Datasets:

Mobiusi
/

Flow-Battery-Energy-Scheduling-Optimization-Code-Dataset

+{
+  "optimization_code": "import cvxpy as cp\nimport numpy as np\n\ndef optimize_profile_tracking():\n    demand = np.array([5.5, 5.2, 5.0, 5.4, 6.2, 6.8, 6.4, 5.8])\n    renewable = np.array([1.0, 1.2, 1.8, 2.4, 1.7, 0.8, 0.3, 0.2])\n    target_import = np.array([4.3, 4.0, 3.6, 3.4, 4.2, 5.0, 5.4, 5.0])\n    n = len(demand)\n    dt = 0.25\n\n    energy_min, energy_max = 0.8, 7.2\n    initial_energy = 4.4\n    charge_max = discharge_max = 2.0\n    eta_charge = eta_discharge = 0.90\n    ramp_limit = 0.75\n    tracking_penalty = 120.0\n    throughput_cost = 0.8\n\n    charge = cp.Variable(n, nonneg=True)\n    discharge = cp.Variable(n, nonneg=True)\n    energy = cp.Variable(n + 1)\n    charging_mode = cp.Variable(n, boolean=True)\n    deviation_above = cp.Variable(n, nonneg=True)\n    deviation_below = cp.Variable(n, nonneg=True)\n\n    battery_net_output = discharge - charge\n    net_import = demand - renewable + charge - discharge\n    constraints = [\n        energy[0] == initial_energy,\n        energy[n] == initial_energy,\n        energy_min <= energy,\n        energy <= energy_max,\n        charge <= charge_max * charging_mode,\n        discharge <= discharge_max * (1 - charging_mode),\n        net_import - target_import == deviation_above - deviation_below,\n    ]\n    for t in range(n):\n        constraints.append(\n            energy[t + 1] == energy[t]\n            + eta_charge * charge[t] * dt\n            - discharge[t] * dt / eta_discharge\n        )\n    for t in range(1, n):\n        constraints += [\n            battery_net_output[t] - battery_net_output[t - 1] <= ramp_limit,\n            battery_net_output[t - 1] - battery_net_output[t] <= ramp_limit,\n        ]\n\n    tracking_cost = tracking_penalty * dt * cp.sum(\n        deviation_above + deviation_below\n    )\n    cycling_cost = throughput_cost * dt * cp.sum(charge + discharge)\n    problem = cp.Problem(cp.Minimize(tracking_cost + cycling_cost), constraints)\n    problem.solve(solver='SCIP')\n\n    if problem.status not in (cp.OPTIMAL, cp.OPTIMAL_INACCURATE):\n        raise RuntimeError(f'Optimization failed: {problem.status}')\n\n    return {\n        'objective_usd': float(problem.value),\n        'net_import_mw': np.round(net_import.value, 3).tolist(),\n        'charge_mw': np.round(charge.value, 3).tolist(),\n        'discharge_mw': np.round(discharge.value, 3).tolist(),\n        'energy_mwh': np.round(energy.value, 3).tolist(),\n        'tracking_deviation_mw': np.round(\n            deviation_above.value + deviation_below.value, 3\n        ).tolist(),\n    }\n\nif __name__ == '__main__':\n    print(optimize_profile_tracking())",
+  "scheduling_problem": "Dispatch an 8 MWh flow battery in eight 15-minute intervals to track a requested site net-import profile. Site demand is [5.5, 5.2, 5.0, 5.4, 6.2, 6.8, 6.4, 5.8] MW, renewable generation is [1.0, 1.2, 1.8, 2.4, 1.7, 0.8, 0.3, 0.2] MW, and the target net import is [4.3, 4.0, 3.6, 3.4, 4.2, 5.0, 5.4, 5.0] MW. The battery starts and ends at 4.4 MWh, must remain between 0.8 and 7.2 MWh, and has 2 MW charge and discharge limits. Use 90% charging and discharging efficiencies. Limit the change in battery net output between adjacent intervals to 0.75 MW. Minimize absolute net-import tracking error weighted at 120 USD/MWh of deviation, plus a throughput cost of 0.8 USD/MWh. Use a 0.25-hour time step, permit net export if the schedule requires it, and prevent simultaneous charging and discharging. Implement the mixed-integer optimization in CVXPY with SCIP."
+}

record_c50f070b-cb20-453f-8019-f256a2703b5c.json ADDED Viewed

	@@ -0,0 +1,4 @@

+{
+  "optimization_code": "import cvxpy as cp\nimport numpy as np\n\ndef optimize_arbitrage():\n    prices = np.array([34, 30, 28, 31, 45, 72, 88, 61], dtype=float)\n    n = len(prices)\n    dt = 1.0\n    capacity = 12.0\n    energy_min, energy_max = 1.2, 10.8\n    initial_energy = 6.0\n    charge_max = discharge_max = 3.0\n    eta_charge = eta_discharge = 0.90\n    throughput_cost = 1.5\n\n    charge = cp.Variable(n, nonneg=True)\n    discharge = cp.Variable(n, nonneg=True)\n    energy = cp.Variable(n + 1)\n    charging_mode = cp.Variable(n, boolean=True)\n\n    constraints = [\n        energy[0] == initial_energy,\n        energy_min <= energy,\n        energy <= energy_max,\n        energy[n] >= 6.0,\n        charge <= charge_max * charging_mode,\n        discharge <= discharge_max * (1 - charging_mode),\n    ]\n    for t in range(n):\n        constraints.append(\n            energy[t + 1] == energy[t]\n            + eta_charge * charge[t] * dt\n            - discharge[t] * dt / eta_discharge\n        )\n\n    energy_cost = dt * cp.sum(cp.multiply(prices, charge - discharge))\n    cycling_cost = throughput_cost * dt * cp.sum(charge + discharge)\n    problem = cp.Problem(cp.Minimize(energy_cost + cycling_cost), constraints)\n    problem.solve(solver='SCIP')\n\n    if problem.status not in (cp.OPTIMAL, cp.OPTIMAL_INACCURATE):\n        raise RuntimeError(f'Optimization failed: {problem.status}')\n\n    return {\n        'objective_usd': float(problem.value),\n        'charge_mw': np.round(charge.value, 3).tolist(),\n        'discharge_mw': np.round(discharge.value, 3).tolist(),\n        'energy_mwh': np.round(energy.value, 3).tolist(),\n    }\n\nif __name__ == '__main__':\n    print(optimize_arbitrage())",
+  "scheduling_problem": "Optimize an eight-hour energy-arbitrage schedule for a vanadium flow battery using the hourly electricity prices [34, 30, 28, 31, 45, 72, 88, 61] USD/MWh. The battery has 12 MWh of usable capacity, must remain between 1.2 and 10.8 MWh, starts at 6 MWh, and must finish with at least 6 MWh. Limit charging and discharging to 3 MW each. Use a 1-hour time step, 90% charging efficiency, and 90% discharging efficiency. Prevent simultaneous charging and discharging. Minimize energy purchase cost minus discharge revenue, with a throughput cost of 1.5 USD/MWh charged or discharged. Implement the mixed-integer optimization in CVXPY using SCIP and return the dispatch and state-of-charge schedule."
+}

record_d8619c25-57b7-4807-85c5-6c5499af5b0c.json ADDED Viewed

	@@ -0,0 +1,4 @@

+{
+  "optimization_code": "import cvxpy as cp\nimport numpy as np\n\ndef optimize_peak_shaving():\n    demand = np.array([4.0, 4.2, 4.5, 5.8, 6.2, 5.0], dtype=float)\n    solar = np.array([0.5, 1.0, 2.5, 3.5, 2.0, 0.3], dtype=float)\n    price = np.array([42, 38, 35, 76, 91, 55], dtype=float)\n    n = len(demand)\n    dt = 1.0\n\n    energy_min, energy_max = 1.0, 9.0\n    initial_energy = 5.0\n    charge_max = discharge_max = 2.5\n    eta_charge = eta_discharge = 0.90\n    import_limit = 8.0\n    demand_charge = 18.0\n    throughput_cost = 1.0\n\n    charge = cp.Variable(n, nonneg=True)\n    discharge = cp.Variable(n, nonneg=True)\n    energy = cp.Variable(n + 1)\n    charging_mode = cp.Variable(n, boolean=True)\n    curtailment = cp.Variable(n, nonneg=True)\n    grid_import = cp.Variable(n, nonneg=True)\n    peak_import = cp.Variable(nonneg=True)\n\n    constraints = [\n        energy[0] == initial_energy,\n        energy[n] == initial_energy,\n        energy_min <= energy,\n        energy <= energy_max,\n        charge <= charge_max * charging_mode,\n        discharge <= discharge_max * (1 - charging_mode),\n        curtailment <= solar,\n        grid_import <= import_limit,\n        grid_import <= peak_import,\n    ]\n    for t in range(n):\n        constraints += [\n            energy[t + 1] == energy[t]\n            + eta_charge * charge[t] * dt\n            - discharge[t] * dt / eta_discharge,\n            grid_import[t] == demand[t] - solar[t] + curtailment[t]\n            + charge[t] - discharge[t],\n        ]\n\n    energy_cost = dt * cp.sum(cp.multiply(price, grid_import))\n    demand_cost = demand_charge * peak_import\n    cycling_cost = throughput_cost * dt * cp.sum(charge + discharge)\n    problem = cp.Problem(\n        cp.Minimize(energy_cost + demand_cost + cycling_cost), constraints\n    )\n    problem.solve(solver='SCIP')\n\n    if problem.status not in (cp.OPTIMAL, cp.OPTIMAL_INACCURATE):\n        raise RuntimeError(f'Optimization failed: {problem.status}')\n\n    return {\n        'total_cost_usd': float(problem.value),\n        'grid_import_mw': np.round(grid_import.value, 3).tolist(),\n        'solar_curtailment_mw': np.round(curtailment.value, 3).tolist(),\n        'charge_mw': np.round(charge.value, 3).tolist(),\n        'discharge_mw': np.round(discharge.value, 3).tolist(),\n        'energy_mwh': np.round(energy.value, 3).tolist(),\n        'peak_import_mw': float(peak_import.value),\n    }\n\nif __name__ == '__main__':\n    print(optimize_peak_shaving())",
+  "scheduling_problem": "Schedule a 10 MWh flow battery over six one-hour intervals to reduce electricity costs and the billing-period import peak at a site with demand [4.0, 4.2, 4.5, 5.8, 6.2, 5.0] MW and on-site solar output [0.5, 1.0, 2.5, 3.5, 2.0, 0.3] MW. The import tariff is [42, 38, 35, 76, 91, 55] USD/MWh, and the demand charge is 18 USD per MW of maximum grid import for this modeled period. The battery starts and ends at 5 MWh, has an allowed energy range of 1 to 9 MWh, and is limited to 2.5 MW charging and 2.5 MW discharging. Use 90% charging and discharging efficiencies, a 1-hour time step, and a cycling cost of 1 USD/MWh throughput. Allow solar curtailment, prohibit grid export, and prevent simultaneous charging and discharging. Minimize energy charges, demand charges, and cycling cost using CVXPY with SCIP."
+}