The paper analyzes the influence of meteorological conditions (air temperature, wind speed, humidity, visibility) and anthropogenic factors (population in cities and in rural areas, road length, number of vehicles, emission of dusts and gases, coal consumption in industrial plants, number of air purification devices installed in industrial plants) on the concentration of PM2.5 and PM10 dusts in the air in the region of Kielce city in Poland. Spearman correlation coefficient was used to evaluate the relationship between the mentioned independent variables and air quality indicators. The calculated values of the correlation coefficient showed statistically significant relationships between air quality and the amount of installed air purification equipment in industrial plants. A statistically significant effect of the population in rural settlement units on the increase in air concentrations of PM2.5 and PM10 was also found, which proves the influence of the so-called low emission of pollutants on the air quality in the studied region. The analyses also revealed a statistically significant effect of road length on the decrease in PM2.5 and PM10 air content. This result indicates that a decrease in traffic intensity on particular road sections leads to an improvement in air quality. The analyses showed that despite the progressing anthropopression in the Kielce city region the air quality with respect to PM2.5 and PM10 content is improving. To verify the results obtained from statistical calculations, parametric models were also determined to predict PM2.5 and PM10 concentrations in the air, using the methods of Random Forests (RF), Boosted Trees (BT) and Support Vector Machines (SVM) for comparison purposes. The modelling results confirmed the conclusions that had been made based on previous statistical calculations.

The issue of leak detection is a problem faced by most water utilities, especially those operating old and failing or poorly managed water supply networks. In such cases, water losses due to failures may reach a dozen or even more percent of produced water, exposing the company to large financial losses. Thus, it is a significant economic problem, but also an environmental one, as water losses deplete water resources. Of course, the remedy can be a properly planned revitalization of the water supply network, changing the pump pressure control in pumping stations and zone pumping stations, but also early detection and removal of failures, which is the subject of the paper. The development of algorithms to solve the problem of water leaks detection is essential for the proper operation of water supply networks, however, it should be borne in mind that their implementation is also associated with high costs and organisational difficulties for waterworks company. The paper discusses methods related to the detection and location of emergency states on water supply networks and presents two algorithms of varying degrees of complexity, developed at the Institute of System Research of the Polish Academy of Sciences.

A mathematical model to simulate and study the interaction between the content of nitrogen compounds in the outflow of a wastewater treatment plant and the efficiency of activated sludge sedimentation in its secondary clarifiers is presented in this paper.The goal of the model is to control the biological reactor settings (e.g. total nitrogen (TN) and sedimentation properties of the sludge) in case of discontinuity of quality indicators at wastewater inflow.Such an approach has not been applied so far.Continuity of calculated numerical values of model-dependent variables (TN, sludge volume index) was obtained by replacing the independent variables with the results of calculations obtained using the statistical models.Data mining methods were used to simulate the content of nitrogen compounds at the wastewater outlet.To simulate active sludge sedimentation a classification model based on the logistic regression method was used.According to the obtained results the proposed comprehensive model of simulation, referring to TN and activated sludge sedimentation, enables optimal selection of bioreactor settings.

One of the key parameters constituting the basis for the operational assessment of stormwater systems is the annual number of storm overflows. Since uncontrolled overflows are a source of pollution washed away from the surface of the catchment area, which leads to imbalanced receiving waters, there is a need for their prognosis and potential reduction. The paper presents a probabilistic model for simulating the annual number of storm overflows. In this model, an innovative solution is to use the logistic regression method to analyze the impact of rainfall genesis on the functioning of a storm overflow (OV) in the example of a catchment located in the city of Kielce (central Poland). The developed model consists of two independent elements. The first element of the model is a synthetic precipitation generator, in which the simulation of rainfall takes into account its genesis resulting from various processes and phenomena occurring in the troposphere. This approach makes it possible to account for the stochastic nature of rainfall in relation to the annual number of events. The second element is the model of logistic regression, which can be used to model the storm overflow resulting from the occurrence of a single rainfall event. The paper confirmed that storm overflow can be modeled based on data on the total rainfall and its duration. An alternative approach was also proposed, providing the possibility of predicting storm overflow only based on the average rainfall intensity. Substantial simplification in the simulation of the phenomenon under study was achieved compared with the works published in this area to date. It is worth noting that the coefficients determined in the logit models have a physical interpretation, and the universal character of these models facilitates their easy adaptation to other examined catchment areas. The calculations made in the paper using the example of the examined catchment allowed for an assessment of the influence of rainfall characteristics (depth, intensity, and duration) of different genesis on the probability of storm overflow. Based on the obtained results, the range of the variability of the average rainfall intensity, which determines the storm overflow, and the annual number of overflows resulting from the occurrence of rain of different genesis were defined. The results are suited for the implementation in the assessment of storm overflows only based on the genetic type of rainfall. The results may be used to develop warning systems in which information about the predicted rainfall genesis is an element of the assessment of the rainwater system and its facilities. This approach is an original solution that has not yet been considered by other researchers. On the other hand, it represents an important simplification and an opportunity to reduce the amount of data to be measured.

The paper analyzes the possibility of using the CFD (Computational Fluid Dynamics) method to predict the amount of sewage remaining in siphons after a full air blast of the pressure sewer system. For this purpose, the results from measurements carried out on a laboratory installation were compared with the results obtained from modelling using a spatial model (3D) and a plain model (2D) of the installation. To determine these models, the structure of the VOF (Volume of Fluid) model was used in the CFD method. The simulation calculations carried out make it possible to state that the use of the plain model with the development of the installation modelled in the plan does not result in significant deterioration of the obtained results. The possibility of using 2D models for modelling pumped sewer systems allows for a significant shortening of the calculation time, which, in practice, results in the possibility of modelling much larger and longer installations than is possible with 3D models.

Our paper presents an analysis of the effect of data selection uncertainty for the teaching and testing sets in the black box model (multilayer perceptron type of artificial neural networks) using the bootstrap method on the accuracy of forecast and control of activated sludge sedimentation (SE) and the sludge volume index (SVI). The calculations show that sludge sedimentation, and hence also the sludge volume index, can be predicted based on the wastewater quality indicators and biological reactor operating parameters. The presented analyses also confirmed the significant influence of the neural network model structure on the uncertainty of estimating biological reactor operating parameters (mixed liquor suspended solids, concentration of oxygen) which, in practical considerations, leads to problems of continuous control of the sludge sedimentation capacity.

This article presents several algorithms for controlling water supply system pumps. The aim of having control is the hydraulic optimisation of the network, i.e., ensuring the desired pressure in its recipient nodes, and minimising energy costs of network operation. These two tasks belong to the key issues related to the management and operation of water supply networks, apart from the reduction in water losses caused by network failures and ensuring proper water quality. The presented algorithms have been implemented in an Information and Communications Technology (ICT) system developed at the Systems Research Institute of the Polish Academy of Sciences (IBS PAN) and implemented in the waterworks GPW S.A. in Katowice/Poland.

The processes that affect sediment quality in drainage systems show high dynamics and complexity. However, relatively little information is available on the influence of both catchment characteristics and meteorological conditions on sediment chemical properties, as those issues have not been widely explored in research studies. This paper reports the results of investigations into the content of selected heavy metals (Ni, Mn, Co, Zn, Cu, Pb, and Fe) and polycyclic aromatic hydrocarbons (PAHs) in sediments from the stormwater drainage systems of four catchments located in the city of Kielce, Poland. The influence of selected physico-geographical catchment characteristics and atmospheric conditions on pollutant concentrations in the sediments was also analyzed. Based on the results obtained, statistical models for forecasting the quality of stormwater sediments were developed using artificial neural networks (multilayer perceptron neural networks). The analyses showed varied impacts of catchment characteristics and atmospheric conditions on the chemical composition of sediments. The concentration of heavy metals in sediments was far more affected by catchment characteristics (land use, length of the drainage system) than atmospheric conditions. Conversely, the content of PAHs in sediments was predominantly affected by atmospheric conditions prevailing in the catchment. The multilayer perceptron models developed for this study had satisfactory predictive abilities; the mean absolute error of the forecast (Ni, Mn, Zn, Cu, and Pb) did not exceed 21%. Hence, the models show great potential, as they could be applied to, for example, spatial planning for which environmental aspects (i.e., sediment quality in the stormwater drainage systems) are accounted.

The last decade has seen the development of complex IT systems to support city management, i.e., the creation of so-called intelligent cities. These systems include modules dedicated to particular branches of municipal economy, such as urban transport, heating systems, energy systems, telecommunications, and finally water and sewage management. In turn, with regard to the latter branch, IT systems supporting the management of water supply and sewage networks and sewage treatment plants are being developed. This paper deals with the system concerning the urban water supply network, and in particular, with the subsystem for detecting and locating leakages on the water supply network, including so-called hidden leakages. These leaks cause the greatest water losses in networks, especially in old ones, with a very diverse age and material structure. In the proposed concept of the subsystem consisting of a GIS (Geographical Information System), SCADA (Supervisory Control and Data Acquisition) system and hydraulic model of the water supply network, an algorithm of leak detection and location based on the neural networks’ MLP (multi-layer perceptron) and Kohonen was developed. The algorithm has been tested on the hydraulic models of several municipal water supply networks.

The article presents a mathematical model for the analysis of operational reliability of a wastewater treatment plant, in which sedimentation of activated sludge and removal of biogenic compounds were taken into account.The presented model allows for continuous control and monitoring of both processes, even in the case of measurement discontinuities.In the presented approach, the values of quality indicators can be determined using selected data mining methods on the basis of wastewater flow and temperature measurements.The paper proposes an innovative indicator that takes into account the interaction between the quantity, the quality of inflowing wastewater expressed by means of physicochemical parameters and the susceptibility of activated sludge for bulking.Based on the presented calculation algorithm, an exemplary concept of controlling the biological process (mixed liquor suspended solids, oxygen concentration and the amount of coagulant dosed) is presented, taking into account the variable conditions at the inflow to the bioreactor.

This article analyzes the effect that the physicochemical parameters of the wastewater flowing into a treatment plant have on the activated sludge settleability. The statistical analysis shows that as far as the technological parameters are concerned, the activated sludge sedimentation capacity is mostly affected by the biomass concentration in the chamber, whereas as for the abiotic factors, settleability is significantly determined by the season of the year and thus the temperature. With regard to the wastewater quality-related parameters, biological oxygen demand has the greatest effect on settleability. The conducted analyzes involved the development of statistical models to predict the activated sludge sedimentation capacity on the basis of multiple linear regression and genetic programming.

This paper presents the concepts of a probabilistic model for storm overflow discharges, in which arbitrary dynamics of the catchment urbanization were included in the assumed period covered by calculations. This model is composed of three components. The first constitutes the classification model for the forecast of storm overflow discharges, in which its operation was related to rainfall characteristics, catchment retention, as well as the degree of imperviousness. The second component is a synthetic precipitation generator, serving for the simulation of long-term observation series. The third component of the model includes the functions of dynamic changes in the methods of the catchment development. It allows for the simulation of changes in the extent of imperviousness of the catchment in the long-term perspective. This is an important advantage of the model, because it gives the possibility of forecasting (dynamic control) of catchment retention, accounting for the quantitative criteria and their potential changes in the long-term perspective in relation to the number of storm overflows. Analyses carried out in the research revealed that the empirical coefficients included in the logit model have a physical interpretation, which makes it possible to apply the obtained model to other catchments. The paper also shows the use of the prepared probabilistic model for rational catchment management, with respect to the forecasted number of storm overflow discharges in the long-term and short-term perspective. The model given in the work can be also applied to the design and monitoring of catchment retention in such a way that in the progressive climatic changes and urbanization of the catchment, the number of storm overflow discharges remains within the established range.

Potential application of artificial neural networks (ANN) to forecast total nitrogen content (TNC) in treated wastewater was presented as a function of selected nitrogen forms present in the secondary effluent. The analyzed data from the period of 2010-2016 covered measurements of the nitrogen content in the effluent from the treatment plant servicing agglomeration with a population equivalent of more than 100,000. The input data set was initially subjected to cluster analysis and then, used to train a neural network in the form of a multilayer perceptron (MLP). The simulations demonstrated that the smallest error values for the forecast of TNC (2-3%) were obtained for the variant, the value of which was a function of all the forms of nitrogen present in the secondary effluent. For the total nitrogen model based on inorganic nitrogen and nitrates data only, the simulation results did not differ significantly from the actual values, as indicated by a very high correlation coefficient (over 97%). In this case, the value of the mean absolute error increased only by nearly 4% to 6.2% (learning process) or 6.9% (testing/validation process), compared to the simulation based on all the nitrogen forms in the sewage.

An approach to forecast the mixed liquor suspended solids (MLSS) and food-to-mass ratio (F/M) of the activated sludge in bioreactor using some methods of statistical modelling has been proposed. The impact of explanatory variables used in the models on the exactness of the models developed has also been analyzed. Those variables are wastewater quality indicators and parameters of activated sludge chambers while the modelling methods used are the support vectors machines, cascade neural networks and boosted trees. Moreover, the possibility of modelling those variables based on the measurements of wastewater flow and temperature in the wastewater inflow to the wastewater treatment plant has been investigated. It was concluded that the MLSS as well as the F/M could be successfully forecasted by variety of statistical models in which the wastewater quality indicators are not measured but modelled. The method is very useful operationally because it makes possible to monitor and correct the values of MLSS and F/M quickly and efficiently while only a limited access to the wastewater quality measurements is available.

The paper presents the results of forecasting the sewage inflow into the municipal wastewater treatment plant in Rzeszow using multilayer perceptron neural networks. For the forecast model, the following independent variables were adopted: the measured inflow volume to the treatment plant from the previous days, the water level in the Wislok River (effluent receiver), the total daily precipitation and the daily water inflow into the network. The calculations led to conclusions that variables substantially affecting the prognostic capacity of the forecast model included the water level in the Wislok River, the volume of precipitation and the sewage inflow to the facility from the previous days. Additionally, the impact of individual structural parameters of the model based on artificial neural networks on forecasting results was analyzed. The research conducted with the use of classification trees demonstrated that number of neurons in the hidden layer was influenced by the number of inputs to the model, while the type of activation function in the hidden and output layer was of minor importance which was confirmed by the data of prognostic value. The applicability of a linear discriminant analysis for assessment of prognostic ability of the constructed forecast models was also investigated. The results obtained demonstrated that the linear discriminant model might be an interesting assessment tool to select variables for the forecast model of sewage inflow to a treatment plant.

The study compares an annual number of weir overflows calculated using a hydrodynamic model by continuous simulations and a probabilistic model. The weir outflow for a single precipitation event was successfully modelled using logistic regression. Performed numerical experiments showed that the calculated number of weir outflows with the hydrodynamic model falls within confidence intervals of the probabilistic model. This suggests that the model of the logistic regression can be used in practice. The probabilistic simulations revealed that a model with a probabilistic description of a number of annual precipitations and a model with an assumed average number of such events are not consistent. The proposed methodology can be applied for the design of outflow weirs and other storm devices.

In this paper, statistical models to forecast based on the sludge volume index (SVI) with the continuous measurements carried out in the period from 2013 to 2016 for waste water treatment Sitkowka-Nowiny was developed at the same, for two variants of analyses. In the first one, a model of SVI predicting based on the quality indicators of wastewater flowing into the treatment plant, i.e. Biochemical (BOD) and chemical oxygen demand (COD), the content of total nitrogen (TN) and ammonia nitrogen (NH4), total suspended solids, total phosphorus (TP) and the operating parameters of the bioreactor (pH, temperature, oxygen concentration in the nitrification chamber). In the second case, the possibility of replacing individual measurements of the quality of wastewater values calculated on the basis of daily sewage flows to the treatment plant was examined. The above mentioned models statistical analysis was performed using the method of k-nearest neighbor (k-NN), cascading neural network (CNN) and boosted tree (BT). To evaluate the predictive ability of these models the average relative error (MAE) and absolute error (MAPE) were used. The conducted analysis showed that based on the above mentioned indicators of effluent quality and technological parameters of the biological reactor it is possible to modeling of sediment volume index with satisfactory accuracy. In the case under consideration methods of lower values of the prediction error of SVI obtained using a cascade neural networks (MAE = 17.49 ml/g and MAPE = 9.80%) than for the method k-nearest neighbor (MAE = 27.85 ml/g and MAPE = 14.50%). Furthermore, based on the performed simulation, it was found that it is possible to model the analyzed work of the quality of waste water on the basis of the daily flow with reasonable accuracy, it is confirmed by the calculated value of the average and absolute and relative error, and the better ability predictive characterized by the models obtained on the basis CNN than k-NN. In examined cases, the MAP in a set of validation did not exceed 10.13%. The simulation results of quality indicators obtained by CNN were substituted in place of the explanatory variables of sludge volume index in the model for prediction index of sediment and conducted simulations SVI, set out the error MAE = 25.15 ml/g and MAPE = 15.26%. On this basis, it is possible to replace the measured values of the quality of the results of their simulation, thereby reducing the cost of testing, but also gives you continuous control of SVI and adjustments discussed in this work of technological parameters of the biological reactor.

The aim of the study was to evaluate the possibility of applying different methods of data mining to model the inflow of sewage into the municipal sewage treatment plant. Prediction models were elaborated using methods of support vector machines (SVM),random forests (RF), k-nearest neighbour (k-NN) and of Kernel regression (K). Data consisted of the time series of daily rainfalls, water level measurements in the clarified sewage recipient and the wastewater inflow into the Rzeszow city plant. Results indicate that the best models with one input delayed by 1 day were obtained using the k-NN method while the worst with the K method. For the models with two input variables and one explanatory one the smallest errors were obtained if model inputs were sewage inflow and rainfall data delayed by 1 day and the best fit is provided using RF method while the worst with the K method. In the case of models with three inputs and two explanatory variables, the best results were reported for the SVM and the worst for the K method. In the most of the modelling runs the smallest prediction errors are obtained using the SVM method and the biggest ones with the K method. In the case of the simplest model with one input delayed by 1 day the best results are provided using k-NN method and by the models with two inputs in two modelling runs the RF method appeared as the best.