Optimal Cleaning Schedule Research Articles

Optimal cleaning schedule in solar PV using the biography-based helianthus optimization enabled coupled deep network

The research aims to develop a system that determines the optimal intervals for cleaning solar PV panels to maximize their efficiency and minimize power loss due to dust accumulation. The research utilizes data related to daily solar photovoltaic tariff rates, power generation, and conditions of dust accumulation on PV panels. To monitor the performance of PV panels, the research involves forecasting power generation. If the power generation falls below a predefined threshold, it indicates a need for cleaning. Dust particle loss conditions are analyzed using an optimized Support Vector Machine (SVM) coupled with a Deep Convolutional Neural Network (CNN) model. This model processes PV panel images to identify dusty conditions. The Deep CNN model extracts features from PV panel images, such as color-based statistical features and local gradient patterns. These features are crucial for identifying dusty conditions. The optimal cleaning schedule is determined based on the detection of dust particle loss conditions. Cleaning is initiated when the conditions warrant it. The novelty of this research lies in the development of a deep learning model for both detecting dust particle loss and forecasting power generation. This model is trained using data that includes daily tariff rates and dust particle loss information. The research proposes a new optimization method called “biography-based Helianthus optimization (HBBHO)” specifically designed to suit the characteristics of sunlight. This method is employed to schedule cleaning activities. The proposed approach achieves impressive accuracy, precision, sensitivity, and specificity, with values reaching 96.53% for most metrics, except for specificity, which is at 96.21%.

Soiling Modelling in Large Grid-Connected PV Plants for Cleaning Optimization

Soiling of PV modules is an issue causing non-negligible losses on PV power plants, between 3 and 4% of the total energy production. Cleaning is the most common way to mitigate soiling. The impact of the cleaning activity can be significant, both in terms of cost and resources consumption. For these reasons, it is important to monitor and predict soiling profiles and establish an optimal cleaning schedule. Especially in locations where raining is irregular or where desert winds carry a high concentration of particles, it is also important to know how precipitation and dust events affect the soiling ratio. This paper presents a new model based on environmental conditions that helps the decision-making process of the cleaning schedule. The model was validated by the analysis of five large grid-connected PV plants in Spain over two years of operation, with a total power of 200 MW. The comparison between the model and soiling sensors at the five locations was included. Excellent results were achieved, the mean difference between sensors and model being 0.71%.

Energy saving and pollution reduction through optimal scheduling of cleaning actions in a heat exchanger network

Heat exchanger fouling may impair energy recovery in Heat Exchanger Networks (HENs), thereby increasing the energy intensity of production processes. Fouling growth on the heat-transfer surfaces of the exchangers can be monitored to mitigate its adverse effects, using measurement data that also enable scheduling exchanger cleaning when economically justified. The research builds on the authors’ previous work devoted to different measurement-aided monitoring methods of fouling thermal resistance, where the method based on the least-square approach to measurement data is the preferred one. The aim is to apply the preferred approach to and perform a case study on the optimization of scheduling of cleaning actions on heat exchangers operated in a HEN. The avoided HEN operating cost is employed as an objective function to be maximized by the optimal cleaning schedule that also has to satisfy a set of constraints related to the timing of cleaning actions. The mathematical formulation of the schedule optimization problem is simplified to make solution finding easier. The case study uses historical operation records covering three years of the real-life HEN composed of 26 shell-and-tube heat exchangers working continuously in connection with a Crude Distillation Unit. Having identified historical time characteristics of the thermal resistance of fouling in all the heat exchangers, one can retroactively determine optimal HEN cleaning schedules. The optimization results are discussed, considering their economic aspects and potential for energy saving and pollution reduction.

Dust deposition effect on photovoltaic modules performance and optimization of cleaning period: A combined experimental–numerical study

The main purposes of this study are: i) evaluating magnitude of error in predicting soiling power loss of photovoltaic (PV) modules based on the measurement of short-circuit current; ii) presenting simple experimental models for estimating dust effect on the PV power generation; and finally iii) finding an optimal cleaning schedule for a utility-scale solar power station located in a semi-arid environment. The results showed that, on average, soiling short-circuit current loss underestimates actual soiling power loss by approximately 9%, and with increasing dust density this value also increases. The soiling rate in the first week of dust deposition was evaluated to be 0.27%/day, meanwhile, for the last week it was 0.13%/day. This observation confirmed that the exponential model is more accurate than linear and power relations for estimating dust effect on PV systems in the case study. The optimal cleaning period of a solar farm was obtained 17.50 and 6.39 days, for the manual and machine-assisted washing methods, respectively. The sensitivity analysis regarding the effect of environmental variables on the optimal cleaning period was addressed at last. The results indicated that such analysis is necessary in order to find a reliable estimate for optimal cleaning period in PV power plants.

Photovoltaic cleaning optimization through the analysis of historical time series of environmental parameters

This work investigates the possibility of using historical environmental parameter data to predict the typical soiling loss profile and the most convenient cleaning schedule for a PV site. The three-year performance of a 1 MW system in Southern Spain is evaluated using different soiling extraction methods. When the rainfall pattern is used to detect natural cleaning events, the best results are obtained if a 1.0 mm/hour threshold is considered. However, despite the optimization, setting a fixed threshold is found to lead occasionally to the over- or under-detection of cleaning events. Similar trends in the modelling results are found if the thresholds are set using the maximum hourly or the cumulative daily rainfall data, but the errors and the optimal values change depending on the rainfall dataset. The study also shows that a soiling extraction method based only on precipitation and particulate matter, calibrated against one year of PV data, is able to generate a soiling profile with a mean absolute error of 0.022 and to recommend a cleaning day within a week of the actual optimal dates. This will make it possible to estimate the soiling losses and the optimal cleaning schedule for a PV site even if no power data are available.

Stability of Optimal Closed-Loop Cleaning Scheduling and Control with Application to Heat Exchanger Networks under Fouling

Heat exchanger networks subject to fouling are an important example of dynamic systems where performance deteriorates over time. To mitigate fouling and recover performance, cleanings of the exchangers are scheduled and control actions applied. Because of inaccuracy in the models, as well as uncertainty and variability in the operations, both schedule and controls often have to be revised to improve operations or just to ensure feasibility. A closed-loop nonlinear model predictive control (NMPC) approach had been previously developed to simultaneously optimize the cleaning schedule and the flow distribution for refinery preheat trains under fouling, considering their variability. However, the closed-loop scheduling stability of the scheme has not been analyzed. For practical closed-loop (online) scheduling applications, a balance is usually desired between reactivity (ensuring a rapid response to changes in conditions) and stability (avoiding too many large or frequent schedule changes). In this paper, metrics to quantify closed-loop scheduling stability (e.g., changes in task allocation or starting time) are developed and then included in the online optimization procedure. Three alternative formulations to directly include stability considerations in the closed-loop optimization are proposed and applied to two case studies, an illustrative one and an industrial one based on a refinery preheat train. Results demonstrate the applicability of the stability metrics developed and the ability of the closed-loop optimization to exploit trade-offs between stability and performance. For the heat exchanger networks under fouling considered, it is shown that the approach proposed can improve closed-loop schedule stability without significantly compromising the operating cost. The approach presented offers the blueprint for a more general application to closed-loop, model-based optimization of scheduling and control in other processes.

Economics of seasonal photovoltaic soiling and cleaning optimization scenarios

The present study analyzes the soiling losses of a 1 MW photovoltaic system installed in the South of Spain. Both the Levelized Cost of Energy and the Net Present Value are used to compare the convenience of different mitigation strategies. It is found that also photovoltaic installations located in moderate regions, where the yearly soiling losses are limited to 3%, can suffer of a severe seasonal soiling, with power drops higher than 20%. In these conditions, an optimized cleaning schedule can be considerably beneficial from an economic perspective. For the given site, an optimal cleaning schedule generates a raise in profits up to 3.6% if one yearly cleaning is performed within a ±31-day window in summer. The convenience of one and multiple cleaning strategies is investigated by considering variable electricity prices and cleaning costs. In addition, the impact of the module efficiency on the cleaning strategy is analyzed. It is found that an optimized cleaning schedule can enhance the benefits of installing high efficiency modules, as it increases the amount of energy recovered through each cleaning and, therefore, the profits.

Oil Refinery Heat Exchanger Network Cleaning Scheduling Strategy with Unit Cleanability Consideration

Heat exchanger networks (HENs) play an important role in the chemical industries. Unfortunately, fouling is inevitable in heat exchangers operation. Therefore, the optimal cleaning procedure is required to restore heat exchangers' performance periodically. A systematic cleaning scheduling strategy for the heat exchanger network in an oil refinery is proposed in this work. There are 11 operating heat exchangers in an oil refinery to be reviewed. Different cleaning decision scenarios based on the overall heat transfer coefficient are explored for optimal cleaning schedule performance. The daily number of exchangers available to be cleaned i.e., the unit cleanability, is investigated while minimizing the energy consumption and the additional heat requirement due to the offline heat exchanger under cleaning procedure. The HEN performance and the energy-saving from the cleaning procedures are benchmarked with the uncleaned HEN. The results indicate that the cleaning procedure significantly increases the HEN performance and simultaneously reduces the heat requirement if compared to the untreated HEN benchmark. The possible conflicting situation is discussed when some heat exchangers are waiting to be cleaned due to the unit cleanability restriction, which allows the overall heat transfer coefficient to be below the allowed limit. Therefore, nonconflicting cleaning scheduling is also addressed in this work by relaxing the unit cleanability limit. Furthermore, the optimal cleaning schedule is also suggested for user reference. In this work, the optimum cleaning schedule with minimum energy consumption and maximum energy saving could be achieved when cleaning decision limit is set at 40% decrease of overall heat transfer coefficient. In the contrast, the lowest number of cleaning procedures is associated with 90% decrease in the overall heat transfer coefficient as the cleaning decision limit.

Optimization of cleaning strategies for heliostat fields in solar tower plants

The reduction due to soiling of the optical efficiency of the heliostats in the solar field is a significant detrimental factor in concentrating solar power (CSP) plants. Artificial cleaning is required to maintain acceptable values of optical efficiency, especially in those areas where CSP tends to be economically viable, i.e. where the yearly available DNI is high and rain is scarce. The optimization of the cleaning activities is then a fundamental step to properly balance the operation and maintenance (O&M) costs of the plant with the revenue losses due to soiled heliostats. In this work the best cleaning schedule for a given solar field is computed through a mixed integer linear programming (MILP) model and compared with the results of a heuristic approach. The optical efficiency reduction is assessed for each sector of the solar field through a physical model. The MILP model accounts for the soiling impact and finds the most economical solution in terms of cleaning trucks number and number of cleanings. The optimal cleaning schedule for each sector of the solar field is obtained by minimizing the total cleaning cost (TCC), which is the sum of direct cleaning costs and monetized losses due to soiling. A few test cases are evaluated to demonstrate the strength and the applicability of the developed algorithm. The TCC improvements span between 0.7% and 19.6%, depending on the different scenarios and cost structures considered. For the case studies considered, the savings due to the MILP optimized cleaning strategy were between 927 kAU$/yr and 4744 kAU$/yr (575 k€/yr and 2941 k€/yr).

Optimal Cleaning Scheduling for Photovoltaic Systems in the Field Based on Electricity Generation and Dust Deposition Forecasting

This article focuses on cleaning scheduling for photovoltaic (PV) systems in the field. A method to design cleaning schedules for PV strings is proposed to optimize the profit of a PV system over an extended period. Previous studies have usually focused solely on the cleaning frequency on experimental systems. Our method designs specific cleaning schedules for PV systems in the field based on the forecasting of environmental conditions, PV power generation, and dust deposition. We propose that the cleaning decision is an infinite scheduling problem. A problem model in the infinite time horizon is developed and a periodic scheduling method based on forecasting is proposed to transform the infinite process to an optimizable problem. The cleaning schedules are amended with time as the forecasting accuracy for future times increases with time. The performance of our proposed method that operates under various conditions is evaluated in a case study and compared with two classical scheduling methods.

Performance Analysis of Photovoltaic Module by Dust Deposition in Western Rajasthan

Background: Dust soiling is a main contributing environmental issue with the climatic condition of Western Rajasthan that influences adversely the performance of solar PV technology. The deposition of dust particles reduces the solar output by hampering the solar radiation. Objectives: To explore the trend of dust soiling rate over different seasons in a year and its impact on the performance of solar photovoltaic (PV) system in Western Rajasthan in India. Especially, dust particle settlement is primarily affecting the optical properties of the PV modules, which results in declining their output electrical energy yield. Thus, it adds significantly to quantify the effect of soiling by measuring transmittance losses and soiling ratio with different dust particles. Methods: This article presents a significant loss in output energy yield of photovoltaic system with the layer by layer deposition of the different dust particles in Western Rajasthan. In the context of physical properties, the XRD and SEMEDS analysis of the collected dust sample from SOTA is carried out, and also the dust particle size study has been performed in order to observe the sediment characteristics of various dust samples in Western Rajasthan. The preliminary results depict the optical (glass transmittance) and electrical (soiling ratio) parameters strongly depend on the dust properties and its deposition density. Findings: With this consideration, the highest power loss is observed in the Bikaner dust sample (S1) due to its high transmittance loss (96.1%) and whiles the minimum for Barmer (78.4%). Applications: This study can be useful to guide the appropriate level of dust soiling mitigation on solar PV systems in an arid or particular region. It can also be decided on the optimal cleaning schedule in actual conditions in the year. Keywords: Dust Soiling, Desert, PV Module, Air Pollution, Solar Energy, Transmittance, Soiling Ratio.

Detection of cleaning interventions on photovoltaic modules with machine learning

Soiling losses are a major concern for remote power systems that rely on photovoltaic energy. Power loss analysis is efficient for the monitoring of large power plants and for developing an optimal cleaning schedule, but it is not adapted for remote monitoring of standalone photovoltaic systems that are used in rural and poor regions. Indeed, this technique relies on a costly and dirt sensitive irradiance sensor. This paper investigates the possibility of a low-cost monitoring of cleaning interventions on photovoltaic modules during daytime. We believe that it can be helpful to know whether the soiling is regularly removed or not, and to decide if it is necessary to carry out additional cleaning operations. The problem is formulated as a classification task to automatically identify the occurrence of a cleaning intervention using a time window of temperature, voltage and current measurements of a photovoltaic array. We investigate machine learning tools based on Logistic Regression, Support Vector Machines, Artificial Neural Networks and Random Forest to achieve such classification task. In addition, we study the influence of the temporal resolution of the signals and the feature extraction on the classification performance. The experiments are conducted on a real dataset and show promising results with classification accuracy of up to 95%. Based on the results, three implementation strategies addressing different practical needs are proposed. The results may be particularly useful for non-governmental organizations, governments and energy service companies to improve the maintenance level of their photovoltaic facilities.

Investigation of soiling effects, dust chemistry and optimum cleaning schedule for PV modules in Lahore, Pakistan

We investigate the photovoltaic (PV) power losses due to soiling for Lahore, Pakistan for solar panels. Optimized cleaning schedules are proposed incorporating the effect of solar panels’ tilt angle and the method (manual vs. automatic) for cleaning. Output power losses and dust accumulation on solar panels were measured at variable tilt angles for a period of 120 days at an open roof top location in Lahore. The relative soiling losses for monofacial vs. bifacial (constructed by stacking two back to back monofacial) solar panels were compared for two different panel orientations, i.e., south faced tilted panels vs. East/West faced vertical panels. We found that the soiling rate for Lahore was consistently around 0.8% per day for 30° tilted panel (for the measurement period between October to January), which is among one of the highest soiling rates reported for various urban locations across South Asian and Gulf regions. A dust accumulation rate of 0.01−0.02mg/cm2 per day was recorded for panels that were fixed at 30° tilt. The variation for soiling/dust deposition rates was found to be negligible for different dry periods spanning between October and January. The chemistry and composition of the dust were analyzed using scanning electron microscopy (SEM), X-ray diffraction (XRD), and, electron dispersive x-ray (EDX) spectroscopy. Large contents of carbon and quartz were found in the dust collected from the samples through EDX and XRD analysis. High carbon contents in the accumulated dust are attributed to air pollutants and could be a contributing factor for the high soiling rate. For manual cleaning, the optimal cleaning schedule was calculated to be about once per week for panels at 30° tilt, and, once every three weeks for panels at 90° tilt.

Online Integration of Optimal Cleaning Scheduling and Control of Heat Exchanger Networks under Fouling

Fouling mitigation is paramount to maintaining the reliable and efficient operation of a heat exchanger network (HEN). From the operational perspective, fouling can be mitigated by changing the flo...

Extracting and Generating PV Soiling Profiles for Analysis, Forecasting, and Cleaning Optimization

The identification and prediction of the daily soiling profiles of a photovoltaic site is essential to plan the optimal cleaning schedule. In this article, we analyze and propose various methods to extract and generate photovoltaic soiling profiles, in order to improve the analysis and the forecast of the losses. New soiling rate extraction methods are proposed to reflect the seasonal variability of the soiling rates and, for this reason, are found to identify the most convenient cleaning day with the highest accuracy for the investigated sites. Also, we present an approach that could be used to predict future soiling losses through the implementation of stochastic weather generation algorithms whose ability to identify in advance the best cleaning schedule is also successfully tested. The methods presented in this article can optimize the operation and maintenance schedule and could make it possible, in the future, to predict soiling losses through analysis based only on environmental parameters, such as rainfall and particulate matter, without the need of long-term soiling data.

Integration of optimal cleaning scheduling and control of heat exchanger networks under fouling: MPCC solution

Fouling is a major source of energy inefficiencies that decreases the performance of process units. Fouling mitigation alternatives are required to ensure a sustainable, profitable, and safe operation. In heat exchanger networks, two mitigation alternatives are: flow distribution control, and periodical cleanings. This paper addresses the optimal control and optimal cleaning scheduling, individually and simultaneously, for heat exchangers in refining operations. The problem is formulated as a MINLP, which uses an accurate model of the process, logic disjunctions, and a continuous time representation. To solve it efficiently, it is reformulated as a mathematical program with complementarity constraints (MPCC). The modelling and solution strategies are demonstrated in several case studies, from small to moderately large networks for realistic applications. The formulation is versatile, and large network problems are solved in a reasonable computational time. The integration of optimal scheduling and control decreases the operational cost substantially relative to independent mitigation alternatives.

Integration of Optimal Cleaning Scheduling and Control of Heat Exchanger Networks Undergoing Fouling: Model and Formulation

The performance and operability of heat exchanger networks (HENs) is strongly affected by fouling, which involves the deposition of unwanted material, which reduces the heat-transfer rate and incre...

Optimization of scheduled cleaning of fouled heat exchanger network under ageing using genetic algorithm

The problem of heat exchanger network (HEN) fouling and ageing has been a costly feature of chemical and petrochemical industries, particularly, the oil refineries. In this paper, a strategy to determine the optimal cleaning schedule of the crude oil refinery's preheat train (PHT) under fouling and different ageing scenarios (i.e., slow and fast ageing) is addressed by a specialized Genetic Algorithm (GA) which uses an array of discrete and positive integers as chromosome representation. Specifically, the effect of key cleaning decision factors (i.e., which, when and how to clean), dictated by the GA, on optimal cleaning schedules based on both thermal and economic behavior of the PHT was studied. The results highlight the efficacy of the GA and a novel chromosome application to the optimization of complex cleaning scheduling. Considerable savings are achieved by the implementation of optimum cleaning schedules for all the ageing scenarios compared with un-optimized cleaning or no cleaning performed in a case study involving a 14-unit crude oil refinery HEN. Fast ageing led to significant reduction in energy losses, and hence, improved heat recovery and fewer number of cleanings due to the higher formation rate of a more thermally conductive coke, compared with slow ageing.

Thermal and hydraulic impacts consideration in refinery crude preheat train cleaning scheduling using recent stochastic optimization methods

Fouling in heat exchanger network (HEN) in a refinery has been identified as a major obstacle for efficient energy recovery. Fouling causes loss in efficiency over the time, additional pumping cost and loss of production due to additional downtime. A complex crude preheat train (CPT) in a petroleum refinery was chosen in this study to represent an industrial HEN experiencing severe fouling and huge economic losses due to fouling issues. The objective of this study was to develop a realistic cleaning schedule optimization problem. An improved optimization problem for the cleaning schedule of the heat exchangers in the CPT was developed which takes into account the hydraulic impact of fouling through the additional pressure drops. The problem fall into the MINLP class, which is very complex and finding the global optimum is a challenging task. Hence, the recent stochastic methods are proposed and used to solve the MINLP problem without introducing any approximations or simplifying assumptions. Optimizations were performed over an operating period of 44months following crude slate variations and operating conditions of the refinery. The solution provided by recent stochastic algorithms is global optimum solution. The results show that ignoring the additional pumping cost in the objective function resulted in an optimal cleaning schedule that provides a less savings (18.09% of maximum potential savings) in the net loss compared to the optimal cleaning schedule that utilizes the additional pumping cost in the objective function, which increases about 19.34% of maximum potential savings.

Simulation of heat exchanger network (HEN) and planning the optimum cleaning schedule

Modeling and simulation of heat exchanger networks for estimating the amount of fouling, variations in overall heat transfer coefficient, and variations in outlet temperatures of hot and cold streams has a significant effect on production analysis. In this analysis, parameters such as the exchangers’ types and arrangements, their heat transfer surface areas, mass flow rates of hot and cold streams, heat transfer coefficients and variations of fouling with time are required input data. The main goal is to find the variations of the outlet temperatures of the hot and cold streams with time to plan the optimum cleaning schedule of heat exchangers that provides the minimum operational cost or maximum amount of savings. In this paper, the simulation of heat exchanger networks is performed by choosing an asymptotic fouling function. Two main parameters in the asymptotic fouling formation model, i.e. the decay time of fouling formation (τ) and the asymptotic fouling resistance (Rf∞) were obtained from empirical data as input parameters to the simulation relations. These data were extracted from the technical history sheets of the Khorasan Petrochemical Plant to guaranty the consistency between our model outputs and the real operating conditions. The output results of the software program developed, including the variations with time of the outlet temperatures of the hot and cold streams, the heat transfer coefficient and the heat transfer rate in the exchangers, are presented for two case studies. Then, an objective function (operational cost) was defined, and the optimal cleaning schedule of the HEN (heat exchanger network) in the Urea and Ammonia units were found by minimizing the objective function using a numerical search method. Based on this minimization procedure, the decision was made whether a heat exchanger should be cleaned or continue to operate. The final result was the most cost effective plan for the HEN cleaning schedule. The corresponding savings by implementing the optimum cleaning schedules are reported.