Search

The chiller energy consumption prediction plays an important role in reducing the building energy consumption. Continuously improving the performance of the prediction model is the key to ensuring the accuracy of the chiller energy consumption prediction. As a result, a new improved ensemble model of the chiller energy consumption prediction is proposed, based on the idea of algorithmic ensemble aided by an attention mechanism (AM). To guarantee that the proposed model can see information from many spatial and structural viewpoints, the improved ensemble model incorporates the benefits of several base prediction methods. The suggested approach chooses crucial features to simplify the model input by utilizing AM, thereby reducing the complexity of the samples. The improved ensemble model was experimentally verified in an actual multi-chiller system, and the results showed that the established model could obtain good prediction results. Finally, the proposed model was compared with existing, well-known prediction models. To evaluate the accuracy of these models, the mean absolute error (MAE), mean absolute percentage error (MAPE), mean square error (MSE), variance accounted for (VAF), and coefficient of determination (R 2) were used as the evaluation indices. With an MAE of 2.858, an MAPE of 1.2755 %, an MSE of 20.1878, a VAF of 93.6058, and an R 2 of 0.9923, the findings demonstrate that the improved ensemble model has a greater prediction accuracy than the other well-known prediction methods. The suggested model enhances the variety of empirical model algorithm libraries. [ABSTRACT FROM AUTHOR]

Mixed refrigerant (MR) Cascade Refrigeration Systems are widely employed in industries requiring both extremely low temperatures and large cooling capacity, such as semiconductors and displays. The composition of the MR has a significant impact on the performance of the cascade refrigeration systems, so optimizing the MR composition is essential. The current studies mainly focus on optimization that changes only the composition ratio with a limited number of refrigerants. Therefore, there may be a configuration and composition ratio that can improve the performance of the system more than existing methods. This study presents the need to find the optimal combination and number of MRs before determining the performance of MRs according to changes in composition. The proposed method selects five types of refrigerants that meet the conditions, then classifies cases according to the number of refrigerants in MR and analyzes the performance of each MR at various HTC evaporation temperatures based on energy and exergy through cycle simulation for each case. The proposed method showed COP up to 14.85 % higher than the conventional method, so it is expected to contribute to improving system performance by finding the optimal refrigerant combination and number before optimizing the composition of the MR.

The refrigerants used to cool the heat generated in semiconductor production are Freons, which are now subject to regulation because of their high ozone depletion potential (ODP) and global warming potential (GWP). Therefore, R-744 is considered an eco-friendly refrigerant. This study experimentally analyzed the effect of the refrigerant charge amount based on the temperature discharged after cooling objects when operating a brine chiller using R-744 under subcritical conditions. The results showed that the optimal charge amount in terms of the coefficients of performance (COP) differed from that in terms of the maximum cooling capacity. In case of the apparatus used in this study, the optimal refrigerant charge amount with maximum cooling capacity was estimated at 8.2 kg when the temperature of the cooling object was in the range of −20 °C and −10 °C and at 9.2 kg when the temperature of the cooling object was 0 °C. However, in terms of COP, the optimal refrigerant charge amount was estimated at 8.2 kg when the temperature of the cooling object was in the range of −20 °C and 0 °C. This difference was attributed to the variation in the required evaporation heat capacity with the target temperature of the cooling object.

Accurate fault diagnosis of heating, ventilation, and air conditioning (HVAC) systems is of significant importance for maintaining normal operation, reducing energy consumption, and minimizing maintenance costs. However, in practical applications, it is challenging to obtain sufficient fault data for HVAC systems, leading to imbalanced data, where the number of fault samples is much smaller than that of normal samples. Moreover, most existing HVAC system fault diagnosis methods heavily rely on balanced training sets to achieve high fault diagnosis accuracy. Therefore, to address this issue, a composite neural network fault diagnosis model is proposed, which combines SMOTETomek, multi-scale one-dimensional convolutional neural networks (M1DCNN), and support vector machine (SVM). This method first utilizes SMOTETomek to augment the minority class samples in the imbalanced dataset, achieving a balanced number of faulty and normal data. Then, it employs the M1DCNN model to extract feature information from the augmented dataset. Finally, it replaces the original Softmax classifier with an SVM classifier for classification, thus enhancing the fault diagnosis accuracy. Using the SMOTETomek-M1DCNN-SVM method, we conducted fault diagnosis validation on both the ASHRAE RP-1043 dataset and experimental dataset with an imbalance ratio of 1:10. The results demonstrate the superiority of this approach, providing a novel and promising solution for intelligent building management, with accuracy and F1 scores of 98.45% and 100% for the RP-1043 dataset and experimental dataset, respectively. [ABSTRACT FROM AUTHOR]

This work proposes a new, efficient, economically, and environmentally viable approach for developing cutting-edge energy systems and assisting the anticipated global green transition with maximal renewable integration. The cogeneration of hydrogen and power is driven by biomass, which in turn drives the vanadium chloride cycle and the proton exchange membrane fuel cells. A cooling absorption unit is powered by waste heat recovered using a passive energy improvement technique to improve performance and cut costs. Energy, exergy, exergo-economic, exergo-environmental impacts, and CO 2 emission rate of the suggested renewable-based model are analyzed using an engineering equation solver tool. Parametric analysis is also used to assess the impact of key operational factors on main performance indicators. With machine learning, a particle swarm method is implemented in MATLAB to find the optimal operating state with high precision and low computing cost. The results show the importance of multi-objective optimization by pointing out a conflicting change in the performance metrics from different angles by picking up the biomass moisture content and fuel cell current density. According to the optimization results, an acceptable total cost, environmental damage effectiveness, and exergy efficiency of 5 $/h, 0.86, and 55% are achieved through the integration of particle swarm optimizer and artificial neural network method. The results further reveal that the gasification temperature is not sensitive; however, changing the fuel cell utilization factor significantly impacts the system's performance from all sides. Finally, the chord diagram of the irreversibility rate indicates that the fuel cell and gasifier have the highest destruction of 6.4 kW and 2.6 kW under the optimum condition, owing to mixing and chemical reactions. As for the environmental aspect, by optimizing the system, the system's CO 2 emission are greatly reduced. • Novel gasification and PEM fuel cell system are suggested. • The thermochemical hydrogen production cycle is added for hydrogen production. • The components' irreversibility is investigated through the Sankey diagram. • Gasifiers and fuel cells are the worst components from the exergy aspect. • The highest efficiency and lowest exergo-environmental factor are 68% and 0.0128. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Refrigeration is the property of Journal of Refrigeration Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

To reduce energy consumption in data center cooling systems and ensure the healthy and rapid development of data centers, this study investigated on the cooling system of a data center in Guangzhou. Aimed at the efficiency variations of chillers at different Partial Load Rates (PLR), a novel model predictive control method was proposed. The study employed Gaussian process regression to establish a chillers' operational efficiency model and utilized Long Short-Term Memory (LSTM) neural networks to create an outdoor wet bulb temperature time series prediction model. The cooling system's chilled water storage and discharge modes, storage and discharge capacity, the number of chillers, and the PLR were adjusted based on the data center cooling load and the prediction of outdoor wet bulb temperature. This adjustment aims to partially decouple the demand and supply sides, resulting in the chillers being operated in high efficiency or turned off, thereby reducing the operational energy consumption of the cooling system. Compared to the traditional rule-based control strategy, the model predictive control strategy reduced the overall energy consumption by 3.13% and operating costs by 3.33%.

The energy consumption of data centers (DCs) is rising year by year, and cooling station accounts for more than 40% of the total DCs' energy consumption, which has a huge energy-saving potential. Building models for whole DCs' cooling station can help predict total power to improve the energy efficiency of the system, before this, establishing a single model for each component is a basic work. This paper mainly studies the chiller and cooling tower models and compares the predictive performance of the empirical model, hybrid model, and neural network model of chillers and cooling towers. Giving the model selection scheme of the chiller and cooling tower for the establishment of the whole system of the refrigeration station. For the chiller model, the Yoshida function model in empirical models has the highest accuracy with a mean square error of 0.0592, followed by the neural network model with a mean square error of 0.2, and the hybrid model has a lower accuracy than the former two models. For the cooling tower model, the empirical models and the neural network model have similar accuracy, and both are higher than the hybrid model. [ABSTRACT FROM AUTHOR]

Owing to the rapid expansion of data science, data-driven methods have emerged as a dominant trend in chiller fault detection and diagnosis (FDD). Most of these methods prioritize feature selection to achieve optimal diagnostic performance. However, on-site research indicates a common installation of a limited number of sensors, coupled with a necessity to minimize diagnostic costs. This discrepancy between existing research's feature selection principles and the current on-site sensor installation status presents a significant challenge. To facilitate the practical implementation of data-driven methods in real chiller units, this study addresses a critical question: under the constraint of limited on-site sensor installations, what is the optimal performance achievable by data-driven methods and their improved versions? To answer this, only features derived from commonly installed sensors on field chillers are chosen as indicators for typical chiller faults. The FDD performance of six frequently used data-driven methods, namely, back-propagation neural network, convolutional neural network, support vector machine, support vector data description, Bayesian network, and random forest, along with their improved versions, is comprehensively evaluated and validated using experimental data, considering four evaluation metrics. The conclusions drawn in this paper provide valuable insights for users/manufacturers with limited or no budget, detailing the best achievable diagnostic performance for each typical fault and offering guidance for those aiming to further enhance FDD performance. [ABSTRACT FROM AUTHOR]

Copyright of Chinese Journal of Refrigeration Technology is the property of Shanghai Society of Refrigeration and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Building automation systems installed in large commercial buildings record sub-hourly measurements from hundreds of sensors. The use of such large datasets are challenging because of missing and erroneous data, which can prevent the development of accurate prediction models of the performance of heating, ventilation, and air-conditioning equipment. The use of the transfer learning (TL) method for building applications attracted researchers to solve the problems created by small and incomplete datasets. This paper verifies the hypothesis that the deep neural network models that are pre-trained for one chiller (called the source chiller) with a small dataset of measurements from July 2013 could be applied successfully, by using TL strategies, for the prediction of the operation performance of another chiller (called the target chiller) with different datasets that were recorded during the cooling season of 2016. Measurements from a university campus are used as a case study. The results show that the initial hypothesis of this paper is confirmed. [ABSTRACT FROM AUTHOR]

Turbine generator (TG) buildings in thermal power plants require chilled water air conditioning systems (CWAS) to control the temperature and humidity of heat dissipated rooms, which are energy-intensive and non-eco-friendly. As an alternative, a novel refrigerant-free 100% fresh air (FA)-based hybrid liquid desiccant air conditioning system (HLDAS) is proposed and presented the methodology to design it for a case study of TG building of 2X660 MW power plant in the most humid climate. The cooling and heating requirements for HLDAS are met with a cooling tower and potential waste heat from the plant. To assess the saving potential of the proposed system, annual energy consumption for the HLDAS, 10% FA-based CWAS (CWAS-1) and 100% FA-based CWAS (CWAS-2) are quantified and compared. The results revealed that HLDAS consume 55% and 31% of CWAS-1 and CWAS-2 power consumption, respectively. Low-grade potential waste heat from the plant is estimated and found to be sufficient for solution regeneration. The proposed system is extremely energy efficient for 100% FA applications where low-grade waste heat is available. In addition to this, since there are no ODP gases associated with the proposed system, the proposed HLDAS is recommendable in view of environmentally friendly technology and indoor air quality (since it is a 100% FA-based system). This work will support HVAC engineers and researchers in designing the LDAS for a given air conditioning application. [ABSTRACT FROM AUTHOR]

Copyright of Chinese Journal of Refrigeration Technology is the property of Shanghai Society of Refrigeration and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Selecting the type of chillers to be used in commercial buildings is critical considering their high energy consumption, which consequently requires the use of chillers with higher efficiency in order to improve the energy efficiency of buildings and reduce their carbon footprint. Through a case study in a commercial resort, this research presents an energy consumption analysis of the replacement of air-cooled chillers with water-cooled chillers. The energy savings were calculated based on not only the simulation software but also the actual energy consumption data after the installation of the water-cooled chillers. The energy analysis indicated that the use of water-cooled chillers to substitute the air-cooled chillers in the resort resulted in energy saving of around 17.5% with a reduction of 2860 tonnes of CO2 equivalent emission per year. Meanwhile, the cost analysis showed that the life-cycle cost of the water-cooled chillers is around 13.4% less. These findings demonstrate the potential of using water-cooled chillers as a replacement for air-cooled chillers in existing commercial buildings to reduce energy consumption and its associated carbon emission without compromising the buildings’ indoor comfort. This option can be an appeal to building owners as it has some evident economic benefits.

The condenser needs to be cleaned so that the performance of the heat exchanger can optimally. This research was conducted on a water cooled chiller machine with a capacity of 650TR and the type of refrigerant used is R134a to determine the performance of the machine after cleaning process. The electric power used by the compressor’s motor before there was a fouling, fouling and after cleaning the condenser tubing were 174,80 kW, 191,22 kW and 174,42 kW respectively. The mass flow rate of cooling water before there is a fouling, fouling and after cleaning the condenser tubing is 98,01 kg/s, 106,74 kg/s and 126,15 kg/s respectively. The actual cooling load before there is a fouling, fouling and after cleaning the condenser tubing is 1352,46 kW, 1343,27 kW and 1363,92 kW respectively. Heat rejected before there is a fouling, fouling and after cleaning the condenser tubing is 1527,26 kW, 1534,49 kW and 1538,34 kW respectively. Coefficient of performance heating before there is a fouling, fouling and after cleaning the condenser tubing is 8,75, 8,04 and 8,82 respectively. Based on the results of the research, the performance of the water cooled chiller machine has increased after cleaning the condenser tubing.

The article is devoted to the study R134a refrigerating machine efficiency and consisting of a centrifugal compressor, a condenser, a temperature-regulating valve and an evaporator. The main purpose of the work is to analyze the centrifugal compressor gas dynamic characteristics effect on the refrigeration machine vapor-compression cycle efficiency. This goal is achieved through the study by actual working process numerical experiment in the refrigeration machine centrifugal compressor with an idealized process for other elements. The object of the study are the refrigeration machine characteristics, expressed by the theoretical refrigeration coefficient COPRt. Single-stage centrifugal compressors with the design conditional flow coefficient in the range from 0.035 to 0.12 are considered. The design of centrifugal compressors was carried out according to a new calculation method to the flow part efficiency increase. The method com-prehensively combines the inviscid and viscous flow calculations with the use of the single-criteria and multiparametric optimization. Previously, the method was tested and compared with experimental data. The most important result is the results of the refrigeration cycle efficiency evaluating through the centrifugal compressors highly efficient flow parts design methodology application. An increase in COPRt was obtained taking into account the centrifugal compressor actual process in the range from 2.6% to 7.2%. The significance of the results obtained lies in the possibility of using high-efficiency centrifugal compressors gas dynamic characteristics for the chillers refrigeration cycles analysis and calculation. The level of the compressors isentropic efficiency ranges from 0.80 to 0.85, depending on the design conditional flow coefficient.

To reduce energy consumption in data center cooling systems and ensure the healthy and rapid development of data centers, this study focuses on the cooling system of a data center in Guangzhou. Addressing the efficiency variations of chillers at different Partial Load Rates (PLR), a novel model predictive control method is proposed. The study employs Gaussian process regression to establish a chillers’ operational efficiency model and utilizes Long ShortTerm Memory (LSTM) neural networks to create an outdoor wet bulb temperature time series prediction model. Based on the data center cooling load and the prediction of outdoor wet bulb temperature, the cooling system’s chilled water storage and discharge modes, storage and discharge capacity, the number of chillers, and the PLR are adjusted. This adjustment aims to partially decouple the demand and supply sides, resulting in the chillers being operated in high efficiency or turned off, thereby reducing the operational energy consumption of the cooling system. To evaluate the performance of this strategy, annual energy consumption simulations were conducted, compared to the traditional rule-based control strategy, the model predictive control strategy reduced the overall energy consumption by 3.13% and operating costs by 3.33%. [ABSTRACT FROM AUTHOR]

Copyright of Revista Inovação, Projetos e Tecnologia is the property of Revista Inovacao, Projetos e Tecnologia and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Selecting the type of chillers to be used in commercial buildings is critical considering their high energy consumption, which consequently requires the use of chillers with higher efficiency in order to improve the energy efficiency of buildings and reduce their carbon footprint. Through a case study in a commercial resort, this research presents an energy consumption analysis of the replacement of air-cooled chillers with water-cooled chillers. The energy savings were calculated based on not only the simulation software but also the actual energy consumption data after the installation of the water-cooled chillers. The energy analysis indicated that the use of water-cooled chillers to substitute the air-cooled chillers in the resort resulted in energy saving of around 17.5% with a reduction of 2860 tonnes of CO2 equivalent emission per year. Meanwhile, the cost analysis showed that the life-cycle cost of the water-cooled chillers is around 13.4% less. These findings demonstrate the potential of using water-cooled chillers as a replacement for air-cooled chillers in existing commercial buildings to reduce energy consumption and its associated carbon emission without compromising the buildings’ indoor comfort. This option can be an appeal to building owners as it has some evident economic benefits. [ABSTRACT FROM AUTHOR]

This paper evaluates the energy operational performance of a system to produce domestic hot water, from the use of waste heat in a centralized air conditioning system in a hotel in Cuba. The hotel facility has two different chillers capacities, where the system recovers the condensed heat, through a heat exchanger at the compression process outlet. These schemes allow the simultaneous production of air conditioning and Domestic Hot Water, also those considering thermal storage tanks. The system modelling starts analyzing its components and their processes, based on the traditional thermodynamics and heat transfer equations and ends with a validation process. This model estimates a wide range of operating conditions, such as thermal demands of the cooling load, the chiller water temperature, the electricity consumed by chillers, and the climatological variables involved. The system is evaluated in the highest tourism season in Cuba during a year under partial load. Results were validated in a hotel as a study case, which allowed us to estimate the thermal potential available in the plant to meet the thermal demand of the hot water required, as well as showing that variations in the chiller operation affect the temperature stability in the hot water system.

A fault diagnosis method of heating, ventilation, and air conditioning (HVAC) systems based on the ReliefF-recursive feature elimination based on cross validation-support vector machine (ReliefF-RFECV-SVM) combined model is proposed to enhance the diagnosis accuracy and efficiency. The method initially uses ReliefF to screen the original features, selecting those that account for 95% of the total weight. The recursive feature elimination based on cross validation (RFECV), based on a random forest classifier, is then applied to select the optimal feature subset according to diagnostic accuracy. Finally, a support vector machine (SVM) model is constructed for fault classification. The method is tested on seven typical faults of the ASHRAE 1043-RP water chiller dataset and three typical faults of an air-cooled self-built air conditioner simulation dataset. The results show that the ReliefF-RFECV-SVM method significantly reduces diagnosis time compared to SVM, shortening it by about 50% based on the ASHRAE 1043-RP dataset, while achieving an overall accuracy of 99.98%. Moreover, the proposed method achieves a comprehensive diagnosis accuracy of 99.97% on the self-built simulation dataset, with diagnosis time the reduced by about 65% compared to single SVM. [ABSTRACT FROM AUTHOR]

Cooling degree-hours (CDH) received the broadest application in evaluation of the ambient air cooling efficiency in power engineering (engine intake air cooling systems) and air conditioning. The current CDH numbers are defined as a drop in air temperature multiplied by associated time duration of performance and their summarized annual number is used to estimate the annual effect achieved due to sucked air cooling in power plants based on combustion engines (fuel saving, power output increment) and in air conditioning (refrigeration energy generation according to needs). A majority of approaches to designing ambient air cooling systems is proceeding from the cooling capacity of the chillers selected to provide a maximum current or annual CDH number with corresponding maximum current or annual effect (additional energy produced or fuel consumption reduction) in site climate location. But such approaches lead to inevitable oversizing the chillers and cooling systems in the whole. The analysis of intake air cooling efficiency in site varying climatic conditions, accompanied by quite a simple numerical simulation, enabled to reveal the potential of its enhancement and evaluate numerically the results of each step of designing in logical sequence. The new approaches to cooling system rational designing were introduced, that enables to synthesize and substantiate innovative principal decisions to exclude unproductive waste of installed (design) cooling capacity in actual operation. The innovative findings of methodological approaches include the use of the rate of annual CDH number increment as an indicator for selecting the optimum and rational values of design cooling capacity. The optimum cooling capacity corresponds to maximum rate of summarized annual CDH increment and maximum level of thermal loading accordingly, which provides minimum sizes of the chiller. In reality, it is a minimum permissible value of cooling capacity of the chiller installed and the overall ambient air cooling system. The rational cooling capacity, that enables to achieve practically maximum value of annual CDH and avoid chiller oversizing, is determined as the second, local, maximum of the rate in the summarized annual CDH over the range above the first one, global, maximum. A rational design cooling capacity determined by applying the novel methodology allows to decrease the ambient air cooling system sizes by 15 to 20% compared with traditional designing issuing from the peaked thermal load during a year. With this practically a maximum annual effect in fuel saving (energy generation or others) can be achieved too. [ABSTRACT FROM AUTHOR]

Computer-empowered detection of possible faults for Heating, Ventilation and Air-Conditioning (HVAC) subsystems, e.g., chillers, is one of the most important applications in Artificial Intelligence (AI) integrated Internet of Things (IoT). The cyber-physical system greatly enhances the safety and security of the working facilities, reducing time, saving energy and protecting humans’ health. Under the current trends of smart building design and energy management optimization, Automated Fault Detection and Diagnosis (AFDD) of chillers integrated with IoT is highly demanded. Recent studies show that standard machine learning techniques, such as Principal Component Analysis (PCA), Support Vector Machine (SVM) and tree-structure-based algorithms, are useful in capturing various chiller faults with high accuracy rates. With the fast development of deep learning technology, Convolutional Neural Networks (CNNs) have been widely and successfully applied to various fields. However, for chiller AFDD, few existing works are adopting CNN and its extensions in the feature extraction and classification processes. In this study, we propose to perform chiller FDD using a CNN-based approach. The proposed approach has two distinct advantages over existing machine learning-based chiller AFDD methods. First, the CNN-based approach does not require the feature selection/extraction process. Since CNN is reputable with its feature extraction capability, the feature extraction and classification processes are merged, leading to a more neat AFDD framework compared to traditional approaches. Second, the classification accuracy is significantly improved compared to traditional methods using the CNN-based approach.

This paper evaluates the operational energy performance of a system to produce domestic hot water using waste heat in a centralised air conditioning system in a hotel in Cuba. The hotel facility has two different chillers capacities, where the system recovers the condensed heat through a heat exchanger at the compression process outlet. These schemes allow the simultaneous production of air conditioning and Domestic Hot Water, also those considering thermal storage tanks. The system modelling starts analyzing its components and processes based on the traditional thermodynamics and heat transfer equations and ends with a validation process. This model estimates a wide range of operating conditions, such as the thermal demands of the cooling load, the chiller water temperature, the electricity consumed by chillers, and the climatological variables involved. The system is evaluated in the highest tourism season in Cuba during a year under partial load. Results were validated in a hotel as a study case, which allowed us to estimate the thermal potential available in the plant to meet the thermal demand of the hot water required and show that variations in the chiller operation affect the temperature stability in the hot water system. [ABSTRACT FROM AUTHOR]

Sensor faults have been observed to negatively impact the operation of the HVAC system. Among these faults is the complexity of multi-source sensor faults, which may result in fault confusion due to multiple fault points and different fault patterns. This paper proposes a fault diagnosis model applicable to single- and multi-source faults of HVAC system sensors. Based on the distribution patterns of chillers sensor data, the ensemble empirical mode decomposition soft threshold denoising Gaussian mixture model (EEMDSTD-GMM) is proposed. The study suggests a K-means-based pre-classification method for potentially confusing types of sensor faults. EEMDSTD-GMM-K-means has shown a better fault diagnosis capability under four single-source sensor faults and five multi-source sensor faults. Under the three examined fault levels, the results indicate a satisfactory performance with an average diagnosis rate of 98.7% for single-source faults and 96.5% for multi-source faults. [ABSTRACT FROM AUTHOR]

This paper presents the development and use of benchmarking grey-box models for the detection and diagnosis of multiple-dependent faults (MDFDD) of a water-cooled centrifugal chiller. Models are developed using data recorded by a Building Automation System (BAS) from a central cooling plant of an institutional building. The forward residual-based fault detection model identifies a fault symptom, when the difference between the measured value of target variable and benchmarking value exceeds the corresponding threshold. For the fault diagnosis, most publications start from a known single fault and establish the impact on following variables in the system. This paper presents a rule-based backward approach. The proposed method identifies if (i) the fault symptom is correct (i.e., a variable has abnormal values), or (ii) the fault symptom is incorrect (i.e., the symptom of target variable is caused by impacts generated by other faulty variables due to the dependency between variables), or (iii) both target and regressor variables are abnormal. For testing the proposed MDFDD model, some artificial faults are inserted into the measurement data file, and results are discussed about the method potential for the application. [ABSTRACT FROM AUTHOR]

In this article, energy optimization of the cooling system of IKEA Budaörs is carried out. The cooling system is served by a centrifugal water chiller and includes a large-volume cooling buffer tank. The facility operates the hydraulic system of the buffer storage tank only during the transitional period. The main goal is to reduce energy consumption by changing the operating strategy of the existing system. To test the operating strategies, the operation and the thermal load of the shopping center during the summer season had to be simulated to find the best operation strategy. A hybrid method (real data and calculated values) was used in the simulation. The three operating scenarios examined show that the annual energy consumption and the number of operating hours of the chiller can be reduced by using the buffer tank with the right strategy. In the examined scenarios, a 30% energy improvement was achieved. The possibility of using a buffer tank is significantly limited by the fact that the heat exchangers were sized for low forward water temperatures. By re-sizing the heat exchangers, the utilization of the buffer tank could be considerably improved in conditions close to peak heat load. [ABSTRACT FROM AUTHOR]

The integrated optimisation modelling for the chiller-air handling units system is developed for increasing the efficiency and energy utilisation of system. A building management system in the chillers network controls the cooling load to ensure the specified desired set temperature of the cooling air within the building can be satisfied. Unfortunately, the desired set point temperature of the cooling air is a fixed value and does not vary with the dynamic change of cooling demand with different ambient temperatures. Therefore, the power consumption of the chillers and the building cooling requirement with a wide range of different ambient temperatures is properly modelled by optimising the performance of chillers, air handling units, cooling towers, and water pumps. The linear programming model for the system is established to model a real representation of the chiller-air handling unit system. The result shows that the optimal coefficient of performance is greater by about 7%–10% than the current chiller system. The optimal power consumption of the chiller system reduces to 3%. Overall, the optimal decision solutions could be used as the potential improvement strategy to control the desired set point values in the building management system for efficient chiller-AHU system.

Building automation systems installed in large commercial buildings record sub-hourly measurements from hundreds of sensors. The use of such large datasets are challenging because of missing and erroneous data, which can prevent the development of accurate prediction models of the performance of heating, ventilation, and air-conditioning equipment. The use of the transfer learning (TL) method for building applications attracted researchers to solve the problems created by small and incomplete datasets. This paper verifies the hypothesis that the deep neural network models that are pre-trained for one chiller (called the source chiller) with a small dataset of measurements from July 2013 could be applied successfully, by using TL strategies, for the prediction of the operation performance of another chiller (called the target chiller) with different datasets that were recorded during the cooling season of 2016. Measurements from a university campus are used as a case study. The results show that the initial hypothesis of this paper is confirmed.

Copyright of Chinese Journal of Refrigeration Technology is the property of Shanghai Society of Refrigeration and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Background. Some quality defects can cause changes in the attributes of meat, among which we can highlight PSE (Pale, Soft and Exudative) meats. PSE results from a sudden drop in pH while the carcass is still at an elevated temperature. The identification of PSE meat has previously been done through the measurement of pH and L* value (Luminosity). Studies suggest that a more accurate assessment of the kinetics of pH and temperature decrease needs to be conducted to better understand the etiology of PSE meat in poultry. The objective of this work was to obtain a glycolytic curve for normal meat and chicken PSE through the analysis of pH, L*, and WHC (Water holding capacity). Materials and methods. A glycolytic curve was obtained for normal meat and PSE through pH, L* and WHC. Samples of breast fillets were obtained from carcasses immediately after sampling from the chiller, and the pH, temperature, and L* were measured between the times of 1h35min and 25h35min postmortem (pm). The WHC analysis was performed at 25h35 pm. Results. As of 8h35 pm, the pH values of the meat were found to have stabilized, with the pH of the PSE meat being 5.69 ±0.07, and of the regular meat being 5.93 ±0.09. The final pH (25h35min pm) was 5.98 ±0.06 and the L* 57.30 ±2.39 for normal meat, while the result for the PSE meat was 5.72 ±0.06 for pH and 59.44 ±1.51 for L*. The WHC results showed a difference between the normal chicken fillets and PSE, which were 67.19 ±3.13 and 64.45 ±2.66 respectively. Conclusion. We conclude that under industrial conditions of slaughter, the resolution of rigor mortis occurs 8h35 pm. [ABSTRACT FROM AUTHOR]

Chillers are the important components across various applications, and understanding the impact of common faults on their performance is crucial for monitoring fault progression and predicting degradation. Consequently, an experimental investigation was conducted on a magnetic levitation-driven chiller across different severity levels. The quantitative influence of common faults on performance was then analyzed, with a specific focus on thermodynamic mechanism parameters. From a data variation perspective, a sensitivity analysis was implemented to identify the fault-sensitive parameters that exhibited noteworthy variations due to faults. From a classification performance perspective, the diagnosis-effective parameters were identified based on their significance in fault diagnosis. The results of this study led to several key conclusions: 1) Trends and magnitudes of changes in 38 parameters related to nine typical faults are identified in comparison to their normal levels; 2) Common faults have a significant influence on thermodynamic mechanism parameters, causing big changes in two to eight thermodynamic mechanism parameters for each fault; 3) Discrepancies between fault-sensitive parameters and diagnosis-effective parameters results in a difference of up to 64% in diagnostic performance; 4) Nine decoupling parameters are identified as pivotal in diagnosing concurrent faults, and an effective set of fault diagnosis rules is recommended. These findings provide valuable insights into the selection of sensitive parameters, identification of fault-decoupling parameters, and the development of fully interpretable and reliable fault diagnosis approaches. [ABSTRACT FROM AUTHOR]

This article provides an insight into the numerical simulation of an ammonia-water solution-based plate heat absorber, which is a crucial part of an absorption chiller. These chillers could be used as part of a stratified chilled-water storage tank configuration for coupling to a small modular reactor. The common approach utilized in the process of designing the chiller is to call the absorber a "black box" without considering the processes inside. It follows that the resulting design must be highly oversized to ensure reliable operation of the system. This article focuses on the obstacles that needs to be overcome in order to simulate the processes inside the absorber, such as stream breakup into separate droplets or absorption of ammonia. Unfortunately, we do not have any experimental data we could use to validate our simulations and thus this article should be looked at as a "proof of concept" and a possible guideline for simulations of similar phenomenon. [ABSTRACT FROM AUTHOR]

An appropriate design of a chiller plant is crucial to guarantee highly performing solutions. However, several design variables, such as type of systems, total cooling capacity, and hydraulic arrangement, need to be considered. On the one hand, at present, different technical criteria for selecting the most suitable design variables are available. Studies that corroborate the influence of the design variables over the operational variables are missing. In order to fill this knowledge gap, this work proposes a statistical analysis of design variables in chiller plants operating in medium- and large-scale applications and evaluates their influence on energy consumption and life cycle cost (LCC) under the same thermal demand conditions. A case study involving 138 chiller plant combinations featuring different arrangements and a Cuban hotel was selected. The results suggested that the total chiller design and cooling capacity distribution among chillers have a significant influence on the energy consumption of the chiller plant with a Spearman's Rho and Kendall Tau (τ) correlation index value of −0.625 and 0.559, respectively. However, with LCC, only the cooling capacity distribution among the chillers had a certain influence with a Kendall Tau correlation index value of 0.289. As for the considered total cooling capacity, the applied statistical test showed that this design variable does not have any influence on performing the chiller plant. [ABSTRACT FROM AUTHOR]

Practical applications of data-driven fault detection (FD) are limited by their portability. The costs of model training and validation are extremely high when each system requires a model retrained on its own fault and fault-free data. Therefore, this paper proposes a statistical-based online cross-system FD method to address the problem of model portability. The proposed FD model can be cross-utilized between building chillers with various specifications while it only needs to update the original fault detection model by the normal operation data of the new chiller system, thus saving huge fault experimental costs for the fault detection of new chiller. First, a theoretical basis for the proposed cross-system fault detection method is presented. Then, experiments were conducted on three building chillers with different specifications. Both fault and fault-free data were collected from the three chillers. The development and validation of the proposed cross-system fault detection method are then conducted. Results show that the cross-system fault detection models perform well when used with different chillers. For instance, when the fault detection model of system #1 was cross-utilized to system #2, the detection accuracies of refrigerant leakage, refrigerant overcharge, and reduced evaporator water flow were 99.73%, 90.17%, and 96.94%, respectively. Compared with original models, the detection accuracies were improved by 33.78%, 84.07%, and 65.56%, respectively. Therefore, the proposed cross-system fault detection method has potential for online application to practical engineering FD. [ABSTRACT FROM AUTHOR]

Abstract The performance of a multistage centrifugal compressor is highly influenced by the ambient conditions, especially during the summer seasons; their capacity shrinks and thus the power requirement for compression will increase. The prime cause of these constraints is the interstage cooling limitations. This study simulates various suction conditions of a multistage compressor on Aspen HYSYS® and suggests its debottlenecking by making the suction temperatures comparable to winter seasons. This is achieved by installing an additional exchanger at the downstream of each interstage cooler, cooling down the gas further by using absorption refrigeration chillers. These chillers are powered up by the waste heat recovered from the exhaust steam coming from the prime mover, steam turbine, of the same compressor. This modification will save a considerable amount of power (663 kW), net savings (Gross Savings – OPEX: 72 289 $/y), and reduce the carbon footprint (954 ton/y) of the overall process.

A fault diagnosis method of heating, ventilation, and air conditioning (HVAC) systems based on the ReliefF-recursive feature elimination based on cross validation-support vector machine (ReliefF-RFECV-SVM) combined model is proposed to enhance the diagnosis accuracy and efficiency. The method initially uses ReliefF to screen the original features, selecting those that account for 95% of the total weight. The recursive feature elimination based on cross validation (RFECV), based on a random forest classifier, is then applied to select the optimal feature subset according to diagnostic accuracy. Finally, a support vector machine (SVM) model is constructed for fault classification. The method is tested on seven typical faults of the ASHRAE 1043-RP water chiller dataset and three typical faults of an air-cooled self-built air conditioner simulation dataset. The results show that the ReliefF-RFECV-SVM method significantly reduces diagnosis time compared to SVM, shortening it by about 50% based on the ASHRAE 1043-RP dataset, while achieving an overall accuracy of 99.98%. Moreover, the proposed method achieves a comprehensive diagnosis accuracy of 99.97% on the self-built simulation dataset, with diagnosis time the reduced by about 65% compared to single SVM.