EndorsedendorsedCompleted

AI Forecasting Tool Implementation for Virtual Power Plant

Services Covered:

AI and Machine Learning
Project hero — replace with case study imagery
Industry
Industry

Energy & Natural Resources

Duration
Duration

2 months

Budget
Budget

Confidential

Start Date
Start Date

1 May 2024

Team Size
Team Size

1-5

Engagement Model
Engagement Model

Nearshore

Project Capability Score (PCS)

gauge meter
73/100
Good

Project Capability Score (PCS) estimates how capable Innowise is of successfully delivering a project like this, based on its past experience, track record, and ability to handle similar work. Learn more about PCS


Client

Confidential

map Switzerland

Project Summary

Innowise developed and implemented an AI forecasting tool for a virtual power plant to predict power consumption using past consumption, weather conditions, and customer behavior. The model improved prediction quality and reduced errors by around 10%. The team was responsive, professional, delivered outcomes on time, and presented results clearly throughout the engagement.

Key Challenges

  • Implementing a forecast model for energy consumption.
  • Integrating multiple data sources (past consumption, weather conditions, customer behaviour) to improve prediction accuracy.

Project Deliverables

  • Forecast tool to predict consumption in a 4-days horizon
  • Detailed analysis of the model's performance
  • Report on feature importance
  • Modified approach for a 2-days horizon
  • Deployment of the model in a pre-production system
  • Configuration and use of a dedicated AWS service to deploy the model to production

Project Solution

Innowise implemented an AI-based forecast tool to predict power consumption for a pool of electrical devices using historical consumption, weather data, and customer behavior. Deliverables included a 4-day forecast model, performance analysis, feature-importance reporting, a modified 2-day approach, deployment to a pre-production environment, and configuration of an AWS service for production deployment.

Project Outcome

  • Reduced prediction errors by around 10%.
  • Demonstrated significant improvement in prediction quality versus previous implementations.
  • Delivered outcomes on time and consistently presented requested results in regular meetings.

Platforms

  • CloudCloud

Tech Stack

  • Amazon Web Services (AWS)Amazon Web Services (AWS)

Client Endorsement

Overall Review Rating

5star5 out of 5 stars

Timeliness

Cost Rating

Willing to Refer

Quality of Deliverables

Our predictions got better and errors dropped by 10% because of Innowise's work. They responded to requests, acted professionally and responsibly, delivered on time, explained ideas clearly, and went beyond what we expected. We were impressed by their customer service.

Anonymous

Head of People Operations

Note: This endorsement is based on publicly available client feedback from external review sources.

More Projects by Innowise