Skip to content

Commit

Permalink
Merge branch 'master' of github.com:dpinney/omf
Browse files Browse the repository at this point in the history
  • Loading branch information
@ASC95
ASC95 committed Nov 6, 2024
2 parents 283bd64 + 0b481ce commit 319a4cc
Show file tree
Hide file tree
Showing 12 changed files with 705 additions and 570 deletions.
2 changes: 2 additions & 0 deletions omf/models/derConsumer.html
View file
Original file line number Diff line number Diff line change
Expand Up @@ -258,13 +258,15 @@
<p class="inputSectionHeader">Distributed Energy Resource (DER) Program Design Inputs</p>
</div>
<hr style="border-style: solid; border-color: #196b12; margin-top: 10px;">
<!--
<div class="shortInput">
<label class="tooltip">Enable DER sharing program<span class="classic">Enabling DER sharing program will perform a cost-benefit analysis of a utility's DER sharing program.</span></label>
<select id="utilityProgram" name="utilityProgram" value="{{allInputDataDict.utilityProgram}}">
<option value="True" {{ 'selected' if allInputDataDict.utilityProgram == 'Yes' }}>Yes</option>
<option value="False" {{ 'selected' if allInputDataDict.utilityProgram == 'No' }}>No</option>
</select>
</div>
-->
<div class="shortInput">
<label class="tooltip">Energy Compensation Rate ($/kWh)<span class="classic">The dollar amount per kWh compensated to the member-consumer.</span></label>
<input type="text" id="rateCompensation" name="rateCompensation" value="{{allInputDataDict.rateCompensation}}" pattern="^\d+\.?\d*?$" min="0" required="required"/>
Expand Down
268 changes: 11 additions & 257 deletions omf/models/derConsumer.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -433,280 +433,34 @@ def work(modelDir, inputDict):
else:
PV = np.zeros_like(demand)

## NOTE: This section for BESS uses REopt's BESS, which we are finding is inconsistent and tends to not work
## unless an outage is specified and no generator is enabled (if we specify a generator, REopt will not build a BESS).
## We are instead going to try to use our own BESS model.
## Currently, the BESS model is only being run when considering a DER utility program is enabled.
## TODO: Change this to instead check for BESS in the output file, rather than input
## Using REopt's Battery Energy Storage System (BESS) output
if 'ElectricStorage' in reoptResults and any(reoptResults['ElectricStorage']['storage_to_load_series_kw']): ## BESS
print("Using REopt's BESS output. \n")
BESS = reoptResults['ElectricStorage']['storage_to_load_series_kw']
grid_charging_BESS = np.asarray(reoptResults['ElectricUtility']['electric_to_storage_series_kw'])
if 'PV' in reoptResults:
grid_charging_BESS += np.asarray(reoptResults['PV']['electric_to_storage_series_kw'])
outData['chargeLevelBattery'] = reoptResults['ElectricStorage']['soc_series_fraction']

## Update the monthly consumer savings
## Add BESS compensation amount to vbatDispatch's thermal BESS savings
monthHours = [(0, 744), (744, 1416), (1416, 2160), (2160, 2880),
(2880, 3624), (3624, 4344), (4344, 5088), (5088, 5832),
(5832, 6552), (6552, 7296), (7296, 8016), (8016, 8760)]
BESS_compensation_monthly = np.asarray([sum(BESS[s:f]) for s, f in monthHours])
outData['savings'] = list(np.asarray(outData['savings'])+np.asarray(BESS_compensation_monthly)) ## NOTE: There is likely a better way to add two lists together, but this works for now

else:
print('No BESS found in REopt: Setting BESS data to 0. \n')
BESS = np.zeros_like(demand)
grid_charging_BESS = np.zeros_like(demand)
outData['chargeLevelBattery'] = np.zeros_like(demand)

#BESS = np.zeros_like(demand)
#grid_charging_BESS = np.zeros_like(demand)
#outData['chargeLevelBattery'] = list(np.zeros_like(demand))

## NOTE: The following 3 lines of code read in the SOC info from a static reopt test file
#with open(pJoin(__neoMetaModel__._omfDir,'static','testFiles','residential_reopt_results.json')) as f:
# static_reopt_results = json.load(f)
#outData['chargeLevelBattery'] = static_reopt_results['outputs']['ElectricStorage']['soc_series_fraction']

########## DER Sharing Program options ######################################################################################################################################################
if (inputDict['utilityProgram']):
print('Considering utility DER sharing program \n')

## Gather TOU rates
#latitude = float(inputDict['latitude'])
#longitude = float(inputDict['longitude'])
## TODO: To make this more modular, require user to input specific urdb label and use the getpage variable in get_tou_rates() function
## NOTE: Temporarily override lat/lon for a utility that has TOU rates specifically
inputDict['latitude'] = 39.986771
inputDict['longitude'] = -104.812599 ## Brighton, CO
rate_info = get_tou_rates(modelDir, inputDict)

## Look at all the "name" keys containing "TOU" or "time of use"
#filtered_names = [item['name'] for item in rate_info['items'] if 'TOU' in item['name'] or 'time-of-use' in item['name'] or 'Time of Use' in item['name']]
#for name in filtered_names:
# print(name) ## Print the filtered names

## Select one of the name keys (in this case, the residential TOU)
TOUdata = []
TOUname = 'Residential Time of Use'
for item in rate_info['items']:
if item['name'] == TOUname:
TOUdata.append(item)
#print(TOUdata)

## NOTE: ad-hoc TOU rates were used when the rate_info was not working.
#tou_rates = get_tou_rates_adhoc()

PV_series = pd.Series(PV, index=timestamps)
demand_series = pd.Series(demand, index=timestamps)
temperature_series = pd.Series(temperatures, index=timestamps)
BESS_series = pd.Series(BESS, index=timestamps)
#grid_serving_load_series = pd.Series(grid_to_load, index=timestamps)
## NOTE: REopt gives grid_to_load, but for now I'm just prioritizing DERs serving the load and buying grid if needed.
## If we don't want to prioritize DERs serving the load before buying from grid, then need to integrate REopt's grid_to_load
## into the code loop below.

## Create weekday and weekend TOU schedules
weekday_tou_schedule = create_tou_schedule(TOUdata[0]['energyweekdayschedule'])
weekend_tou_schedule = create_tou_schedule(TOUdata[0]['energyweekendschedule'])

## Combine the weekday and weekend schedules
tou_schedules = {'weekday': weekday_tou_schedule, 'weekend': weekend_tou_schedule}
tou_structure = TOUdata[0]['energyratestructure']
fixed_monthly_charge = TOUdata[0]['fixedmonthlycharge']
compensation_rate = float(inputDict['rateCompensation'])
subsidy_amount = float(inputDict['subsidy'])

## Identify the on- and off-peak periods
on_peak_periods = []
off_peak_periods = []
for period, rate in enumerate(tou_structure):
if rate[0]['rate'] == max(tou_structure, key=lambda x: x[0]['rate'])[0]['rate']:
on_peak_periods.append(period)
if rate[0]['rate'] == min(tou_structure, key=lambda x: x[0]['rate'])[0]['rate']:
off_peak_periods.append(period)

## Create dataframe to store the schedule
schedule = pd.DataFrame({
'Solar Generation (kW)': PV_series,
'Household Demand (kW)': demand_series,
'Grid Serving Load (kW)': np.zeros(len(timestamps)), ## Placeholder; will get updated in the loop
'Battery State of Charge': np.zeros(len(timestamps)), ## Placeholder
}, index=timestamps)

## BESS physical parameters
## TODO: Create user inputs on the HTML side if keeping this battery model
battery_energy_capacity = 13.5 ## kWh; Tesla Powerwall 2
battery_max_charge_rate = 3.3 ## kW
battery_max_discharge_rate = 3.3 ## kW
#battery_efficiency = 0.9 # Efficiency of the battery ## NOTE: maybe add this in later?
battery_soc = 0 ## Initial battery state of charge
battery_soc_min = 0.2 * battery_energy_capacity ## Min= 20% of battery capacity

## BESS allowed for utility use
## NOTE: I just realized that this code is set up to only use 80% of the battery to do peak shaving, rather than
## giving/selling the 80% to the utility. TODO: how do we modify this? Lisa and I are thinking of ways to do (and plot) this
max_utility_usage_percentage = 1 #float(inputDict['maxBESSDischarge']) ## Up to 80% of the total battery charge
max_utility_usage = battery_energy_capacity * max_utility_usage_percentage

## Initial variables to be updated in the loop
economic_benefit = 0 ## Economic benefit of storing excess solar. NOTE: Is this useful?
total_economic_benefit = 0
total_energy_cost = 0 ## The total cost of energy
total_der_compensation = 0 ## The total DER compensation amount (when DERs are discharged)
max_demand_periods = {period: 0 for period in range(len(tou_structure))} ## The max demand during each period (for calculating the demand cost)

## Initial monthly cost tracking variables
monthly_energy_cost = [0] * 12
monthly_der_compensation = [0] * 12
monthly_economic_benefit = [0] * 12

for i in range(1, len(timestamps)):
timestamp = timestamps[i] ## Grab the hour timestamp
month = timestamp.month - 1 ## grab month information for cost savings analysis (Index 0=Jan, 11=Dec)
tou_rate, tou_period = calc_tou_rate(timestamp, tou_schedules, tou_structure) ## Grab the specific TOU info for that hour

net_load = demand_series[i] - PV_series[i] ## Net load after PV is accounted for
if net_load > 0: ## If the PV did not serve all of the load, then:
## Discharge the battery during on-peak hours; update BESS and economic variables
if tou_period in on_peak_periods:
## Discharge BESS to household first
discharge = min(battery_max_discharge_rate, net_load, battery_soc - battery_soc_min, net_load) ## NOTE: this doesn't keep track of which variable is being used though
battery_soc -= discharge
net_load -= discharge

## If BESS > 20% battery energy capacity after serving household load, then utility can use it
if battery_soc > battery_soc_min:
utility_discharge = min(battery_max_discharge_rate, battery_soc - battery_soc_min, max_utility_usage)
battery_soc -= utility_discharge
der_compensation += utility_discharge * compensation_rate
monthly_der_compensation[month] += utility_discharge * compensation_rate
max_utility_usage -= utility_discharge # Reduce the remaining utility usage allowance

else: ## buy from the grid
grid_usage = net_load ## grid_usage is how much energy needs to be bought from the grid
total_energy_cost += grid_usage * tou_rate
monthly_energy_cost[month] += grid_usage * tou_rate

else: ## If off-peak period, use grid if necessary
battery_soc = min(battery_energy_capacity, battery_soc + battery_max_charge_rate)
grid_usage = net_load
total_energy_cost += grid_usage * tou_rate
monthly_energy_cost[month] += grid_usage * tou_rate
der_compensation = 0

else: ## If net_load <= 0 (PV has served all load)
excess_solar = -net_load
## Store any excess PV in the BESS during off-peak periods
if tou_period in off_peak_periods:
charge = min(battery_max_charge_rate, excess_solar)
battery_soc = min(battery_energy_capacity, battery_soc + charge)
#print('battery soc in offpeak 2: \n',battery_soc)
grid_usage = 0 ## No energy being bought from the grid
economic_benefit = charge * compensation_rate ## Economic benefit from storing excess solar.
monthly_economic_benefit[month] += charge * compensation_rate
der_compensation = 0 ## No DER compensation for charging
else:
## If not off-peak, sell excess solar to the grid
der_compensation = excess_solar * compensation_rate
monthly_der_compensation[month] += excess_solar * compensation_rate

## Update the total economic benefit and DER compensation variables
total_economic_benefit += economic_benefit
total_der_compensation += der_compensation

## Track the maximum demand for the current TOU period
max_demand_periods[tou_period] = max(max_demand_periods[tou_period], demand_series[i] + grid_usage)

## Update the DER schedule
schedule.loc[timestamp, 'Grid Serving Load (kW)'] = grid_usage
schedule.loc[timestamp, 'Battery State of Charge'] = battery_soc

## Calculate total demand cost (including fixed monthly charge)
demand_cost = sum(max_demand_periods[period] * tou_structure[period][0]['rate'] for period in max_demand_periods) + fixed_monthly_charge

## Calculate total DER compensation amount (including any subsidies)
total_compensation = total_economic_benefit + total_der_compensation - total_energy_cost + subsidy_amount - demand_cost

#print(f"Total Energy Cost: ${total_energy_cost:.2f}")
#print(f"Total DER Compensation: ${total_der_compensation:.2f}")
#print(f"Total Compensation: ${total_compensation:.2f}")

## Add variables to outData
outData['energyCost'] = list(np.asarray(outData['energyCost']) + monthly_energy_cost)
outData['monthlyDERcompensation'] = monthly_der_compensation
outData['monthlyEconomicBenefit'] = monthly_economic_benefit

## DER Dispatch Plot ######################################################################################################################################################
fig = go.Figure()
fig.add_trace(go.Scatter(x=schedule.index,
y=schedule['Solar Generation (kW)'],
yaxis='y1',
mode='none',
fill='tozeroy',
fillcolor='yellow',
name='Solar Generation (kW)'))
fig.add_trace(go.Scatter(x=schedule.index,
y=schedule['Household Demand (kW)'],
yaxis='y1',
mode='none',
fill='tozeroy',
fillcolor='black',
name='Household Demand (kW)'))
fig.add_trace(go.Scatter(x=schedule.index,
y=schedule['Grid Serving Load (kW)'],
yaxis='y1',
mode='none',
fill='tozeroy',
fillcolor='gray',
name='Grid Serving Load (kW)'))
fig.add_trace(go.Scatter(x=schedule.index,
y= battery_energy_capacity - schedule['Battery State of Charge'],
yaxis='y1',
mode='none',
fill='tozeroy',
fillcolor='green',
name='Battery Discharge (kW)'))
fig.add_trace(go.Scatter(
x=schedule.index,
y=(schedule['Battery State of Charge'] / battery_energy_capacity) * 100,
yaxis='y2',
mode='none',
fill='tozeroy',
fillcolor='rgba(0, 0, 255, 0.3)',
name='Battery State of Charge (%)',
visible='legendonly' ## Initially hide this trace
))
fig.add_trace(go.Scatter(
x=schedule.index,
y=np.asarray(vbat_discharge_flipsign),
yaxis='y1',
mode='none',
fill='tozeroy',
fillcolor='rgba(128, 0, 128, 1)',
name='Thermal BESS'
))


## Plot layout
fig.update_layout(
xaxis=dict(title='Timestamp'),
yaxis=dict(
title='Power (kW)',
range=[0, battery_energy_capacity] # Set range for y1 axis
),
yaxis2=dict(
title='Battery State of Charge (%)',
overlaying='y',
side='right'
),
legend=dict(
orientation='h',
yanchor='bottom',
y=1.02,
xanchor='right',
x=1
)
)

#fig.show()

## Encode plot data as JSON for showing in the HTML side
outData['derBESSplot'] = json.dumps(fig.data, cls=plotly.utils.PlotlyJSONEncoder)
outData['derBESSplotLayout'] = json.dumps(fig.layout, cls=plotly.utils.PlotlyJSONEncoder)


## Create DER overview plot object ######################################################################################################################################################
fig = go.Figure()
Expand Down Expand Up @@ -1111,7 +865,7 @@ def new(modelDir):

## Financial Inputs
'demandChargeURDB': 'Yes',
'demandChargeCost': '25',
'demandChargeCost': '0', ## Utility does not usually charge this for residential consumers (but could for commercial consumers)
'projectionLength': '25',

## vbatDispatch inputs:
Expand Down
8 changes: 4 additions & 4 deletions omf/models/derUtilityCost.html
View file
Original file line number Diff line number Diff line change
Expand Up @@ -235,7 +235,7 @@
<input type="text" id="resistance" name="resistance" value="{{allInputDataDict.resistance}}" pattern="^\d+\.?\d*?$" required="required">
</div>
<div class="shortInput">
<label class="tooltip">COP<span class="classic">Coefficient of performance of each device. Must be between 1 and 3.5</span></label>
<label class="tooltip">Coefficient of Performance<span class="classic">Coefficient of performance of each device. Must be between 1 and 3.5</span></label>
<input type="text" id="cop" name="cop" value="{{allInputDataDict.cop}}" pattern="^\d+\.?\d*?$" required="required">
</div>
<div class="shortInput">
Expand Down Expand Up @@ -285,7 +285,7 @@
<div id="thermalDevicePlotData" class="tightContent" style="width: 1000px; height: 600px;"></div>
-->

<p class="reportTitle" style="page-break-before:always">Battery Charge Percentage</p>
<p class="reportTitle" style="page-break-before:always">Chemical BESS Charge Percentage</p>
<div id="batteryChargePlotly" class="tightContent" style="width: 1000px; height: 600px;"></div>

<p class="reportTitle" style="page-break-before:always">Resilience Overview</p>
Expand Down Expand Up @@ -341,8 +341,8 @@
insertMetric("monthlySummaryTable","Total Cost ($)", allOutputData.totalCost)
insertMetric("monthlySummaryTable","Adjusted Total Cost ($)", allOutputData.totalCostAdjusted)
insertMetric("monthlySummaryTable","Utility Savings ($)", allOutputData.savings)
insertMetric("monthlySummaryTable","1 kW Consumer Savings ($)", allOutputData.savingsSmallConsumer)
insertMetric("monthlySummaryTable","10 kW Consumer Savings ($)", allOutputData.savingsLargeConsumer)
insertMetric("monthlySummaryTable","1 kW Consumer Savings ($)", allOutputData.totalSavingsSmallConsumer)
insertMetric("monthlySummaryTable","10 kW Consumer Savings ($)", allOutputData.totalSavingsLargeConsumer)
</script>
</div>
</div>
Expand Down
Loading

0 comments on commit 319a4cc

Please sign in to comment.