Humid air gas turbine systems that are regenerative cycle using humidified air can achieve higher thermal efficiency than gas turbine combined cycle (GTCC) power plant even though they do not require a steam turbine, a high combustion temperature, or a high pressure ratio. In particular, the advanced humid air gas turbine (AHAT) system appears to be highly suitable for practical use because its composition is simpler than that of other systems. Moreover, the difference in thermal efficiency between AHAT and GTCC is greater for small and medium-size gas turbines. To verify the system concept and the cycle performance of the AHAT system, a 3MW-class pilot plant was constructed that consists of a gas turbine with a two-stage centrifugal compressor, a two-stage axial turbine, a reverse-flow-type single-can combustor, a recuperator, a humidification tower, a water recovery tower, and other components. As a result of an operation test, the planned power output of 3.6MW was achieved, so that it has been confirmed the feasibility of the AHAT as a power-generating system. Moreover, running tests on the AHAT pilot plant were carried out over a few years so that various characteristics such as the effect of changes in ambient temperature, and start-up characteristics were clarified by analyzing the data obtained from the running tests.

Smoke emissions during startup of a standby gas turbine generator set were reduced by improvements of fuel control logics. A tested engine was mainly composed of two- stage centrifugal compressor, three-stage axial turbine and a single can combustor. A fully electronic controlled liquid fuel valve is used to adjust the flow rate of heavy fuel oil A. Smoke emissions in the exhaust gas were measured using an opacity smoke meter. Traditionally, a fuel control valve is instantaneously opened to avoid ignition failures at engine startup period. And high fuel pressure is required to obtain fine fuel spray. However, they cause fuel rich combustion during the startup, and thereby much black smoke emission. At our test rig, smoke emissions can be seen at exhaust duct exit during the period. For reducing the smoke emissions, two modifications were tested. First, a start point of fuel injection was delayed to wait for increase of air flow. And the fuel flow rate was reduced at the initial injection. They could shift to lean combustion condition at the ignition point. Secondly, fuel flow rate was gradually ramped with an increase of compressor rotational speed. Therefore, fuel flow rate could be kept under ignitable region even if ignition delay occurred. Smoke emissions reduction during engine startup was achieved by the two modifications without ignition stability loss.

This paper describes a newly-developed simulation method for pulverized coal combustion, and examples of commercial power generation boiler optimization by using the method. Recently, a number of pulverized coals, especially low rank coal, are mixed and supplied for commercial pulverized coal firing boiler to reduce power generation costs. However, it causes ash deposition, and overheating of furnace wall and heat exchangers. Experience-based boiler optimization is difficult, because much type of coals is used and changed in short time, so simulation method for boiler optimization is needed. Simulation accuracy is strongly affected by the kinetic parameters for the pulverized coal oxidation. However, the kinetic parameters based on Arrhenius plots, which uses experimental data obtained in a simple drop-tube furnace, do not have enough accuracy to optimize practical boiler. A newly-developed method, called Inverse Analysis (IA) method, has developed to estimate the kinetic parameters more accurately, and used for numerical simulation of a test furnace and practical boilers. Simulations based on Arrhenius plots and IA methods are performed for CRIEPI's coal combustion test furnace, and are compared with experimental results. The IA method shows better prediction of unburned Carbon in Ash (CIA) compared with Arrhenius plots method. Simulations with the kinetic parameters from the IA method are performed for two pulverized coal-fired boilers, in which the low rank coal is mixed and fired with bituminous coal. An optimization method is proposed to reduce CO and CIA as well as ash deposition under high mixture ratio condition of low rank coal. It is found that optimization based on numerical simulation with IA method is valid procedure.

Mitsubishi Heavy Industries, Ltd. (MHI) has developed and added the new KU30GSI and CM (Central Mixing)-MACH models to its lineup of KU gas series engines. The KU30GA gas engine, formerly the Micro Pilot Ignition (MP) -type model, has delivery of more than 150 units since 2001. The experience and know-how accumulated from their on-going operations have been fed back into the development process to ensure even higher reliability and performance.

The KU30GSI, whose ignition concept has been modified to a Spark Ignition (SI) system, was developed in order to meet the demand for a simple gas engine that does not require liquid pilot fuel and an engine with improved energy utilization efficiency. The KU30GSI has optimized its exhaust temperature and consequently reached a total efficiency of 66% - combined with generation efficiency and steam efficiency, the world's highest for this class of engine. Moreover, the KU30GSI start-up time has been reduced to less than 7 minutes from activation to 100% loading, meeting the requirements for peak application. Intricate details combining optimum control and the diagnosis techniques for combustion greatly contribute to this performance achievement.

Further, the concept of CM-MACH (MP-type) was developed to expand the utilization of low calorie gases and other specialty gases as operational fuel. With the CM-MACH, low calorie gas has been achieved by means of gas supply features in both the intake port at each cylinder and the suction port before the turbocharger. This feature offers an additional safety advantage in that it keeps an appropriate concentration of air-fuel mixture in the intake system to prevent auto ignition. MHI believes that through our expanded lineup of KU gas engines, we are able to meet an unprecedented diversity of customer needs.

In order to mitigate the global warming issue, we promoted the construction of nuclear power and the electric power production using renewable energy such as biomass, wind power, photovoltaic power, etc. as well as made an effort to decrease the fuel consumption of the thermal power through the raise of thermal efficiency by improving steam conditions or the adoption of gas turbines with high inlet gas temperature. Moreover, in order to make thermal power another pillar of near-zero-emission energy sources than nuclear power and renewable energy, the development of the CCS (Carbon Dioxide Capture and Storage) technology has been advanced. In such circumstance, from the end of 1980's, we paid attention to and have been developing the oxyfuel combustion technology as one of CO₂ capture technologies from the thermal power plant firing coal, which was expanded to be used as an alternate fuel of oil at that time. On the other hand, since nuclear power that is an important and largest near-zero-emission electric power source is now in the situation that cannot be operated temporarily, it can be said that the importance of the thermal power with the CCS technology increases by coming here. In this paper, the outline of our R&D supported by Japanese government on the oxyfuel combustion technology to capture CO₂ from coal-fired power station and the demonstration project in Australia with support from Japanese and Australian governments.

The objective of this study is to suggest a new method to predict the stability limit of high temperature air spray combustion. For the first stage of this study, behaviors of the high temperature air spray combustion at the stability limit are investigated experimentally. Kerosene was used as fuel and was supplied into the furnace through a fuel-air spray nozzle. Mixtures of air and nitrogen were used as oxidizer and were preheated up to 1023K. The O₂ concentration in the oxidizer was changed from 21 % to 9 %. Heat loss in the furnace was controlled by cooling tubes up to 2.3kW. The stability limit was determined on the basis of CO concentrations measured at the exit of furnace. An increase in heat loss shifts the stability limit to higher preheated temperature and O₂ concentration conditions. When an experimental condition approaches to the stability limit, the temperature of the recirculated burned gas decreases to consequently lead to the lifted flame. A further decrease in the temperature in the recirculated burned gas delays the ignition of the unburned mixture and increases the liftoff height. This fact indicates that the liftoff height is determined by the temperature and O₂ concentration in the recirculated burned gas. At conditions below the stability limit, the liftoff height becomes larger than the position of the core of recirculation vortex. The vortices recirculate both the burned gas and the unburned mixture of vapor fuel and oxidizer. The recirculation of the unburned mixture decreases the temperature in the recirculated gas, thus the ignition does not occur. It has been therefore founded that the use of the temperature and O₂ concentration in recirculated burned gas provides a generalization of the stability limit regardless of the variation of heat loss.

Numerical simulation on the counterflow flame of methane-air rich-premixed gas with opposing high-temperature air is carried out by using rather complex chemistry. A sinusoidal fluctuation is added to the spout velocity in order to investigate unsteady flame behavior and to clarify the effects of various parameters on flame structure. First, the influences of average velocity and the fluctuation frequency of the spout velocity on flame structure are examined with Smooke's Skeletal chemical kinetics model. It is shown that both premixed and diffusion flames exist together and the phase lag increases with the increase in the fluctuation frequency of the spout velocity. Second, the effects of premixed flame and diffusion flame on unsteady behavior are examined by using “extended IYH-Skeletal chemical kinetics model” proposed in the present study. This novel model discriminates premixed-gas-originated oxygen atom X from opposing- air-originated oxygen atom Y, so that the consumption rates of oxygen molecules X₂ and Y₂ correspond to premixed and diffusion flames. It is found that the strength of diffusion flame monotonically changes corresponding to the increase of flame stretch, but the strength and location of premixed flame intricately changes corresponding to the change of spout velocity.