IEEE-30 energy system data of multi-period market with intertemporal constraints

This is the dataset that is used for the original article: "Locational marginal pricing in multi-period AC OPF environment"

The dataset consists of the following files

• Case1.zip
• Case2.zip
• Case3.zip
• case_modifications.py
• data_spec.py
• OPF_formulation.pdf

Multiperiod AC OPF is given in OPF_formulation.pdf. Modifications of traditional IEEE 30-node case are given in case_modifications.py

The case files incorporate input and output multiperiod AC OPF and LMP decomposition data in csv and pickle formats. Data structure of case files is given in data_spec.py.

For python users pickle files are given. Nevertheless, python environment is not required. Specification can be read as a text file. All necessary data are repeated in csv format.

Step 1 are to define LMPs of  limited energy resources or storage resources that are formed by actual marginal resources from all time periods (first LMP definition in fig. 6 in the paper).

Step 2 are to define all other LMPs at price-taking nodes (second LMP definition in fig. 6 in the paper).

The following interrelation between Lagrange multipliers, LMP components, and price-bonding factors holds true:

assert np.max(np.abs(output_ramp.sensitivities.dot(output_ramp.offer_gen_data.price).tolist() -
                     output_ramp.ramping_gen_data.price)) < 1e-2
if output_pt_step1.components.shape[0]:
    step1_pf_filter = (~output_pf.is_limited_energy) & (~output_pf.is_storage)
    assert np.max(np.abs(output_pt_step1.components.node_price -
                         (output_pt_step1.components.f + output_pt_step1.components.tc_sum + output_pt_step1.components.vc_sum))) < 1e-2
    assert (output_pt_step1.components.f - output_pt_step1.w_f.dot(output_pf.node_price[step1_pf_filter])).abs().max() < 1e-2
    assert (output_pt_step1.components.tc_sum -
            pd.concat(
            (w.dot(output_pf.offer_price[step1_pf_filter]) for w in output_pt_step1.w_tc_list), axis=1
        ).sum(axis=1)).abs().max() < 1e-2

assert np.max(np.abs(output_pt_step2.components.node_price -
                     (output_pt_step2.components.f + output_pt_step2.components.tc_sum + output_pt_step2.components.vc_sum))) < 1e-2
assert (output_pt_step2.components.f - output_pt_step2.w_f.dot(output_pf.node_price)).abs().max() < 1e-2
assert (output_pt_step2.components.tc_sum -
            pd.concat(
                (w.dot(output_pf.offer_price) for w in output_pt_step2.w_tc_list), axis=1
            ).sum(axis=1)
        ).abs().max() < 1e-2