ANN represents an application based on artificial intelligence which aims to determine the output of a system based on its inputs. It is the most useful in the case of stochastic systems for which it is very difficult to define the relation between the input and output values. Therefore, it is necessary to apply a method that has the ability to estimate the output of the system without knowing the exact definition of the system transfer function. [10]
In this study ANN were implemented for supervised machine learning where the inputs x and outputs y are already known and mapping between them is being conducted. For the model to be created the machine is being fed a great amount of data so the network could be trained. The greater the number of data is the more accurate the network prediction will be. [11]
The accuracy of the network will be analyzed through the value of the calculated errors MAPE (Mean Average Percentage Error), MAE (Mean Absolute Error), MSE (Mean Square Error), RMSE (Root Mean Square Error). According to Yadav and Chandel [12], the model is considered high precision if MAPE <10%, good if 10% <MAPE <20%, reasonable if 20% <MAPE <50% and inaccurate for MAPE> 50%. Multiple errors were calculated to ensure an accurate representation of the models' deviations.
![](data:image/png;base64,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)
The logic of ANN is based on imitating behavior of neurons in the human brain. From one neuron a layer of neurons is formed and from multiple layers a network is formed. One layer is made out of similar input data where every input data has its own synaptic weight which defines the strength of synapses. Neural network consists of an input layer, hidden layers and output layer which is shown in picture 2.1. Input payer is constructed from input data. Hidden layers process data that they got from the previous layer and their output is input for the next layer. The final layer in the network is the output layer and it is responsible for prediction based on the input data. [13,14]
In this study input data consists of production from different power plants and consumption that was measured by Serbian Electric Power Company. Production included in ANN data are values from thermal, hydro and wind power plants. Besides these parameters that influence production such as temperature, wind speed, season, time of the day, etc. were included as well. In the fig. 2.2 algorithm that shows the process of forming ANN and concluding its precision is presented.
The data needs to be processed and prepared for processing. In this part the data is separated on the input x and output y part. Final database is formed when data goes through the normalization process so the values could be compared.
Testing and training of the neural network was done through three phases. Firstly, the data was separated on the training part that is 70% of the input database and testing part that is the remaining 30% of the database. Then the parameters of ANN are chosen such as number of hidden layers, types of activation function, regularization parameters and optimal time limit for the training. Lastly the ANN is trained. It has been observed that ANN gives best predictions when there are five hidden layers containing 24, 12, 12, 12 and 12 neurons and 25 epochs of training.
The trained ANN is used for consumption forecasting. Result of yearly consumption forecasting for data used in this study is shown in figure 2.3. Calculated MAPE for this prediction is 5.64 %, which puts it in the MAPE range where MAPE <10% is defined as high precision prediction.
For easy review daily consumption forecasting is present in fig 2.4. Considering it comes from the same ANN, prediction error MAPE is the same, 5.64 %.This daily prediction graph can be made for any random day. In table 2.2these results were given through their values. It shows exact inputs, predicted consumption, real consumption and absolute percentage error that is made while predicting every value.
Table 2.2: Results of part of prediction for one day in winter
![](data:image/png;base64,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)
Considering values of prediction error, it can be concluded that consumption prediction errors are lowered and progress was made. The developed ANN gives very good results that can also be seen through graphs on fig 2.3. and 2.4. where the differences between real values and predicted values are almost minor.
The comparison between the ANN and ARIMA models is illustrated by the errors calculated for each model, as presented in Table 2.3.
Table 2.3: Error comparison for ANN and ARIMA model
|
ANN
|
ARIMA
|
MAE
|
229.91
|
401.50
|
MSE
|
8.7290⸱104
|
16.6254⸱104
|
RMSE
|
295.45
|
407.74
|
MAPE
|
5.64 %.
|
10.0025 %
|
After analyzing the variance in prediction errors, it becomes evident that the predictions generated by artificial neural networks (ANNs) are more accurate. These predictions will be basis for further calculations and forecasts concerning power production, leveraging adaptive neuro-fuzzy inference systems (ANFIS)