Water desalination plants are significant consumers of electric power, making them some of the largest energy users in power grids. Their electricity consumption presents an urgent challenge for efficient and sustainable operation, and they are among the most power-quality-threatening customers for utilities. This study presents three distinct strategies to enhance prediction accuracy and extend forecasting horizons, aiming to reduce algorithmic and hardware delays. Additionally, it suggests effective methods to compensate for voltage fluctuations, voltage flicker, and dips arising from desalination plants. The paper also discusses forecasting harmonics and interharmonics in current signals. Furthermore, it integrates the above techniques into a comprehensive computing system, along with an active power filter (APF) scheme, within the Simulink framework. A comparison is drawn between the performance of predictive techniques in an APF and a conventional, non-predictive APF. The proposed data augmentation method successfully increases prediction accuracy. By effectively forecasting upcoming waveforms, it reduces algorithmic and hardware delays. These techniques are designed to address multiple power quality issues simultaneously, including harmonics, interharmonics, flicker, and voltage dips, which often coexist in the spectrum as interharmonics. The suggested approach employs Long Short-Term Memory (LSTM) networks combined with the Jetson TX2 embedded artificial intelligence computer to accelerate machine learning applications. This method has proven effective in predicting and classifying time series data, including harmonics, interharmonics, and raw current signals, achieving 100% accuracy. This eliminates the need for designing specific low-pass filters. The evaluation results for this time-domain deep learning-based technique will be reported in the subsections below. The implementation is conducted in Python using the KERAS deep learning framework and TensorFlow backend, and it is evaluated on a workstation equipped with an Intel i7 processor running at 4.0 GHz and 48 GB of RAM.

A multi-energy resource aims to maintain a balance between energy output and load consumption and to ensure power continuity during different operating conditions. The harmonic distortions can be estimated from the output current of a harmonic source, which may not fully reflect its true harmonic distortions due to the interactions between the state changes at the power network level and the harmonic sources. System operators monitor each system's harmonic performance under different conditions of operation to find the actual contribution of grid-connected systems to harmonic-related issues. Development of machine learning algorithms leads to effective progress in the harmonic prediction and computation. In this paper, the combined data sequencing, and Bidirectional Long-Short Term Memory (Bi-LSTM) network has been exploited for the real-time harmonic prediction of future events in multi-energy sources. The validity of the proposed Model including the applications of ANFIS, ANNs, MLRA and LSTM is conducted on the two standard systems as IEEE 9-bus and IEEE 34-bus multi energy resources system that is associated with PV systems. The simulation results, based on climate changes of solar irradiance and ambient temperature in PV systems, demonstrate that the proposed methods can accurately forecast changes in total harmonic distortion (THD) as well as the voltage profile at the point of common coupling. The performance of Bi-LSTM, original LSTM, Machine Linear Regression (MLR), and Artificial Neural Networks (ANNs) techniques were assessed. These findings provide valuable insights. Four performance validation indices, RMSE, R-squared and MSE are considered to assess the performance of the competitive learning algorithms. The results showed that in the model IEEE 9-bus, Bi-LSTM outperformed all the applied methods as its RMSE value was 0.000019 while its MSE value was 3.61e-10 and finally, the Bi-LSTM had a higher value squared error (R²) was equal 1 which indicates the effectiveness of Bi-LSTM for predicting sequential total harmonic distortion. On the other hand, in case study of IEEE 34-bus, the RMSE, MSE and R² are 0, 3.276e-30 and 1 using Bi-LSTM which means that the Bi-LSTM leads to the best performance validation indices compared to other competitive algorithms for the tested multi-energy systems.