The high-resolution Energy use data of a large scale complex facility was analyzed. This data included the energy consumption of each air-conditioning zone in the office area, energy consumption of each tenant in the retail zone, and separate energy consumption data for lighting, plugs, fans, cooling, and heating. The result for the change in energy consumption in the retail zone for four years shows that the energy consumption had been decreased by visualizing the energy use of each tenant. The accuracy of estimation of the total energy consumption in the office zone with initial low occupancy is evaluated. The breakdown of energy end-use for the office zone and retail zone are quantified separately. Regarding the variations in energy consumption intensity, there are large differences depending on the type of tenant. With regard to the variations among the same kind of tenant, retail shop is the smallest, and restaurant, which consumes a large amount of energy, is the largest. The difference in energy consumption intensity in the office is about 3 times, and the variation in energy consumption for lighting and appliances is the largest among energy end-uses. The annual energy consumption for lighting and appliances is influenced by the energy use during operation rather than the operating time.

In a combined generator system, it is difficult to decide the optimum operating parameters of the generator for a particular purpose due to the fluctuation in energy unit price and electricity / heat load. Therefore, in this research, we propose a method for the monthly or daily optimization of the operating parameters of generators, and estimate the effect of cost and energy reduction. Moreover, in order to verify the cost savings effect of the proposed method, monthly optimization and daily optimization were applied to an actual general hospital. As a result, with monthly optimization, there was no cost savings effect compared to that of the previous year due to the differences in load and operation. With daily optimization, the cost reduction effect was estimated to be about 250,000 yen / year and the reduction in primary energy consumption was estimated to be about 150 GJ / year, thus confirming the effectiveness of the method.

As the COP 21 Paris agreement was ratified by the Japanese government and the nuclear plant was shut down after the great East Japan earthquake, low carbonization and peak shaving have become serious social issues. Universities and research facilities with high energy consumption have become an issue. This research provides a method for energy conservation by analyzing the energy data of each campus of a university using the clustering method. This research also aims at providing a simplified and automated method for classifying the energy consumption at the university into base energy, energy consumption by human activity factor ,and energy consumption by air conditioning factor. Using the k-means clustering method, the electrical consumption data for one whole year is classified into 6 groups depending upon the size of the data. Each cluster represents the central value of the cluster group. The initial value of each cluster is determined using the percentile method. On comparing with the university schedule, it is found that cluster 1 usually falls on Sundays and national holidays. Since there is no use of lighting, office equipment and air conditioning, cluster 1 represents the base energy consumption of the university. Clusters 2 and 3 fall on Saturdays and the days on which there is usage of lighting and office equipment but no air conditioning. Clusters 4 ~ 6 fall on the days with air conditioning usage, with cluster 6 falling on the days with peak energy consumption. Subtracting cluster 1 from clusters 2 and 3 gives the energy consumption by human activity factor, whereas subtracting cluster 3 from clusters 4 ~ 6 gives the energy consumption by air conditioning factor. The result is verified by comparing the result of clustering with the actual measurement of base energy, energy consumption by human activity factor, and energy consumption by air conditioning factor measured at Building no. 52 of “College of Life and Health Science”. It is found that the energy consumption obtained by the actual measurement and the clustering method are the same for the air conditioning factor, and is almost similar for the base energy and energy consumption by human activity factor. This shows that the energy consumption at a university can be classified using the k-means clustering method. We expect that by 2030 the emission intensity unit will be reduced by 40% in comparison to 2011. It is possible to reduce the consumption by human activity factor by 40% by replacing the existing lighting with LED bulbs and increasing the energy efficiency of office equipment. The energy consumption for conditioning is expected to be reduced by 15% through equipment renewal. According to the target of the Energy White Paper，the emission factor for electricity until 2030 is 0.37 kgCO₂/kWh, which is 24% less than the current value of 0.489 kgCO₂/kWh. In order to achieve 40% reduction in emission in the civilian sector, the electrical consumption is targeted to be reduced by 21%; for this, the base energy consumption needs to be reduced by 15.6%.

This paper describes the long term repair plan that was planned after the grand opening. The plan estimated the year of the first major repair and the first replacement, in addition to the cost per year. This was planned for each building and each work. The actual maintenance records for ten years after the grand opening were analyzed. As a result, it was found that only three seawater intake pumps were replaced during the period of ten years after the grand opening. The life cycles of the pumps were prolonged. It was estimated that proper maintenance helped in extending life cycle. This paper describes the results of the investigation on the deterioration of the Aqua Marine Fukushima equipment 18 years after the grand opening.