scholarly journals Optimal Design of an On-Grid MicroGrid Considering Long-Term Load Demand Forecasting: A Case Study

2021 ◽  
Author(s):  
Bing Han ◽  
Mingxuan Li ◽  
Jingjing Song ◽  
Junjie Li ◽  
Jamal Faraji
Keyword(s):  
Demand Forecasting ◽  
Load Prediction ◽  
Load Profile ◽  
Demand Prediction ◽  
Demand Growth ◽  
Load Demand ◽  
Cost Of Energy ◽  
Artificial Neural Network Ann

In this article, an optimal on-grid MicroGrid (MG) is designed considering long-term load demand prediction. Multilayer Perceptron (MLP) Artificial Neural Network (ANN) has been used for time-series load prediction. Yearly demand growth has also been considered in the optimization process based on the forecasted load profile. Two case studies have been performed with the forecasted and historical load profiles, respectively. It has been shown that by applying the forecasted load profile, realistic results of net present cost (NPC), cost of energy (COE) and MG configuration would be achieved. Moreover, it has been demonstrated that utilizing battery storage systems (BSSs) are not economic in the proposed system. The introduced MG also produces lower emission compared to the system with the historical load profile.

scholarly journals Ultra-Short-Term Load Demand Forecast Model Framework Based on Deep Learning

Energies ◽  
2020 ◽  
Vol 13 (18) ◽  
pp. 4900 ◽  
Author(s):  
Hongze Li ◽  
Hongyu Liu ◽  
Hongyan Ji ◽  
Shiying Zhang ◽  
Pengfei Li
Keyword(s):  
Time Series ◽  
Deep Learning ◽  
Meteorological Data ◽  
Demand Forecasting ◽  
Dense Layer ◽  
Load Prediction ◽  
Demand Forecast ◽  
Short Term ◽  
Model Framework ◽  
Load Demand

Ultra-short-term load demand forecasting is significant to the rapid response and real-time dispatching of the power demand side. Considering too many random factors that affect the load, this paper combines convolution, long short-term memory (LSTM), and gated recurrent unit (GRU) algorithms to propose an ultra-short-term load forecasting model based on deep learning. Firstly, more than 100,000 pieces of historical load and meteorological data from Beijing in the three years from 2016 to 2018 were collected, and the meteorological data were divided into 18 types considering the actual meteorological characteristics of Beijing. Secondly, after the standardized processing of the time-series samples, the convolution filter was used to extract the features of the high-order samples to reduce the number of training parameters. On this basis, the LSTM layer and GRU layer were used for modeling based on time series. A dropout layer was introduced after each layer to reduce the risk of overfitting. Finally, load prediction results were output as a dense layer. In the model training process, the mean square error (MSE) was used as the objective optimization function to train the deep learning model and find the optimal super parameter. In addition, based on the average training time, training error, and prediction error, this paper verifies the effectiveness and practicability of the load prediction model proposed under the deep learning structure in this paper by comparing it with four other models including GRU, LSTM, Conv-GRU, and Conv-LSTM.

Computational Performance Analysis of Neural Network and Regression Models in Forecasting the Societal Demand for Agricultural Food Harvests

2022 ◽  
pp. 1287-1300
Author(s):  
Balaji Prabhu B. V. ◽  
M. Dakshayini
Keyword(s):  
Neural Network ◽  
Crop Production ◽  
Demand Forecasting ◽  
Forecasting Model ◽  
Ann Model ◽  
Demand Prediction ◽  
Computational Performance ◽  
Societal Needs ◽  
Ann Models ◽  
Artificial Neural Network Ann

Demand forecasting plays an important role in the field of agriculture, where a farmer can plan for the crop production according to the demand in future and make a profitable crop business. There exist a various statistical and machine learning methods for forecasting the demand, selecting the best forecasting model is desirable. In this work, a multiple linear regression (MLR) and an artificial neural network (ANN) model have been implemented for forecasting an optimum societal demand for various food crops that are commonly used in day to day life. The models are implemented using R toll, linear model and neuralnet packages for training and optimization of the MLR and ANN models. Then, the results obtained by the ANN were compared with the results obtained with MLR models. The results obtained indicated that the designed models are useful, reliable, and quite an effective tool for optimizing the effects of demand prediction in controlling the supply of food harvests to match the societal needs satisfactorily.

Computational Performance Analysis of Neural Network and Regression Models in Forecasting the Societal Demand for Agricultural Food Harvests

2020 ◽  
Vol 12 (4) ◽  
pp. 35-47
Author(s):  
Balaji Prabhu B. V. ◽  
M. Dakshayini
Keyword(s):  
Neural Network ◽  
Crop Production ◽  
Demand Forecasting ◽  
Ann Model ◽  
Demand Prediction ◽  
Computational Performance ◽  
Societal Needs ◽  
Ann Models ◽  
Artificial Neural Network Ann ◽  
Agricultural Food

Demand forecasting plays an important role in the field of agriculture, where a farmer can plan for the crop production according to the demand in future and make a profitable crop business. There exist a various statistical and machine learning methods for forecasting the demand, selecting the best forecasting model is desirable. In this work, a multiple linear regression (MLR) and an artificial neural network (ANN) model have been implemented for forecasting an optimum societal demand for various food crops that are commonly used in day to day life. The models are implemented using R toll, linear model and neuralnet packages for training and optimization of the MLR and ANN models. Then, the results obtained by the ANN were compared with the results obtained with MLR models. The results obtained indicated that the designed models are useful, reliable, and quite an effective tool for optimizing the effects of demand prediction in controlling the supply of food harvests to match the societal needs satisfactorily.

Freight Transport Demand Forecasting in Urban Logistics Planning: A Case Study of Yiwu City

2014 ◽  
Vol 505-506 ◽  
pp. 915-921
Author(s):  
Shi Chao Sun ◽  
Zheng Yu Duan ◽  
Chuan Chen
Keyword(s):  
Demand Forecasting ◽  
Freight Transport ◽  
Key Indicator ◽  
Development Policies ◽  
Logistics Demand ◽  
Exponential Smoothing Method ◽  
Coefficient Method ◽  
Logistics Planning

Freight transport demand forecasting as one of the basis in urban logistics planning, is not only an important premise of designing a variety of logistics development policies and infrastructure constructions, but also a key indicator to measure whether the logistics planning is reasonable. This paper addresses the methods of the freight transport demand forecasting in urban logistics planning based on a case study of Yiwu city. Considering the change of long-term trend emphatically, conventional trend extrapolation method, regression analysis method, elasticity coefficient method, linear exponential smoothing method and grey model are applied to predict the logistics demand of Yiwu city respectively. Then the results of five kinds of forecasting methods are analyzed to obtain the final forecasting logistics demand.

scholarly journals Model Prakiraan Harga dan Permintaan pada Rantai Pasok Karet Spesifikasi Teknis Menggunakan Jaringan Syaraf Tiruan

2012 ◽  
Vol 1 (3) ◽  
pp. 116
Author(s):  
Nofi Erni ◽  
M. Syamsul Maarif ◽  
Nastiti S.Indrasti ◽  
Machfud Machfud ◽  
Soeharto Honggokusumo
Keyword(s):  
Expert Knowledge ◽  
Price Change ◽  
Demand Forecasting ◽  
Raw Material ◽  
Market Condition ◽  
Ann Model ◽  
Demand Growth ◽  
Demand Pattern ◽  
Artificial Neural Network Ann ◽  
Hidden Layer

<p style="text-align: justify;">Karet spesifikasi teknis (TSR) merupakan jenis karet alam yang penting, dengan pertumbuhan permintaan yang tinggi dibanding jenis karet alam yang diproduksi dan diekspor oleh Indonesia. TSR paling banyak digunakan sebagai bahan baku untuk industri ban, sehingga dengan tumbuhnya indutri otomotif mendorong peningkatan permintaan terhadap TSR. Namun permasalahan muncul dalam produksi TSR, dimana tingkat  fluktuasi baik karena kelebihan maupun kekurangan produksi sangat berpengaruh terhadap perubahan harga TSR di pasar Internasional. Untuk mengurangi fluktuasi tersebut diperlukan suatu metode untuk memperkirakan tingkat permintaan dan harga. Penelitian ini bertujuan untuk merancang suatu metode prakiraan yang dapat merperkirakan tingkat harga dan volume permintaan untuk TSR 20.  Prakiraan dilakukan dengan Jaringan Syaraf Tiruan (JST) dengan algoritma propagasi balik, menggunakan data perkembangan pasar TSR di bursa berjangka SICOM. Model JST yang dirancang  mempertimbangkan pola harga, pola permintaan dan interaksi kedua faktor.  Hasil simulasi menunjukkan penggunaan 5 input neuron yaitu: 1) harga tertinggi, 2) harga terendah, 3) harga penutupan, 4) volume permintaan awal, 5) volume permintaan penutupan, 15 neuron pada lapisan tersembunyi dan 2 output yaitu harga dan volume permintaan pada lapisan output. Tingkat akurasi hasil prakiraan harga mencapai 91% dan akurasi prakiraan permintaan 87%. Berdasarkan hasil prakiraan ditentukan status harga dan permintaan. Harga tinggi jika perbedaan antara nilai maksimum dan nilai tengah lebih tinggi dari 47%, harga rendah jika perbedaan antara nilai minimum dan nilai tengah lebih dari 20%.  Prakiraan permintaan dinyatakan tinggi atau rendah jika terjadi peningkatan maupun penurunan sebesar 50 % dari rata-rata permintaan<em>.</em></p><h6 style="text-align: center;"><strong><em>Abstract</em> </strong></h6><p style="text-align: justify;">Technically Specified Rubber (TSR) is the most important of natural rubber type which has a high demand growth which is produced and exported by Indonesia. TSR is mostly used as raw material for tire industries, as the world’s automotive industries grow up the demand for TSR is also rise up. However, the problem appears in the production of TSR, which is fluctuative production rate in the form of over and under production correlated to the price change in International market.  Therefore, a method to forecast the price and demand level is needed to design in order to reduce fluctuation. The result is a forecasting that used as an input for preparing and adjusting TSR rubber production planning that working adaptively with market condition by utilising the expert knowledge. This research aimed to design a method that can forecast the changes in price level and demand volume. Artificial Neural Network (ANN) which is  backpropagation algorithm that has been designed according to data TSR market condition in SICOM is used in this research, the ANN model is modified by observing the price pattern, demand pattern and the connection between both of them together. Experiments have shown that the optimal architecture network for price and demand forecasting can be obtained by using 5 different neuron parameter, there are: 1) the highest price, 2) the lowest price, 3) the closing price, 4) demand volume interest, 5) demand volume close for input layer, 15 neuron for hidden layer and 2 different neuron there are price and demand volume for output layer. The accuracy of forecasting price had reached 91% and 87% for forecasting demand.  Based on forecasting result had determined the state of price and demand. The price is high if the differences between maximum and mean score is higher than 47% and the price is low if the differences between the minimum and mean score is higher than 20%. The demand is high if the demand forecasting is higher than 50% and it is low if smaller than 50% of average demand volume.</p>

scholarly journals Application of LEAP model on long-term electricity demand forecasting in Indonesia, period 2010-2025

2018 ◽  
Vol 49 ◽  
pp. 02007 ◽  
Author(s):  
Jaka Windarta ◽  
Bambang Purwanggono ◽  
Fuad Hidayanto
Keyword(s):  
Demand Forecasting ◽  
Electricity Consumption ◽  
Electricity Demand ◽  
Planning System ◽  
Large Country ◽  
Electricity Demand Forecasting ◽  
Electricity Planning ◽  
Business As Usual

Electricity demand forecasting is an important part in energy management especially in electricity planning. Indonesia is a large country with a pattern of electricity consumption which continues to increase, therefor need to forecasting electricity demand in order to avoid unbalance demand and supply or deficit energy. LEAP (Long-range Energy Alternative Planning System) as a tool energy model and Indonesia as a case study. Basically, electricity demand is influenced by population, economy and electricity intensity. The purpose of this study is to provide understanding and application of electricity demand forecasting by using LEAP. The base year is 2010 and end year projection is 2025. The scenarios of simulated model consist of two scenarios. They are Business as Usual (BAU) and Government policy scenario. Results of both scenarios indicate that end year electricity demand forecasting in Indonesia increased more than two fold compared to base year.

scholarly journals STG2Seq: Spatial-Temporal Graph to Sequence Model for Multi-step Passenger Demand Forecasting

2019 ◽  
Author(s):  
Lei Bai ◽  
Lina Yao ◽  
Salil S. Kanhere ◽  
Xianzhi Wang ◽  
Quan Z. Sheng
Keyword(s):  
State Of The Art ◽  
Demand Forecasting ◽  
Short Term ◽  
Demand Prediction ◽  
On Demand ◽  
Passenger Demand ◽  
Temporal Correlations ◽  
Output Module ◽  
Real World Datasets

Multi-step passenger demand forecasting is a crucial task in on-demand vehicle sharing services. However, predicting passenger demand is generally challenging due to the nonlinear and dynamic spatial-temporal dependencies. In this work, we propose to model multi-step citywide passenger demand prediction based on a graph and use a hierarchical graph convolutional structure to capture both spatial and temporal correlations simultaneously. Our model consists of three parts: 1) a long-term encoder to encode historical passenger demands; 2) a short-term encoder to derive the next-step prediction for generating multi-step prediction; 3) an attention-based output module to model the dynamic temporal and channel-wise information. Experiments on three real-world datasets show that our model consistently outperforms many baseline methods and state-of-the-art models.

Sign in / Sign up

Export Citation Format

Share Document