Method for predicting biological slime growth based on circulating water quality characteristic parameters driven by data

Abstract

The invention discloses a method for predicting the growth state of biological slime based on circulating water quality parameters driven by data. Dynamic correlation analysis is carried out on water quality characteristic parameters and biological slime growth, internal correlation between water quality variables is excavated, and the relation between key water quality indexes and biological slime growth and development is captured, so that biological slime formation is accurately predicted in time. Based on data-driven circulating water key water quality index and biological slime growth prediction research, the defect and limitation that a pipe needs to be opened to measure biological slime are avoided, the technical problem that a process quantification model is imperfect is effectively solved, a new thought normal form is provided for achieving early warning of biological slime growth through circulating water quality diagnosis, and the method is suitable for popularization and application. And automatic control and intelligent operation of the circulating water system are facilitated.

Description

Technical Field

[0001] This invention relates to the field of circulating water system monitoring technology, and in particular to a method for predicting the growth of biological slime through changes in circulating water quality conditions. Background Technology

[0002] With industrial and urban development, the demand for water resources is increasing, while water pollution and waste are also becoming increasingly serious. How to rationally utilize water resources and improve their efficiency has become a crucial issue for current social development. Circulating water systems are an efficient and environmentally friendly method of water resource utilization, widely used in industry, construction, and agriculture. Through circulating water systems, wastewater is treated and reused, reducing the consumption of fresh water and the discharge of wastewater. In industrial circulating water systems, open-loop systems are typically used. Due to heat exchange and sufficient contact with the external environment, these systems provide a suitable living environment (such as dissolved oxygen) and sufficient nutrients for microbial growth. However, under conditions of improper operation, untimely maintenance, and large fluctuations in water quality, the aggregation of large numbers of microorganisms can easily produce rubbery deposits, viscous substances, gelatinous films, or colored slime clumps, i.e., biological slime. Slime adheres extremely quickly, with the amount adhering within a month reaching several millimeters. When biofilm forms in large quantities, it slows down the flow rate of circulating water, increases the pressure on water pumps, and causes pipe blockage. Biofilm also adheres to the pipe walls of circulating water equipment, leading to poor heat transfer performance and increased heat transfer resistance. Furthermore, as a significant component of fouling, biofilm affects the electrochemical properties of the metal surfaces of circulating water system pipes, exacerbating scaling and corrosion, and increasing production and maintenance costs for enterprises. Therefore, timely and accurate feedback on the growth dynamics of biofilm is a primary prerequisite for ensuring the stable and safe operation of industrial circulating water systems.

[0003] Currently, the amount of biological slime in industrial circulating water systems is manually detected using a biofilter method. This involves first collecting a certain volume of water, passing it through a filter, then rinsing the filter to flush the sludge into a biological slime collection tank, and finally manually measuring the volume of biological slime in the circulating water. While the manual detection technique is relatively mature, it is cumbersome, requires specialized personnel, further increasing labor costs, and sometimes necessitates opening the circulating water transport pipeline, disrupting normal production processes. In fact, the frequency of biological slime quantity testing in industrial circulating water systems is low, and the data is delayed, making it unsuitable as a timely basis for judging the overall growth status of biological slime in industrial circulating water systems. Furthermore, the manual detection method using a biofilter in circulating water places high demands on sample flow rate, velocity, and water volume, and subjective factors significantly influence the test results. If the detection is untimely or inaccurate, leading to excessive biological slime in the circulating water system, it can cause irreparable production safety hazards. Therefore, how to solve the problems of cumbersome operation, low accuracy, and significant human influence on the detection results of biological slime quantity is an urgent issue that needs to be addressed by those skilled in the art. Summary of the Invention

[0004] To address the problems of existing technologies, this invention proposes a data-driven method for predicting biofilm growth based on circulating water quality characteristic parameters. This method predicts the growth and development status of biofilm by correlating circulating water quality characteristic parameters with changes in biofilm thickness.

[0005] The technical solution adopted in this invention is as follows: A data-driven method for predicting biofilm growth based on circulating water quality characteristic parameters includes the following steps: (1) Monitoring specific water quality environmental factors in circulating water: measuring ammonium nitrogen (NH4) in pipeline water + -N), nitrite nitrogen (NO2) - -N), nitrate nitrogen (NO3) - -N), reactive phosphate (PO4) 3- Changes in P and total organic carbon (TOC) content; measurements of dissolved oxygen (DO), pH, conductivity, and UV in pipeline water. 254 Changes in absorbance, turbidity, and suspended solids (SS) under 254nm ultraviolet light; changes in the concentrations of calcium (Ca), magnesium (Mg), sodium (Na), and potassium (K) metal ions in pipeline water;

[0006] Based on the above technical solutions, further methods were employed to determine NH4 in the pipeline water, including salicylic acid spectrophotometry, N-(1-naphthyl)-1,2-diaminoethane dihydrochloride spectrophotometry, aminosulfonic acid-UV spectrophotometry, and ammonium molybdate spectrophotometry. + -N, NO2 - -N, NO3 - -N and PO4 3- Changes in -P content; changes in TOC content in pipeline water were determined using a total organic carbon analyzer.

[0007] Based on the above technical solutions, further steps can be taken to detect NH4. + -N, NO2 - -N, NO3 - -N,PO4 3- Before determining the P and TOC content, the water sample was filtered through a 0.45 µm filter membrane.

[0008] Based on the above technical solutions, further, a digital portable dissolved oxygen meter, pH meter, conductivity meter, turbidimeter, and UV spectrophotometer were used to measure DO, pH, conductivity, turbidity, and UV in the pipeline water. 254 Changes in values; changes in SS content in pipeline water were determined using the gravimetric method.

[0009] Based on the above technical solutions, further detection of DO, pH, conductivity, turbocharger, and UV is performed. 254 Before determining the SS content, the water sample requires no further treatment and can be directly tested.

[0010] Based on the above technical solutions, inductively coupled plasma atomic emission spectrometry (ICP-AES) was further used to determine the changes in the concentrations of Ca, Mg, Na, and K metal ions in the pipeline water.

[0011] Based on the above technical solution, further, before detecting the concentration of Ca, Mg, Na, and K metal ions, the water sample is first filtered with a 0.22 µm filter membrane, and then acidified with nitric acid to 0.2%~5% w / w (preferably 1% w / w) HNO3.

[0012] Based on the above technical solutions, the filter membrane is further selected from polyethersulfone filter membrane, mixed cellulose ester membrane, or hydrophilic polyvinylidene fluoride membrane.

[0013] Based on the above technical solutions, the detection interval for environmental factors of pipeline water quality is further set at 2 days.

[0014] Based on the above technical solutions, further, by simulating circulating water pipeline loop experiments, water quality environmental factors in the circulating water are monitored.

[0015] Based on the above technical solution, the operating conditions for the pipeline loop experiment are as follows: 1) The circulating water is taken from surface water, groundwater, seawater, or recycled water; 2) The pipe material is polyvinyl chloride, carbon steel, stainless steel, galvanized steel, or cast iron; 3) The temperature of the circulating water is maintained at 27~30℃; 4) The flow velocity of the circulating water is maintained at 1.4 ~ 1.5 m / s; 5) The pipeline loop test adopts a continuous operation mode with an operation cycle of 135 to 140 days.

[0016] Based on the above technical solutions, the circulating water continues to operate in the pipeline.

[0017] (2) Periodically measure the thickness of the biofilm on the pipe wall: use scanning electron microscopy to visualize the outline of the growth and development of biofilm; use image analysis software to measure the thickness of the biofilm layer.

[0018] Based on the above technical solutions, the further image analysis software is Image-Pro Plus image analysis software.

[0019] Based on the above technical solutions, the measurement time interval for the thickness of the bio-slime is further 30 to 35 days.

[0020] (3) Spearman correlation analysis and Mantel correlation analysis were used to correlate water quality environmental factors with changes in the thickness of biological slime layer, and water quality environmental factors that are correlated with the growth and development of biological slime at different stages were identified.

[0021] Based on the above technical solutions, Spearman correlation analysis and Mantel correlation analysis are existing technologies in this field. Both are implemented using R language and Rstudio. By installing the necessary R packages (such as dplyr, linkKEt, ggplot2) and running the relevant programming code, the above analysis can be further realized.

[0022] Based on the above technical solutions, the growth and development stages of bio-slime further include the initial stage of bio-slime formation, the stage of bio-slime accumulation and thickening, the stage of bio-slime maturity and stabilization, and the stage of bio-slime aging.

[0023] Based on the above technical solutions, further analysis shows that in the early stage of biofilm formation (0-34 days), TOC and DO are key water quality environmental factors; in the biofilm accumulation and thickening stage (35-86 days), DO, metal ions (Ca, Mg, K, Na), SS, and NO2 are also important factors. - -N, NO3 --N and TN are key water quality environmental factors; during the maturation and stabilization stage of biological slime (87-113 days), their correlation with various water quality environmental factors is generally low; during the aging stage of biological slime (114-140 days), DO, pH, TOC, metal ions (Ca, Mg, K, Na), SS, and NH4+ are significantly higher. + -N, NO2 - -N, NO3 - -N and TN are key water quality environmental factors.

[0024] (4) An ensemble learning model based on random forest was used to evaluate the ranking importance of water quality environmental factors on the growth and development of biofilm in the circulating water system. The system quantifies the overall impact of different water quality environmental factors on biofilm accumulation and identifies key driving factors. By comparing the ranking of the importance of water quality environmental factors at different stages, the dynamic evolution of the contribution of different water quality environmental factors was elucidated, and the changes in the influencing mechanism with the growth stage of biofilm were revealed. The integrated data of all growth stages and the data of individual independent stages were used as the training set and the test set, respectively. The predictive performance of the model on the test set was observed (R²). 2 The stability and generalization of the prediction framework are comprehensively evaluated to assess its accuracy, robustness, and universality.

[0025] Based on the above technical solutions, the Random Forest Ensemble Learning Model is a prior art in this field. It is implemented using R language and Rstudio. The above analysis can be further realized by installing the necessary R packages (such as randomForest, party, caret) and running the relevant programming code.

[0026] Based on the above technical solutions, further research shows that changes in different water quality environmental factors (especially Conductivity, TN, and DO) have significant predictive importance for the growth and development of biofilm. During the initial stage of biofilm formation (0–34 days) and the subsequent accumulation and thickening stage (35–86 days), Turbidity, SS, TOC, and PO4 levels are significantly influenced by biofilm formation. 3- Changes in pH are relatively important water quality factors for predicting the growth and development of biofilm; while changes in DO, pH, and metal ions (Ca, Mg, K, Na) are relatively important water quality factors for predicting the entire process of biofilm growth and development; the dynamic changes of water quality factors can accurately predict (R... 2 The accuracy of the prediction for the growth and development of biogenic slime was 0.974; although the prediction accuracy for individual biogenic slime growth and development decreased, the overall accuracy remained relatively stable (R = 0.974). 2 The predictive performance is 0.890 ~ ​​0.994. In this invention, by long-term monitoring of changes in specific water quality environmental factors and biofilm thickness in pipeline water, Spearman correlation analysis and Mantel correlation analysis are used to correlate these changes, identifying water quality characteristic parameters highly correlated with different growth and development stages of biofilm. Furthermore, the accuracy, robustness, and universality of the water quality environmental factor monitoring and prediction framework for biofilm growth are evaluated using a random forest machine learning model. The dynamic changes of these key water quality parameters over time can serve as indicative features for predicting the initial formation and subsequent accumulation of biofilm. Predicting biofilm growth through changes in the most direct and simplest pipeline water quality parameters avoids the need for pipe-opening measurements of biofilm, effectively solving the cumbersome steps of manual biofilm quantity detection, saving manpower and resources, and reducing subjective errors caused by manual detection.

[0027] By applying this invention to existing online water quality monitoring instruments on relevant platforms or by adding a few key sensors, the system can analyze water quality data in real time, predicting the risk of accelerated growth of biological slime in pipelines days or even weeks in advance and issuing tiered warnings to ensure the long-term operational safety of critical equipment. Furthermore, this invention can also be applied to intelligent early warning and optimization of the system. Deploying a predictive system at key system nodes allows for real-time optimization of pretreatment processes (such as filtration accuracy and deoxygenation efficiency) and bactericide injection strategies by dynamically predicting slime growth trends, ensuring that circulating water quality is controlled from the source. Therefore, this invention facilitates transforming post-event treatment into pre-event warning and experience-based judgment into data-driven decision-making. Through data simulation, it uncovers early characteristics hidden in the dynamic changes of water quality parameters (such as dissolved oxygen, turbidity, and ion concentration), achieving accurate and advanced prediction of biological slime growth trends. In summary, this invention breaks away from the lag of traditional methods that rely on periodic disassembly and inspection, manual observation, or single-indicator alarms. It helps to innovate the operation and maintenance management paradigm of circulating water systems, provides a new approach and paradigm for early warning of biological slime growth through circulating water quality diagnosis, and facilitates the automatic control and intelligent operation of circulating water systems.

[0028] The beneficial effects of this invention are: During the entire circulating water pipeline loop experiment, changes in specific water quality environmental factors in the pipeline water showed a high degree of positive correlation. The interaction between different water quality environmental factors jointly promoted the initial colonization and subsequent growth of biofilm on the pipe wall surface. Dynamic analysis of the circulating water quality environmental factors can accurately predict the overall growth and development status of biofilm and effectively identify different growth and development stages of biofilm, providing a data-driven theoretical prediction framework for early warning and control of biofilm in circulating water networks. In summary, this invention, through dynamic correlation analysis between circulating water quality environmental factors and biofilm growth, explores the internal correlations between water quality environmental factor variables, captures the relationship between key water quality factors and biofilm growth and development, and thus predicts biofilm formation in a timely and accurate manner. This invention can predict the growth and development status of biofilm by dynamically analyzing the changes in specific water quality environmental factors in the pipeline water, avoiding the drawbacks and limitations of requiring open-pipe measurement of biofilm, effectively solving the technical problem of incomplete process quantification models, and providing a simple and novel method for timely and accurate feedback of biofilm hazards. This method requires no additional manpower or material resources and can achieve long-term intensive monitoring of all the above-mentioned environmental factors through an online real-time monitoring system. It provides a new paradigm for realizing early warning of biological slime growth through circulating water quality diagnosis, which is conducive to the automatic control and intelligent operation of the circulating water system and ensures the long-term stable and safe operation of the circulating water system. Attached Figure Description

[0029] Figure 1 A represents the experimental setup for the pipeline loop; B represents the microscopic morphology of the biological slime on the pipe wall on day 30.

[0030] Figure 2 This represents the dynamic changes of different water quality environmental factors during the growth and development of biological slime. A represents changes in DO content; B represents changes in pH value; C represents changes in conductivity value; D represents changes in UV. 254 Value change; E represents TN content change; F represents NH4+. + -N content changes; G is NO2 - Changes in -N content; H is NO3 - Changes in -N content; I represents changes in TOC content; J represents PO4 content. 3- -P content change; K represents Turbidity value change; L represents SS content change; M represents Na ion concentration change; N represents Mg ion concentration change; O represents K ion concentration change; P represents Ca ion concentration change.

[0031] Figure 3A shows the outline morphology of the biofilm layer at different growth and development stages (from left to right: day 33 (Colonization, Col), day 85 (Accumulation, Acc), day 110 (Maturation, Mat), and day 135 (Old)); B shows the biofilm layer thickness; C shows the Spearman and Mantel correlation analysis between changes in water quality environmental factors and changes in biofilm thickness.

[0032] Figure 4 This section validates and robusts the prediction framework based on the random forest machine learning model. A represents the quantitative ranking of the importance of different water quality environmental factors; B represents the stage-specific contribution of different water quality environmental factors; C compares the prediction accuracy of the framework on the training set and the colonization stage test set; D compares the prediction accuracy of the framework on the training set and the accumulation stage test set; E compares the prediction accuracy of the framework on the training set and the mature stage test set; F compares the prediction accuracy of the framework on the training set and the aging stage test set. Detailed Implementation

[0033] The present application will be further described in detail below with reference to examples and accompanying drawings.

[0034] Example 1 like Figure 1 As shown in (A), the pipe loop experiment consisted of a pipe ring (outer diameter 32 mm, inner diameter 25.6 mm, wall thickness 3.2 mm) connected by elbows. All pipe materials were polyvinyl chloride (PVC). The total length of the pipe ring was approximately 2 m. The pipes were transparent, facilitating direct observation of biofilm growth on the pipe walls. Removable and replaceable PVC sheets (1 × 1 cm) were installed on the pipes to accumulate biofilm for subsequent characterization. A CNC thermostatic water bath with a working volume of 15 L provided a water flow velocity of approximately 1.41 m / s. The water circulated continuously within the pipes, rather than passing through in a single pass. The entire pipe loop experiment ran continuously for 135 days. The temperature of the water bath was maintained at 28 ± 1 °C. The water supply for the pipe loop was taken from a natural surface lake.

[0035] Example 2 Pipeline water samples were collected every two days using a sampling tap to determine specific water quality parameters. Dissolved oxygen (DO), pH, conductivity, turbidity, and UV concentrations in the pipeline water were measured using a digital portable dissolved oxygen meter, pH meter, conductivity meter, turbidity meter, UV spectrophotometer, and gravimetric method. 254Direct detection of SS content was performed; after filtering the water sample through a 0.45 µm polyethersulfone membrane, the NH4 content in the pipeline water was analyzed using salicylic acid spectrophotometry, N-(1-naphthyl)-1,2-diaminoethane dihydrochloride spectrophotometry, aminosulfonic acid-UV spectrophotometry, ammonium molybdate spectrophotometry, and a total organic carbon analyzer. + -N, NO2 - -N, NO3 - -N,PO4 3- -P and TOC contents were detected; the water sample was filtered through a 0.22 µm polyethersulfone filter membrane, acidified with nitric acid to 1% w / w HNO3, and the concentrations of Ca, Mg, Na, and K metal ions in the water were detected by inductively coupled plasma atomic emission spectrometry.

[0036] Example 3 Biofilm was collected every 30 days, and the outline morphology of biofilm growth and development was visualized using a scanning electron microscope; the thickness of the biofilm layer was determined using Image-Pro Plus image analysis software.

[0037] Example 4 Spearman correlation analysis was used to analyze the correlations between changes in specific water quality environmental factors in pipeline water; Mantel correlation analysis was used to analyze the correlations between specific water quality environmental factors in pipeline water and changes in biofilm thickness.

[0038] Example 5 An analytical and predictive framework based on a random forest ensemble learning model is constructed to reveal the impact of water quality environmental factors on the growth and development of biofilm in a circulating aquatic system. The overall impact of different water quality environmental factors on biofilm growth and development is quantified; the stage-specific contributions of specific water quality environmental factors are analyzed; and the generalization ability, accuracy, and robustness of the "water quality factor predicts biofilm growth" prediction framework are comprehensively evaluated.

[0039] like Figure 1 As shown in (B), brownish-yellow biofilm is dispersed on the inner surface of the pipe wall, forming a flat film. Spherical, rod-shaped, and filamentous microorganisms coexist in the biofilm, indicating that planktonic microorganisms in the pipe water have successfully colonized the pipe wall surface rapidly.

[0040] like Figure 2As shown, the characteristic parameters of the pipeline water quality show obvious dynamic changes at different stages of the growth and development of biological slime. For example, the concentrations of Ca and Mg ions in the pipeline water gradually decrease during the growth and development of biological slime, which may mean that Ca and Mg ions react with some inorganic components in the water to form scale deposits on the pipe wall, providing attachment points for microorganisms and accelerating the stacking of biological slime. The concentrations of TOC and TN in the pipeline water gradually decrease from the initial formation of biological slime to subsequent accumulation and then to the mature stage, indicating that carbon and nitrogen are limiting substrates for the growth and development of biological slime before aging, while the concentration of PO4 3- -P decreases continuously during the entire growth and development process of biological slime, which may be related to the synthesis of microbial phospholipids in biological slime. The Turbidity and SS contents in the pipeline water show a highly consistent change trend, first gradually decreasing and then slowly increasing, indicating that the inert particles in the pipeline water are likely to first attach to the pipe wall surface and form a "bio-abiotic" thick layer through interaction with the extracellular polymers and soluble microbial products secreted by the microorganisms colonized on the pipe wall surface. As the biological slime ages and gradually detaches from the pipe wall surface, some inert substances may disperse into the water, resulting in an increase in the Turbidity and SS of the pipeline water.

[0041] As Figure 3 shown in (A) and (B), the growth and development of biological slime show obvious stage (colonization - accumulation - maturity - aging) characteristics, with the thickness gradually increasing, but the thickness significantly decreases after entering the aging stage. As Figure 3 shown in (C), the Spearman correlation analysis shows that there is a strong positive correlation among different water quality environmental factors as a whole, which means that the interaction among different water quality environmental factors is the driving force for the growth and development of biological slime. Further, the Mantel correlation analysis identifies the water quality environmental factors significantly related to different growth and development stages of biological slime. When P the value is smaller and the Mantel’s r value is larger, it indicates that there is a significant correlation between this water quality environmental factor and the growth and development of biological slime. For example, at the colonization stage, DO (0.01 < P < 0.05 and 0.25 < Mantel’s r < 0.5) and TOC (0.001 < P<0.01 and Mantel's r>0.5) were the strongest correlation factors for biofilm growth. During the accumulation phase, biofilm is constrained by a wider and more diverse range of aquatic environmental factors, including DO, TN, metal ions (Ca, Mg, Na, K), and SS. Simultaneously, the biofilm layer thickness increases significantly. In the maturity phase, the community structure stabilizes, and the growth and development rate of biofilm slows (thickness increase is slow). Therefore, mature biofilm may be less affected by external aquatic environmental factors and its association with all aquatic environmental factors is not significant. Conversely, when biofilm enters the aging stage and detaches from the pipe wall surface, the remaining basal biofilm layer may regrow by recruiting new planktonic microorganisms and interacting with aquatic environmental factors, thus exhibiting a high degree of significant correlation with most aquatic environmental factors again.

[0042] like Figure 4 As shown in (A), changes in all water quality environmental factors have a significant impact on the growth and development of biological slime. P <0.05) importance, among which Conductivity, TN and DO showed extremely significant ( P The importance of parameters <0.001) underscores the necessity of multi-parameter integrated monitoring for predicting the growth and development of biofilm. For example... Figure 4 As shown in (B), TOC, TN, and PO4 3- -P, Turbidity, and SS contribute more significantly to the early and mid-stages of biofilm growth and development, indicating the crucial role of nutrient substrates and inert particulate matter in promoting initial colonization and subsequent accumulation of biofilm. In contrast, DO, pH, and metal ions (Ca, Mg, K, Na) contribute equally significantly to all stages of biofilm growth and development, demonstrating the sustained influence of these aquatic environmental factors during biofilm growth and development. For example, Ca and Mg ions may promote the formation of inorganic sedimentary layers on pipe walls, providing a favorable substrate for planktonic microorganism attachment; furthermore, the random forest model's performance on the training set (R... 2 = 0.974) and test set (R 2 The model demonstrated good predictive accuracy (ratios 0.890 to 0.994). Although the predictive difficulty increased slightly with the growth and development of the biofilm, the model maintained high predictive stability and robustness at all stages. These results indicate that the monitoring framework based on changes in environmental factors of circulating water can accurately distinguish different growth and development stages of biofilm, especially in the initial formation stage (R0.890 to R0.994). 2 = 0.994).

Claims

1. A data-driven method for predicting biofilm growth using circulating water quality characteristic parameters, characterized in that, Includes the following steps: (1) Monitoring water quality environmental factors in circulating water: Measuring ammonium nitrogen (NH4) in pipeline water + -N), nitrite nitrogen (NO2) - -N), nitrate nitrogen (NO3) - -N), reactive phosphate (PO4) 3- Changes in P and total organic carbon (TOC) content; measurements of dissolved oxygen (DO), pH, conductivity, and UV in pipeline water. 254 Changes in turbidity and suspended solids (SS) content; changes in the concentrations of calcium (Ca), magnesium (Mg), sodium (Na), and potassium (K) metal ions in pipeline water; (2) Periodically measure the thickness of the biofilm on the pipe wall: use scanning electron microscopy to visualize the outline morphology of biofilm growth and development; use image analysis software to measure the thickness of the biofilm layer; (3) Spearman correlation analysis and Mantel correlation analysis were used to correlate water quality environmental factors with changes in the thickness of biological slime layer, and water quality environmental factors that are correlated with the growth and development of biological slime at different stages were identified.

2. The method according to claim 1, characterized in that, In step (1), the NH4 in the pipeline water was determined by salicylic acid spectrophotometry, N-(1-naphthyl)-1,2-diaminoethane dihydrochloride spectrophotometry, aminosulfonic acid-UV spectrophotometry, and ammonium molybdate spectrophotometry, respectively. + -N, NO2 - -N, NO3 - -N and PO4 3- Changes in -P content; changes in TOC content in pipeline water were measured using a total organic carbon analyzer; DO, pH, conductivity, turbidity, and UV levels in pipeline water were measured using a digital portable dissolved oxygen meter, pH meter, conductivity meter, turbidity meter, and UV-Vis spectrophotometer, respectively. 254 The changes in values ​​were measured; the changes in SS content in the pipeline water were determined by gravimetric method; and the changes in the concentrations of Ca, Mg, Na, and K metal ions in the pipeline water were determined by inductively coupled plasma atomic emission spectrometry.

3. The method according to claim 1, characterized in that, In step (1), during the detection of NH4 + -N, NO2 - -N, NO3 - -N and PO4 3- Before measuring P and TOC content, the water sample was filtered through a 0.45 µm filter membrane; before measuring the concentrations of Ca, Mg, Na, and K metal ions, the water sample was filtered through a 0.22 µm filter membrane and then acidified with nitric acid to 0.2%–5% w / w HNO3.

4. The method according to claim 1, characterized in that, In step (1), the filter membrane is one of polyethersulfone filter membrane, mixed cellulose ester membrane or hydrophilic polyvinylidene fluoride membrane.

5. The method according to claim 1, characterized in that, In step (1), the detection time interval for environmental factors of pipeline water quality is 2 days.

6. The method according to claim 1, characterized in that, In step (1), the water quality environmental factors in the circulating water are monitored by simulating the circulating water pipeline loop experiment.

7. The method according to claim 6, characterized in that: The operating conditions for the pipeline loop experiment are as follows: (1) The circulating water is taken from surface water, groundwater, seawater or recycled water; (2) The pipe material is polyvinyl chloride, carbon steel, stainless steel, galvanized or cast iron; (3) The temperature of the circulating water is maintained at 27 ~ 30℃; (4) The flow velocity of the circulating water is maintained at 1.4 ~ 1.5 m / s; (5) The pipeline loop test adopts continuous operation mode, and the operation cycle is 135 ~ 140 days.

8. The method according to claim 1, characterized in that, In step (2), the image analysis software is Image-ProPlus image analysis software; the time interval for measuring the thickness of the biofilm is 30 to 35 days.

9. The method according to claim 1, characterized in that, In step (2), the stages of bio-slime growth and development include the initial stage of bio-slime formation, the stage of bio-slime accumulation and thickening, the stage of bio-slime maturity and stabilization, and the stage of bio-slime aging.

10. The method according to claim 1, characterized in that, Also includes: (4) Use an ensemble learning model based on random forest to evaluate the importance of water quality environmental factors on the growth and development of biological slime in the circulating water system, quantify the overall impact of different water quality environmental factors on biological slime accumulation, and identify key driving factors.

Images