Icon

Irradiance LSTM Forecast Model

GLOBAL HORIZONTAL IRRADIANCE LSTM FORECAST MODEL

The University of Saskatchewan
Ph.D. in Interdisciplinary Studies

Created by: Carlos Enrique Diaz, MBM, P.Eng.
Email: carlos.diaz@usask.ca

Supervisor: Lori Bradford, Ph.D.
Email: lori.bradford@usask.ca

This workflow forecasts time series (irradiance) using an LSTM network with lagged values as input. The trained model is then used for out-of-sample forecasting. The predicted values are compared to the actual values, and the forecast's performance is reported via scoring metrics and a line plot in a dashboard. This workflow used GHI data from CWEEDS for Saskatoon and was adapted from the following KNIME workflow:

https://hub.knime.com/-/spaces/-/~Qo3OuYyVQV58k9cq/

URL: "Once Upon A Time..." by LSTM Network https://www.knime.com/blog/text-generation-with-lstm
URL: KNIME Workflow Template https://hub.knime.com/-/spaces/-/~Qo3OuYyVQV58k9cq/
URL: CWEEDS Data https://climate.weather.gc.ca/prods_servs/engineering_e.html

Loading Cleaning Network Architecture Input Vector Filtering GLOBAL HORIZONTAL IRRADIANCE LSTM FORECAST MODELThe University of SaskatchewanPh.D. in Interdisciplinary StudiesCreated by: Carlos Enrique Diaz, MBM, P.Eng.Email: carlos.diaz@usask.caSupervisor: Lori Bradford, Ph.D.Email: lori.bradford@usask.caThis workflow used GHI data from CWEEDS for Saskatoon and was adapted from thefollowing KNIME workflow:https://hub.knime.com/-/spaces/-/~Qo3OuYyVQV58k9cq/ Training & Testing Visualization Optimization SaskatoonCWEEDS Data(UTC)LinearInterpolationGlobalHorizontalIrradianceSelectionTrainingMSE Loss Function64 Tensors for64 Lagged Inputs80/20 Split1 OutputTensor128 UnitsActivation ReLUCombineInto List64 LagsLast12 Months64 UnitsActivation ReLU10 IterationsBestIterationFilteringPlotsTrainingMSE Loss Function50/50 Split CSV Reader Preprocessing Missing Value Column Filter Timestamp Alignment ParameterOptimization Loop End Table Rowto Variable Keras NetworkLearner Keras Input Layer Partitioning Keras Dense Layer Numeric Scorer Keras LSTM Layer Column Aggregator Lag Column Deployment Loop Row Filter Keras LSTM Layer Variable ConditionLoop End Processing Row Filter Dashboard Generic Loop Start Deployment Loop Keras NetworkLearner Numeric Scorer Parameter OptimizationLoop Start (Table) Parameters Partitioning Loading Cleaning Network Architecture Input Vector Filtering GLOBAL HORIZONTAL IRRADIANCE LSTM FORECAST MODELThe University of SaskatchewanPh.D. in Interdisciplinary StudiesCreated by: Carlos Enrique Diaz, MBM, P.Eng.Email: carlos.diaz@usask.caSupervisor: Lori Bradford, Ph.D.Email: lori.bradford@usask.caThis workflow used GHI data from CWEEDS for Saskatoon and was adapted from thefollowing KNIME workflow:https://hub.knime.com/-/spaces/-/~Qo3OuYyVQV58k9cq/ Training & Testing Visualization Optimization SaskatoonCWEEDS Data(UTC)LinearInterpolationGlobalHorizontalIrradianceSelectionTrainingMSE Loss Function64 Tensors for64 Lagged Inputs80/20 Split1 OutputTensor128 UnitsActivation ReLUCombineInto List64 LagsLast12 Months64 UnitsActivation ReLU10 IterationsBestIterationFilteringPlotsTrainingMSE Loss Function50/50 Split CSV Reader Preprocessing Missing Value Column Filter Timestamp Alignment ParameterOptimization Loop End Table Rowto Variable Keras NetworkLearner Keras Input Layer Partitioning Keras Dense Layer Numeric Scorer Keras LSTM Layer Column Aggregator Lag Column Deployment Loop Row Filter Keras LSTM Layer Variable ConditionLoop End Processing Row Filter Dashboard Generic Loop Start Deployment Loop Keras NetworkLearner Numeric Scorer Parameter OptimizationLoop Start (Table) Parameters Partitioning

Nodes

Extensions

Links

Download

To use this workflow in KNIME, download it from the below URL and open it in KNIME:

Download Workflow
On NodePit since: 2024-06-14
Last update: 2024-09-16
Created with KNIME version: v5.2.4
Tags: Time SeriesPredictionLSTMDeep LearningNeural NetworkKerassolarphotovoltaicPVenergyelectricsystemelectrical engineeringweathermeteorologyforecasting

Deploy, schedule, execute, and monitor your KNIME workflows locally, in the cloud or on-premises – with our brand new NodePit Runner.

Try NodePit Runner!