نرخ نفوذ در بازار فن آوری های جدید انرژی

The market penetration rates of 11 different new energy technologies were studied covering energy production and end-use technologies. The penetration rates were determined by fitting observed market data to an epidemical diffusion model. The analyses show that the exponential penetration rates of new energy technologies may vary from 4 up to over 40%/yr. The corresponding take-over times from a 1% to 50% share of the estimated market potential may vary from less than 10 to 70 years. The lower rate is often associated with larger energy impacts. Short take-over times less than 25 years seem to be mainly associated with end-use technologies. Public policies and subsides have an important effect on the penetration. Some technologies penetrate fast without major support explained by technology maturity and competitive prices, e.g. compact fluorescent lamps show a 24.2%/yr growth rate globally. The penetration rates determined exhibit some uncertainty as penetration has not always proceeded close to saturation. The study indicates a decreasing penetration rate with increasing time or market share. If the market history is short, a temporally decreasing functional form for the penetration rate coefficient could be used to anticipate the probable behavior.

The inertia of energy systems against changes is large, among others because of the long investment cycles of energy infrastructures or production plants. Energy is a basic commodity for which reason the price often dominates the competition over innovative features. Earlier studies have shown that the build up of present primary energy sources on a global scale from an embryonic to a significant market position took about a century (Shell International, 2001; Davis, 2001). On the other hand, new renewable energy sources have shown remarkably high market growth rates during the recent years (International Energy Agency, 2004a). Moving across the whole energy chain brings forward end-use and consumer products, which affect the total energy demand through their specific energy consumption. Here the capital turnover times are much shorter than in the energy production end, which could result in faster energy impacts through the energy-saving features. Traditionally, energy system modeling has shed light on the above questions through scenarios on the energy future. The underlying methodology is often based on forecasting the cost development of new technologies and consequent penetration based on their cost competitiveness (European Commission, 2003). Technology learning or experience curves that describe the decrease of unit costs with increasing production volumes have proved to be useful aids for macro models (Wene, 2002). Statistical models of technology adoption or technology diffusion models are examples of approaches in which the penetration is mathematically described as a diffusion process resembling epidemic growth. Diffusion theory is a well-established field of science and the literature on diffusion of different technologies is ample. Logistic diffusion curves have been previously applied to modeling of changes in the global energy system (Häfele, 1981). Energy efficiency indicators or energy intensity coefficients are a way to describe the development in energy use (Schipper and Meyers, 1992) and indirectly the diffusion of more efficient end-use technologies. These can also be applied even on the level of a single appliance (Enerdata, 2004). The number of patents has been used to describe diffusion of environmentally responsive technology (Lanjouw and Mody, 1996). The main objective here is to investigate how fast different new energy technologies penetrate to the market. In present approach, a diffusion model is applied to long-term real market data of 11 different technologies to reveal penetration rates and to enable comparison of these. The 20 data sets used in this study cover different geographical scope from single countries to global data, technologies with varying market maturity, and also a few established reference cases or benchmarks (nuclear energy, oil) for comparison. Both energy production and energy end-use technologies are considered.

This paper analyzed the market penetration of new energy technologies. Twenty technology datasets were studied covering energy production and end-use technologies at different stage of maturity and different geographical areas. The penetration rates were determined by fitting real data to a S-shaped technology diffusion model. The analyses show that the true penetration rates may vary by an order of magnitude from 4 to over 40%/yr. The take-over times from a 1% to 50% share of the estimated potential of the new technology could take from less than 10 to 70 years. The lower build-up rate is often associated with higher energy impacts and the high penetration with more local or smaller impacts. The shorter take-over times less than 25 years seem to be linked with energy-saving consumer products or energy end-use technologies whereas energy production has in general longer values. Policies and subsides may have an important effect on the penetration rate as demonstrated for example by the market growth of wind power or photovoltaics in Germany. The German wind penetrates with a rate (β) of 30.6%/yr and would reach half of its estimated potential in 21 years. But there are new technologies, which penetrate fast even without major public financial support, often explained by sufficient technology maturity and competitive price. For example compact fluorescent lamps show a 24.2%/yr exponential growth. Industrial biomass in Finland or class A cold appliances in EU are other examples of build-up without public volume support. The penetration rates and take-over times determined in this study contain some uncertainty as the diffusion of the new technologies have not necessarily proceeded close to saturation and the pace of future penetration is unknown. The time intervals of the measured data sets used were quite varying in range, but typically around 20–30 years. The study clearly indicates a decreasing penetration rate with increasing time interval of the sample set or increasing market share. The penetration rate seems to start leveling off after some 30 years of diffusion. For younger new technologies with short market history, a temporally decreasing functional form instead of a constant value, for example a power or exponential curve, could be used to anticipate the observed behavior. Using such a model to anticipate diffusion on a 40 years time scale, reduced the observed penetration rates of the new technologies in average by 42% but the rates would still be comparable or even exceed those of the reference technologies.