Verfahren zur messung der temperatur im brennraum einer brennkraftmaschine

The invention concerns a procedure for the measurement of the temperature in the combustion chamber of an internal-combustion engine. The development of the internal-combustion engine aims at a high burn temperature for the improvement of the thermodynamic efficiency. Besides however the point temperature levels must be reduced, in order to decrease the education from nitrogen oxide missions to. In addition the exact temperature distribution must during the burn in the combustion chamber admits to be. The exact temperature distribution regulation is however very complex and requires extensive manipulations at the internal-combustion engine. It is the task of the invention to make a simple procedure available for the measurement of the temperature in the combustion chamber during the burn for an foreignignited internal-combustion engine. This is reached according to invention by the following steps: - Making a fiber optics available flowing into the combustion chamber of the internal-combustion engine including spectral evaluation unit, - seizing at least a spectral gang at least a radical of the flame or the incineration gases in the combustion chamber during a clock rate, - acceptance of a Boltzmannverteilung of the gang spectrum of the flame and/or the incineration gases, - interpolating a Boltzmannverteilung of exhibiting gang spectrum due to the measured spectral lines, - assigning from temperature levels to the Emissionsbanden. The emission spectra of hydrocarbon flames contain a multiplicity of alternating spectral lines of radicals, which develop during the burn process. Using Boltzmann statistics the plasma temperature from the measured line intensities can be computed. Actually it is well-known to use the line spectrum of a material and/or a body for the determination of its temperature. A condition for is it however that a so-called Boltzmann distribution of the spectral line intensity is present. Since the burn with internal-combustion engines is subject to high-grade, stochastic influences, by the professional world it was so far accepted that the spectral intensity runs likewise stochastically, and Boltzmann statistics for the determination of the temperature distribution thus not to be used to be able. For the available invention it was all the more surprising from there that under certain conditions with the burn in an internal-combustion engine a Boltzmann well-behaved distribution of the emission spectra! lines is present. A condition for it is that the plasma is in a local, thermodynamic equilibrium. If this is present, einzelAT the 001 7 7 U1 nen particles show a Maxwell Boltzmann distribution of velocity and the total number of the lively atoms or molecules exhibits thereby a Boltzmann well-behaved distribution. A further condition is that the plasma must be optically thin. The measured intensities of the spectral lines correspond thereby with the total number of the particles. Preferably the procedure according to invention plans that the emission spectral lines are seized at least a radical from the group of OH, CH and C2. The radicals OH, CH and C2 rank most frequent radicals developing during the Verbrennungsablaufes among. The OH-radical for example is a predominating intermediate product of each hydrocarbon flame with strong emission volumes close of the UV-RADIATION. It is formed both in the Flammenfront and in the Verbrennungsprodukten behind the Flammenfront. The most suitable spectral gang of OH is with 306 Nm you shows many suitable spectral lines, which do not overlap themselves with the lines of other molecules. With foreignignited internal-combustion engines one finds a local, thermodynamic equilibrium with gas pressure levels within ranges between 0,5 to 7Mpa and with middle temperature levels within the range of 1500 to 3000K. In this printing temperature region the free distance of the Gasmoleküle amounts to between and 70nm. Therefore this plasma dominated from collisions among themselves, systematic deviations from the local thermodynamic equilibrium are very improbable. Within the life span of lively OH-radicals (l, 6 s) formed lively OH-radicals are received more than i0000 collisions among themselves, before the radiation of a light quantum takes place. Therefore the observed spectral volumes decrease/go back to lively OH-radicals, which are in the thermodynamic equilibrium with their environment. The invention is more near described on the basis of examples and a comparison with analytically won combustion chamber temperatures. Fig show. la a Boltzmannverteilung of emission volumes, during the burn, Fig. lb the temperatures resulting from it, Fig. 2 a temperature crank angle diagram of a further Meßbeipieles. In order to measure the optical density of a flame, their self-absorption in an attempt internal-combustion engine equipped with Glasfenster was measured. Self-absorption with wavelengths of 310nm is subject to typical losses from 15 to 20% of the light intensity over the combustion chamber diameter. With an internal-combustion engine with a combustion chamber diameter of 80mm the measurement was measured with 15° Kurbelwinkel after the upper dead center of the piston. At this time the flame filled out the combustion chamber completely. Since the current temperature measurements RK 001 7 7 U1 with 15mm of optical distance were accomplished, losses could be neglected by self-absorption. The Fig. la an example of a Boltzmannverteilung of emission volumes shows, whereby the intensity I of the spectral lines over the energy levels E is laid on. In Fig. Ib are represented the resulting temperature levels T for a complete engine cycle, whereby APPROX. means the Kurbelwinkel. The spectrum was taken up the close beginning of burn forming spark plug. The measuring accuracy amounted to 100K. The spectroscopically measured temperature levels Ts lie importantly over the middle temperatures Ttd computed due to the Zylinderdruckes, which was won on the way of the thermodynamic analysis. In order to explain this discrepancy, a simplified burn model was developed. If one considers that the first part of the burn takes place at a relatively low pressure, and the incineration gas of far by the following increase of pressure due to the continued burn of the fuel is heated up, then the place of the spark plug should exhibit the hottest gases, if one assumes that no important gas flow takes place. Then the maximum temperatures in the combustion chamber represent spectrally the temperature levels determined within the range of the spark plug. This hypothesis could be verified with some many zone model the combustion chamber on the following assumption: i. Each zone exhibits the same mass. First the mass of the zone 1 burns completely, then those the zone 2, etc. the zones consists not of firm volumes, but of specific molecules. Diffusion is not intended in the model. The temperature is accepted in each zone as homogeneous. 2. The Zylinderdruck and the burn rate are determined by current measurements. 3. The temperature in each zone affected by the heat transmission and the adibate compression in accordance with the increase of pressure. Fig. the spectroscopically won temperature levels Ts for a complete engine cycle shows 2. The middle temperature TM and the maximum temperature T1 were computed in accordance with the descriptive many zone model. A comparison of the results confirms that the local flame temperature in foreignignited internal-combustion engines can be measured using the plasma diagnostics techniques. The measured temperatures fit very well into the results of the combustion chamber model.