Category: Team’s Status Reports

Risks

• We are training LSTM/CNN models based off of data polled from sensors.  Retraining the model every request (if the cache fails) can cause lag

• Home Assistant API latency spikes can delay forecasts. Plotting charts via matplotlib adds overhead (200–300ms)

• Caching logic failing silently if timestamps are misaligned

• Infeasible LP results due to overlapping constraints or underforecasted solar

• LP objective may optimize cost but ignore user experience (e.g., clustering devices)

• Forecast errors can propagate into LP, yielding poor schedules

Changes

The initial approach used an LSTM network to forecast solar generation based on time-series data from sensor.maya_solar_panel. Main challenge was high latency due to recurrent nature and sequential computation and overfitting on short term trends rather than daily cycles. Transitioned to a 1D Convolutional NN. CNNs capture short-term temporal correlations ( sunrise to peak curve) better.

Progress

• Successfully polled data from all sensors into backend

• CNN model produces realistic solar forecasts aligned with production hours

• Performed testing on API response times

Verification/Validation

• Used curl requests to profile API timing in the terminal while polling data from sensors

• Unit tested LP constraints to ensure feasibility and logic correctness

• Compared cached vs. uncached outputs for CNN model for correctness and stability

• Compared forecast vs. actual solar power to assess prediction alignment
• To validate the 15% energy savings, we’ll compare historical baseline energy usage (from fixed schedules) against SmartWatt’s optimized schedules using the same input conditions. Using sensor data (from sensor.maya_fan_power, sensor.maya_solar_panel), we’ll simulate both scenarios over a 1–2 week period and calculate daily energy costs. The optimized case will use CNN-based solar forecasts and LP scheduling. If the average cost reduction exceeds 15% compared to the baseline, the savings target is met.
• To verify the robust assembly of the house, I will carry the house around and apply reasonable force to ensure it will remain intact during the demo.
• To validate the integration of the hardware components with the house, I will perform a fit test with the interior components. I will ensure that all feedthrough holes are wide enough and that the components can be neatly and securely placed inside their respective rooms.
• For validation, we plan to conduct user testing sessions where participants interact with the model house and dashboard to simulate their household energy consumption. This will help us evaluate whether SmartWatt creates a coherent and immersive experience. We will analyze user feedback to determine if the system effectively communicates the intended purpose of the project, the scaled energy usage and passage of time, and feels intuitive to interact with.

 

Risks

• Backend performance under real-time load
  With multiple device schedules and real-time data polling, system lag or crashes may occur if not optimized.
• Home Assistant integration 
  Integration with HA might face compatibility or API syncing issues depending on setup.
• Edge-case handling for devices
  Unresponsive or offline devices could cause unexpected failures if not properly handled

Changes

• System currently loads price data from CSV files, need to change it to connect to a SQLite database, for storing device state, energy usage, and forecasts is in place
• Add some more loads such as speakers for an audio component in the model house

Progress

• Frontend and backend frameworks set up
• User preferences taken into the consideration when performing constrained optimization
• Model house for demo set up
• Sensors integrated into Home Assistant

Significant Risks & Mitigation Strategies

Potential Risks:

1. Material Delays: The arrival of the acrylic sheets is critical for completing the assembly of the model house. Any delays could push back integration.

   • Mitigation: The order was placed early this week, about two weeks before the interim demo to allow enough time for laser cutting and assembly.

Contingency Plan: If the acrylic sheets do not arrive on time, we will use tape to secure the wiring and move forward with integrating the electronics.

2.

Risk: Optimization does not accurately reflect the assigned priority levels of individual devices, leading to suboptimal or unfair load scheduling.

Mitigation:

• Integrate device priority directly into the optimization objective and constraints as weighted penalties or scheduling preferences.

• Validate post-optimization results by checking if higher-priority devices are scheduled within their preferred windows or receive preferential treatment.

• Implement unit tests to ensure the priority values are parsed and utilized correctly in the solver.

---

Design Changes & Justification

1. Change: Slight modifications to the DXF files were made to improve the fit between the basswood and acrylic layers. I have also decided to not place acrylic sheets over the wooden floors. This reduces materials, lowers cost, and removes unnecessary complexity with laser cutting.

• Reason for Change: Initial design testing revealed minor misalignments in slots for acrylic insertion. Adjusting the DXF ensures a more precise assembly.

• Costs Incurred: Additional time spent in CAD revision and verification.

• Mitigation Strategy: These changes were made early enough, before laser cutting the wood and ordering the acrylic, that they do not impact the schedule. The updated files are ready for cutting as soon as materials arrive.

Risks:

Gradient Step Update Issue:

• In SGD and L-BFGS optimization algorithms, when forecasted energy demand closely matches actual load, the gradient of the cost function is near-zero.
• This results in very small updates to optimization variables, reducing adaptability and making it harder for the model to improve scheduling decisions.
• Mitigation: Introduce regularization techniques, perturbations in forecast data (white Gaussian noise)

Misinterpreting Transients as Long-Term Load Changes

Short-term power spikes (when a fan starts) can be mistaken for sustained high energy demand. The optimizer may overcompensate by reducing battery discharge or increasing grid imports unnecessarily.

Changes:

No major changes happened this week. We are thinking of adding a user tracking functionality to improve future recommendations based on past actions to make the system more interactive.

Progress:

• Optimization Model Debugging: Refined solver selection and fixed SOC linearity issues.
• Created CAD model of the model house
• Algorithm Development: Partial success with SGD & L-BFGS, but linear programming constraints need debugging.
• Recommendation Engine: Classifies insights by priority and impact, with user tracking integration.
• Frontend: Built an interactive energy visualization dashboard.
  Pending:
• Hardware Integration: Waiting for power sensor polling to complete before full backend integration.
• Building the model house in which to house all of our electrical components.

• Risk: Hardware Integration Challenges
  • Issue: Ensuring seamless communication between sensors, microcontrollers, and the backend system is critical. Incompatibility between components or unexpected electrical issues could cause delays.
  • Mitigation: We are testing each hardware module separately before full integration and maintaining thorough documentation for troubleshooting.
• Risk: Energy Monitoring Accuracy
  • Issue: If the system fails to accurately track and optimize energy consumption, the value of SmartWatt is diminished.
  • Mitigation: We are calibrating sensors early in the process, running validation tests, and comparing data against trusted benchmarks..

Contingency Plans:

• If hardware integration delays occur, we will simulate system behavior in software to continue progress.
• If energy monitoring accuracy issues arise, we will refine our algorithms and conduct more extensive testing.

2. Design Changes and Their Impact

This week, we made the following adjustments to our system design:

• Change: Updated Physical Model Design
  • Reason: To improve wiring organization and ease of sensor placement, we modified the demo house’s wall structure to include a layered design (wood exterior, acrylic interior).
  • Costs: This change requires additional material procurement and minor CAD redesign work.
  • Mitigation: We are sourcing materials efficiently and ensuring that fabrication remains within our timeline.

By proactively managing these risks and changes, we aim to keep SmartWatt on track for a successful completion.

ARIMA models from sklearn have been added to the load forecasting module as they have better accuracy on testing data.

Progress

• Frontend of webapp scoped out
• ML models working with simulated data
• API and communication requests between backend and ESPHome and Home Assistant established

Additional Weekly Questions

3. Considerations of Global Factors:

SmartWatt addresses the growing global need for energy efficiency and sustainability by optimizing household energy consumption. As the world transitions to renewable energy, challenges such as grid instability, peak demand management, and high electricity costs are becoming more pressing. SmartWatt provides a data-driven solution by integrating smart meters, IoT-enabled devices, and AI-driven analytics to monitor and optimize energy use in real time. This system empowers homeowners to reduce their carbon footprint, lower electricity bills, and contribute to overall grid stability—an issue that affects both developed and developing regions as energy demands rise.

Beyond local energy concerns, SmartWatt aligns with broader global initiatives such as the United Nations Sustainable Development Goals (SDGs), particularly Goal 7: Affordable and Clean Energy. By enabling users to shift energy consumption to off-peak hours and integrate renewable sources like solar, SmartWatt helps reduce reliance on fossil fuels and supports cleaner energy adoption worldwide. Additionally, in regions where energy access is limited or expensive, SmartWatt’s optimization algorithms can help households maximize their use of available power, ensuring more efficient distribution of resources. As energy prices fluctuate due to geopolitical and economic factors, SmartWatt provides a scalable, intelligent approach to energy management that benefits users across diverse global contexts.

Part B (written by anyab)

Cultural differences influence energy consumption behaviors, regulations, and sustainability priorities. In Europe, where time-of-use tariffs are common, the system automates energy usage during off-peak hours (we use Nordpool API which provides spot electricity prices in Europe) . In regions with flat-rate pricing, it prioritizes renewable energy and battery storage. These region-specific energy insights make the system adaptable to diverse communities, ensuring relevance and effectiveness. Beyond energy consumption habits, cultural attitudes toward technology and automation play a significant role in system adoption. In regions where smart home technology is widely embraced, users may prefer fully automated energy management, while in areas with privacy concerns or skepticism toward automation, the system can offer manual control options to align with user preferences.

Part C (written by sdoshi2)

Environmental factors play a crucial role in the development of SmartWatt.
It is expressly designed to reduce household energy consumption. By optimizing energy usage and promoting demand-side management, SmartWatt helps lower reliance on fossil fuel-based power generation, which remains a major contributor to greenhouse gas emissions. Additionally, by integrating renewable energy sources such as solar panels, the system encourages sustainable energy practices and reduces strain on non-renewable resources.

We also take into account the fact that the users environment will discate the renewable power production and the power consumption based on climate/weather

Beyond energy reduction, SmartWatt also considers energy efficiency in hardware selection. The sensors and microcontrollers used in the system are also chosen to lower power consumption to ensure that the monitoring and optimization processes do not inadvertently contribute to excessive energy use.

Furthermore, SmartWatt aids in environmental conservation by mitigating energy wastage. By identifying inefficiencies such as poorly insulated spaces, users can make informed decisions to enhance their home’s energy efficiency. This proactive approach not only saves money but also reduces unnecessary energy demand, contributing to a broader reduction in environmental degradation. In regions where power generation relies heavily on coal or other environmentally harmful methods, SmartWatt’s ability to shift energy consumption to cleaner energy sources can make a significant difference in reducing ecological impact.

By addressing these environmental factors, SmartWatt aligns with sustainability goals and contributes to a more energy-efficient future, making it an essential tool for environmentally conscious consumers and policymakers alike.

 

Risks

1. Need to test how different LP solvers (CVX, PuLP etc) output different results for the optimization and choose which one is the most accurate at reducing energy consumption cost

Changes

Direct communication between the ESP32 and Django backend for storing load consumption data

Schedule & Progress

• Optimization algorithm framework finalized
• Design specifications (including components in the embedded layer, ML model, backend and mobile app and communication protocols between these finalized)
• Design presentation done
• external API libraries and datasets (and data collection mechanisms for ML models) identified
• Sketched up a model of the demo house with dimensions and ordered materials accordingly.
  
  Additional Weekly Questions (A written by anyab, B written by Erika , C written by Maya)
• Part A :
• Health: SmartWatt contributes to community health by reducing reliance on non-renewable energy sources (increasing reliance on solar panel energy, which will help decrease air pollution in the long run through all of the fossil fuels that haven’t burnt). Additionally, SmartWatt has an optimization function that maximizes consumption of solar power, thereby reducing the need for drawing power off the grid
• Welfare: By optimizing energy costs and enhancing efficiency, SmartWatt supports economic welfare, reducing household energy expenses and ensuring consistent power availability for essential needs.
• Safety : Users can see what their power consumption for different appliances are, and can determine if an appliance is faulty based off the power profile.
• B : SmartWatt, our household energy optimization and monitoring system, addresses the growing need for sustainable energy management within residential communities. Rising energy costs and increasing awareness of climate change have created a strong societal demand for solutions that reduce consumption, lower utility bills, and promote environmentally responsible behavior. SmartWatt empowers homeowners to monitor and optimize their energy use through real-time analytics, personalized recommendations, and automated control of appliances. By reducing energy waste, SmartWatt not only decreases household expenses but also contributes to broader efforts to lower carbon emissions and combat climate change—issues that affect society as a whole.Socially, SmartWatt fosters a culture of energy-conscious living within neighborhoods and communities. The platform can be extended to include social features that allow users to compare their energy savings with neighbors, participate in local energy-saving challenges, and share tips, encouraging collective action toward sustainability. Additionally, SmartWatt’s affordability and user-friendly interface ensure accessibility for diverse socioeconomic groups, helping low-income households reduce utility costs and alleviate energy poverty. By promoting sustainable habits and supporting energy equity, SmartWatt advances social and environmental responsibility, creating a positive impact on both individuals and their communities

C: Consideration of Economic Factors

SmartWatt is designed to optimize energy consumption in residential homes, directly impacting economic factors by reducing household electricity costs. By integrating real-time monitoring, predictive analytics, and automation, SmartWatt ensures that homeowners can minimize unnecessary energy expenditure, ultimately leading to lower utility bills. The optimization algorithms leverage data from smart sensors to shift power usage to off-peak hours, allowing users to take advantage of lower electricity rates where time-of-use pricing is available.

On a broader scale, SmartWatt contributes to economic efficiency by reducing strain on power grids, decreasing the demand for expensive energy production during peak times. The system also enables better load balancing, which can prevent overuse and extend the lifespan of household appliances, reducing replacement and maintenance costs over time. By making energy management more accessible and cost-effective, SmartWatt supports economic sustainability and affordability, ensuring long-term financial benefits for users while promoting responsible energy consumption at a larger scale.

However, the initial cost may be high as many appliances will need to be wired into Home Assistant for both control and monitoring. With stuff like the ESPHome it is a lot cheaper given some technical knowledge

Presented the proposal to the class

Risks : 

1. Right now, our load forecasting model is trained on synthetic data. Main risk is overfitting to synthetic data. Once we switch to real-time power consumption logs from ESP32 sensors, our model may struggle to adapt to real-world variations such as sudden spikes in energy usage (a heater turning on) or unusual load drops (someone turning off all appliances).

Mitigation : Collect real IoT data logs from ESPHome devices to improve model accuracy. Implement outlier detection to ignore unusual spikes or drops in power usage. Use XGBoost & RandomForest models, which handle nonlinear patterns better. Will plan to use multiple forecasting models (XGBoost, ARIMA, LSTMs) and use ensemble methods to combine predictions.

2. The reliability and quality of the optimization algorithms are directly dependent on the quality of the forecast data (for load forecasting + solar power + grid prices). Lower forecast errors lead to more accurate optimization results, whereas high forecast errors can lead to bad optimization outcomes

Changes : 

No significant design changes have been made to the system yet.

Progress:

• Solcast API integration for access to real-time weather forecasts
• Load forecasting module that predicts load for next 24h within household
• Set up home assistant instance
• Connect ESPHome to home assistant
• We have researched and validated the necessary electronics components to execute our plan