1、 AbstractWith increasingly serious environmental problems,improving the energy utilization and reducing the emission of pollution has become an important direction for smart grid.The distributed CCHP system becomes a promising DG with its high energy efficiency.In this paper,firstly the traditional
2、CCHP structure is optimized by heat storage system and power driven heat pump/electric refrigerator,in which the fixed relationship between power supply and heating is broken.Secondly,this paper analyses the characteristics of the flexible load.Considering the advantages of grid power supply,DG and
3、demand response collaboration,this paper puts forward a multiple energy complementary strategy considering CCHP and FL.The method considers many indicators including the economic indicator,the environmental indicator,the energy efficiency indicator and the interests of users.Further,this paper trans
4、forms the multi-objective optimization problem into an integrated single objective problem by analytic hierarchy process.PSO algorithm is used in this strategy and the fitness function is constructed by penalty function method.Index TermsCombined cooling heating and power(CCHP),distributed generator
5、(DG),flexible load(FL),multi-objective optimization,Particle swarm optimization(PSO)I.INTRODUCTION ITH the increasingly serious environmental problems,improving the energy efficiency and reducing the emission of pollution has become a direction for smart grid.The DG connected to the grid and the dev
6、elopment of micro-grid and active distribution network(ADN)provides a solution to these problems.The distributed CCHP system becomes a promising DG with its high energy efficiency and stable power quality,which has been widely studied and applied.The CCHP can achieve 90%energy efficiency,but the Man
7、uscript received June 20,2018.J.W.ZHOU is with the Suzhou Electric Power Design Institute Co.,Ltd(e-mail:).Y.WANG is with the Suzhou Electric Power Design Institute Co.,Ltd.C.L.WANG is with the Suzhou Electric Power Design Institute Co.,Ltd.research in 1 shows that energy saving is not an inherent c
8、haracteristic of CCHP.The CCHP needs to be well controlled so as to get good energy efficiency and emission reduction effects.2 optimizes the traditional CCHP structure and researches an optimization strategy based on operable power and heat load.The superiority of CCHP is proved from various aspect
9、s in 2.In addition,with the development of DG,CCHP combined with DG has an excellent application prospect 3 4.5 combines the CCHP with solar energy to design a combined power supply system.The integration of CCHP and micro-grid,heat network and power grid mentioned in 6 has attracted much attention
10、both at home and abroad.How to break through the power determined by the heat mode and combine with DG is the focus of research on CCHP now.At present,load on the electricity demand side is incorporated into the range of power grid control.These electricity consumptions that can be changed or transf
11、erred within a defined time zone are flexible load.The full use of FL can realize peak-load shifting and balance the fluctuation of intermittent DG so as to improve the stability of the power grid and save energy.7 proposes a class of electrical equipment with thermostatic control which real-time re
12、sponses to the price of electricity and help users get better economy.8 puts forward to an economic scheduling method with DG and FL based on real-time electricity price and the FL plays a decisive role.In this paper,we take CCHP,DG and FL into consideration to form a multiple energy complementary s
13、trategy.Firstly,the traditional CCHP structure is optimized by heat storage system and power driven heat pump/electric refrigerator to break through the power determined by the heat mode.Secondly,different kinds of FL are analyzed.Thirdly,this paper puts forward a multiple energy complementary strat
14、egy considering CCHP,FL and DG in ADN.The method considers many indicators including the economic indicator,the environmental indicator,the energy efficiency indicator and the interests of users.The effectiveness and validity of this strategy is verified A Multiple Energy Complementary Strategy Cons
15、idering CCHP and Flexible Load in Active Distribution Network J.W.ZHOU,Y.WANG,C.L.WANG W 2018 China International Conference on Electricity Distribution Tianjin,17-19 Sep.2018CICED2018 Paper No.201803310000002 Page1/61991by the simulation of study case.II.CCHP STRUCTURE OPTIMIZATION A.Traditional CC
16、HP structure CCHP is usually composed of power system,gas supply system,power supply and distribution system,waste heat utilization system,monitoring system,etc.The working principle is to use natural gas as the main fuel to drive internal combustion engine(or gas turbine,micro-combustion engine)and
17、 generators to run.The power is generated to meet the power demand of the user and the waste heat with power generating is supplied to the heat and cool users by heat recycle equipment(or waste heat boiler,waste heat turbine).According to the CCHP structure,we can get a general idea of how CCHP work
18、s.As Fig.1 shown,it draws the core and represents a simple energy flow of CCHP system.Generation equipment,such as internal combustion engine or gas turbine,burns natural gas F pgu to generate power E pgu.E pgu provides power to users along with the power grid E grid.The waste heat Q pgu is recycled
19、 by the heat recycle equipment then CCHP supplies cooling by absorption refrigerating or supplies heating by heating unit.The afterburning equipment in the dashed line indicates that complementary combustion equipment burns natural gas F b to provide power when F pgu doesnt meet the users needs.Gas
20、Power GenerationHeat Recycle EquipmentAbsorption RefrigeratingHeating UnitComplementary Combustion EquipmentPower DemandHeat DemandCold Demand+gridEpguFbFpguQpguE Fig.1.System structure of CCHP B.Improved CCHP structure According to the energy flow shown in Figure 1,its easy to find that the heat-to
21、-electric ratio is fixed.That means the amount of electricity is determined,and the amount of heat is determined only,and vice versa.So CCHP is usually operated by power determined by the heat or heat determined by the power mode.In practice,however,the load of cooling,heating and power changes in h
22、igh frequency in ADN.Neither of aforementioned two modes can make CCHP work at the best output point.In order to improve the power regulation ability of CCHP and adapt to the development of ADN,we improve traditional CCHP structure.Storage system and Electric heat transfer equipment are used to bala
23、nce the cooling,heating and power output so that power and heating(cooling)is decoupled.The power driven heat pump/electric refrigerator,which can consume the electricity and produce heat/cool,is added to the system.At the same time,heat and cold storage system is used to control the heat and cool s
24、upply of CCHP.Through the above two ways achieve the changeable reasonable heat-to-electric ratio and the improved structure is shown as Fig.2.Gas Power GenerationHeat Recycle EquipmentHeat and Cold Storage SystemPower Driven Heat Pump/Electric RefrigeratorAbsorption RefrigeratingHeating UnitComplem
25、entary Combustion EquipmentPower DemandHeat DemandCold Demand+gridEpguFpguFpguQpguEN Fig.2.Improved CCHP system structure In addition,flue gas waste heat recycling from prime mover of traditional CCHP structure is insufficient in winter.Storage system can also recycle flue gas waste heat which serve
26、s as low temperature heat source of power driven heat pump.Improved CCHP structure make the output of CCHP more flexible.III.CHARACTERISTICS OF FLEXIBLE LOAD A.Adjustable load Adjustable load refers to the load in which the amount of electricity used can be changed in a given interval.Nowadays,many
27、adjustable loads are involved in the scheduling in ADN.Air conditioning is a typical adjustable load because we can change the setting temperature within allowed range to adjust power.B.Transferable load Transferable load refers to the load that can be transferred in a certain time zone.When a large
28、 number of transferable loads can be involved in the power grid scheduling,peak-load shifting will be remarkably effective and the stability of power grid will be improved.In the residential load,water heater and washing machine can be classified as this kind of FL.For example,smart water heater can
29、 transfer power consumption time based on power grid requirements within the time zone set by users.According to grid prediction information or real-time pricing curve,heating period of water heater is set at the most economical time.After heating to the set temperature,water heater maintains the he
30、at with its maintenance power.Washing machine acts as transferable load too but its control is relatively simple.The use of washing machine does not need the follow-up maintenance power,so intelligent device can choose the lowest price period to use washing machine.C.Electric vehicles Electric vehic
31、les(EV)as a special kind of FL attracted great attention.EV can charge and discharge with its battery and participate in the control of the ADN while completing the 2018 China International Conference on Electricity Distribution Tianjin,17-19 Sep.2018CICED2018 Paper No.201803310000002 Page2/61992cha
32、rging requirements.We can optimize algorithm for intelligent charging pile to make better use of electricity.There are two modes for EV.One is called grid to vehicle(G2V),another is called vehicle to grid(V2G).In G2V mode,EV acts as buffer load to absorb the redundant power in grid.In V2G mode,EV su
33、pports the lack capacity of grid.We can consider that EV is energy storage battery but it needs to charge to the specified capacity at a specified time.IV.MULTIPLE ENERGY COMPLEMENTARY STRATEGY A.Objective function The optimization strategy for ADN should combine the power generation side and the de
34、mand side.As shown in Fig.3,according to the daily load forecasting curve,power demand,forecasting renewable energy output and real-time electricity price,optimization strategy sets the objective function to control DG from several aspects including economy,energy efficiency,environment and the inte
35、rests of users.As the strategy takes the interests of users into account,it is necessary to adjust the FL to coordinate the operation of the entire ADN control.Daily load forecasting curvePower,heating and cooling demandReal-time electricity priceForecasting renewable energy outputMultiple energy co
36、mplementary strategyEconomyEnergy efficiencyEnvironmentInterests of users Fig.3.Integrated optimization strategy for ADN Economic indicator Economic indicator is shown as(1).C w is total cost to meet cooling,heating and power demand each unit time.C ele is power purchase cost.C gas is gas purchase c
37、ost.C om is operation and maintenance cost for DG.(1)w ele gas omCC C C=+The specific calculation of each cost is as follows.Power purchase cost is shown as(2).C grid is real-time electricity price.P load is the ADN power load.P wt is wind turbine(WT)output.P pv is PV output.W is power of the intern
38、al combustion engine of CCHP.N is power consumption of power driven heat pump/electric refrigerator.P flex is reduced power after FL response to grid control.P grid is power from external power grid.The unit time t is 15 minutes.()(2)ele grid load wt pv flexgrid gridC C P P P WN P tCP t=Gas purchase
39、 cost is shown as(3).C g is the unit price of natural gas.k is coefficient of unit conversion and it equals 3.6.q is the heating value of natural gas and it equals 35.53Mj/m3.(/)*(/)(3)gas g pguC CW kq t=Operation and maintenance cost is shown as(4).C omwt,C ompv,C omW and C omN respectively are ope
40、ration and maintenance cost of wind turbine,PV,internal combustion engine and power driven heat pump/electric refrigerator.Their value respectively is 0.03,0.04,0.03 and 0.01.(4)om omwt wt ompv pvomW omNC C P tC P tC Wt C Nt=+Environment indicator With increasingly serious environmental problems,red
41、ucing greenhouse gas emissions is also an important direction for optimization.Environment indicator is shown as(5).C co2 is emission of CO 2 in per unit time.j gco2 is CO 2 emission coefficient for natural gas and it equals 2.16kg m3.j eco2 is CO 2 emission coefficient for power grid and it equals
42、0.684kg/kWh.222(/)/(5)co gco pgu eco gridC j W kq t j P t=+Energy efficiency indicator In this paper,energy efficiency indicator mainly represents the energy efficiency of CCHP.Energy efficiency indicator is shown as(6).Q huser,Q wuser and Q cuser respectively are heating load,domestic hot water loa
43、d and air conditioning cooling load.Q x is energy change of heat and cold storage system.The greater is,the higher energy efficiency is.()(6)/huser wuser cuser xpguQ Q Q WN QW+=Interests of users The strategy takes FL into account which may affect the interests of users.In order to achieve the balan
44、ce between FL control effect and interests of users,we use(7)to quantify the relationship between interests of users and adjustment value of FL.S,whose value varies from 0 to 1,is defined as the interests of users.0 represents the most satisfactory and 1 represents the least satisfactory.P r is tota
45、l controllable power of FL.k 1 and k 2 are satisfaction parameter.k 1 represents the dissatisfaction degree of users.k 2 represents the rate of dissatisfaction degree of users change with adjustment increase.212/(/)(7)0 1 flex r flex rS k PP k PPS=+B.Weight determined by analytic hierarchy process A
46、ccording to the analysis of multi-objective in the previous 2018 China International Conference on Electricity Distribution Tianjin,17-19 Sep.2018CICED2018 Paper No.201803310000002 Page3/61993section,the integrated objective function is shown in(8).2(min(,(1)(8)co wf SC C=The above formula is a mult
47、i-objective function that takes four goals into account and it is very difficult to solve.Therefore,we determine the weight of each objective by analytic hierarchy process(AHP)and a single objective function is concluded.Before weight determining,each sub-objective function needs to be normalized.Th
48、at is to say,the dimension of each sub-objective function is unified.S and are between 0 and 1 so they dont need to deal with.C w can get its normalized value by dividing the cost of daily total load as(9).C co2 can get its normalized value by dividing the daily total CO 2 emission as(10)./()/(9)/)w
49、 grid load h w hcC C P Q Q COPQ COP t=+222/(10)()/gco g eco grideco h w h c c loadj Qk q j Pj Q Q COP Q COP P+=+The normalized multi-objective optimization function can be shown as(11).12 3 4min(1)(11)fS=+C.Constraint condition In the process of solving the problem,each variable is constrained by a
50、series of constraints,such as user demand and generator output.The constraint condition of the objective function is mainly considered in the following aspects.Power constraint Power constraint is shown as(12).It means the power generated by the power generation is equal to the load on the demand si
