No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Experiment and Prediction Analysis of Thermal Energy Storage for Heat Load Balancing in Domestic Hot Water System
Ice storage systems can be used as an efficient cooling source during summer, as well as a heat source for heat pumps during winter. The non-linear behavior of the heat exchange process in storage makes formulations for optimizing the design and operation of these technologies complex. In this work, we propose a quadratically-constrained mixed-integer programming formulation, that can capture the latent and sensible behavior of the storage and its impact on the delivery of heating and cooling. A building demonstrator integrating an ice storage device was used as a case study. Monitoring data were used to validate the simplified ice storage model employed in the optimization. Results showed that the most common optimal storage cycle requires freezing the water during late winter and when the air temperature falls below 0 ∘C. Increasing the storage volume increases both storage efficiency and the amount of free cooling available during summer. For these reasons, economies of scale can make larger systems more competitive than smaller ones. Storage size and thermal insulation level affect the duration of the charging and discharging phases. Thermal insulation improves seasonal efficiency and free cooling significantly. A higher CO2 emissions price does not yield significant benefits in terms of emissions reductions. High investment costs and the seasonal variation in CO2 intensity of electricity reduce the economic and environmental competitiveness of long-term ice storage systems, respectively.
Thermal energy storage (TES) has been widely applied in buildings to shift airconditioning peak loads and to reduce operating costs by using time-of-use (ToU) tariffs. Meanwhile, TES control strategies play a vital role in maximizing the benefits of their application. To this end, an optimization framework that integrates data-driven cooling load prediction model, system physical model, and advanced optimization algorithm was proposed and applied to a district cooling system (DCS) coupled with an ice-based TES in Beijing, China. Operational strategy of the DCS was optimized based on the predicted cooling load to minimize operating cost under the current ToU tariff. The superior economic performance of the proposed optimization framework was verified by comparing it with two conventional operational strategies-that is, the optimal strategy reduced the operating cost over a two-month cooling period by approximately 8 %. Furthermore, the robustness of the proposed framework to weather forecast uncertainty was tested using three hypothetical weather forecasts of different accuracies as inputs. Although the operating cost decreased with the cooling load prediction accuracy, a diminishing return was evident. The impact of the ToU tariff adjustments was also considered for long-term strategic planning. It was evident that under the current ToU tariff structure, each 0.1 CNY increase in the off-peak rate led to a 9.2 % reduction in absolute cost-savings for the TES system over the regular system. The findings of this study have important implications for optimizing the application of TES for better building energy management and flexibility.
High-temperature reservoir thermal energy storage (HT-RTES) has the potential to become an indispensable component in achieving the goal of the net-zero carbon economy, given its capability to balance the intermittent nature of renewable energy generation. In this study, a machine-learning-assisted computational framework is presented to identify HT-RTES site with optimal performance metrics by combining physics-based simulation with stochastic hydrogeologic formation and thermal energy storage operation parameters, artificial neural network regression of the simulation data, and genetic algorithm-enabled multi-objective optimization. A doublet well configuration with a layered (aquitard-aquifer-aquitard) generic reservoir is simulated for cases of continuous operation and seasonal-cycle operation scenarios. Neural network-based surrogate models are developed for the two scenarios and applied to generate the Pareto fronts of the HT-RTES performance for four potential HT-RTES sites. The developed Pareto optimal solutions indicate the performance of HT-RTES is operation-scenario (i.e., fluid cycle) and reservoir-site dependent, and the performance metrics have competing effects for a given site and a given fluid cycle. The developed neural network models can be applied to identify suitable sites for HT-RTES, and the proposed framework sheds light on the design of resilient HT-RTES systems.
Heat demand forecasting is in one form or another an integrated part of most optimisation solutions for district heating and cooling (DHC). Since DHC systems are demand driven, the ability to forecast this behaviour becomes an important part of most overall energy efficiency efforts.
Low temperature district heating (LTDH) can increase the integration of renewable and waste energy sources that have lots of reliability and acceptance issues. This article aims to identify necessary measures to enable connection of the distributed heat sources into the LTDH. An energy balance model for an area was developed on annual level. The heat supply model included a central plant and distributed plants. A multiobjective, genetic optimization was used. The results showed that the investment cost, electricity price, and heat pump performance influenced higher share of the distributed energy sources into the central district heating (DH) system. Further, bigger building area showed to be more suitable to export heat to the central DH system.
In this study, an alternative bend design is proposed to overcome the aging problem in piping bends. In this design, the foam pad is not included. Finite element analysis was performed based on the total pipe diameter. From this analysis, the shape of the Shear Control Ring (SCR) was determined. Temperature, stress, and other data of the proposed reinforced pipe were acquired and analyzed after the test was performed. The value of the thermal stress for the reinforced steel pipe satisfied the required standard without the foam pad based on the manufacturing of the reinforced fitting and construction site of the test. The reinforcement provided a shear strength level for the foam pad that resulted in maximum shear stress less than stress based on the original foam pad applied at the pipe bend . Additionally, an increasing factor of safety effect for the reinforced fitting application was discovered.
In the present study, we have developed an optimal heat supply algorithm which minimizes the heat loss through the distribution pipe line in a group energy apartment. Heating load variation of a group energy apartment building according to the outdoor air temperature was predicted by a correlation obtained from calorimetry measurements of all households in the apartment building. Supply water temperature and mass flow rate were simultaneously controlled to minimize the heat loss rate through the distribution pipe line. A group heating apartment building located in Hwaseong city, Korea, which has 1473 households, was selected as the object building to test the present heat supply algorithm. Compared to the original heat supply system, the present system adopting the proposed control algorithm reduced the heat loss rate by 10.4%.
High level of accuracy in forecasting heat demand of each district is required for operating and managing the district heating efficiently. Heat demand has a close connection with the demands of the previous days and the temperature, general demand forecasting methods may be used forecast. However, there are some exceptional situations to apply general methods such as the exceptional low demand in weekends or vacation period. We introduce a new method to forecast the heat demand to overcome these situations, using the linearities between the demand and some other factors. Our method uses the temperature and the past 7 days` demands as the factors which determine the future demand. The model consists of daily and hourly models which are multiple linear regression models. Appling these two models to historical data, we confirmed that our method can forecast the heat demand correctly with reasonable errors.
The optimal operation of thermal energy storage in modern district heating networks is crucial for enhancing the system efficiency and economics to better suit the increasing energy demand, especially since thermal storage is lacking in these networks in France. This necessitates accurate and reliable tools that account for the non-linearities and thermal inertia of the system. Most of the proposed optimization schemes in the literature intentionally diminish those non-linearities by linearizing the problem to reduce the computational burden at the expense of the recorded accuracy. In the present study, a heat production plant, comprising biomass and backup gas boilers, as well as thermal energy storage, is examined. The storage unit is a stratified hot water tank, simulated using a multi-layer model, which was validated against experimental observations. The hourly flow rates of different connections of the system are optimized via constrained non-linear programming to minimize the operating costs over several days ahead. A novel partial optimization framework is proposed, which offers equal or even better-quality solutions, compared to the conventional full optimization technique, while cutting down the computational costs by up to 95% and scaling well to longer optimization periods. Thermal storage was found to shift and shave the peak heat production profiles, hence reducing or canceling the reliance on the backup boiler. For most days, the use of thermal energy storage enabled reducing operating costs by up to 21.77%. However, during the annual peak-demand days, it increased the operating costs by 1.9%, compared to a system without storage, due to the relatively small storage volume and low initial temperature. The proposed cost-effective approach enables optimizing these system parameters and oth2ers in future works with a reasonable computational burden. Besides, the proposed framework can be easily opted for other energy sources.
Planning and managing the operation of cogenerative plants (CHP) is increasingly becoming an industrial necessity because of the participation of CHP plants in the day-ahead energy market and the need to deal with heat demand fluctuations. Short-term operational planning usually considers power and heat demand forecast, which can widely fluctuate daily and seasonally, to maximise the net revenue and fulfil the heat requirements. In this framework, a Thermal Storage (TS) can balance day-night fluctuations due to outdoor temperature, as well as unexpected energy surplus and deficiencies caused by heat demand forecast errors, thus giving the CHP plant more operational flexibility. In this paper, the capability of a TS to balance thermal load fluctuations and forecast errors is investigated when a Machine Learning (ML) thermal load forecast for short-term predictions up to 48 h is used. The TS is modelled as a layered storage tank with perfectly mixed layers. Weather and consumption data from 2018 to 2020 related to a large greenhouse powered by a CHP plant located in Tuscany were used as a case study to perform the training and validation of the forecast model as well as to analyse the TS capability in balancing fluctuations with volumes ranging from 500 up to 10,000 m 3. Several ML algorithms were used and compared against a naive prediction based on load persistency. Support Vector Machine (SVM) resulted as the best-performing algorithm. Using SVM leads to better exploitation of the TS capacity, compared to persistence, leading to a more regular State of Charge (SOC) trend and allowing the system to operate within expected conditions up to 80 % of the year. In contrast, a more naïve forecast approach brings to relevant volume size increase to achieve equal performance. Finally, a more accurate forecast reduced the TS size to 50 %, potentially cutting the investment and operational costs compared to the load persistency forecast strategy.
The district cooling system (DCS) is generally considered to be a key efficient cooling technology to cope with the increasing cooling demand. However, the energy consumption of pumps in DCS accounts for a large proportion. This work demonstrated a novel variable flow control strategy based on measured operating parameters. An existing DCS in a subtropical climate zone was used as the baseline model, denoted as Case A. Furthermore, using TRNSYS software, the variable flow control of the chilled water pumps and the cooling water pumps in Case A was performed in turn, and two variable flow models were established, denoted as Case B and Case C, respectively. The operation strategies capable of dealing with the diversified cooling loads were developed and simulated. The results indicated that compared with Case A, the newly proposed models Case B and Case C reduced the energy consumption by 9.50% and 14.15%, cut down CO2 emissions by 9.45% and 14.18%, saved costs by 10.34% and 15.16%, respectively. Moreover, the efficiency of the two models was also improved by 10.70% and 16.67%, respectively, compared with the Case A. Effective operation strategies can further improve the performance of DCS, thereby providing energy-saving and efficient cooling solutions.
This paper presents the numerical simulations and machine learning-based prediction of the transient melting
process of phase change material (PCM) in latent heat thermal storage (LHTS) units. The storage units are
rectangular enclosures equipped with fins of different heights and numbers. For all enclosures, the volume of fins
and PCM are kept constant. Melting processes of PCM in different storage units are simulated using computational
fluid dynamics (CFD) to determine the impacts of fin parameters on the thermal behavior of the LHTS unit.
Transient variation of liquid fraction and stored energy in the different storage units are obtained. Then, the
group method of data handling (GMDH) type of artificial neural networks (ANNs) is employed and trained
through numerical findings to develop correlations for predicting the instantaneous liquid fractions and stored
energy in the finned enclosures. To evaluate the effectiveness of the prediction model, mean square, root mean
square, and standard deviation errors as well as correlation coefficient have been calculated and proved the
accuracy of the proposed correlations.
Heating load prediction based on machine learning algorithms has received increasing attention, especially the Long Short Term Memory (LSTM) network, have been shown to have a superior performance in predicting the heat load consumption. However, most of the current research reports on load prediction models using LSTM models are focused on the unidirectional (Uni-LSTM) network. In this paper, a bidirectional (Bi-LSTM) network for heat load prediction is proposed to make full use of the model hyperparameters to obtain the optimal model and to fully compare with the Uni-LSTM model, and the Bi-LSTM model can improve the prediction accuracy of heat load in a district heating system by using both past and future weather information. In addition, the two types of models are set up with different depth-stacked layers, and for each of the proposed models, a hyperparametric optimization tool has been used to obtain the best model. The results indicate that the increase in depth-stacked LSTM layers has no significant improvement in the prediction accuracy. The input time series length reflects the inertia influence duration of the district heating system, and the optimal model can be obtained for different settings of input time series length. The best optimally configured models were compared, and the single-layer Bi-LSTM model outperformed the single-layer Uni-LSTM model by 19.56%, 16.43%, 14.16%, and 20.69% in terms of the Root Mean Square Error (RMSE), Mean Average Percentage Error (MAPE), Mean Absolute Error (MAE), and the coefficient of variation of the RMSE (CV-RMSE), respectively.
District heating has been widely implemented in residences because of its environmental and economic advantages. The hot water load of residences generally peaks in the morning and at night because of the lifestyle of the residents. An imbalance in the heat load presents high capital costs and low operating efficiency of the district heating facilities. In this study, a latent heat storage system using a solid-liquid phase change material was developed for the peak load shifting control of hot water supplied from district heating in an apartment complex. The latent heat storage system was designed using a shell-tube structure and can store heat for more than 100 h using an insulation of 50 mm. A field demonstration was performed by installing the developed latent heat storage system in a machinery room in the basement of the apartment complex. The demonstration showed that the total heat flow of hot water during a day and its standard deviation with the latent heat storage are lower than those without latent heat storage by 2.69% and 59.9%, respectively. Therefore, the latent heat storage system is crucial in the peak load shifting of hot water in the apartment complex.
In this study, a renewable energy-assisted poly-generation system with energy storage systems has been investigated with a newly developed integrated operation strategy (IOS). The IOS was designed to simultaneously decide the operation of the energy generation unit and energy conversion systems by considering the state of electrical and thermal storage systems. Subsequently, the developed strategy was applied to the poly-generation system for hospital building applications and it was compared with the system under typical operation strategies, including the minimum distance (MD) strategy and the following maximum electrical efficiency load (FML) strategy. A simulation model with validated and verified components was developed, and the performance of the system was evaluated in terms of multi-criteria (energy, environment, and economy). The results showed that poly-generation systems with the new operation strategy showed the best performance among those using the different strategies; 30.79% of primary energy saving ratio, 28.35% of carbon dioxide emission reduction ratio, and 36.86% of operating cost saving ratio with optimal operation variables, as compared to the conventional system. The results of the economic feasibility study indicated a payback period of 6.6 years for system under IOS, which was 1.4 and 3.9 years less compared to the payback period of the systems under MD and FML, respectively. The IOS flexibly controlled the electric cooling/heating ratio and prime mover operation to prevent energy storage systems from overcharging and over-discharging, which led to maximum utilization of the energy produced by the renewable energy system.
The article concerns a theoretical investigation on the ground thermal energy storage (GTES) applied to a low-temperature district heating (LTDH) distribution network. A general formula describing the time-dependent GTES temperature variation was introduced. The highest values of the GTES operating parameters were obtained for the second model. The ground temperature of a varying liquid fraction content (0, 0.1, 0.2) reached a value of 33.42 °C, 33.29 °C and 33.07 °C, respectively. After the GTES was applied, the heat flux transferred to the end-users increased by 14.5 kW. Deactivation of the GTES caused that the heat flux transferred to the consumers had a constant value of 149.5 kW. The LTDH heat supplier was able to provide the heat flux which decreased in time by 8.26% when the GTES was activated. The impact of the GTES insulation on the ground temperature was studied. The results have shown that reducing the insulation by 1/3 results in the GTES temperature decrease by about 3%. It was also found out that inappropriate GTES operation mode may cause larger thermal depletion of the GTES.
The realization of refined management in the heating station can not only meet the comfortable indoor, but also improve the energy efficiency, reduce the heating consumption, and alleviate air pollution. Previous studies ignored the indoor temperature and building thermal inertia (BC), as a result, the prediction models of secondary supply temperature have poor energy saving and thermal comfort. This paper adopts the support vector machine to compare and analyze the influence when adding BC and indoor temperature as input parameters. The results show that with temperature automatic monitor indoor, and when Tout, Tin, Th are taken as input parameters, the maximum error between the actual and predicted is 3% with BC, and 4% without. When there is no temperature monitor indoor and only Tout, Th are taken as input parameters, the BC can be calculated by manual indoor temperature measurement, and the maximum error between the actual and predicted is 3.5% when considering BC, and 4.75% without. To validate the universality of this proposed model, four models are applied to the heat stations in different cities, their performance all show that the BC and indoor temperature has a great impact on the accuracy of predict models, and BC has the greater.
With the development of data-driven techniques, building energy prediction on a district level has attracted increasing attention in recent years for revealing energy use patterns and reduction potentials. However, large-group building data acquisition is more difficult as buildings interact with each other at the city level. To reduce data cost and consider the inter-building impact on the data-driven building energy model, this study proposes a deep learning predictive approach to integrate building networks with a long short-term memory model to create building energy model on a district scale. The building network was calculated via correlations between the energy use intensity of buildings; buildings with high correlation were applied in predictive model to reduce the required number of buildings. To validate the proposed model, five typical building groups with energy use from 2015 to 2018 on two institutional campuses were selected to develop and test the proposed model with the TensorFlow tool. Based on error assessments on the predicted building group energy use, the results suggest that for total building energy use intensity prediction, the proposed model can achieve a mean absolute percentage error of 6.66% and a root mean square error of 0.36 kWh/m2 , compared to 12.05% and 0.63 kWh/m2 for artificial neural network model and to 11.06% and 0.89 kWh/m2 for support vector regression model. Therefore, the proposed model integrating building network and long short-term memory approaches shows good accuracy in predicting building energy use on a district scale.
Optimal operation of district heating (DH) systems usually relies on the forecast of thermal demand profiles of the connected buildings. Depending on the purpose of the analysis, thermal request can be required at various levels, from building level to thermal plant level. In the case of demand response for example, thermal request is necessary at a building level to evaluate its applicability and at a plant level to determine the effects. Thermal request profiles are quite different, depending on the observation point. Total requests are not just the summation of the downstream requests, mainly because of the thermal transients. The heat losses also contributes to modify the curves, although generally in a smaller way. In this work, a multi-level thermal request prediction is proposed. This approach has the aim of evaluating the thermal request in the various sections of DH network with reduced computational resources. This includes a compact model for the prediction of building demand and a network model in order to compose together the requests at the various levels. The application to a portion of the Turin district heating network is proposed. This shows that the network dynamics significantly affects the evolution, especially at peak load.
District heating networks play an important role in the heating and cooling sector, serving up to 60% of the citizens in some countries. The availability of a thermal network supplying multiple users allows producing heat from different sources and multiple technologies. The possibility of relying on different solutions allows the system manager to optimize the heat generation by choosing the best unit for each operation condition. This choice is based on a deep knowledge of heat load profiles, that are related to users’ behavior, network performances and control logics.
Ensemble weather predictions are introduced in the operation of district heating systems to create a heat load forecast with dynamic uncertainties. These provide a new and valuable tool for time-dependent risk assessment related to e.g. security of supply and the energy markets. As such, it is useful in both the production planning and the online operation of a modern district heating system, in particular in light of the low-temperature operation, integration of renewable energy and close interaction with the electricity markets. In this paper, a simple autoregressive forecast model with weather prediction input is used to showcase the new concept. On the study period, its performance is comparable to more complex forecast models. The total uncertainty of the heat load forecast is divided into a constant model uncertainty plus a time-dependent weather-based uncertainty. The latter varies by as much as a factor of 18 depending on the ensemble spread. As a consequence, the total forecast uncertainty varies significantly. The forecast model is applied to the operation of three heat exchanger stations. Applying an optimized temperature control can significantly lower supply temperatures compared to current operation. Improving the temperature control with dynamic time-dependent weather-based uncertainties can lower the supply temperature further and reduce heat losses to the ground. The potential benefit of using dynamic uncertainties is larger for systems with relatively smaller pumping capacities.
This paper defines the concept of 4th Generation District Heating (4GDH) including the relations to District Cooling and the concepts of smart energy and smart thermal grids. The motive is to identify the future challenges of reaching a future renewable non-fossil heat supply as part of the implementation of overall sustainable energy systems. The basic assumption is that district heating and cooling has an important role to play in future sustainable energy systems – including 100 percent renewable energy systems – but the present generation of district heating and cooling technologies will have to be developed further into a new generation in order to play such a role. Unlike the first three generations, the development of 4GDH involves meeting the challenge of more energy efficient buildings as well as being an integrated part of the operation of smart energy systems, i.e. integrated smart electricity, gas and thermal grids.
This paper aims at providing sizing information concerning a thermal storage system in the case of a low
energy consumption building (<50 kW h/m2 y). Numerical simulations for a reference individual building
were run for 23 different cities in Europe. Results show a clear correlation between annual heat demand
Qy and annual heating degree-day (HDD): Qy = f1(HDD). There is also a good correlation between power
demand and HDD. But as heat coverage of the entire heating period may be too ambitious, the analysis
goes further. It presents a correlation between heating demand Q and HDD as a function of the storage
system autonomy t of the form Q = 1 -� exp (-�t/tau) * Qy with tau = f2(HDD). It also gives the absolute
distribution of sorted power demand values as a function of HDD in the form of a power sizing chart.
The purpose of this chart is to provide specifications for the sizing of a thermal storage system intended
to partially cover energy needs and power demand of a low energy residential house.
This article refers to development of ANN modeling for geothermal district heating systems, and the novel and optimal control strategy for exergy efficiency maximization by using these systems. As a real case study, the geothermal district heating system in Afyonkarahisar (AGDHS), Turkey is considered. Its actual thermal data as of average weekly data are collected in heating seasons during the period 2006–2010 for ANN model and control strategy. In this study, a novel control strategy-based PID controller is proposed to ensure the maximum exergy efficiency via the flow rate control of the AGDHS. ANN model of the AGDHS is used as a test system to demonstrate the effectiveness of the proposed control strategy under various operating conditions. The results of this study show that the network yields a good statistical performance with respect to maximum correlation coefficient (0.9986) with minimum RMSE and MAPE values. Moreover, the proposed PID controller has better control performance compared to the manual control even in the presence of the AGDHS. Energy efficiency and cost saving of the system are increased by 13% by the proposed control strategy. Thus, the proposed control strategy has the potential for creating more comfortable thermal environments for district heating systems.
The development of a mechanistic model for the thermal plant in the district heating system is addressed. Energy balance is composed of the change of enthalpy and the generation of the heat by a fuel combustor, on the basis of several assumptions. Effective thermal capacitance is estimated using the data measured in the heating system of Su-Seo area in Korea, and the validity of the developed model is clearly shown by the comparison of outlet temperatures from the boiler between the model prediction and data measured during the period other than that for the estimation. The model performs satisfactorily, and its usefulness has been illustrated by the application of the model-based controller for the regulation of outlet temperature. The controlled case shows that the temperature is maintained close to the set-point, and the energy saving is expected to be about 10 million USD for a year in the area.
A hybrid method (HM) is proposed for the prediction of convective enthalpy transport and the boiling boundary location. The momentum equation is solved with the SIMPLE procedure, while the energy equation is solved with the Method of Characteristics (MOC) and with the application of the Lagrange interpolation polynomial (LIP). The MOC method is applied on an equidistant grid. The initial values at the starting points of characteristic paths are calculated with the LIP. The accuracy of the method is demonstrated on tests of the transient boiling boundary prediction and the well known problem of a propagating discontinuity. The dependence of the hybrid method on the spatial integration step size and the degree of the LIP is analyzed. The LIP of the third degree is recommended, by which practically “exact” solutions with acceptable number of nodes are obtained.
This paper deals with an artificial neural network (ANN) modeling to predict the exergy efficiency of geothermal district heating system under a broad range of operating conditions. As a case study, the Afyonkarahisar geothermal district heating system (AGDHS) in Turkey is considered. The average daily actual thermal data acquired from the AGDHS in the 2009–2010 heating season are collected and employed for exergy analysis. An ANN modeling is developed based on backpropagation learning algorithm for predicting the exergy efficiency of the system according to parameters of the system, namely the ambient temperature, flow rate and well head temperature. Then, the recorded and calculated data conducted in the AGDHS at different dates are used for training the network. The results showed that the network yields a maximum correlation coefficient with minimum coefficient of variance and root mean square values. The results confirmed that the ANN modeling can be applied successfully and can provide high accuracy and reliability for predicting the exergy performance of geothermal district heating systems.
A prediction of thermal transients in district heating systems is important in order to adjust in an energy efficient manner the loads of heat power plants and pump stations with dynamic consumption needs during changes of the outside air temperature, the wind intensity, the solar radiation, or system start-ups and shut-downs. A model and corresponding computer code are developed with the aim to simulate the thermal transients in district heating systems. They are based on the high-order accurate numerical solution of the transient energy equation, and the hydraulic prediction of pressure and fluid flow rates within the complex pipe network. Thermal transients caused by an increase and decrease of the heat power plant load are simulated for real operating conditions of the district heating system. Predicted temperature front propagations show a good agreement with data measured at three consumer substations located in different parts of the district heating network and at different distances from the heat plant. The developed computational tool provides reliable information about time periods of temperature fronts propagation and heat distribution from the heat source to consumers within the whole district heating network.
In order to improve the operation of district-heating systems, it is necessary for the energy companies to have reliable optimization routines, both computerized and manual, implemented in their organizations. However, before a production plan for the heat-producing units can be constructed, a prediction of the heat demand first needs to be determined. The outdoor temperature, together with the social behaviour of the consumers, have the greatest influence on the demand. This is also the core of the load prediction model developed in this paper. Several methodologies have been proposed for heat-load forecasting, but due to lack in measured data and due to the uncertainties that are present in the weather forecasts, many of them will fail in practice. In such situations, a more simple model may give as good predictions as an advanced one. This is also the experience from the applications analyzed in this paper.
Predicting building energy load is important in energy management. This load is often the result of steam heating and cooling of buildings. In this paper, a data-driven approach for the development of a daily steam load model is presented. Data-mining algorithms are used to select significant parameters used to develop models. A neural network (NN) ensemble with five MLPs (multi-layer perceptrons) performed best among all data-mining algorithms tested and therefore was selected to develop a predictive model. To meet the constraints of the existing energy management applications, Monte Carlo simulation is used to investigate uncertainty propagation of the model built by using weather forecast data. Based on the formulated model and weather forecasting data, future steam consumption is estimated. The latter allows optimal decisions to be made while managing fuel purchasing, scheduling the steam boiler, and building energy consumption.
Current methodologies for the optimal operation of district heating systems use model predictive control. Accurate forecasting of the water temperature at critical points is crucial for meeting constraints related to consumers while minimizing the production costs for the heat supplier. A new forecasting methodology based on conditional finite impulse response (cFIR) models is introduced, for which model coefficients are replaced by coefficient functions of the water flux at the supply point and of the time of day, allowing for nonlinear variations of the time delays. Appropriate estimation methods for both are described. Results are given for the test case of the Roskilde district heating system over a period of more than 6 years. The advantages of the proposed forecasting methodology in terms of a higher forecast accuracy, its use for simulation purposes, or alternatively for better understanding transfer functions of district heating systems, are clearly shown.
Prediction of heat load in district heating systems by Support Vector Machine with Firefly searching algorithm
- Al-Shammari
Stochastic modelling and operational optimization in district heating system
- Arvaston