ALLSHIFT/docs/02-specifications/rule-based-controller.md

RULE-BASED EMS

Rule-Based Energy Management System to Enhance PV Self-Consumption in a Building: A Real Case

“The EMS presented in this article is a rule-based EMS (RB-EMS) and its objective is to maximize the self-consumption rate (SCR) of PV energy actuating on the heating, ventilation and air conditioning (HVAC) of ESTIA2 building involved in the CSC, taking into account the comfort of users.”

HVAC: heat, ventilation and air-conditioning

CSC: collective self-consumption

“At each time step, the priority is to self-consume the building’s energy production. Then, any surplus is used to charge the battery and, finally, it is exported to the grid in case the storage system is full. Import from the grid is the last resort in case of low local production and not enough stored energy inside the battery. ”

This paper uses HVAC as the flexible load

rather than a battery. When there is excess

PV, instead of storing electricity in a battery,

it runs the heating/cooling system harder

(pre-heating or pre-cooling the building).

The building itself acts as a thermal "battery."

Step 1 / Read data
It read PV power being generated (PPV) and building consumption
(PC). Then calculates e = PPV-PC (the energy surplus or deficit)

Step 2 / Compare e against a threshold h
e < -h: More consumption than PV (deficit)
HVAC units are sorted by room temperated, if a room is above
comfort threshold, the corresponding unit is switched OFF to reduce
consumption, then checks if any OFF units can be turned ON for
comfort maintenance
|e| < h: Roughly balanced
PV and consumption are close enough, so no action needed
e > h: More PV than consumption (surplus)
Excess solar energy is available, the controller looks for HVAc units that are
OFF and could be switched ON. Switches ON the unit in the coldest room
to absorb the extra PV energy. If more surplus remains, it may switch OFF a unit and ON another to redistribute

Step 3 / Wait 5 minutes, then repeat until the end of day (5 minutes is the sampling time)

*Techno Park in France

---

RB-EMS vs AI-EMS

tibo energy

Rule Based EMS How
A rule-based EMS runs on predefined rules. Think: “if the solar array produces more than 50 kW, switch on the battery” or “only charge the EV fleet between 22:00 and 06:00.” The system follows these instructions literally, regardless of what’s happening in the energy market or on the grid at that moment.
This model works fine in stable, predictable environments. But today’s energy market is anything but stable. Dynamic energy prices change every quarter of an hour, renewable generation fluctuates with the weather, and grid congestion is forcing companies to actively steer their power flows.

AI Energy Management
AI energy management takes a fundamentally different approach. Instead of fixed rules, the system uses machine learning and predictive models to continuously learn from data: historical consumption, weather forecasts, energy prices, grid load and the behaviour of every connected asset.
Based on that data, an AI-driven EMS makes autonomous decisions every few minutes. When to charge? When to store? When to feed back to the grid? The system doesn’t optimise based on a static rule, but based on what is smartest right now.

---

HOW WOULD THIS LOOK FOR US?

1 Objective

The purpose of the rule-based controller is to manage energy flows within the hospital energy system using a deterministic set of priorities. This baseline strategy ensures:

• Reliable supply to hospital demand
• Efficient use of on-site renewable energy (PV)
• Rational use of storage systems (battery and hydrogen)
• Reduced reliance on the grid where possible

2 System Components

The controller manages the following energy sources and storage systems:

• Photovoltaic (PV) generation
• Battery storage system
• Hydrogen system
  • Electrolyzer (electricity → hydrogen)
  • Fuel cell (hydrogen → electricity)
• Grid connection
• Hospital load demand

3 Control Philosophy

The controller follows a priority-based dispatch strategy, where energy is allocated based on predefined rules rather than optimization.

The general philosophy is:

1. Use local renewable energy first
2. Store excess energy for future use
3. Use stored energy during shortages
4. Use the grid as a last resort (or strategically based on price)

4 Energy Allocation Logic

4.1 Normal Operation

At each time step, the controller compares PV generation with hospital demand.

Case 1: PV Generation Exceeds Demand

When on-site renewable generation exceeds the hospital load:

1. PV energy is first used to fully satisfy the hospital demand
2. The excess energy is allocated as follows:
   • Charge the battery until it reaches maximum capacity
   • If the battery is full, use excess energy to produce hydrogen via the electrolyzer
   • If both battery and hydrogen storage are full, excess energy is exported to the grid

Case 2: PV Generation is Less Than Demand

When PV generation is insufficient:

1. PV energy is used as much as possible
2. The remaining demand is supplied:
   • Battery discharge (if sufficient charge is available)
   • Hydrogen fuel cell (if hydrogen is available)
   • Grid import as the final fallback

5 Cost-Aware Operation (Optional Enhancement)

To improve economic performance, electricity price signals can be incorporated:

• During peak price periods:
  • Prioritize battery discharge and hydrogen usage
  • Minimize grid import
• During off-peak periods:
  • Prefer grid electricity if it is cheaper
  • Preserve stored energy for future high-cost or emergency periods

6 Reliability and Hospital Constraints

Given the critical nature of hospital operations, additional constraints are introduced:

6.1 Minimum Battery Reserve

A minimum state of charge (for example 20–30%) is maintained to ensure backup availability in case of emergencies.

6.2 Backup Operation (Grid Failure Scenario)

In the event of a grid outage:

1. The battery is used as the primary energy source
2. The hydrogen fuel cell acts as a secondary backup
3. Energy is prioritized for critical hospital loads

6.3 Critical vs Non-Critical Loads (Optional)

If load segmentation* is available:

• Critical loads (ICU, operating rooms, emergency systems) are always prioritized
• Non-critical loads may be reduced or shed during severe shortages

*dividing the hospital’s total electricity demand into different categories based on priority, function, or flexibility, so the energy system can make smarter decisions during shortages or optimization.

7 Hydrogen System Role

Due to lower round-trip efficiency compared to batteries:

• Hydrogen is primarily used for:
  • Long-term energy storage
  • Backup during extended shortages
• It is only produced when:
  • Battery storage is fully utilized
  • Excess renewable energy is available

8 Summary of Control Priorities

Energy Supply Priority (Deficit):

1. PV generation
2. Battery storage
3. Hydrogen fuel cell
4. Grid import

Energy Allocation Priority (Excess):

1. Serve load
2. Charge battery
3. Produce hydrogen
4. Export

9 Limitations of Rule-Based Approach

• Does not anticipate future demand or generation
• Cannot optimally respond to dynamic electricity prices
• May lead to suboptimal use of storage systems
• Lacks adaptability to changing conditions