Support for PV systems with multiple arrays

ci/requirements-py36-min.yml

@@ -13,6 +13,7 @@ dependencies:
     - pytz
     - requests
     - pip:
+        - dataclasses
         - numpy==1.12.0
         - pandas==0.22.0
         - scipy==1.2.0

ci/requirements-py36.yml

@@ -27,5 +27,6 @@ dependencies:
     - siphon  # conda-forge
     - statsmodels
     - pip:
+        - dataclasses
         - nrel-pysam>=2.0
         - pvfactors==1.4.1

docs/sphinx/source/api.rst

@@ -20,6 +20,7 @@ corresponding procedural code.
 
    location.Location
    pvsystem.PVSystem
+   pvsystem.Array
    tracking.SingleAxisTracker
    modelchain.ModelChain
 

docs/sphinx/source/pvsystem.rst

@@ -10,13 +10,25 @@ PVSystem
     from pvlib import pvsystem
 
 
-The :py:class:`~pvlib.pvsystem.PVSystem` class wraps many of the
-functions in the :py:mod:`~pvlib.pvsystem` module. This simplifies the
-API by eliminating the need for a user to specify arguments such as
-module and inverter properties when calling PVSystem methods.
-:py:class:`~pvlib.pvsystem.PVSystem` is not better or worse than the
-functions it wraps -- it is simply an alternative way of organizing
-your data and calculations.
+The :py:class:`~pvlib.pvsystem.PVSystem` represents one inverter and the
+PV modules that supply DC power to the inverter. A PV system may be on fixed
+mounting or single axis trackers. The :py:class:`~pvlib.pvsystem.PVSystem`
+is supported by the :py:class:`~pvlib.pvsystem.Array` which represents the
+PV modules in the :py:class:`~pvlib.pvsystem.PVSystem`. An instance of
+:py:class:`~pvlib.pvsystem.PVSystem` has a single inverter, but can have
+multiple instances of :py:class:`~pvlib.pvsystem.Array`. Arrays can have
+different tilt, orientation, and number or type of modules.
+
+The :py:class:`~pvlib.pvsystem.PVSystem` class methods wrap many of the
+functions in the :py:mod:`~pvlib.pvsystem` module. Similarly,
+:py:class:`~pvlib.pvsystem.Array` wraps several functions with its class
+methods.  Methods that wrap functions have similar names as the wrapped functions.
+This practice simplifies the API for :py:class:`~pvlib.pvsystem.PVSystem`
+and :py:class:`~pvlib.pvsystem.Array` methods by eliminating the need to specify
+arguments that are stored as attributes of these classes, such as
+module and inverter properties. Using :py:class:`~pvlib.pvsystem.PVSystem`
+is not better or worse than using the functions it wraps -- it is an
+alternative way of organizing your data and calculations.
 
 This guide aims to build understanding of the PVSystem class. It assumes
 basic familiarity with object-oriented code in Python, but most
@@ -40,26 +52,55 @@ data that influences the PV system (e.g. the weather).
 The data that represents the PV system is *intrinsic*. The
 data that influences the PV system is *extrinsic*.
 
-Intrinsic data is stored in object attributes. For example, the data
-that describes a PV system's module parameters is stored in
-`PVSystem.module_parameters`.
+Intrinsic data is stored in object attributes. For example, the parameters
+that describe a PV system's inverter is stored in
+`PVSystem.inverter_parameters`.
 
 .. ipython:: python
 
-    module_parameters = {'pdc0': 10, 'gamma_pdc': -0.004}
-    system = pvsystem.PVSystem(module_parameters=module_parameters)
+    inverter_parameters = {'pdc0': 5000, 'eta_inv_nom': 0.96}
+    system = pvsystem.PVSystem(inverter_parameters=inverter_parameters)
+    print(system.inverter_parameters)
+
+In the case of a PV system with a single array, the parameters that describe the
+system's modules can be provided directly to `PVSystem.module_parameters`:
+
+.. ipython:: python
+
+    module_parameters = {'pdc0': 5000, 'gamma_pdc': -0.004}
+    system = pvsystem.PVSystem(module_parameters=module_parameters,
+                               inverter_parameters=inverter_parameters)
     print(system.module_parameters)
 
-Extrinsic data is passed to a PVSystem as method arguments. For example,
+In the case of a PV system with several arrays, the module parameters are
+provided for each array, and the arrays are provided to
+:py:class:`~pvlib.pvsystem.PVSystem` as a tuple or list of instances of
+:py:class:`~pvlib.pvsystem.Array`:
+
+.. ipython:: python
+
+    module_parameters = {'pdc0': 5000, 'gamma_pdc': -0.004}
+    array_one = pvsystem.Array(module_parameters=module_parameters)
+    array_two = pvsystem.Array(module_parameters=module_parameters)
+    system_two_arrays = pvsystem.PVSystem(arrays=[array_one, array_two],
+                                          inverter_parameters=inverter_parameters)
+    print(system_two_arrays.module_parameters)
+    print(system_two_arrays.inverter_parameters)
+
+Note that in the case of a PV system with multiple arrays, the
+`module_parameters` attribute contains a tuple with the `module_parameters`
+for each array.
+
+Extrinsic data is passed to a PVSystem instance as method arguments. For example,
 the :py:meth:`~pvlib.pvsystem.PVSystem.pvwatts_dc` method accepts extrinsic
 data irradiance and temperature.
 
 .. ipython:: python
 
-    pdc = system.pvwatts_dc(1000, 30)
+    pdc = system.pvwatts_dc(g_poa_effective=1000, temp_cell=30)
     print(pdc)
 
-Methods attached to a PVSystem object wrap corresponding functions in
+Methods attached to a PVSystem object wrap the corresponding functions in
 :py:mod:`~pvlib.pvsystem`. The methods simplify the argument list by
 using data stored in the PVSystem attributes. Compare the
 :py:meth:`~pvlib.pvsystem.PVSystem.pvwatts_dc` method signature to the
@@ -95,66 +136,92 @@ as well as the incidence angle modifier methods.
 PVSystem attributes
 -------------------
 
-Here we review the most commonly used PVSystem attributes. Please see
-the :py:class:`~pvlib.pvsystem.PVSystem` class documentation for a
-comprehensive list.
-
-The first PVSystem parameters are `surface_tilt` and `surface_azimuth`.
-These parameters are used in PVSystem methods such as
-:py:meth:`~pvlib.pvsystem.PVSystem.get_aoi` and
-:py:meth:`~pvlib.pvsystem.PVSystem.get_irradiance`. Angle of incidence
+Here we review the most commonly used PVSystem and Array attributes.
+Please see the :py:class:`~pvlib.pvsystem.PVSystem` and 
+:py:class:`~pvlib.pvsystem.Array` class documentation for a
+comprehensive list of attributes.
+
+The first parameters which describe the DC part of a PV system are the tilt
+and azimuth of the modules. In the case of a PV system with a single array,
+these parameters can be specified using the `PVSystem.surface_tilt` and
+`PVSystem.surface_azimuth` attributes. In the case of a PV system with
+several arrays, the parameters are specified for each array using
+the attributes `Array.surface_tilt` and `Array.surface_azimuth`.
+
+The `surface_tilt` and `surface_azimuth` attributes are used in PVSystem
+(or Array) methods such as :py:meth:`~pvlib.pvsystem.PVSystem.get_aoi` or
+:py:meth:`~pvlib.pvsystem.Array.get_aoi`. The angle of incidence (AOI)
 (AOI) calculations require `surface_tilt`, `surface_azimuth` and also
-the sun position. The :py:meth:`~pvlib.pvsystem.PVSystem.get_aoi` method
-uses the `surface_tilt` and `surface_azimuth` attributes in its PVSystem
-object, and so requires only `solar_zenith` and `solar_azimuth` as
-arguments.
+the extrinsic sun position. The `PVSystem` method :py:meth:`~pvlib.pvsystem.PVSystem.get_aoi`
+uses the `surface_tilt` and `surface_azimuth` attributes from the
+:py:class:`pvlib.pvsystem.PVSystem` instance, and so requires only `solar_zenith`
+and `solar_azimuth` as arguments. The `Array` method :py:meth:`~pvlib.pvsystem.Array.get_aoi`
+operates in a similar manner. These two methods differ only in scope: the
+`Array` method operates only on the `Array` instance, whereas the `PVSystem`
+method operates on all `Array` instances.
 
 .. ipython:: python
 
-    # 20 deg tilt, south-facing
-    system = pvsystem.PVSystem(surface_tilt=20, surface_azimuth=180)
-    print(system.surface_tilt, system.surface_azimuth)
+    # single south-facing array at 20 deg tilt
+    system_one_array = pvsystem.PVSystem(surface_tilt=20, surface_azimuth=180)
+    print(system_one_array.surface_tilt, system_one_array.surface_azimuth)
 
     # call get_aoi with solar_zenith, solar_azimuth
-    aoi = system.get_aoi(30, 180)
+    aoi = system_one_array.get_aoi(solar_zenith=30, solar_azimuth=180)
     print(aoi)
 
+.. ipython:: python
+
+    # two arrays each at 30 deg tilt with different facing
+    array_one = pvsystem.Array(surface_tilt=30, surface_azimuth=90)
+    array_two = pvsystem.Array(surface_tilt=30, surface_azimuth=220)
+    system_multiarray = pvsystem.PVSystem(arrays=[array_one, array_two])
+    print(system_multiarray.num_arrays)
+    # call get_aoi with solar_zenith, solar_azimuth
+    aoi = system_multiarray.get_aoi(solar_zenith=30, solar_azimuth=180)
+    print(aoi)
+
+Note that when the PV system includes more than one array, the output of the
+`PVSystem` method :py:meth:`~pvlib.pvsystem.PVSystem.get_aoi` is a *tuple* with
+the order of the elements corresponding to the order of the arrays. If the AOI
+is desired for a specific array, the `Array` method :py:meth:`~pvlib.pvsystem.Array.get_aoi`
+returns the AOI for the specific array.
+
+.. ipython:: python
+
+    aoi = array_one.get_aoi(solar_zenith=30, solar_azimuth=180)
+    print(aoi)
 
 `module_parameters` and `inverter_parameters` contain the data
 necessary for computing DC and AC power using one of the available
-PVSystem methods. These are typically specified using data from
-the :py:func:`~pvlib.pvsystem.retrieve_sam` function:
+PVSystem methods. Values for these attributes can be obtained from databases
+included with pvlib python by using the :py:func:`~pvlib.pvsystem.retrieve_sam` function:
 
 .. ipython:: python
 
+    # Load the database of CEC module model parameters
+    modules = pvsystem.retrieve_sam('cecmod')
     # retrieve_sam returns a dict. the dict keys are module names,
     # and the values are model parameters for that module
-    modules = pvsystem.retrieve_sam('cecmod')
     module_parameters = modules['Canadian_Solar_Inc__CS5P_220M']
+    # Load the database of CEC inverter model parameters
     inverters = pvsystem.retrieve_sam('cecinverter')
     inverter_parameters = inverters['ABB__MICRO_0_25_I_OUTD_US_208__208V_']
-    system = pvsystem.PVSystem(module_parameters=module_parameters, inverter_parameters=inverter_parameters)
+    system_one_array = pvsystem.PVSystem(module_parameters=module_parameters,
+                                         inverter_parameters=inverter_parameters)
 
 
 The module and/or inverter parameters can also be specified manually.
-This is useful for specifying modules and inverters that are not
-included in the supplied databases. It is also useful for specifying
-systems for use with the PVWatts models, as demonstrated in
-:ref:`designphilosophy`.
-
-The `losses_parameters` attribute contains data that may be used with
-methods that calculate system losses. At present, these methods include
-only :py:meth:`PVSystem.pvwatts_losses
-<pvlib.pvsystem.PVSystem.pvwatts_losses>` and
-:py:func:`pvsystem.pvwatts_losses <pvlib.pvsystem.pvwatts_losses>`, but
-we hope to add more related functions and methods in the future.
+This is useful for modules or inverters that are not
+included in the supplied databases, or when using the PVWatts model,
+as demonstrated in :ref:`designphilosophy`.
 
 The attributes `modules_per_string` and `strings_per_inverter` are used
 in the :py:meth:`~pvlib.pvsystem.PVSystem.scale_voltage_current_power`
 method. Some DC power models in :py:class:`~pvlib.modelchain.ModelChain`
 automatically call this method and make use of these attributes. As an
-example, consider a system with 35 modules arranged into 5 strings of 7
-modules each.
+example, consider a system with a single array comprising 35 modules
+arranged into 5 strings of 7 modules each.
 
 .. ipython:: python
 
@@ -165,6 +232,60 @@ modules each.
     data_scaled = system.scale_voltage_current_power(data)
     print(data_scaled)
 
+The `losses_parameters` attribute contains data that may be used with
+methods that calculate system losses. At present, these methods include
+only :py:meth:`PVSystem.pvwatts_losses
+<pvlib.pvsystem.PVSystem.pvwatts_losses>` and
+:py:func:`pvsystem.pvwatts_losses <pvlib.pvsystem.pvwatts_losses>`, but
+we hope to add more related functions and methods in the future.
+
+.. _multiarray:
+
+PVSystem with multiple Arrays
+-----------------------------
+
+It is possible to model a system with multiple arrays by passing a list of
+:py:class:`~pvlib.pvsystem.Array` to the :py:class:`~pvlib.pvsystem.PVSystem`
+constructor. The :py:class:`~pvlib.pvsystem.Array` class includes those 
+:py:class:`~pvlib.pvsystem.PVSystem` attributes that may vary from array
+to array. These attributes include `surface_tilt`, `surface_azimuth`,
+`module_parameters`, `temperature_model_parameters`, `modules_per_string`,
+`strings_per_inverter`, `albedo`, `surface_type`, `module_type`, and
+`racking_model`.
+
+.. ipython:: python
+
+    array_one = pvsystem.Array(surface_tilt=30, surface_azimuth=90)
+    array_two = pvsystem.Array(surface_tilt=30, surface_azimuth=220)
+    system = pvsystem.PVSystem(arrays=[array_one, array_two])
+    system.num_arrays
+
+When instantiating a :py:class:`~pvlib.pvsystem.PVSystem` with a tuple or list
+of :py:class:`~pvlib.pvsystem.Array`, each array parameter must be specified individually
+when each instance of :py:class:`~pvlib.pvsystem.Array` is constructed.
+For example, if all arrays are at the same tilt you must specify that tilt for
+every array. When using :py:class:`~pvlib.pvsystem.Array` you shouldn't
+also pass any array attributes to the `PVSystem` attributes; these values
+are ignored.
+
+The output of `PVSystem` methods and attributes changes when the system has
+multiple arrays. Accessing any of Array attributes on the PVSystem object returns
+return a tuple with the value of the attribute for each array, in
+the same order as the `PVSystem.arrays` parameter. For example, using the system
+constructed above:
+
+.. ipython:: python
+
+    system.surface_tilt
+    system.surface_azimuth
+
+Similarly, other `PVSystem` methods expect tuples as input and return tuples:
+
+.. ipython:: python
+
+    aoi = system.get_aoi(solar_zenith=30, solar_azimuth=180)
+    print(aoi)
+    system.get_iam(aoi)
 
 .. _sat:
 
@@ -176,4 +297,4 @@ The :py:class:`~pvlib.tracking.SingleAxisTracker` is a subclass of
 includes a few more keyword arguments and attributes that are specific
 to trackers, plus the
 :py:meth:`~pvlib.tracking.SingleAxisTracker.singleaxis` method. It also
-overrides the `get_aoi` and `get_irradiance` methods.
+overrides the `get_aoi` and `get_irradiance` methods.

docs/sphinx/source/whatsnew/v0.9.0.rst

@@ -0,0 +1,66 @@
+.. _whatsnew_0810:
+
+v0.9.0 (MONTH DAY YEAR)
+-----------------------
+
+Breaking changes
+~~~~~~~~~~~~~~~~
+
+
+Deprecations
+~~~~~~~~~~~~
+* The following ``ModelChain`` attributes are deprecated. They have been moved
+  to the :py:class:`~pvlib.modelchain.ModelChainResult` class that is
+  accessible via ``ModelChain.results``
+  * ``ModelChain.ac``
+  * ``ModelChain.airmass``
+  * ``ModelChain.aoi``
+  * ``ModelChain.aoi_modifier``
+  * ``ModelChain.cell_temperature``
+  * ``ModelChain.dc``
+  * ``ModelChain.diode_params``
+  * ``ModelChain.effective_irradiance``
+  * ``ModelChain.spectral_modifier``
+  * ``ModelChain.total_irrad``
+  * ``ModelChain.tracking``
+
+Enhancements
+~~~~~~~~~~~~
+* In :py:class:`~pvlib.modelchain.ModelChain`, attributes which contain
+  output of models are now collected into ``ModelChain.results``.
+  (:pull:`1076`, :issue:`1067`)
+* Added :py:class:`~pvlib.pvsystem.Array` class to represent an array of
+  modules separately from a :py:class:`~pvlib.pvsystem.PVSystem`.
+  (:pull:`1076`, :issue:`1067`)
+* Added capability for modeling a PV system with multiple arrays in
+  :py:class:`~pvlib.pvsystem.PVSystem`. Updates the ``PVSystem`` API
+  to operate on and return tuples where each element of the tuple corresponds
+  to the input or output for a specific ``Array``. (:pull:`1076`,
+  :issue:`1067`)
+* Support for systems with multiple ``Arrays`` added to
+  :py:class:`~pvlib.modelchain.ModelChain`. This includes substantial API
+  enhancements for accepting different weather input for each ``Array`` in the
+  system. (:pull:`1076`, :issue:`1067`)
+* Support for :py:func:`~pvlib.inverter.sandia_multi` added to
+  :py:class:`~pvlib.pvsystem.PVSystem` and
+  :py:class:`~pvlib.modelchain.ModelChain` (as ``ac_model='sandia_multi'``).
+  (:pull:`1076`, :issue:`1067`)
+
+Bug fixes
+~~~~~~~~~
+
+Testing
+~~~~~~~
+
+Documentation
+~~~~~~~~~~~~~
+
+Requirements
+~~~~~~~~~~~~
+
+
+Contributors
+~~~~~~~~~~~~
+* Will Holmgren (:ghuser:`wholmgren`)
+* Cliff Hansen (:ghuser:`cwhanse`)
+* Will Vining (:ghuser:`wfvining`)