Energy consumption forecasting for buildings has immense value in energy efficiency and sustainability research. Accurate energy forecasting models have numerous implications in planning and energy optimization of buildings and campuses. For new buildings, where past recorded data is unavailable, computer simulation methods are used for energy analysis and forecasting future scenarios. However, for existing buildings with historically recorded time series energy data, statistical and machine learning techniques have proved to be more accurate and quick. This study presents a comprehensive review of the existing machine learning techniques for forecasting time series energy consumption. Although the emphasis is given to a single time series data analysis, the review is not just limited to it since energy data is often co-analyzed with other time series variables like outdoor weather and indoor environmental conditions. The nine most popular forecasting techniques that are based on the machine learning platform are analyzed. An in-depth review and analysis of the ‘hybrid model’, that combines two or more forecasting techniques is also presented. The various combinations of the hybrid model are found to be the most effective in time series energy forecasting for building.

A review on time series forecasting techniques for building energy consumption

A combined model based on CEEMDAN and modified flower pollination algorithm for wind speed forecasting

Energy planning models (EPMs) play an indispensable role in policy formulation and energy sector development. The forecasting of energy demand and supply is at the heart of an EPM. Different forecasting methods, from statistical to machine learning have been applied in the past. The selection of a forecasting method is mostly based on data availability and the objectives of the tool and planning exercise. We present a systematic and critical review of forecasting methods used in 483 EPMs. The methods were analyzed for forecasting accuracy; applicability for temporal and spatial predictions; and relevance to planning and policy objectives. Fifty different forecasting methods have been identified. Artificial neural network (ANN) is the most widely used method, which is applied in 40% of the reviewed EPMs. The other popular methods, in descending order, are: support vector machine (SVM), autoregressive integrated moving average (ARIMA), fuzzy logic (FL), linear regression (LR), genetic algorithm (GA), particle swarm optimization (PSO), grey prediction (GM) and autoregressive moving average (ARMA). In terms of accuracy, computational intelligence (CI) methods demonstrate better performance than that of the statistical ones, in particular for parameters with greater variability in the source data. However, hybrid methods yield better accuracy than that of the stand-alone ones. Statistical methods are useful for only short and medium range, while CI methods are preferable for all temporal forecasting ranges (short, medium and long). Based on objective, most EPMs focused on energy demand and load forecasting. In terms geographical coverage, the highest number of EPMs were developed on China. However, collectively, more models were established for the developed countries than the developing ones. Findings would benefit researchers and professionals in gaining an appreciation of the forecasting methods, and enable them to select appropriate method(s) to meet their needs.

/pdf/forecasting-methods-in-energy-planning-models-443zr1kmvi.pdf

Forecasting methods in energy planning models

A review on renewable energy and electricity requirement forecasting models for smart grid and buildings

Short-term electric load forecasting based on singular spectrum analysis and support vector machine optimized by Cuckoo search algorithm

This paper proposes a novel combined electricity demand forecasting method, which based on Back Propagation (BP) neural network, Adaptive Network-based Fuzzy Inference System (ANFIS) and Difference Seasonal Autoregressive Integrated Moving Average (diff-SARIMA).The combined method eliminates drawbacks and incorporates in the merits of the individual methods. It has the capability to deal with the linearity, nonlinearity and seasonality data.Experimental case study shows that the proposed combined method performed better than the other three individual methods and had a higher accuracy. And the proposed method also performed better than the method ESPLSSVM that I proposed before. Electricity demand forecasting, as a vital tool in the electricity market, plays a critical role in power utilities, which can not only reduce production costs but also save energy resources, thus making the forecasting techniques become an indispensable part of the energy system. A novel combined forecasting method based on Back Propagation (BP) neural network, Adaptive Network-based Fuzzy Inference System (ANFIS) and Difference Seasonal Autoregressive Integrated Moving Average (diff-SARIMA) are presented in this paper. Firstly, the combined method uses all the three methods (BP, ANFIS, diff-SARIMA) to forecast respectively, and the three forecasting results were obtained. By multiplying optimal weight coefficients of the three forecasting results respectively and then adding them up, in the end the final forecasting results can be obtained. Among the three individual methods, BP and ANFIS had the ability to deal with the nonlinearity data, and diff-SARIMA had the ability to deal with the linearity and seasonality data. So the combined method eliminates drawbacks and incorporates in the merits of the individual methods. It has the capability to deal with the linearity, nonlinearity and seasonality data. In order to optimize weight coefficients, Differential Evolution (DE) optimization algorithm is brought into the combined method. To prove the superiority and accuracy, the capability of the combined method is verified by comparing it with the three individual methods. The forecasting results of the combined method proved to be better than all the three individual methods and the combined method was able to reduce errors and improve the accuracy between the actual values and forecasted values effectively. Using the half-hour electricity power data of the State of New South Wales in Australia, relevant experimental case studies showed that the proposed combined method performed better than the other three individual methods and had a higher accuracy.

Yanhua Chen

Papers

Modelling a combined method based on ANFIS and neural network improved by DE algorithm

Mixed kernel based extreme learning machine for electric load forecasting

A hybrid application algorithm based on the support vector machine and artificial intelligence: An example of electric load forecasting

A novel combined model based on echo state network for multi-step ahead wind speed forecasting: A case study of NREL

A novel wind speed forecasting model based on moving window and multi-objective particle swarm optimization algorithm