汽車電控液壓助力轉(zhuǎn)向系統(tǒng)設(shè)計,汽車,液壓,助力,轉(zhuǎn)向,系統(tǒng),設(shè)計 Utilization of excess wind power in electric vehicles 1. Introduction Electric vehicles (EVs) raise the expectation that they can also be used as storage for intermittent electricity production from renewable energies (wind and solar) .This was mentioned as early as 2002( Kempton and Letendre ,2002) and also found its way into the energy concept of the German Federal Government (BMWi and BMU,2010).To investigate the effects of a fleet of EVs in Germany,the NET-ELAN project was initiated,funded by the German Federal Ministry of Economics and Technology.It covers abroad field of topics: ● Development trends of the electric grid and the power plant pool, ● development trends of future EV designs and determination of battery requirements and energy demand (Waldowski et al., 2010), ● scenarios of future energy supply and build-up of an EV fleet, ● assessment of spatial and temporal distributions of EVs connected to the grid (Linssen et al., 2011), ● grid integration of EVs with regard to feasibility, energy demand (Hennings and Linssen, 2010), emissions, and cost aspects (Bickert et al., 2011), including battery durability (Gunther et al., 2010). The project final report is published as a book(in German) (Linssen et al.,2012). This article describes the assessment of future wind power availability for charging EVs. These assessments also rely on results from the other project parts which are not described in detail here . We start with the general grid load and wind power production. The energy demand and the usage and charging of EVs are determined,and finally the energy balance for the scenario years 2020 and2030 is assessed . The potential for wind energy production and usage is first assessed with the assumption of unlimited transfer capabilities of the grid. In chapter 7 the grid limitations are addressed, the details of which are published separately (Mischinger et al., 2012). 2.Scenario This publication aims to assess the effects of a given fleet of EVs rather than predicting the probable EV deployment, therefore the build-up of a fleet of EVs is postulated.The total number of EVs is assumed tobe1millionin2020and6millionin2030,as aimed at by the German Federal Government(BMU,2011). The assumed development of the energy system is based on the objectives of the Energy Concept 2010 of the German Federal Government (BMWi and BMU, 2010), supplemented by the nuclear energy phase out decreed in 2011. Given these objectives, the power plant and wind turbine capacities installed in 2020 and 2030 are derived from calculations with the energy system model IKARUS (described e. g. in (Linssen et al., 2012)).The offshore capacities assumed in the NET-ELAN scenario are approximately reached if each offshore project for which an application was submitted(dena, 2011) will be finished within two years after planned start of construction.The wind turbine capacities in the NET-ELAN scenario are shown in Table 1. For comparison, Table 1 also shows the capacities in the base scenario in the concretisation of the Energy Concept 2010 by (Nagl et al., 2010; Schlesinger et al., 2010). The dena website also provides the locations of the existing and planned wind parks. For each wind park assumed to be in operation in 2020 (2030), the wind speeds measured at the nearest of the measuring platforms FINO 1 to 3 (FINO platforms, 2012) are used to assess the potential wind power production. Table 1 also shows the allocation of installed wind park power to the three FINO locations. 3.General grid load and wind power production To calculate the fluctuations of the wind power production,the electricity demand and the charging demand of EVs,a time dependent model with at least hourly resolution is required. To assess the time series of the grid load and electricity generation in 2020 and 2030, the exact approach is separately assessing the time series of the power generation from photo- voltaic, wind, other renewables, and power consumption, and extrapolating each of them to 2020 and 2030. As some of these values were not available, an approximate approach is chosen: The assessment is based on data from the years 2007 and 2010, since the year 2007 is an example of a good wind year(wind index 106%)and 2010 of a weak wind year(windindex74%,source: (Bundesverb and WindEnergie,2012), verified by our own calculations with the data from German transmission system operators(TSOs)). The time series of onshore wind power production for 2007 and 2010 are taken from the data supplied by the German TSOs, available from their websites(50Hertz Transmission,2011a; amprion, 2011a; TenneT,2011a; TransnetBW,2012a). The offshore wind power production in 2007and 2010 was neglectable.The time series of “vertical grid load” for 2007and2010aretakenfrom the data supplied by the German TSOs,available from their websites (50Hertz Transmission, 2011b; amprion, 2011b; TenneT, 2011b; TransnetBW, 2012b). The vertical grid load is defined as the total power transferred from the transmission grid to distribution grids and consumers.Upto2012 nearly all renew able power sources(including wind farms)were connected to distribution grids (110kVandlower, Table2, source: our own evaluation of (Engel, 2012)), so the vertical grid load is the consumption minus production from renewables (minus production from small scale conventional plants).By adding the wind power to the vertical grid load, time series are derived which are independent from wind power production. Because charging at night is the focus of the analysis, photovoltaic production can be neglected. The time series of onshore wind power in 2020 and 2030 are extrapolated from the 2007 (2010) time series by the ratio of installed onshore wind turbine capacity.The time series of the offshore wind power in 2020 and 2030 are derived from the time series of the wind speed measured in 2007 and 2010 on the offshore measuring platforms FINO 1 to3 (FINO platforms, 2012) in 90m above sea level (hub height of typical offshore wind turbines) which are available from the FINO database1 (BSH, 2011). To each offshore wind park the wind speed of the nearest FINO measuring platform is assigned. Table1 shows how the wind park capacities areas signed to the three FINO locations.The electrical power available from the wind turbines is calculated from the wind speeds using a typical power curve for offshore wind turbines and multiplied by the off shore wind turbine capacity in 2020 and 2030. This derivation of the future wind power production in the NET-ELAN project is basically similar to the derivation in the dena Grid Study II (dena, 2010). Although the dena Grid Study II uses a more detailed modelling, the duration curve of the wind power production modelled in NET-ELAN is in quite good agreement with that in the dena Grid Study II (Fig. 1). 4.Energy demand of electric vehicles The specific energy consumptions of the EVs were estimated using detailed mathematical models of the cars, performing the Artemis and some measured real-world driving cycles,including consumption of ancillary systems and losses in the battery and the charger. The estimated energy demands for the Artemis driving cycle are close to those for the measured real-world driving cycle “commuter” and used for assessing the energy drawn from the grid. Three EV sizes were modelled:mini, subcompact, and compact, and three drive train concepts: pure battery vehicles (BEV) with a driving range of 120 km,EV with range extender (REEV) with an electrical driving range of 50 km (in charge depleting mode, CDM) and plug-in hybrid EV (PHEV) with an electrical driving range of 30 km (CDM). The BEV is limited to a daily driving distance of 120 km,the REEV and PHEV cover distances above the electrical range by their internal combustion engines(i.e.in charge sustaining mode).The energy demands(Table3) include a decrease over the manufacturing year due to technical improvements and the penetration of new vehicles in the fleet. The average daily driving distance of the EV was derived from the distribution of daily driving distances of privately used passenger cars from the statistical survey “Mobility in Germany 2008” ( infas and DLR, 2009). Some investigations make assumptions on future EV usage,e.g. (Metz and Doetsch,2012) assume that only cars with a yearly mileage of 12,500 to 20,000km will be substituted by an EV.However,future EV usage is influenced by more than just economic criteria and could be higher or lower than today's, so it is here assumed that driving distances of future EVs will besimilar to today's average cars. With these assumptions the total energy demand of 6 million BEVs in 2030 is 10.7 TWh/a, in contrast to 17 TWh/a in (Metz and Doetsch, 2012), partly because all cars are taken into account, including those with lower yearly mileage, partly because of the lower energy demand per km. As the REEV and PHEV cover daily distances above 30 km or 50 km with their ICE,the fleet energy demand with shares of BEV, REEV and PHEV as in Table3 is lower,about 9.8TWh/a. 5.Time dependent usage and charging of electric vehicles In the project, only home charging is modelled. This was decided for the following reasons: Grid interaction is only possible when the EV is parked. Although charging on-the-road seems technically possible (Yu et al., 2011; Shwartz, 2012), costs are expected to be prohibitive. A charging connection must be available where the EV is parked. The car user must connect the EV to the grid. Also here, wireless (inductive) charging is technically possible (BBC News, 2012) but assumed not to be generally applied because of high costs. The evaluation of the German nationwide survey of driving habits “MiD 2008” revealed that 92% of the daily driving distances can be covered purely electrically with a BEV and 75% with a REEV if the battery is only charged once a day after returning from the last trip of the day. In the case of urban Driving profiles measured in the project,these shares are 95% and 88%. Additional charging during the day increases these shares only marginally. The survey “MiD 2008” also indicates that the cars are parked at home for the majority of time((Metz and Doetsch, 2012) come to the same conclusion), and that the majority of privately used cars have a dedicated parking or garage near the home. That allows a private charging connection to be established with low costs, whereas public charging stations are costly (Schroeder and Traber, 2012). Connecting the EV to the grid is an extra effort for the user, so the user may not be willing to do this if it is not required for his own driving requirements.Several publications assume that the EV is connected to the grid whenever it is parked (Capion, 2009; Ekman, 2011), however the same article(Capion, 2009) admits that this is unrealistic.At least the benefit from connecting must justify the effort,therefore (Rehtanz and Rolink, 2009) assume that the EV is connected to the grid only if parked for longer than 1 h. Itis therefore assumed here that the EV is connected to the grid only after returning home from the last trip of the day and disconnected just before starting the first trip of the next day. For example(Dallinger etal.,2011) make a similar assumption. While being connected, charging can either be uncontrolled (“dumb”), which means that charging starts as soon as the EV is connected to the grid and ends when the battery is fully charged, or it can be controlled in various ways. Up to 2020, the charging power at home is assumed to be 3.3 kW which is the maximum active power available at a standard 230 V 16 A connection with a power factor of 0.9 allowing for the non-sinusoidal current drawn by the charger. In 2030, the availability of three-phase charging with 9.9 kW is assumed. In all of the modelled car types this charging power is within the design limits of the battery. For uncontrolled charging, only a part of the EVs are charging at the same time,because they return home at different times, therefore the maximum grid load for 1 million EV in 2020 is 700 MW (Fig. 2), in contrast to 3300 MW if charging of all EVs would start at the same time. But this maximum of the charging load will occur at about 6 pm,when at winter time also the other grid loads are at maximum. That can cause problems particularly for the distribution grid,described in detail in the final report (Linssen etal.,2012).The simplest mode of controlled charging is shifting the charging into off-peak times.The grid load minimum in Germany is between about midnight and 6 am.In order to achieve a nearly constant charging load between 0 and 6 am the statistically distributed charging times ranging from some minutes to 6.4 h (for the BEV full driving range of 120 km) must be taken into account. A very simple control algorithm could be as follows: The 15% of the EVs having a charging need of 3 h or more start charging at midnight. Charging of each of the 27% of EVs needing 1.5 to 3 h starts at a time so that charging is finished at 6 am.The charging times of the 58% of EVs needing less than 1.5 h are evenly distributed between midnight and 4 am.The resulting course of the share of simultaneously charging EVs over time is not perfectly even, ranging from 22% to 31% (Fig. 3), but much better than a simultaneous start of all charging (100%) at midnight. With a charging power of 9.9 kW (assumed for 2030), all charging times are below 2 h and can be suitably distributed between 0 and 6 am, giving a smooth grid load. 6.Energy balance without grid restrictions Because of its statutory priority, all electric power from renew- able energies is consumed unless its production exceeds the consumption (including possible export) or the grid stability requires its limitation. As long as all renewable power is already utilized in other loads,the additional load of EVs must be satisfied by increasing the production of conventional power plants. Renewable power is only available for charging EVs if this power could not be used otherwise. The amount of excess renewable power depends not only on the renewable power production and the power consumption but also on the amount of conventional power required to stabilize the grid (so-called must-run capacity). The dena Grid Study I (dena, 2005) assessed that in 2020 with an installed wind power capacity of 48 GW, 20 to 30 GW of conventional power plant production are needed to be able to supply negative control power (that is power which can be reduced when needed for balancing production and consumption). The FGH assessed that currently (2012 to 2014) in the German grid 8 to 25 GW of conventional power plant production must run to be able to control the active power balance and 4 to 20 GW for delivering reactive power (FGH et al., 2012). It is assumed here that 20 GW of conventional power production are needed to ensure grid stability. To show the impact of the must-run capacity, the results for zero must-run are pointed out, too. Fig. 4 shows the distribution of potential excess energy from wind power in the hours between midnight and 6 am of each night, for the year 2030. In a good wind year (solid lines) with 20 GW must-run power, excess energy is available in 70% of the nights,and it meets the daily(Mo–Fr)energy demand of 6 million EVs (dotted line) in 50% of the nights.On the other hand, if no must-run power is required, excess energy is available in only 20% of the nights and meets the EV demand in 8% of the nights.The available excess energy is even less for a weak wind year (dashed lines). The utilization of wind power is limited by the charging demand of the vehicles,i. e. each night only as much wind power can be charged into the batteries of the vehicles as was discharged by driving during the past day (dotted line).The capability of the vehicles to utilize wind power could be increased if the battery would not be fully charged in nights with low wind, but that would mean a decrease of available driving range which will probably not be accepted by the vehicle users. Fig. 4 shows only the energy balance of each night. As the excess wind power is not evenly distributed over the night hours and the charging power of the EVs is limited, the excess wind energy which actually can be utilized by EVs is even lower. The yearly sums of utilized and non-utilized wind power are shown in Fig. 5. In 2020, the effect of 1 million EVs is so small that it can hardly be seen. 6 million EV in 2030 have a noticeable but not dramatic effect and can be powered without extensions of the electric power system (described in detail in (Linssen et al., 2012)). Note that in Fig. 5 “other power” only includes the power directly fed into the transmission grid (380 and 220 kV voltage level) and the total energy in this figure is lower than the total electricity consumption. In 2030 about 50% of the energy need of the EVs can be met by utilizing excess wind power, but only if 20 GWof must-run power are required for grid stabilization. If zero must-run power would be required, even in 2030 most wind power could be utilized in other loads and hardly any excess wind power is available for charging the EVs. 7.Energy balance with grid restrictions For assessing the capabilities of the transmission grid to deliver EV charging power and to absorb wind power, a model of the German transmission grid including the power plant portfolio was developed for 2020 and 2030. The details are published, see (Mischinger et al., 2012). The calculations with the grid model show that in the scenario year 2030 the limitation is more significant, 8%compared to 15% without grid restrictions,because the grid capacity does not keep up with the increased installed wind turbine power.The share of charging energy supplied by wind energy in 2030 is limited by grid bottlenecks to 30%,compared to a potential of 50% without grid restrictions. 8.Conclusions 1 million EV have hardly any effect on the energy balance, 6 million have a noticeable but not dramatic effect. In the scenario,significant excess wind power is only available if it is assumed that 20 GW of conventional power is required for grid stabilization. If no minimum of conventional power is required,required,all wind power–as far as it can be transported by the transmission grid–can be utilized by other consumers,so that all charging power of the EVs must be delivered by increased production from conventional power plants. Without grid restrictions and a must-run power plant capacity of 20 GW, in the model year 2030 about 15% of the excess wind power can be utilized for charging EVs and can supply up to 50% of the energy needed by the EVs. The utilization of wind power is limited by the daily charging demand of the cars. Taking bottlenecks of the transmission grid into account, in the model year 2030 a significant amount of wind power can not be transported to the consumers, reducing the share of EV charging supplied from wind power from 50% to 30%. 在電動車中利用過剩的風能 1.引言 電動汽車的出現(xiàn)，提升了人們的預期：人們希望電動汽車可作為貯存器，貯存可再生能源（比如風能和太陽能）間歇發(fā)電所產(chǎn)生的電能。早在2002年，就有人提及上述想法（科普頓和勒讓德，2002年），并且，這一設(shè)想，已被引入德國聯(lián)邦政府的能源理念當中（寶馬公司下屬子公司和BMU,2010年）。為了調(diào)研德國境內(nèi)電動汽車的整體情況，啟動了歐洲局域網(wǎng)網(wǎng)絡(luò)項目。該項目，由德國聯(lián)邦經(jīng)濟技術(shù)部提供資金支持，涉及諸多領(lǐng)域的課題： ●電力系統(tǒng)網(wǎng)絡(luò)和電廠聯(lián)營體的發(fā)展趨勢 ●未來電動汽車設(shè)計的發(fā)展趨勢以及對電池（相關(guān)指/參數(shù)）要求及能源需求的測定 ●未來能源供給以及建立電動汽車車隊的設(shè)想 ●與電網(wǎng)連接的電動汽車的空間分布與時間分布（林森等，2011年） ●電動汽車并網(wǎng)涉及的相關(guān)事項，包括可行性，能源需求（海寧思和林森，2010年），排放及包括電池耐久性（甘特等，2010年）在內(nèi)的成本（伯克特等，2011年）問題。 該項目的最終報告已（在德國）印刷成書并出版發(fā)行（林森等，2012年）。 本文記述了未來風力發(fā)電用于電動汽車充電的有效性的評估。評估的參照標準，是其他項目中的部分結(jié)果，而這些內(nèi)容并未在本文中有詳細地描述。 開始階段，我們采用的是常規(guī)的電網(wǎng)負荷和風力發(fā)電。我們測定了能源需求，確定了電動汽車的使用和充電方法。在最后階段，我們對2020年和2030年設(shè)想所涉及到的能源平衡問題進行了評估。 我們假定電網(wǎng)的傳輸能力是無限的，在這一前提下，首先對風能發(fā)電的潛力和利用進行了評估。在第七章中，對電網(wǎng)的局限性進行了探討，其中的細節(jié)問題并未收錄，而是另行發(fā)布。（米思閣等，2012年） 2. 設(shè)想 本文旨在對設(shè)定的電動汽車車隊所產(chǎn)生的影響進行評估，而不是為了預測電動汽車的部署與規(guī)劃，因此我們會假定建立一支電動汽車車隊。該車隊的汽車數(shù)量，到2020時，假定為100萬輛，到2030年時，假定為600萬輛，而這兩個數(shù)值，也正是德國聯(lián)邦政府的預期