi-parameter sensors. Background Technology
[0002] With increased efforts to protect water resources, the regulatory requirements for sewage treatment plants and industrial park discharge outlets are becoming increasingly stringent. Online water quality monitoring has become the core of environmental protection efforts. In practical applications, multi-parameter water quality sensors are widely deployed to obtain continuous data on key indicators such as chemical oxygen demand and ammonia nitrogen.
[0003] However, wastewater environments are characterized by high biological activity, high turbidity, and complex composition. Sensor probes, constantly immersed in microbial-rich water, readily attract bacteria, algae, and organic debris to their optical windows, forming biofilms. Biofilm growth is a non-linear dynamic process, typically involving a slow initial attachment phase, a rapid logarithmic growth phase, and a stable phase of dynamic equilibrium. As the biofilm thickness increases, its obstruction and scattering of the light path gradually intensifies, leading to a non-linear positive zero-point drift in the sensor's measurement baseline. This drift manifests as a slow, monotonous increase in the time-series data, representing a typical soft fault. Traditional solutions rely on periodically initiating mechanical scraping or chemical cleaning. However, mechanical structures are susceptible to wear, and the drift caused by biofilm regeneration persists between cleaning cycles, contributing to errors in the monitoring data. Existing data processing methods often employ variational mode decomposition algorithms to extract these low-frequency trend terms, eliminate zero-point drift, and achieve data correction.
[0004] When existing variational mode decomposition algorithms extract the zero-point drift trend caused by biofilm, the secondary penalty factor is usually set to a fixed value. However, if the fixed secondary penalty factor is set too large, it will result in a narrow bandwidth, which will not be able to capture the rapidly changing zero-point drift during the logarithmic growth period of rapid biofilm reproduction. If it is set too small, it will result in a wide bandwidth, which will easily mistake the low-frequency fluctuations of real water quality for zero-point drift and filter them out. Summary of the Invention
[0005] To address the technical problem of existing variational mode decomposition algorithms using fixed-size secondary penalty factors that are difficult to adapt to actual water conditions, leading to false positives or false negatives, this invention provides a real-time wastewater quality monitoring system based on a multi-parameter sensor. The system includes the following modules: a data acquisition module for acquiring chemical oxygen demand (COD) and ammonia nitrogen (AM) data sequences from wastewater discharge outlets; and a data analysis module for acquiring the monotonicity trend factor of the COD data sequence, wherein the monotonicity trend factor is positively correlated with the ratio of the absolute value of the net change in COD data sequence to the sum of the absolute values of the total changes; and based on the relative change in COD at each sampling point in the COD data sequence and the relative change in AM at each sampling point in the AM nitrogen data sequence... A multi-source synchronization coefficient is obtained for the chemical oxygen demand (COD) data sequence, which is positively correlated with the sum of the products of the relative changes in COD and ammonia nitrogen. Based on the magnitude of the multi-source synchronization coefficient, a multi-source synchronization damping coefficient is obtained for the COD data sequence, which is negatively correlated with the multi-source synchronization coefficient. A drift confidence index for the COD data sequence is obtained based on the monotonicity factor and the multi-source synchronization damping coefficient. An optimized secondary penalty factor for the COD data sequence is obtained based on the drift confidence index. A data correction module is used to decompose the COD data sequence using the optimized secondary penalty factor and to correct and clean the COD data.
[0006] This invention highlights the monotonicity of biofilm drift by calculating a monotonic trend factor; it analyzes the synchronous correlation between chemical oxygen demand (COD) and ammonia nitrogen (AM) data by calculating multi-source synchronization coefficients and multi-source synchronization damping coefficients, thus enabling the identification of biofilm drift; it adaptively optimizes the secondary penalty factor by combining the monotonic trend factor and the multi-source synchronization damping coefficient to calculate the drift confidence index, thereby enhancing the anti-interference capability of the variational mode decomposition algorithm under complex noise; and it achieves the correction and cleaning of COD data by decomposing and correcting based on the optimized secondary penalty factor, enabling the identification of zero-point drift caused by biofilm accumulation and improving the accuracy and robustness of wastewater quality monitoring.
[0007] Preferably, the monotonic trend factor satisfies the expression: In the formula, This is a monotonic trend factor for the chemical oxygen demand (COD) data series. This represents the number of sampling points for the chemical oxygen demand (COD) data sequence. and These represent the first and second data points within the chemical oxygen demand (COD) data sequence. The sampling point and the first The value of each sampling point To prevent tiny constants with a denominator of zero, It is an absolute value.
[0008] This invention achieves the evaluation of the monotonic trend factor by constructing a function that is the ratio of the absolute value of the net change to the sum of the absolute values of the total change. This ratio term reflects the trend consistency of the chemical oxygen demand data sequence. Biofilm drift usually shows a monotonically increasing trend. The calculated monotonic trend factor with a large value provides a reliable trend characteristic basis for the subsequent drift confidence index analysis.
[0009] Preferably, the method for obtaining the relative change of chemical oxygen demand (COD) at each sampling point in the COD data sequence and the relative change of ammonia nitrogen at each sampling point in the ammonia nitrogen data sequence is as follows: the difference between each sampling point in the COD data sequence and its immediately preceding sampling point is recorded as a first value; the ratio between the first value and the range in the COD data sequence is taken as the relative change of COD at each sampling point in the COD data sequence; the difference between each sampling point in the ammonia nitrogen data sequence and its immediately preceding sampling point is recorded as a second value; the ratio between the second value and the range in the ammonia nitrogen data sequence is taken as the relative change of ammonia nitrogen at each sampling point in the ammonia nitrogen data sequence.
[0010] Preferably, the multi-source synchronization coefficients satisfy the expression: In the formula, This refers to the multi-source synchronization coefficient for the chemical oxygen demand (COD) data sequence. This represents the number of sampling points for the chemical oxygen demand (COD) data sequence. The number of sampling points for the ammonia nitrogen (NHD) data sequence is the same as that for the COD data sequence. The first in the chemical oxygen demand data sequence The relative change in chemical oxygen demand at each sampling point The first in the ammonia nitrogen sequence The relative change in ammonia nitrogen at each sampling point.
[0011] This invention achieves the evaluation of multi-source synchronization coefficient by constructing a function of the sum of the products of the relative changes in chemical oxygen demand (COD) and ammonia nitrogen (MN). This product term reflects the coordinated changes of multiple parameter data. The actual discharge time usually causes COD and MN to fluctuate synchronously, resulting in a larger multi-source synchronization coefficient. In contrast, the zero-point drift time has a smaller value due to the lack of directional correlation, thus providing a reliable basis for the subsequent analysis of the multi-source synchronization damping coefficient.
[0012] Preferably, the multi-source synchronization damping coefficient satisfies the following expression: In the formula, The multi-source synchronicity damping coefficient for the chemical oxygen demand (COD) data sequence. This refers to the multi-source synchronization coefficient for the chemical oxygen demand (COD) data sequence. To obtain a larger value, It is a natural exponential function.
[0013] Preferably, the drift confidence index satisfies the expression: In the formula, The drift confidence index for the chemical oxygen demand (COD) data series. This is a monotonic trend factor for the chemical oxygen demand (COD) data series. is the multi-source synchronicity damping coefficient of the chemical oxygen demand (COD) data sequence.
[0014] This invention constructs a composite function that multiplies a monotonic trend factor and a multi-source synchronicity damping coefficient to evaluate the drift confidence index. The monotonic trend factor term amplifies the trend consistency response of biofilm drift, while the multi-source synchronicity damping coefficient term amplifies the multi-source data non-correlation response of biofilm drift. This results in a larger drift confidence index being calculated for the chemical oxygen demand data sequence affected by biofilm drift events, thus providing a reliable basis for the subsequent optimization of the secondary penalty factor.
[0015] Preferably, the optimized quadratic penalty factor satisfies the expression: In the formula, The secondary penalty factor is used to optimize the chemical oxygen demand (COD) data sequence. This is the lower limit of the quadratic penalty factor. This is the upper limit of the secondary penalty factor. The drift confidence index for the chemical oxygen demand (COD) data series. It is a natural exponential function.
[0016] This invention achieves adaptive optimization of the quadratic penalty factor by constructing a function with the drift confidence index as the linear interpolation coefficient. This index term reflects the confidence of biofilm drift. When the drift confidence index is high, the quadratic penalty factor approaches the lower limit to broaden the frequency band and capture fast drift. When the drift confidence index is low, the quadratic penalty factor approaches the upper limit to narrow the frequency band and suppress noise, thereby effectively balancing the sensitivity and stability of zero-point drift trend extraction.
[0017] Preferably, the correction of the chemical oxygen demand (COD) data includes: identifying the modal component with the smallest center frequency after decomposing the COD data sequence as the biofilm drift component, and subtracting the biofilm drift component from the COD data sequence to obtain the corrected COD data.
[0018] Preferably, the step of obtaining the chemical oxygen demand (COD) data sequence and ammonia nitrogen data sequence of the sewage discharge outlet includes: obtaining COD data of the sewage discharge outlet using a COD sensor, obtaining ammonia nitrogen data of the sewage discharge outlet using an ammonia nitrogen sensor, and segmenting the COD data and ammonia nitrogen data using a sliding window technique to obtain the COD data sequence and ammonia nitrogen data sequence of the sewage discharge outlet.
[0019] Preferably, the system is further configured to: trigger a physical cleaning warning and send it to relevant personnel when the terminal amplitude of the biofilm drift component is greater than a determination threshold.
[0020] The beneficial effects of this invention are as follows: By introducing an adaptive quadratic penalty factor optimization mechanism based on multi-parameter fusion, this invention solves the technical problem that traditional variational mode decomposition algorithms are difficult to adapt to actual water conditions in wastewater quality monitoring, leading to false detections or missed detections.
[0021] This invention establishes the intrinsic mapping relationship between monotonic trend factors and biofilm drift by analyzing the numerical variation characteristics of chemical oxygen demand (COD) data sequences. Based on this, it further integrates multi-source synchronization coefficient and multi-source synchronization damping coefficient, coupling the indicators reflecting trend consistency with the indicators reflecting parameter synergy, thereby realizing the identification of biofilm drift.
[0022] This invention can calculate a secondary penalty factor matching the drift confidence index for each chemical oxygen demand (COD) data sequence. Through adaptive adjustment of the secondary penalty factor and frequency band, it achieves accurate matching between decomposition and noise interference. For COD data sequences affected by biofilm drift events, the frequency band is broadened to capture trends. Ultimately, this invention improves the accuracy and anti-interference capability of wastewater quality monitoring data, enhances the robustness of data correction, and enables timely and accurate identification of sensor contamination, providing a solid technical guarantee for the optimization of wastewater treatment system operation and environmental pollution prevention and control. Attached Figure Description
[0023] Figure 1 This is a schematic diagram illustrating a real-time wastewater quality monitoring system based on a multi-parameter sensor according to the present invention. Figure 2 A comparison chart showing the correction effects of existing technology using a fixed secondary penalty factor and the present invention using an optimized secondary penalty factor on chemical oxygen demand data sequences. Detailed Implementation
[0024] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0025] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.
[0026] This invention provides a real-time wastewater quality monitoring system based on multi-parameter sensors. For example... Figure 1 As shown, a real-time wastewater quality monitoring system based on multi-parameter sensors includes a data acquisition module 100, a data analysis module 200, and a data correction module 300, which are described in detail below.
[0027] The data acquisition module 100 is used to acquire chemical oxygen demand (COD) data sequences and ammonia nitrogen data sequences from the wastewater discharge outlet.
[0028] Specifically, a chemical oxygen demand (COD) sensor is used to acquire COD data at the wastewater discharge outlet, and an ammonia nitrogen sensor is used to acquire ammonia nitrogen data at the wastewater discharge outlet within the same time period. A sliding window technique is used to segment the COD and ammonia nitrogen data to obtain COD and ammonia nitrogen data sequences at the wastewater discharge outlet. In this embodiment, the acquisition frequency of both COD and ammonia nitrogen data is set to once per minute, and the length of the sliding window is set to 1500. In other embodiments, implementers can set the acquisition frequency of COD and ammonia nitrogen data and the length of the sliding window according to the actual implementation situation.
[0029] The data analysis module is used to obtain the monotonicity trend factor of the chemical oxygen demand (COD) data sequence; to obtain the multi-source synchronization coefficient of the COD data sequence based on the relative change of COD at each sampling point and the relative change of ammonia nitrogen at each sampling point in the ammonia nitrogen data sequence; to obtain the multi-source synchronization damping coefficient of the COD data sequence based on the magnitude of the multi-source synchronization coefficient; to obtain the drift confidence index of the COD data sequence based on the monotonicity trend factor and the multi-source synchronization damping coefficient; and to obtain the optimized secondary penalty factor of the COD data sequence based on the drift confidence index.
[0030] It should be noted that existing variational mode decomposition algorithms typically use fixed values for the secondary penalty factor, making it difficult to adapt to and match the differentiated frequency band characteristics exhibited by biofilms at different growth stages. This leads to over- or under-decomposition when extracting the zero-point drift trend caused by biofilms, resulting in a decrease in the correction accuracy of wastewater quality monitoring data. Based on the physical mechanism that the nonlinear cumulative growth process of biofilms on the surface of optical sensors forces chemical oxygen demand (COD) data to exhibit a significant and continuous unidirectional drift over time, this invention determines a COD monotonicity trend factor to characterize the intensity of monotonic changes within COD data segments.
[0031] Specifically, the monotonic trend factor satisfies the expression: In the formula, This is a monotonic trend factor for the chemical oxygen demand (COD) data series. This represents the number of sampling points for the chemical oxygen demand (COD) data sequence. and These represent the first and second data points within the chemical oxygen demand (COD) data sequence. The sampling point and the first The value of each sampling point To prevent tiny constants with a denominator of zero, It is an absolute value.
[0032] In the formula, The larger the value, the stronger the monotonicity of the chemical oxygen demand (COD) data series, indicating a greater likelihood that the COD data series is affected by zero-point drift. Therefore, the larger the monotonicity trend factor of the COD data series, the stronger the monotonicity trend factor.
[0033] It should be noted that, due to the high isomorphism in the univariate temporal morphology of the sustained increase in organic matter concentration caused by real sewage discharge events and the light path obstruction caused by biofilm accumulation, both types of events exhibit significant monotonic responses, making it difficult to accurately pinpoint the biofilm drift component while preserving the real sewage discharge signal. Based on the multi-parameter correlation mechanism of water quality, real sewage discharge events typically involve a systemic, coordinated deterioration of all indicators, resulting in a high degree of gradient synchronicity between chemical oxygen demand (COD) and ammonia nitrogen (AM) data over time. However, biofilm growth, as a specific physical adhesion phenomenon of optical sensors, only independently affects the COD monitoring optical path and does not cause synchronous changes in AM data. Therefore, this invention determines a multi-source synchronization coefficient, a multi-source synchronicity damping coefficient, and a drift confidence index to characterize the reliability of the impact of biofilm zero-point drift events on the COD data sequence.
[0034] Specifically, the difference between each sampling point in the chemical oxygen demand (COD) data sequence and its immediately preceding sampling point is calculated to determine the first value. The ratio between the first value and the range in the COD data sequence is calculated to determine the relative change in COD at each sampling point in the COD data sequence. The difference between each sampling point in the ammonia nitrogen (AM) data sequence and its immediately preceding sampling point is calculated to determine the second value. The ratio between the second value and the range in the ammonia nitrogen (AM) data sequence is calculated to determine the relative change in ammonia nitrogen (AM) at each sampling point in the ammonia nitrogen (AM) data sequence.
[0035] Specifically, the multi-source synchronization coefficients satisfy the expression: In the formula, This refers to the multi-source synchronization coefficient for the chemical oxygen demand (COD) data sequence. This represents the number of sampling points for the chemical oxygen demand (COD) data sequence. The number of sampling points for the ammonia nitrogen (NHD) data sequence is the same as that for the COD data sequence. The first in the chemical oxygen demand data sequence The relative change in chemical oxygen demand at each sampling point The first in the ammonia nitrogen sequence The relative change in ammonia nitrogen at each sampling point.
[0036] In the formula, The larger the value, the stronger the positive correlation between the chemical oxygen demand (COD) data sequence and the ammonia nitrogen (AM) data sequence in their respective numerical change directions. Therefore, the larger the multi-source synchronization coefficient of the COD data sequence, the greater the correlation.
[0037] Specifically, the multi-source synchronicity damping coefficient satisfies the following expression: In the formula, The multi-source synchronicity damping coefficient for the chemical oxygen demand (COD) data sequence. This refers to the multi-source synchronization coefficient for the chemical oxygen demand (COD) data sequence. To obtain a larger value, It is a natural exponential function.
[0038] In the formula, The larger the value, the greater the likelihood that the chemical oxygen demand (COD) data sequence is affected by real pollution events and the less likely it is to be affected by biofilm zero-point drift events. Therefore, the larger the multi-source synchronicity damping coefficient of the COD data sequence is.
[0039] Specifically, the drift confidence index satisfies the following expression: In the formula, The drift confidence index for the chemical oxygen demand (COD) data series. This is a monotonic trend factor for the chemical oxygen demand (COD) data series. is the multi-source synchronicity damping coefficient of the chemical oxygen demand (COD) data sequence.
[0040] In the formula, The larger the value, the stronger the monotonicity of the chemical oxygen demand (COD) data series, and the greater the possibility that the COD data series is affected by zero-point drift. Therefore, the larger the drift confidence index of the COD data series, the greater the drift confidence index. The larger the value, the weaker the positive correlation between the chemical oxygen demand (COD) data series and the ammonia nitrogen (AM) data series in their respective directions of change. This indicates that the COD data series is less likely to be affected by real pollution discharge events and more likely to be affected by biofilm zero-point drift events. Therefore, the larger the drift confidence index of the COD data series, the stronger the correlation.
[0041] It should be noted that biofilm growth is a nonlinear dynamic process, typically going through a slow initial attachment phase, a rapid logarithmic growth phase, and a stable phase of dynamic equilibrium. During the slow phase, the drift is extremely slow, while during the logarithmic growth phase, the drift is relatively fast. Existing variational mode decomposition algorithms usually use a fixed secondary penalty factor, which cannot adapt to this dynamic process and the interference of real water quality fluctuations. Therefore, this invention analyzes the magnitude of the biofilm drift confidence index and dynamically adjusts the secondary penalty factor to adapt to different drift stages.
[0042] Specifically, the optimized quadratic penalty factor satisfies the expression: In the formula, The secondary penalty factor is used to optimize the chemical oxygen demand (COD) data sequence. This is the lower limit of the quadratic penalty factor. This is the upper limit of the secondary penalty factor. The drift confidence index for the chemical oxygen demand (COD) data series. The quadratic penalty factor is a natural exponential function. In this embodiment, the lower limit of the quadratic penalty factor is set to 500. In other embodiments, implementers can set it according to the actual implementation situation. For example, when the biofilm in the wastewater environment is in a stage of explosive growth, resulting in an extremely fast drift rate, the lower limit of the quadratic penalty factor can be appropriately reduced to broaden the bandwidth limitation, thereby improving the algorithm's ability to track rapidly changing trend terms. When the high-frequency random noise mixed in the sensor signal is extremely strong, the lower limit of the quadratic penalty factor can be appropriately increased to prevent the noise from being incorrectly extracted as drift components due to an excessively wide bandwidth. In this embodiment, the upper limit of the quadratic penalty factor is set to 5000. In other embodiments, implementers can set it according to the actual implementation situation. For example, when the baseline stability of the chemical oxygen demand data sequence is required to be high and strong suppression of small fluctuations in water quality is needed, the upper limit of the quadratic penalty factor can be appropriately increased to forcibly narrow the modal bandwidth and enhance the shielding effect against background noise. When it is necessary to improve the sensitivity to weak drift signals in the initial slow phase of biofilm, the upper limit of the quadratic penalty factor can be appropriately reduced to avoid excessive smoothing of modal components due to an excessively large quadratic penalty factor.
[0043] In the formula, The larger the value, the greater the reliability of the chemical oxygen demand (COD) data sequence being affected by the biofilm zero-point drift event. The closer the optimized COD data sequence's secondary penalty factor is to the lower limit of the secondary penalty factor, the larger the bandwidth corresponding to the COD data sequence, the more accurately the drift trend can be extracted, and under-decomposition can be avoided. The smaller the value, the less reliable the chemical oxygen demand (COD) data sequence is affected by the biofilm zero-point drift event. The closer the optimized COD data sequence's secondary penalty factor is to the upper limit of the secondary penalty factor, ensuring that the bandwidth corresponding to the COD data sequence is smaller and avoiding the mis-extraction of real water quality fluctuations.
[0044] The data correction module is used to decompose the chemical oxygen demand (COD) data sequence using an optimized secondary penalty factor, and to correct and clean the COD data.
[0045] Specifically, the chemical oxygen demand (COD) data is corrected and cleaned, including: decomposing the COD data sequence using an optimized secondary penalty factor; identifying the modal component with the smallest center frequency after decomposition as the biofilm drift component; subtracting the biofilm drift component from the COD data sequence to obtain the corrected COD data; using the value at the last sampling point of the biofilm drift component as the terminal amplitude of the biofilm drift component; triggering a physical cleaning warning and sending it to relevant personnel when the terminal amplitude of the biofilm drift component exceeds a judgment threshold. In this embodiment, the judgment threshold is set to 1. In other embodiments, the implementer can set it according to the actual implementation situation. For example, when the accuracy requirement of water quality monitoring data is extremely high and measurement errors need to be strictly controlled, the judgment threshold can be appropriately reduced to improve the timeliness of sensor maintenance and ensure data quality. When the tolerance for slight drift in the COD data sequence is high, the judgment threshold can be appropriately increased to reduce the trigger frequency of physical cleaning and thus reduce the workload of operation and maintenance.
[0046] like Figure 2As shown in the figure, this diagram compares the correction effects of the existing technology using a fixed secondary penalty factor and the present invention using an optimized secondary penalty factor on the chemical oxygen demand (COD) data sequence. Within the time series range of 50 to 200, for complex signals including biofilm drift and sudden pollution events, the significant difference in data correction effects between the present invention and the existing technology is evident. The horizontal axis represents the sampling time series, and the vertical axis represents the COD value. The figure covers two key monitoring scenarios: First, within the sequence range of 50 to 120, the zero-point drift phenomenon caused by biofilm contamination of the sensor is observed. It can be seen that the corresponding curve of the COD data sequence exhibits a clear monotonic upward trend. In this stage, although the corresponding curve of the existing technology decreases somewhat, it still retains some of the upward trend. The upward trend indicates that the drift component was not completely eliminated. In contrast, the curve of the present invention maintains a relatively stable baseline, demonstrating that the present invention effectively removes the drift trend. Secondly, in the interval between sequences 120 and 160, a discharge event occurred, resulting in a high peak in the corresponding curve of the chemical oxygen demand (COD) data sequence. The fluctuation of the curve in the prior art is significantly weaker than that of the COD data sequence, exhibiting over-decomposition and distorting the crucial discharge signal. The fluctuation of the curve in the present invention closely matches that of the COD data sequence, maintaining a constant vertical distance between them. This distance represents the amount of background drift removed, thus more completely replicating the relative amplitude of the peak in the COD data sequence. This is thanks to the present invention's adaptive increase of the secondary penalty factor to narrow the frequency band, effectively avoiding the false deletion of high-frequency discharge signals. In summary, the curve of the present invention exhibits excellent adaptive capability in both scenarios, proving that the present invention can completely retain true water quality anomaly information while filtering out biofilm interference, effectively solving the technical problems faced by traditional methods.
[0047] While this specification has shown and described numerous embodiments of the invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will occur to those skilled in the art without departing from the spirit and essence of the invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in the practice of this invention.
Claims
1. A real-time wastewater quality monitoring system based on multi-parameter sensors, characterized in that, include: The data acquisition module is used to acquire chemical oxygen demand (COD) and ammonia nitrogen (AM) data sequences from wastewater discharge outlets. The data analysis module is used to obtain the monotonic trend factor of the chemical oxygen demand (COD) data sequence. The monotonic trend factor is positively correlated with the ratio of the absolute value of the net change in the COD data sequence to the sum of the absolute values of the total change. Based on the relative change of chemical oxygen demand at each sampling point in the chemical oxygen demand data sequence and the relative change of ammonia nitrogen at each sampling point in the ammonia nitrogen data sequence, a multi-source synchronization coefficient of the chemical oxygen demand data sequence is obtained. The multi-source synchronization coefficient is positively correlated with the sum of the products of the relative changes of chemical oxygen demand and the relative changes of ammonia nitrogen. Based on the magnitude of the multi-source synchronization coefficient, the multi-source synchronization damping coefficient of the chemical oxygen demand data sequence is obtained, and the multi-source synchronization damping coefficient is negatively correlated with the multi-source synchronization coefficient. Based on the monotonic trend factor and the multi-source synchronicity damping coefficient, the drift confidence index of the chemical oxygen demand (COD) data sequence is obtained; based on the drift confidence index, the optimized secondary penalty factor of the COD data sequence is obtained. The data correction module is used to decompose the chemical oxygen demand (COD) data sequence using the optimized secondary penalty factor, and to correct and clean the COD data.
2. The wastewater quality real-time monitoring system based on a multi-parameter sensor according to claim 1, characterized in that, The monotonic trend factor satisfies the expression: In the formula, This is a monotonic trend factor for the chemical oxygen demand (COD) data series. This represents the number of sampling points for the chemical oxygen demand (COD) data sequence. and These represent the first and second data points within the chemical oxygen demand (COD) data sequence. The sampling point and the first The value of each sampling point To prevent tiny constants with a denominator of zero, It is an absolute value.
3. The wastewater quality real-time monitoring system based on a multi-parameter sensor according to claim 1, characterized in that, The method for obtaining the relative change of chemical oxygen demand (COD) at each sampling point in the COD data sequence and the relative change of ammonia nitrogen at each sampling point in the ammonia nitrogen data sequence is as follows: the difference between each sampling point in the COD data sequence and its immediately preceding sampling point is recorded as the first value; the ratio between the first value and the range in the COD data sequence is taken as the relative change of COD at each sampling point in the COD data sequence; the difference between each sampling point in the ammonia nitrogen data sequence and its immediately preceding sampling point is recorded as the second value; the ratio between the second value and the range in the ammonia nitrogen data sequence is taken as the relative change of ammonia nitrogen at each sampling point in the ammonia nitrogen data sequence.
4. The wastewater quality real-time monitoring system based on a multi-parameter sensor according to claim 1, characterized in that, The multi-source synchronization coefficients satisfy the expression: In the formula, This refers to the multi-source synchronization coefficient for the chemical oxygen demand (COD) data sequence. This represents the number of sampling points for the chemical oxygen demand (COD) data sequence. The number of sampling points for the ammonia nitrogen (NHD) data sequence is the same as that for the COD data sequence. The first in the chemical oxygen demand data sequence The relative change in chemical oxygen demand at each sampling point The first in the ammonia nitrogen sequence The relative change in ammonia nitrogen at each sampling point.
5. The wastewater quality real-time monitoring system based on a multi-parameter sensor according to claim 1, characterized in that, The multi-source synchronous damping coefficient satisfies the following expression: In the formula, The multi-source synchronicity damping coefficient for the chemical oxygen demand (COD) data sequence. This refers to the multi-source synchronization coefficient for the chemical oxygen demand (COD) data sequence. To obtain a larger value, It is a natural exponential function.
6. The wastewater quality real-time monitoring system based on a multi-parameter sensor according to claim 1, characterized in that, The drift confidence index satisfies the expression: In the formula, The drift confidence index for the chemical oxygen demand (COD) data series. This is a monotonic trend factor for the chemical oxygen demand (COD) data series. is the multi-source synchronicity damping coefficient of the chemical oxygen demand (COD) data sequence.
7. A real-time wastewater quality monitoring system based on a multi-parameter sensor according to claim 1, characterized in that, The optimized quadratic penalty factor satisfies the expression: In the formula, The secondary penalty factor is used to optimize the chemical oxygen demand (COD) data sequence. This is the lower limit of the quadratic penalty factor. This is the upper limit of the secondary penalty factor. The drift confidence index for the chemical oxygen demand (COD) data series. It is a natural exponential function.
8. A real-time wastewater quality monitoring system based on a multi-parameter sensor according to claim 1, characterized in that, The correction of the chemical oxygen demand (COD) data includes: identifying the modal component with the smallest center frequency after decomposing the COD data sequence as the biofilm drift component, and subtracting the biofilm drift component from the COD data sequence to obtain the corrected COD data.
9. A real-time wastewater quality monitoring system based on a multi-parameter sensor according to claim 1, characterized in that, The process of obtaining the chemical oxygen demand (COD) data sequence and ammonia nitrogen data sequence of the wastewater discharge outlet includes: using a COD sensor to obtain COD data of the wastewater discharge outlet, using an ammonia nitrogen sensor to obtain ammonia nitrogen data of the wastewater discharge outlet, and using a sliding window technique to segment the COD data and ammonia nitrogen data to obtain the COD data sequence and ammonia nitrogen data sequence of the wastewater discharge outlet.
10. A real-time wastewater quality monitoring system based on a multi-parameter sensor according to claim 1, characterized in that, The system is also used to: trigger a physical cleaning warning and send it to relevant personnel when the terminal amplitude of the biofilm drift component is greater than the judgment threshold.