EE461-Special-Topics-AI-Project-Load-Forecasting

  Week 11 Progress: Generation Scheduling GENERATION SCHEDULE Total Load Forecasted: Total load = Load_forecast/720; MonthlyLoads = 1×12 4200 4100 4300 4400 4500 4600 4700 4650 4550 4450 4350 4250 This load forecast table is for testing purposes only. The proposed model load forecast will be used once the best model is chosen. Define Generation Manually: Format:  (sites, 'VariableNames', {'Name','Type','Capacity', 'Priority'}) origTable = 8×5 table   Name Type Capacity Priority Allocated 1 'Solar Farm Maama Mai' 'solar' 1.4120 1 0 2 'Solar Farm Mata o e Laa' 'solar' 1.3000 1 0 3 'Solar Farm Singyes' 'solar' 2.1300 1 0 4 'Solar Farm Sunergise 1' 'solar' 2.3000 1 0 5 'Solar Farm Sunergise 2' 'solar' 2.3000 1 0 6 'Wind Farm I o Manumataongo' 'solar' 1.3750 1 0...

  Week 10 Progress: Generation Scheduling Objectives For Week 10 Creating models for Generation Scheduling An additional task is provided to accomplish, which is to have a planning script that can easily plan the maintenance scheduling of the generators or stations. MANUAL INPUT SECTION Total load demand for the month (in MWh) Total Load = 1400/720 Define generation sites manually Format: {'SiteName', 'Type', 'Capacity_MW'} Table 1: Generation Sites The table above shows the Renewable sites with its Capacity in MW Generation Scheduling remainingLoad = totalLoad; for i = 1:height(genTable) cap = genTable.Capacity(i); alloc = min(cap, remainingLoad); genTable.Allocated(i) = alloc; remainingLoad = remainingLoad - alloc; if remainingLoad <= 0 break; end end OUTPUT Generation Scheduling  Table 2: Generation Scheduling As evident from the table above, the load demand is met and being supplied from two solar farm generation site. Remark So far the first priority for...

  Week 9 Progress: Feature Engineering Objectives For Week 9 Creating models to test our data and predict values Also Redoing our Engineered features in accordance to our new "Response Variable" and the addition of the new variable "Temperature" EDA (Secondary) Principal Component Analysis Figure 1: 2D biplot of our new Dataset. The 2D Biplot above is of normalized data with respect to the principle components (PC1 and PC2) and the variables Billed, Generated, Sent out and Tmp2m_C_ which represents the power that is generated from renewable energy which in this case is solar energy, wind energy and BESS. It can be seen that Tmp2m_C_ has a very high positive correlation on PC1. Billed can be said to have least influence on PC1 and PC2. Due to its weak correlation, billed may require feature engineering to improve forecasting accuracy. Meanwhile, there is very little separation between the Generated and Sent out variables due to the difference from line loss. Figure 2...