How It Works

Closing the feedback loop on the electrode cycle

Electrochemical DLE runs on intercalation and elution phases with different optimal potential windows. EELI monitors electrode state and incoming brine chemistry continuously, then adjusts setpoints to keep both phases near their efficiency ceiling.

The Control Sequence

Four stages, one continuous loop

EELI operates as a real-time process, not a batch optimization. Each stage feeds the next, creating a feedback loop that responds to conditions as they change rather than on a fixed schedule.

1

Upstream brine characterization

Inline ion-selective electrode sensors and conductivity probes at the brine feed inlet measure Li, Mg, Ca, and total dissolved solids continuously. These readings enter EELI's feed-forward model 12 to 40 minutes before the brine front reaches the electrode stack, providing the lead time needed for proactive setpoint adjustment.

2

Electrode state estimation

On-stack electrochemical impedance spectroscopy measurements feed a continuous model of Li+ loading fraction in the intercalation material. The state estimator distinguishes between normal loading variation, Mg co-intercalation signatures, and early-stage fouling responses in the impedance spectrum, updating every 30 seconds throughout the cycle.

3

Cycle optimization

The cycle optimizer combines the brine feed forecast with the real-time electrode state estimate to compute optimal potential setpoints for the intercalation, elution, and rinse phases of the next cycle. The optimization balances recovery rate against purity targets that operators set within EELI's dashboard interface. Cycle duration adjusts dynamically based on actual loading fraction rather than a fixed timer.

4

PLC setpoint delivery

Updated potential setpoints are sent to the site PLC at cycle phase boundaries via Modbus TCP or OPC-UA. EELI also monitors for fouling indicator thresholds and can trigger targeted rinse cycles when the impedance signature crosses early-warning limits. All setpoint changes are logged with timestamp and causal state data for operator review and retrospective analysis.

Control Architecture

The closed-loop control architecture

EELI's control loop runs continuously alongside the electrode cycle. The feedback path closes at the state estimator: what the electrode actually did informs the setpoints for the next phase.

BRINE FEED Li, Mg, pH Continuous Feed-fwd FEED-FORWARD BRINE MODEL Composition forecast 12-40 min horizon CYCLE OPTIMIZER Setpoint compute Recovery vs purity Setpoints PLC Modbus TCP OPC-UA ELECTRODE STACK Intercalation / Elution / Rinse STATE ESTIMATOR EIS analysis State feedback WHAT EELI ADDS Feed-forward model Anticipates composition changes before they reach the stack State estimator Tracks electrode loading in real time via EIS WHAT STAYS YOURS PLC safety interlocks All emergency stops and manual overrides remain authoritative Existing electrode hardware No hardware replacement required; EELI is software-only
Common Questions

Technical questions on EELI integration

At minimum, EELI requires inline Li concentration and conductivity measurements at the brine feed inlet. Adding Mg and pH sensors improves selectivity model accuracy. Sensors do not need to be EELI-supplied; the platform accepts standard 4-20mA analog signals and Modbus-enabled digital sensor interfaces from any major industrial supplier.
The state estimator's EIS model is parameterized during initial commissioning against your specific electrode material. We currently have validated parameter sets for lambda-MnO2 and FePO4-based intercalation materials. For other electrode chemistries, we conduct a 2-4 week characterization campaign at site commissioning to build the electrode-specific model parameters before handing off to autonomous control.
On communication loss, the PLC reverts to the last confirmed setpoints from EELI or falls back to its onboard fixed setpoints if a watchdog timer expires. EELI does not hold the PLC in a suspended state on communication loss. The control handoff design ensures the process continues safely at reduced efficiency rather than stopping, which is the operator preference in most DLE deployments we have worked with.
No. All state estimation and cycle optimization runs on an on-site edge node. Cloud connectivity is optional and is only used for remote monitoring, historian data backup, and software updates. The control loop operates entirely on the local network. Sites with secure air-gap requirements can run EELI in fully offline mode.

Ready to close the control loop?

We work through a structured pilot process: brine chemistry review, sensor placement assessment, PLC communication mapping, and a commissioning timeline that fits your operational schedule.

Request a Process Audit Platform details