《[工学]基于遗传算法pid控制.doc》由会员分享，可在线阅读，更多相关《[工学]基于遗传算法pid控制.doc（32页珍藏版）》请在三一办公上搜索。
1、Genetic algorithm based PID controller design for a multi-area AGC schemein a restructured power systemSandeep Bhongade*1, Barjeev Tyagi*2, H. O. Gupta*3*Electrical Engineering Department, Indian Institute of Technology, Roorkee, INDIA*Corresponding Author: e-mail: 1bhongadesandeep, Tel +91-1332-285
2、555, Fax.+911332-286691AbstractIn this paper, a multi-area Automatic Generation Control (AGC) scheme suitable in a restructured interconnected powersystem has been proposed. Developed scheme utilizes a proportional, integral and derivative (PID) controller to control theoutput of the generators. The
3、 parameter of PID controller has been tuned according to Genetic Algorithm (GA) basedperformance indices. Developed model also include the Superconducting Magnetic Energy Storage (SMES) units to inject orabsorb the active power of an interconnected power system. The functioning of Genetic Algorithm
4、based PID controller hasbeen tested on a 39-bus New England system and 75-bus Indian power system network. The results of GAPID controller havebeen compared with those obtained by using the Least Square Minimization method. Compliance with North American ElectricReliability Council (NERC) standards
5、for AGC has also been established in this work.Keywords: Genetic Algorithms, Automatic Generation Control, Area control error, Superconducting magnetic energy storage(SMES), Control Performance Standards.1. IntroductionIn interconnected power systems the main goal of the AGC is to maintain zero stea
6、dy state errors for frequency deviation andgood tracking load demands. With time, the operating point of a power system changes and hence, these systems may experiencedeviations in nominal system frequency and scheduled power exchanges to other areas, which may yield undesirable effects. Inconventio
7、nal AGC model the variations of frequency and tie-line power exchanges are weighted together by a linear combinationto a single variable called the area control error (ACE). ACE is used as an input to the controller. Many investigations in the areafrequency and tie line control of isolated and inter
8、connected power systems have been reported in the past. The concept ofconventional AGC is discussed in Elgerd et al. (1970) and in Jaleeli et al. (1992).Around the world, the electric power industry has been undergoing reforms from the traditional regulated, vertically integratedutility (VIU) into a
9、 competitive, deregulated market. Market deregulation has caused significant changes not only in the generationsector, but also in the power transmission and distribution sectors. A detailed discussion on Load Frequency Control issues inpower system operation after deregulation is reported in Christ
10、ie and Bose (1996). The load frequency control in a deregulatedelectricity market should be designed to consider different types of possible transactions such as Poolco-based transactions,bilateral transactions, and a combination of these two.After the deregulation of the electricity sector, North A
11、merican Electric Reliability Council (NERC) has modified the controlperformance standard (CPS) for AGC. Maojun et al. (2000) have proposed a new AGC logic which is specifically designed towork under NERC performance standards. In Sasaki and Enomoto et al. (2002), the NERC standard to the Japanese po
12、wer systemand analyzed the compliance of their AGC scheme to these standards.The reliability of electric power supply during peak load period can be improved by using a battery energy storage system(BES). Energy is stored into the BES during off-peak load period and released from the BES during peak
13、 load period. In Shayeghiet al. (2008) the SMES units in each area of the two-area system for AGC has been considered. With the use of SMES units,frequency deviations in each area are effectively suppressed. However, it may not be economically feasible to use SMES unit inBhongade et al. / Internatio
14、nal Journal of Engineering, Science and Technology, Vol. 3, No. 1, 2011, pp. 220-236221every area of a multi-area system. Therefore, it is advantageous if an SMES unit located in an area is available for the control offrequency of other interconnected areas. In Automatic Generation Control (AGC) PID
15、 controller is widely used to control thefrequency and tie-line power. Many researchers (Khamsum et al., 2006; Tyagi et al., 2008) have proposed different methods totune the PID controller; one of them is the least square minimization method. An optimal value of PID controller using LeastSquare Mini
16、mization problem has been proposed in Al-Saggaf et al. (1991). Genetic algorithms are more likely to converge toglobal optima than conventional Least Square Minimization Techniques: since they search from a population of points and arebased on probabilistic transition rules. This minimization techni
17、que is ordinarily based on gradient descent methods, which, bydefinition, will only find local optima. Genetic algorithms can also tolerate discontinuities and noisy function evaluations. In thepresent work effect of SMES unit and GRC are also included. This introduces the non-linearity in the syste
18、m for such a systemconventional minimization technique does not give the effective results. Therefore, GA based PID controller tuning is consideredin the present work.In this work, first a multi-area AGC scheme suitable in a restructured power system has been developed then a GeneticAlgorithm based
19、PID (GAPID) controller has been proposed for this multi area AGC scheme. The proposed method of controllertuning implemented in an interconnected two areas and four area power systems. MATLAB SIMULINK has been used forsimulation studies. By minimizing the fitness function we get the optimal paramete
20、rs of PID controller. Integral of the square ofthe area control error (ISACE) have been utilized to select the fitness function for genetic algorithm. The population size 50 hasbeen chosen for genetic algorithm to obtain the optimal values of PID controller.The proposed GAPID based AGC scheme has be
21、en tested on a practical 39-bus New England system divided into two controlareas and a 75-bus Indian power system divided into four control areas. A deregulated electricity market scenario has beenassumed in both systems. The effect of generator rate constraint (GRC) has also been considered in the
22、multi area AGC model. Acombination of bilateral transactions and Poolco-based transactions has been considered, and it has been assumed that both thegenerators and the consumers are participating in the frequency regulation market. Simulation results show that the proposedGAPID Controller complies w
23、ith NERCs standards. The performance studies have been carried out by using the MATLABSIMULINK for transactions within and across the control area boundaries.2. System ModelingElectricity reforms are being brought to introduce commercial incentives in generation, transmission, distribution and retai
24、ling ofelectricity, with resultant efficiency gain, in many cases. Introduction of competition in electricity market may cause emergence ofseveral new entities, such as Generating companies (Gencos), Transmission companies (Transcos), Distribution companies(Discos) and system operator (SO). The syst
25、em operator is an entity entrusted with the responsibility of ensuring the reliability andsecurity of the power system. It is an independent entity and does not participate in the electricity trading. In order to maintain thesystem security and reliability, the SO procures various services, such as
26、supply of emergency reserves, frequency regulation andreactive power from the other entities in the system. These services are known as the ancillary services (Jayant Kumar et al,1997).A. Poolco based transactionIn Poolco based transaction, the Discos and Gencos of the same area participate in the f
27、requency regulation through systemoperator. System operator (SO) accepts bids (volume and price) from power producers (Gencos) who are willing to quickly (within about 10-15 minutes) increase or decrease their level of production. Consumers (Discos) also can submit bids to SO forincreasing or decrea
28、sing their level of consumption. In each hour of operation, the SO activates the most favorable bid. If thefrequency is lower than nominal value, up regulation bids are activated by the System Operator in steps and the highest activatedbid becomes the regulation price, uniformly paid to all the prov
29、iders of upward regulation service. If the frequency is higher thannominal, down regulation is activated by the System Operator in steps and the lowest activated bid price becomes the uniformprice, to be paid by all the down regulation service providers. Thus, the hourly regulating price is fixed as
30、 the price for the mostexpensive measure (regulating up) or least expensive measure (regulating down) utilized during the hour. At the end of scheduledinterval, the net energy balance of each entity is calculated and financial settlements are carried out.B. Participation factor of a Genco and Disco
31、in Frequency Regulation MarketLet there be n number of power producers and m number of consumers in area-i participating in the market. Assume that thebids submitted by the power producers and consumers, for frequency regulation are (pg(1),cg(1), (pg(2),cg(2),.,(pg(n),cg(n)and (pl(1),cl(1), (pl(2),c
32、l(2),.,(pl(m),cl(m), respectively given in Tyagi et al (2008) , Where, pg (i) is the price forregulating power quoted by ith Genco for upward regulation, cg (i) is the capacity quoted by ith Genco for upward regulation, i=1,2n, pl (j) is the price for regulating power quoted by jth Disco for upward
33、regulation, cl (j) is the capacity quoted by jth Discofor upward regulation, j=1, 2m. If Tdem is the total extra demand that arises in the hour of operation in any area for upwardregulation, the participation factor of each Genco and Disco in that area can be calculated by minimizing the cost of reg
34、ulatingpower,Bhongade et al. / International Journal of Engineering, Science and Technology, Vol. 3, No. 1, 2011, pp. 220-236222. . . . . . . . . .(1)Subject to a set of constraints. . . . . (2). . . (3). . . (4)Where, gen (i) is the change in the power generated by the ith Genco, load (j) is the lo
35、ads curtailed by the jth Disco. Although theprice for the up regulating power is the maximum bid price selected to generate the power for frequency regulation, but the Gencosquoting the minimum price area allowed generating the maximum power. Participation factor of the ith Genco for up regulation c
36、anbe defined as,. .(5)And the participation factor of the jth Disco for up regulation can be defined as,. .(6)For down regulation, the participation factor of each Genco as well as Disco in any area can be calculated by maximizing the costof the regulating power defined as,. . . . . . . . . .(7)Wher
37、e, regn (i) is the reduction in the power output of the ith Genco, uload (j) is the increase in the load by the jth Disco, Tdem isthe reduction in the total load demand in the area. Participation factor of the ith Genco for down regulation can be defined as,. .(8)And the participation factor of the
38、jth Disco for down regulation can be defined as,. .(9)C. Bilateral transactionsIn bilateral transaction, Gencos and Discos negotiate bilateral contracts among each other and submit their contractualagreements to a system operator (SO). The players are responsible for having a communication path to e
39、xchange contract data aswell as measurements to do load following in real-time. In such an arrangement, a Disco sends a pulse to Genco to follow thepredicted load as long as it does not exceed the contracted value. The responsibility of the Disco is to monitor its load continuouslyand ensure the loa
40、ds following requirements are met according to the contractual agreement. A detailed discussion on bilateraltransactions is given in Donde et al (2001).In this work, bilateral transactions within the area and across the area have been considered. Disco of one area can contract to theGenco of same ar
41、ea or other area to supply a certain amount of power in a specified time interval. These bilateral contracts can berepresented in the matrix form in which the number of rows equal to the number of Gencos and column equal to the number ofDiscos in the system. The elements of this Contract Matrix (CM)
42、 represent the percentage load demand of one Disco to differentGencos. Let us consider a Contract Matrix as given below:Bhongade et al. / International Journal of Engineering, Science and Technology, Vol. 3, No. 1, 2011, pp. 220-236223. . . . .0 10 20 10 . . .10.0.0 . . . .For example, the first col
43、umn of CM represents the Disco D1 bilateral contract with different Gencos. Element CM21 is 20 whichmeans 20% of total demand of Disco D1 in the schedule time interval will be supplied by the Genco G2. Sum of the elements ofany column represents the percentage of total demand of that Disco which wil
44、l be supplied by the bilateral contracts. Rest of thedemand will be supplied by the Poolco transactions.In case of Poolco transaction tie-line power between area-i and area-j is settled at zero value. But in case of bilateral transition thetie-line power is not settled at zero value but settled acco
45、rding to the bilateral contract between Gencos of one area and Discos ofother area.D. Calculation of Area Control Error (ACE)In a practical multi area power system, a control area is interconnected to its neighboring areas with tie lines, all forming part ofthe overall power pool. If . is the tie li
46、ne real power flow from an area-i to another area- j and m is the total number of areas, thenet tie line power flow from area-i will be. . . .(10)In a conventional AGC formulation, . is generally maintained at a fixed value. However, in a deregulated electricity market, aDisco may have contracts with the Gencos in the same area as well as with the Gencos in other areas, too. Hence, the scheduledtie-line power of any area may change as the demand of the Disco changes.Thus, the net change in the scheduled steady-state power flow on the tie line from an area- i can be expressed as. . . . . . . .
