Connections¶

Connections define how power flows between elements in your network with support for bidirectional flow, efficiency losses, and transmission costs.

Configuration¶

Use input number helpers to configure constant values.

Connection Endpoint Selection¶

The Source and Target fields show a dropdown of available elements that can be used as connection endpoints. The list of available elements is filtered based on connectivity level and your hub's Advanced Mode setting.

Why filtering? Standard elements (Grid, Battery, Solar, Load) create implicit connections automatically. Explicit connections between these elements are usually unnecessary and can lead to configuration errors. The filtering hides these elements by default to prevent common mistakes.

Filtering behavior:

• Advanced elements that require manual connection setup always appear in the selector regardless of Advanced Mode.
• Standard elements that create implicit connections automatically only appear when Advanced Mode is enabled.
• Connection elements never appear as endpoints to prevent invalid connection topologies.

This filtering ensures that connection endpoints are appropriate for your configuration level. Each element's documentation describes its connectivity level and when it appears in connection selectors.

Configuration Example¶

Bidirectional connection between two network nodes:

Physical Interpretation¶

Bidirectional flow: Both directions are available for optimization. The optimizer will choose the most cost-effective direction at each time step.

Efficiency modeling: Power leaving a node is measured before losses. Power arriving at a node is reduced by efficiency. Example: 10kW leaves source with 95% efficiency → 9.5kW arrives at target.

Asymmetric efficiency: Configure different efficiencies for each direction to model real-world devices.

Transmission costs: Connection pricing models fees for using a power transfer path (wheeling charges, connection fees, peak demand charges).

Common Patterns¶

Unlimited Bidirectional Connection¶

Leave both power limits unset for unlimited flow in both directions:

Unidirectional Connection¶

Set one direction's limit to 0 to prevent flow:

Bidirectional Connection with Efficiency¶

Model power conversion with efficiency losses:

Availability Windows¶

Use time-varying sensor to model device availability. Example: EV only available for charging 6 PM to 8 AM.

Create a template sensor:

template:
  - sensor:
      - name: EV Charging Availability
        unit_of_measurement: W
        state: >
          {% set hour = now().hour %}
          {% if hour >= 18 or hour < 8 %}
            7200
          {% else %}
            0
          {% endif %}

Then configure the connection:

The optimizer will only schedule charging when the sensor value is non-zero.

Input Entities¶

Each configuration field creates a corresponding input entity in Home Assistant. Input entities appear as Number entities with the config entity category.

Input entities include a forecast attribute showing values for each optimization period. See the Input Entities developer guide for details on input entity behavior.

Sensors Created¶

Sensor Summary¶

A Connection element creates 1 device in Home Assistant with the following sensors. Not all sensors are created for every connection - only those relevant to the configuration.

The sensor display names use the actual source and target element names configured for the connection. For example, a connection between two elements would show their actual names in the sensor name instead of generic "Source to Target".

Source to Target Power¶

The optimal power flowing from the source element to the target element.

Values are always positive or zero. A value of 0 means no power is flowing in the forward direction (may be flowing in reverse or not at all). The direction is determined by the connection configuration (source → target).

Example: A value of 3.5 kW means 3.5 kW is flowing from the source element to the target element at this time period.

Target to Source Power¶

The optimal power flowing from the target element to the source element.

Values are always positive or zero. A value of 0 means no power is flowing in the reverse direction (may be flowing forward or not at all). This represents reverse flow through the connection (target → source).

Example: A value of 2.0 kW means 2.0 kW is flowing from the target element back to the source element at this time period.

Max Source to Target Power¶

The configured maximum forward power limit from the sensor configuration. Only created when a forward power limit is configured.

Max Target to Source Power¶

The configured maximum reverse power limit from the sensor configuration. Only created when a reverse power limit is configured.

Max Source to Target Power Shadow Price¶

The marginal value of additional forward capacity (source → target). See the Shadow Prices modeling guide for general shadow price concepts.

This shadow price shows how much the total system cost would decrease if the forward power limit were increased by 1 kW at this time period. Only created when a forward power limit is configured.

Interpretation:

• Zero value: Connection has spare capacity in the forward direction (not at limit)
• Positive value: Connection is at maximum forward capacity and constraining power flow
  • The value shows how much system cost would decrease per kW of additional forward capacity
  • Higher values indicate the forward capacity limit is causing significant cost increases
  • Helps identify bottlenecks where more forward capacity would be valuable

Example: A value of 0.08 means that if the connection could transfer 1 kW more in the forward direction, the total system cost would decrease by $0.08 at this time period.

Max Target to Source Power Shadow Price¶

The marginal value of additional reverse capacity (target → source). See the Shadow Prices modeling guide for general shadow price concepts.

This shadow price shows how much the total system cost would decrease if the reverse power limit were increased by 1 kW at this time period. Only created when a reverse power limit is configured.

Interpretation:

• Zero value: Connection has spare capacity in the reverse direction (not at limit)
• Positive value: Connection is at maximum reverse capacity and constraining power flow
  • The value shows how much system cost would decrease per kW of additional reverse capacity
  • Higher values indicate the reverse capacity limit is causing significant cost increases
  • Helps identify bottlenecks where more reverse capacity would be valuable

Example: A value of 0.12 means that if the connection could transfer 1 kW more in the reverse direction, the total system cost would decrease by $0.12 at this time period.

Time Slice Shadow Price¶

The marginal value of relaxing the time-slicing constraint. See the Shadow Prices modeling guide for general shadow price concepts.

This shadow price shows how much the total system cost would decrease if simultaneous bidirectional power flow were less restricted at this time period. Only created when both forward and reverse power limits are configured.

The time-slicing constraint prevents full power flow in both directions simultaneously: P_forward/P_max_forward + P_reverse/P_max_reverse ≤ 1.0. This models real-world devices that share capacity between directions (e.g., an inverter that can't operate at full charge and discharge simultaneously).

Interpretation:

• Zero value: Connection is not constrained by time slicing (operating in only one direction or well below limits)
• Positive value: Time slicing is constraining bidirectional operation
  • The value shows how much system cost would decrease if the constraint were relaxed by 1% (allowing the sum to reach 1.01)
  • Higher values indicate the connection could benefit from being able to operate more simultaneously in both directions
  • Helps identify devices where increased bidirectional capacity would be valuable

Example: A value of 0.15 means that if the connection could operate slightly more simultaneously in both directions (sum ≤ 1.01 instead of ≤ 1.0), the total system cost would decrease by $0.15 at this time period.

All sensors include a forecast attribute containing future optimized values for upcoming periods.

Troubleshooting¶

See troubleshooting guide for connection issues.

Next Steps¶