IEMS Solution Help

1. Predictive Analytics:

Renewable Generation Forecasting: IEMS uses deep neural networks (DNN) to forecast the day-ahead hourly output of renewable generation sources. Our algorithms are robust and flexible enough to provide renewable generation predictions anywhere in the world. The generated photovoltaic (PV) power mainly depends on the amount of irradiation, temperature, PV technology and the plant size. While using the IEMS’ PV power forecasting application, the user: first, selects the city where the PV plant is located in. Based on the location of the PV plant, the solar irradiation and temperature data of the next 24 hours are gathered. By selecting the PV technology, the corresponding Deep Neural Network is selected. Second, the PV plant capacity is entered by the user. After submitting the information, the DNN predictions are performed in the IBM Watson cloud, and then a couple of graphs are provided to the user. The first graph shows the prediction of irradiation factors that influence the solar PV output power, while the second graph shows the expected output power of the target solar PV plant in the next 24 hours.

Load forecasting: Temperature, humidity and time of day are the main factors that determine hourly electricity demand in a city. IEMS’ state-of-the-art Machine Learning algorithms predict day-ahead hourly load of any distribution network in a city. While using the IEMS’ load forecasting application, the user: first, selects the city where the target distribution feeder is located in. Based on the location of the target feeder, the meteorological data for the next 24hours are gathered; also the corresponding Deep Neural Network is selected. Second, the user should enter the target distribution feeder capacity (or peak demand). After submitting the information, the DNN predictions are performed in the IBM Watson cloud, and then a couple of graphs are provided to the user. The first graph shows the prediction of meteorological factors that influence the distribution feeder load, while the second graph shows the expected feeder load in the next 24 hours. IEMS’ predictive analytics modules are developed within the IBM Watson computational cloud.

2. Distributed Energy Resource Management System (DERMS):

New evolving smart distribution grids with thousands of distributed energy resources (DER) such as solar PV, Wind and Battery storage, and flexible loads create computational challenges to power system operators for planning, operations and optimal control of distributed energy sources. IEMS’ DERMS solution is a novel smart grid simulator and optimizer that addresses these computational challenges. IEMS’s DERMS software has three sub-modules for different network simulation and optimization scenarios including:

– Load Flow: for non-objective based power flow analysis and network situational awareness with and without integrated renewable and distributed energy resources (DER/DG) to the distribution network.

– Optimal Power Flow (OPF): for objective based power flow analysis and network situational awareness with and without integrated renewable and distributed energy resources (DER/DG) to the distribution network. The existing objective functions for the OPF module include network’s: cost minimization, loss minimization and price-based load shedding.

– Contingency (Optimal) Power Flow: for observing the network situational awareness and bottlenecks based on predicted contingency scenarios (i.e. line outage, asset overloading and line congestion); and utilizing DER/DG and demand response (DR) assets to minimize the customers’ interruption costs (CIC) via sustainable islanded microgrids.

After each DERMS run/simulation two summary tables are presented to the user. Table 1 (top one), summarizes the usage of PV and Battery storage resources, grid energy, total load, network losses and loss cost, GHG reduction and CO2 emission cost reduction. The purpose of Table 1 is to show the user (i.e. distribution system operator) the effect of utilizing DER/DG on network’s loss and GHG reductions, and their respective cost reductions.

Table 2 (bottom one), summarizes the network’s reliability indices including SAIFI (system average interruption frequency index), SAIDI (system average interruption duration index), EENS (expected energy not supplied) and VOLL (value of lost loads) per contingency scenario. The purpose of Table 2 is to show the user (i.e. distribution system operator and/or planner) the effect of utilizing DER/DG and price-based DR on improving the above mentioned reliability indices, which directly translates into having sustainable islanded microgrids after the contingency (i.e. line outage) that could reduce the CIC considerably, and prevent complete system blackout.

3. Transactive Energy Platform (BC2E):

Future smart grids need distributed, cyber-secure and efficient market platforms to allow renewable and distributed energy resources, flexible loads, and power system operators engage in trading energy and ancillary services. This will allow these market participants to meet future power system needs such as load balancing, loss minimization, peak-demand reduction, CO2 emission control, and overall system reliability improvement. IEMS’ Blockchain-based peer-to-peer Transactive Energy platform provides the infrastructure for electricity trading among prosumers, consumers, and the power system operator, at vastly reduced costs and complexity. IEMS’ Transactive Energy platform uses IBM-backed Hyperledger Blockchain technology, the gold standard in enterprise blockchain.