Mitsuo Tsurugai

Security Constrained Integrated Unit Commitment Using Quadratic Programming

We have developed an innovative power generation scheduling method using quadratic programming (QP). The advantage of using our method is that it simultaneously solves unit commitment and economic load dispatch. We relax the binary variables of the unit state into continuous variables to apply QP to this problem. We also add the penalty term to converge the value of those variables to 0 or 1 to the objective function; the sum of fuel costs and start-up costs. This penalty term depends on per-unit fuel cost. The possibility of its variable converging to zero increases as the cost increases. This method was applied to a test system of daily generation scheduling that consisted of 29 thermal units, two pumped-storage units, four cascaded-hydro units, and one transmission. The schedule satisfied all constraints, i.e., load-power balance, operation reserve, power flow, minimum up/down-times, and fuel consumption. This result shows that the developed method is effective.

Trial of carbon-free micro-grid project in Japan

A 100% CO2-free micro-grid project using cutting-edge technologies in Japan is presented. At first, the overview of new project located in Rokkasho in Japan is introduced. Then the project results are evaluated on self-sufficiency rate in the community using a kind of dispatchable demand response and direct load control, the direct control of heat pump water heaters with a tank. The ratio is also evaluated when battery and control system are demonstrated. The analysis results between the battery capacity and ratio are introduced at last.

Demand and supply simulations considering detailed forecast, scheduling and control functions for Japanese Power System with a massive integration of renewable energy sources

A demand and supply simulator is developed to evaluate the effectiveness of various countermeasures against the balancing problems caused by the large integration of variable renewable energy sources (VRES). In our study, photovoltaic and wind power generation output pattern is modeled based on the actual data obtained at physical sites and forecast data from Japanese Meteorological Agency (JMA), to realistically reflect their fluctuation characteristics. One of the benefits of our digital simulator is that it reflects the typical Japanese central dispatching control centers methodologies including week-ahead, day-ahead, and hour-ahead scheduling process based on the latest forecast data available on each time horizon. Above all, not only PV output curtailment but also demand shifting and utilization of battery storages are taken into consideration. In conclusion, PV output curtailment is found to be the most cost-effective in many cases, still there is a room for batteries to be economical when CO2 emission cost spikes.

Boiler-Dynamics-Model-Based Economic Load Dispatch

Abstract This paper presents an economic dispatching method, BEST method , where BEST is short for Boiler-dynamics-based Economic load diSpaTch. Because this method is based on dynamics models more accurate than the conventional quadratic ones, it can reduce boiler fuel cost. First, to reduce the model deviations from the actual data, the autoregressive moving average (ARMA) model is employed. Second, an economic dispatching problem based on this model is formulated as a quadratic programming (QP) problem. Finally, the cost reduction is evaluated by computer simulations.