A steel processing plant outside Casablanca, Morocco cut diesel consumption from 4.2 million liters per year to 0.6 million liters per year with a 12 MW PV + 25 MW / 100 MWh BESS + 8 MW diesel backup hybrid — achieving an LCOE of $0.087/kWh over 25 years. The project eliminates 86% of the site's fuel consumption while maintaining 100% supply reliability for a continuous industrial process that tolerates zero downtime.
This case study walks through the project context, system design, EMS dispatch strategy, financial outcomes, and the key lessons learned from a simulation-driven engineering approach. All results presented are simulation estimates based on configured parameters in Energy Optima using the Casablanca Steel Factory Off-Grid demo project (project ID: 8757af9c) and should be validated against site-specific measurement data before final investment decisions.
What You'll Learn
Project Context: Why Off-Grid Steel?
The facility is a 40,000 m² steel processing plant located 45 km southeast of Casablanca, Morocco, in the industrial zone of Berrechid. The plant operates continuous electric arc furnace (EAF) and rolling mill processes, drawing a base load of approximately 14 MW with peak demand reaching 22 MW during full-production shifts. The existing power supply consisted entirely of diesel generators — 18 MW of installed capacity across six units — running 8,400 hours per year.
The problems with the all-diesel baseline were significant:
- Fuel consumption: 4.2 million liters of diesel per year, delivered by truck convoy on a weekly schedule
- Effective energy cost: $0.18/kWh (diesel-only generation), roughly 2.5× the Moroccan grid tariff and 3-4× typical utility-scale solar LCOEs in North Africa
- Supply chain risk: Fuel deliveries were disrupted twice in the previous year due to road closures, forcing production curtailment with estimated losses of $120,000 per day of downtime
- Carbon exposure: The site emitted approximately 11,300 metric tons of CO&sub2; annually from diesel combustion, with no offset strategy
- Maintenance burden: Generator overhauls every 12,000 operating hours at $85,000 per unit, plus unscheduled repairs averaging $140,000/year
The plant operators had three options: negotiate a new grid connection (estimated 18-month timeline with $4.2M in grid extension costs), continue with diesel-only generation (escalating fuel cost trajectory), or build a behind-the-meter hybrid renewable system. The third option — a solar + battery + diesel hybrid — offered the fastest deployment timeline (8-10 months) and the strongest financial case, provided the system could maintain 100% supply reliability for a process intolerant of power interruptions longer than 200 milliseconds.
Key insight: Off-grid industrial sites with continuous processes present a unique opportunity for hybrid renewable systems. The high baseline cost of diesel generation ($0.18/kWh) creates a large headroom for PV and BESS investment that would not exist at grid-connected sites. In this case, the diesel LCOE was more than double the hybrid system's projected LCOE of $0.087/kWh, giving the project a $0.093/kWh margin to absorb capital costs.
System Design: PV, BESS, and Diesel Configuration
The system design was sized using Energy Optima's hybrid optimization engine, which iterated through 1,200+ configuration combinations across PV capacity, BESS power/energy ratio, and diesel generator commitment strategy. The final design balances renewable penetration against dispatch flexibility and capital efficiency.
PV Array: 12 MW DC (Jinko Tiger Neo 650W Bifacial)
The solar array consists of 18,462 Jinko Tiger Neo 650W bifacial modules arranged across 180 hectares of single-axis tracking. The bifacial gain is modeled at 10% (conservative for the high-albedo semi-arid Moroccan terrain), giving an effective DC capacity of 13.2 MW at peak albedo conditions. Sungrow 250 kW string inverters are deployed in a 50-unit configuration, with a DC/AC ratio of 1.33 to extend the production curve into the late afternoon when the plant's evening shift ramps up.
The annual solar generation is estimated at 23.4 GWh (1,950 kWh/kWp), based on the NREL NSRDB typical meteorological year dataset for the Berrechid region. The capacity factor after inverter clipping, bifacial gain, and soiling losses (3% annual, with monthly washing schedule) is 19.6%.
Notably, the PV array is oversized by 20% relative to the average daytime load of approximately 10 MW. This deliberate oversizing ensures that even on partially cloudy days, solar generation covers the full daytime load — and on clear days, the surplus charges the battery while the diesel generators remain off.
BESS: 25 MW / 100 MWh (CATL EnerOne Plus LFP)
The battery energy storage system uses CATL EnerOne Plus LFP cells in a 25 MW / 100 MWh configuration (4-hour duration at rated power, 0.25C average C-rate). The EnerOne Plus product provides containerized LFP storage with integrated liquid thermal management, delivering a round-trip efficiency of 92% at 0.25C and a warranted cycle life of 8,000 cycles to 70% SOH under standard operating conditions (25°C, 90% DoD per cycle).
The system is configured with a 10% minimum SOC (reserve for black-start capability) and 90% maximum SOC (to reduce calendar aging at high voltage). The effective usable capacity is therefore 80 MWh out of 100 MWh nameplate, or 3.2 hours at rated power. The BESS operates through Sungrow SC2500HV-MV inverter stations configured for grid-forming operation — essential for the off-grid scenario where the BESS must establish the voltage and frequency reference when the diesel generators are off.
Diesel Backup: 8 MW (Low-Load Operation)
Four 2 MW Caterpillar C32 diesel generators provide backup power. The diesel fleet is operated with a 30% minimum loading constraint per unit — a critical operational requirement that shaped the EMS dispatch logic. Below 30% load, cylinder glazing and wet stacking accelerate maintenance intervals by up to 60%.
The total diesel capacity (8 MW) is less than the peak load (22 MW) by design: the hybrid system relies on the battery to supply the gap between diesel output and load during periods of low solar generation. The diesel-only mode is reserved for extended periods of low solar resource (3+ consecutive overcast days), a scenario with less than 2% annual probability in the Berrechid climate zone.
| Component | Specification | Key Details |
|---|---|---|
| PV Modules | 12 MW DC (18,462 units) | Jinko Tiger Neo 650W bifacial, single-axis tracking |
| Inverters | 50 × 250 kW | Sungrow string inverters, 1.33 DC/AC ratio |
| BESS Power | 25 MW | CATL EnerOne Plus LFP, liquid thermal management |
| BESS Energy | 100 MWh | 4-hour duration, 92% RTE, 0.25C average |
| Diesel Backup | 8 MW (4 × 2 MW) | Caterpillar C32, 30% min load constraint |
| Useful BESS Capacity | 80 MWh | 10% min / 90% max SOC operating envelope |
EMS Dispatch Strategy: PV First, Always
The energy management system implements a three-tier priority dispatch: PV → BESS → Diesel. At each 5-minute dispatch interval, the EMS evaluates the net load (load minus PV generation) and dispatches resources in priority order.
Tier 1 — PV (always priority): Solar generation is always accepted. If PV exceeds load, the surplus charges the battery until the battery reaches 90% SOC, at which point the inverters curtail PV output. There is no diesel minimum-run constraint that forces curtailment of PV — a deliberate design choice that maximizes renewable utilization.
Tier 2 — BESS (economic dispatch): The BESS follows an economic dispatch with 24-hour lookahead model predictive control. The optimizer solves a linear program that minimizes total operating cost (diesel fuel cost + battery degradation cost + generator startup cost) over the rolling horizon, subject to SOC dynamics, power limits, and diesel minimum runtime constraints. The economic dispatch formulation automatically decides whether to discharge the battery now or reserve capacity for a future high-load or low-solar period.
Tier 3 — Diesel (curtailment with minimum runtime): Diesel generators are the last resort. When the BESS alone cannot cover the net load (typically during extended overcast periods or battery SOC floor events), the EMS commits the minimum number of diesel units. The diesel generators are constrained by a 30% minimum load per unit and a 120-minute minimum runtime once started — meaning once a generator turns on, it must run for at least two hours before being allowed to shut down, even if solar generation recovers during that window.
Key insight: The 120-minute diesel minimum runtime constraint proved to be one of the most consequential parameters in the simulation. Because diesel generators cannot power-cycle every 15 minutes based on cloud movement, the BESS must absorb the full solar variability during the diesel's mandatory runtime window. This constraint increased BESS cycling by 30% compared to an idealized system with zero minimum diesel runtime — a finding that directly informed the battery degradation and augmentation projections. Operators evaluating similar hybrid systems should model this interaction explicitly rather than assuming unconstrained diesel dispatch.
Financial Results: $18.2M NPV at 8% Discount
The financial model covers a 25-year project life with an 8% discount rate, reflecting the risk profile of a Moroccan off-grid industrial project with currency exposure (MAD-denominated revenue, USD-denominated capital costs). The model assumes diesel fuel cost escalation at 2.5% per year, PV module degradation at 0.5%/year (linear), and O&M escalation at 2%/year.
Capital expenditures:
- PV system (12 MW): $9.6M ($0.80/W installed)
- BESS (25 MW / 100 MWh): $28.0M ($280/kWh installed)
- Balance of plant, EMS, grid-forming inverters: $3.8M
- Engineering, procurement, construction management: $2.1M
- Total CAPEX: $43.5M
Operating expenditures (first year):
- Diesel fuel (0.6M L @ $0.80/L): $0.48M
- PV O&M ($12/kW/yr): $0.14M
- BESS O&M ($20/kW/yr): $0.50M
- Generator maintenance: $0.12M
- Total OPEX (first year): $1.24M (vs. $3.90M diesel-only baseline)
Key financial metrics:
| Metric | Value |
|---|---|
| 25-year NPV (8% discount rate) | $18.2M |
| Internal Rate of Return (IRR) | 14.7% |
| Levelized Cost of Energy (LCOE) | $0.087/kWh |
| Simple Payback Period | 6.8 years |
| Diesel savings (annual) | 3.6M L/yr |
| Fuel cost savings (annual) | $2.88M/yr (at $0.80/L) |
| Renewable fraction (energy) | 86% |
| Carbon reduction (annual) | 9,720 metric tons CO&sub2; |
The $2.88 million annual fuel savings is the primary revenue driver — it represents a 74% reduction in the site's energy cost line item. The 25-year NPV of $18.2 million assumes zero carbon revenue or green certificate monetization, which would add approximately $2-4 million to the NPV at current carbon market prices in Morocco's emerging voluntary carbon market.
The payback period of 6.8 years is within the acceptable range for industrial energy projects in North Africa (typical threshold is 7-9 years). The IRR of 14.7% exceeds the weighted average cost of capital of 8% by a comfortable margin, providing an equity return spread of 6.7 percentage points.
BESS Degradation and Augmentation Planning
Battery degradation is a critical concern for a 25-year project that cycles the BESS daily. The CATL EnerOne Plus LFP cells are modeled with a combined calendar + cycle aging model calibrated to the manufacturer's warranty curve. The degradation simulation parameters are based on the cell-level test data published by CATL for the EnerOne Plus series, scaled to system-level assuming uniform cell temperature and SOC distribution with a 0.5°C imbalance allowance.
Degradation assumptions:
- Average daily cycles: 0.5 cycles/day (annual throughput of 18,250 MWh through the 100 MWh nameplate)
- Average C-rate: 0.25C (consistent with 4-hour duration design)
- Operating temperature: 28°C average (liquid thermal management holds cells within 25-32°C)
- Calendar aging contribution: 1.2% SOH loss/year in the first 5 years, decelerating to 0.6%/year by year 20
- Cycle aging contribution: 0.003% SOH loss per equivalent full cycle at 0.25C
Simulated degradation trajectory:
| Year | Estimated SOH | Usable Capacity (MWh) | Cycles Accumulated |
|---|---|---|---|
| 0 (Commissioning) |
100.0% | 80.0 | 0 |
| 5 | 91.4% | 73.1 | 913 |
| 10 | 85.2% | 68.2 | 1,825 |
| 14 | 80.4% | 64.3 | 2,555 |
| 14 (After Augmentation) |
80.4% | 79.3 | 2,555 |
| 20 | 77.9% | 76.9 | 3,650 |
| 25 | 76.4% | 75.4 | 4,563 |
Augmentation event — Year 14: The simulation identifies a single augmentation event in year 14, when the BESS SOH drops to 80.4% — below the 80% threshold at which the system can no longer meet the 80 MWh minimum dispatch requirement during the evening peak window. The augmentation adds 15 MWh of new LFP cells (increasing the nameplate capacity to 115 MWh), restoring usable capacity to 79.3 MWh (accounting for the degraded existing cells). The augmentation cost is estimated at $3.15M ($210/kWh for 15 MWh, including installation, commissioning, and thermal management integration).
The year-25 SOH of 76.4% is above the 70% end-of-life threshold commonly used for LFP systems, meaning the battery could continue operation beyond the 25-year financial horizon without full replacement. The relatively gentle degradation is a result of the conservative 0.25C average C-rate and the moderate 0.5 cycles/day utilization — both consequences of the 4-hour duration design and the PV-oversized dispatch strategy that reduces BESS cycling on sunny days.
Key insight: The single augmentation in year 14 rather than a mid-life battery replacement is made possible by two design choices: (1) the 0.25C LFP chemistry with 8,000-cycle warranty, and (2) the PV oversizing strategy that reduces average daily BESS cycles from a potential 0.8 to 0.5. For the project's 25-year horizon, this saves approximately $8-10M compared to a full battery replacement scenario at year 12-13. Augmentation planning should be incorporated into the initial financial model — not treated as an afterthought during operations.
Key Lessons and Design Trade-Offs
The Casablanca Steel simulation reveals several design principles that generalize to other off-grid industrial hybrid projects. These lessons emerged from sensitivity analysis across the 1,200+ configurations tested during the design phase.
Lesson 1: Diesel minimum runtime constraints are a major driver of BESS cycling. The 120-minute minimum runtime on diesel generators increased total BESS cycling by 30% compared to a theoretical system with instant diesel start/stop capability. When a diesel generator must stay on for two hours, the BESS absorbs solar ramps and load fluctuations during that window, accumulating cycles that would not exist if the diesel could shut down immediately upon solar recovery. Operators should model this interaction explicitly — a common mistake is to assume diesel generators can cycle freely, which leads to a 15-20% underestimate of BESS degradation.
Lesson 2: Oversizing PV by 20% relative to load reduces diesel runtime further. The standard sizing rule for off-grid hybrids is PV capacity equal to 80-100% of peak load. This project pushes to 120% (12 MW PV for a 10 MW average daytime load). The extra 20% means that on a partially cloudy day receiving 80% of clear-sky irradiance, the PV array still produces 9.6 MW — enough to cover the entire daytime load without diesel or battery. On clear days, the surplus charges the BESS to 90% SOC by 14:00, enabling the BESS to carry the entire evening load until 21:00-22:00 without diesel support. The incremental cost of adding 20% more PV capacity ($1.6M) is recovered in 3.2 years through reduced diesel runtime and BESS cycle preservation.
Lesson 3: BESS augmentation timing is sensitive to the operating SOC window. Widening the SOC window from 10-90% (80 MWh usable) to 5-95% (90 MWh usable) delays the first augmentation from year 14 to year 18, but increases calendar aging by 15% due to the sustained high-SOC operation. For the Casablanca project, the 10-90% window was preferred because the augmentation cost ($3.15M) was lower than the incremental degradation cost of the wider SOC window over 25 years ($4.1M in earlier replacement costs).
Lesson 4: The economic dispatch with 24-hour lookahead is essential for an off-grid system with time-coupled diesel constraints. A simple rule-based dispatch (e.g., "discharge battery after sunset until SOC reaches 30%") would not account for the diesel minimum runtime. Under rule-based dispatch, the battery could discharge to 30% SOC before diesel starts, then the diesel would be forced to run for 120 minutes even if load drops — wasting fuel. The economic dispatch with MPC sees the minimum runtime constraint and reserves additional battery capacity to minimize diesel runtime, a net benefit of approximately $0.7M NPV over the project life.
Lesson 5: The diesel-only LCOE of $0.18/kWh is the most important enabling condition. This case study works financially because the alternative (diesel-only) is expensive. For projects where the baseline energy cost is lower (grid-connected sites at $0.06-0.10/kWh), the LCOE advantage of the hybrid system narrows significantly. The Casablanca project's 14.7% IRR depends on the $2.88M annual fuel savings — a number that is specific to the site's diesel dependency and local fuel pricing.
Key insight: The most surprising finding from the sensitivity analysis was that PV oversizing had a larger impact on diesel runtime reduction than BESS capacity expansion. Adding 2 MW of PV (from 10 MW to 12 MW) reduced annual diesel runtime by 340 hours, while adding 25 MWh of BESS (from 75 MWh to 100 MWh) reduced diesel runtime by only 115 hours. This asymmetry is a function of the Moroccan solar resource profile — extended clear-sky periods mean that marginal PV capacity displaces fuel more effectively than marginal battery capacity. The reverse would be true in a higher-latitude, more overcast climate.
How the Simulation Was Built
The Casablanca Steel project was modeled entirely within Energy Optima's simulation platform using the following workflow:
- Project setup. The off-grid hybrid project was created in Energy Optima (project ID: 8757af9c) with Morocco as the geographic location. The Berrechid industrial zone coordinates (33.26°N, 7.58°W) were used for solar resource data via the NREL NSRDB API integration.
- Component selection. PV modules were selected from Energy Optima's database of 136 PV modules (Jinko Tiger Neo 650W bifacial selected). Batteries were chosen from 112 battery models in the database (CATL EnerOne Plus LFP). Diesel generators were drawn from the database of 158 diesel gensets across 19 manufacturers. Inverters were selected from 200 inverter models across 20 manufacturers (Sungrow string inverters selected).
- EMS configuration. The dispatch strategy was set to economic dispatch with 24-hour MPC lookahead. SOC thresholds: min 10%, max 90%. Priority: PV → BESS → Diesel. Diesel minimum runtime: 120 minutes. Diesel minimum load: 30%. The degradation penalty coefficient was calibrated to the CATL cell-level cycle aging data.
- Financial modeling. CAPEX inputs were entered per component. Fuel cost was set at $0.80/L with 2.5% annual escalation. Discount rate: 8%. Project horizon: 25 years. Augmentation was configured as a single event in year 14 at $210/kWh.
- Simulation and sensitivity. The base case simulation was run, followed by sensitivity sweeps across PV capacity (8-16 MW in 1 MW steps), BESS duration (2-6 hours), and diesel capacity (4-12 MW). The optimized configuration was selected based on maximum NPV.
For a detailed walkthrough of hybrid diesel-solar system design, see our guide: Diesel-Solar Hybrid System Design: Sizing, Dispatch, and Economics. For more on the simulation tools used in this analysis, see Microgrid Simulation Software: A Practical Comparison for Engineers. For a deeper dive into the dispatch logic, see EMS Dispatch Strategies for BESS: Maximizing Revenue with Economic Optimization.
Disclaimer: All results presented in this case study are simulation estimates based on configured parameters in Energy Optima. Actual project outcomes will depend on site-specific conditions including solar resource variability, diesel fuel quality and pricing, battery cell manufacturing variance, load profile changes, and regulatory developments. Energy Optima's simulation platform is designed for preliminary feasibility analysis and system optimization — final investment decisions should be supported by detailed engineering design and site-specific measurement data.