Given the unity feedback system with G(s) = K/((s + 4)(s + 6) (Coa) emt emt ak Cun UChr tt Pech Ese ec ie ue RES Cac st To determine the gain K that will yield a 15% overshoot in the step response, we can Pesca ere Re eM Cec Oe Mc e na) nese Re ees Step 1: Determine Damping Ratio ¢ from Overshoot Requirement The percentage overshoot M, in a second-order system is related to the damping ratio va A=) eRe R ua} RECT ee STM ME Uo a anes) Seamer Cai ieee (a The desired pole location for a certain damping ratio and natural frequency w,, is given Ly ee a eee See teria m urate Comet er nen Onn kurt tei ckeu nulls Rem Cs Ry Step 3: Apply the Root Locus to Find AC For the open-loop transfer function: K eal ial CCC CSTD) The poles are at s = —4, 8 = —6,ands = —12 Using the root locus plot, find the location on the root locus where the damping ratio is approximately ¢ = 0.516. From this location, determine the corresponding value of for the system to yield the desired 15% overshoot. Would you like me to plot the root locus and calculate K’ for this damping ratio? ie mee) onGiven the following open loop plant G(s) = 10/s(s + 2)(s + 4)) Design a controller to yield 15% overshoot and atime of 0.4 ee) eer aes gear eee ets To desig a controller forthe plant G(s) De eee Dee eee eects overshoot and setting ime specifications Peer Tener any The specifications ae: Porras + Setting Time T, = 0.4 seconds Lt eee eee ec eee et et See mane) Creer caad, ‘Solving for ¢, we get approximately ¢ = 0.516. Reo ors Ce Cu eens qi es Given T, = 0.4 seconds and ¢ = 0.516: A eae SCTE Deemer PaO TS Bee aay Meee Ee ot ne cur Peer omen Ree eo eee PEs Pa tam me EE TEATY Se ene The third pole should be 10 times farther from the imaginary axis than the dominant eee Pee eg ay To achieve these closetoop poles, we can consider aPID controller a PD controller with pole-zero placement . ee Orr eee) Cre) Pec ted To place the poles as required et go through a design process often using root locus Ce ee ee eee ee ee ee ee tee ey tere ea iFor the digital system with forward transfer function G(2) = 0.56/(z-2)(2-3)(2-0.5)) find the static error constants and the steady state error if the inputs are u(t), u(t) and 7/2 * Oe US hu et a To find the static error constants and steady-state errors for the discrete-time system Rr Ce ara 0.56 ae — 2)(z — 3)(z — 0.5) and sampling time T’ = 0.1, welll proceed as follows: DCU un Can a eee) corresponds to the integral action in discrete systems). The system type dictates See ie ee at Inthis case, the transfer function G(z) has no pole at z = 1, so this is a type 0 Peo ee ee eee Ree eu Eee ei elec SOL EL PR eer ee en A) ine Pe eee ene ec Te ee ee eee CE) ENCED) ne) a 56 (-1)(-2)(0.5) ~ 1 ORs re Ae Sse een Oe ere Ae eT Ee oy eet) See eee eT Mn et) based on the static error constants. Perec aia) 1 i = Ee = ares 1+K, 14056 1.56 Seu aan ace ee eee Cer as) ee as (OH Peo eee en SUS)© _Tofind the static error constants and steady-state errors for the discrete-time system, ST CaS ea 2(z— 3)(2—0.5) and sampling time T = 0.1, well proceed as follows: eee cnn Inthe z-domain, system type is determined by the number of poles at z = 1 (which corresponds to the integral action in discrete systems). The system type dictates Cee ee cate eo eee eee Pcs Se ee ees ane Dee ieee ae ration error constant K, willbe zero. Peete ee stants ate defined based on the type of cs Se ae 0.56 Pate > = lim G2) = Gaya aa 08) en ee arse) aaa Re ee acy co) SET) Peres eee ne ee eet eT) eee ae 3. Steady-State Errors: The steady-state eror ¢,, for each type of input is calculated eee ee eee eer} 1 a 1+K, 14056 1.56 + Fora Ramp input tu():Since Ky = 0 the steady-state eo fora ramp input ae + Fora Parabolic Input u(t): Since K = 0 the steady-state eor fora pes ec Tract eee eet) Pe sec ee) Oe cee Cio Se art et ae! Ae ae ee ee ee ee enaFor a unity feedback system with G(s) = K/(s + 2)(s + 6)(s + 8) design a lag compensator using bode plot so that the system operates with a 10% overshoot and a static error Ceca To design a lag compensator for the unity feedback system with the plant: ie G(s) a (s+ 2)(s + 6)(s +8) Prue 1. Determine Specifications and Requirements: Se eee oe eee este en Tg reece eRe Setting My = 0.1 Ese SON ee Eo + Static Error Constant of 100: For a unity feedback system, the position error Cenc e enor ns 4 (2)(6)(8) eC es rT} K 100 = — Ti Line 9600, + Natural Frequency w,,: To place the dominant poles in the desired location, we assume a standard second-order approximation and use the following formula oats T)=—— where wa =unV1—-@ Sen een ee se focus on the gain and phase margin requirements set by the damping ratio to SRE Pe Ren ie ae eRe em ee without significantly affecting the transient response or stability. Its transfer er D(s) aa to add low-frequency gain, which increases the position error constantconstant Kis given by: K < 2)(6)(8) eee Setting K, = 100: Thus, K = 9600. + Natural Frequency w,,: To place the dominant poles in the desired location, we Pee Sect a Rouen RS Sennen for peak time: Since we haven't been given an explicit settling time or speed requirement, we'll focus on the gain and phase margin requirements set by the damping ratio to Cena Pec cic s ua Re io RoE Cee without significantly affecting the transient response or stability Its transfer Mian ewe D(s) . s+p TU ee ECCI CMe SRL LMU Ce without greatly affecting the phase margin. or PS RU Oem rma MR are] CCU aan eke) OR Ge ease i Eames + Step 2: Position the pole of D(s) (lag compensator) close to the origin, for Oe ee eS A Re aU en aC —0.01, to provide additional gain at low frequencies. BSC eM set eae ee tea werner aioe Chee a gece escic error requirements and ensure phase margin aligns with the desired damping ratio (¢ © 0.591). Would you like a more detailed step-by-step Bode plot analysis to finalize the Ce Ieee Cea a Ae tee RU LT) configuration to approximate the required gain and phase margin. Ot 6 SageS at eo RP cha Cea CR Meu represented in parallel form. Design an observer with a transient response described by zeta = 0.45 and wn = 100 Ce Re Sn Te SU ee Ue a) Re eRe mo my ‘transform the design back to parallel form, eae acne at 20 is CCE) G+ Se ne nena eT Tc natural frequency w, = 100, we will proceed with the following steps: See TD ea ee a ae ee ee Ed faster than the poles of the plant to ensure rapid convergence of the state SET) The dominant poles of the observer are specified by ¢ = 0.45 and w, = 100. This tee een eng ce Ee ter a Res Substituting ¢ = 0.45 and w, = 100: Sees Esra Vem ec Calculating these values: 812 = —45 + 789.44 Tee ene ae a ey axis than the dominant poles. Therefore, we place the third pole at: Pe ee eet) So, the desired observer poles are: 8,2 =—45 489.44, 83 = —450 Pe ek eer a eee Maen Pie 20 a CECE CE) eT nS ac oe eect representation and then rearranging it to place the characteristic polynomial of the PEE eae eee In observer canonical form, the plant dynamics can be expressed with a system matrix A, input matrix B, and output matrix C. The observer canonical form has the Rrtee eg ee aeThe third pole of the observer should be placed 10 times farther from the imaginary CeO ne mm eR ee 83 = —10 x 45 = —450 ROR ee Seca a Pe oe ee ae Pen te eke ee I eR eu 20 a cry CECE Converting this to the observer canonical form involves rewriting it as a state-space Cee Run Ui ene ecient gt EE NCE Cg ee od NeW ae ee UT ee eC) matrix A, Input matrix B, and output matrix C.. The observer canonical form has the eee auc eee ot) Peters Ree Cee Mea Come nis ees ca cAc Rog SECS ees 3. Design the Observer Gains Using Pole Placement: Let L be the observer gain vector that places the poles of the observer at the desired locations. In observer canonical COR cme Nu Rem ee eee cod aU Scent Pee cau em ce ecu Rec calculate the observer gain I to achieve the desired observer poles —45 -+ j89.44 and —450. Pee ne a ee ce ic Cm RT observer gain Lin the observer canonical form, we convert it back to the original (parallel) form by transforming through similarity transformations that revert the BCU e cece Re cee) This process involves several state-space manipulations and transformations, including calculating the observer gain L via pole placement, which can be simplified with computational tools such as MATLAB or Python. Would you like a step-by-step example Gee RR Tne Dassen Ue ® Oo OnFind G(z) for G(s)= 20/(s*5) In cascade witha sampler and See UE ee ee te @ Tofind the discrete-time transfer function G(2) for the continuous-time system eee in cascade with a sampler and a zero-order hold (ZOH) with a sampling petiod T Deon er Dee a ee eee Cres G(s) Pe OTE eee ae eT erent Cerca te oe eee eta) the zero-order hold. The ZOH method considers the effect of hold the input constant between sampling intervals. This conversion involves finding the me Se ee ay eee 5 isthe sampling period. é vie) For simplicity, we use a standard result fr converting a first-order system with a Fa G(2) = —— -G(0) cn ee ae ee Pee eee +K=20 Pee a er eee? 4, Form the Discrete-Time Transfer Function G (2): Substituting into the formula eee) TY AY ree FTC eet rr aco Seen ae par area) This represents the equivalent discrete-time model forthe continuous system G(s) ee ee tdIT ee Cues) COR ACD CL ne ee ec ruck ens Cee ec Rak So ae Reo aC reduce the settling time by a factor of 2 BOs ae gee oC a function: A esr ears CE) welll first evaluate the current dominant poles that yield a 15% overshoot, then design a PD controller to reduce the setiling time by a factor of 2. SORES CIR St Pe see nue a so Locations cre Seu ers Srey overshoot, we can calculate the damping ratio ¢ using the overshoot Ne ec ce TA (£2) RON Rum eee sd Step 2: Current Settling Time and Dominant Poles The settling time T,, for a second-order system is given by Pee CR noe eee Ree ce cA) evaluating the current dominant poles, welll place the dominant poles of the system to ee SC Cn ie neo Decor Cee Scns a ra ue Cn a ee ie oe Naa Pee eto Mt Crna ag ec ame that the root locus of the system has poles at these locations, with the damping Ce SCF Tee aU eke eee Pom locus analysis or direct observation. Suppose the system's dominant poles are found at Pe eee Ree eh Cuan cassStep 2: Current Settling Time and Dominant Poles The settling time T,, for a second-order system is given by: ri ron ay Ree REM ere Rok oa eee eee ce A evaluating the current dominant poles, we'll place the dominant poles of the system to Bee Cee rune Leta ee Reto 1, Assume the open-loop transfer function has dominant poles at: a OVS ee 2. Using the root locus technique, we would determine the values of K and w,, such Ta eR ea Nee) ratio ¢ = 0.516. Let's proceed under the assumption that we can calculate or estimate w, from root ete Dee ete ee es SC a eet 5 Re Eee a Cue tenets Step 3: Design the PD Controller To reduce the settling time by a factor of 2, the new settling time 7 should be: cs PJ Peon De Cn nk e Sne iceeem RC Ree CU) CUM CUG CeCe Mae De ee For a PD controller, the transfer function D(s) is given by: pol) eer e9 Deen ae orn ee eee ee rue ema} place this zero to adjust the effective damping and natural frequency, thereby achieving NIRS Cae ey 1. Place the Zero of the PD Controller: We want the zero of D(s) to be located in such ON ecco es ACC a RC ret} 2. Adjust Gain to Achieve New Natural Frequency: Using the new desired poles (which are located at approximately twice the distance from the origin as the original Peo menu Rem cg ee en een a aera ad Cre eae See adNene au Ee OSC) CeCe uC) @ _Toanalyze the stability of the system with a sampler and zero-order hold (ZOH) in ore Une ee a cy Ey rm) co we need to determine the range of the sampling period T’that ensures the stability of the resulting discrete-time system. Roves etre 1. Continuous-Time System and Discretization: The continuous-time transfer function, of the system is: Os With a zero-order hold, the discrete equivalent transfer function G(z) can be obtained using the formula for converting a first-order system under ZOH sampling: lee? G(z) G(0) os roe + G(s) = *,,with K = 3 anda =3. oe + G0) =k =3=1 ae Deca ea et eed Roum Sten re 0 era 3. Stability Condition: For the discrete-time system to be stable, all poles of G(2) must lie inside the unit circle in the z-plane. The pole of Gz) is at: re For stability, this pole must satisfy: ca a Sa EO ace ee Rn mn Be RE es yea) This condition is always true for positive T,, but for stability, we need to ensure that Oa 4, Range of T° In this case, any positive T will result in a stable discrete-time system ae Peete See ae eicity 3 we need to determine the renge of the sampling period T'that ensures the stabilty of RT etc cun Step-by-Step Approach 1. Continuous-Time System and Discretization: The continuous-time transfer function Cae cules Oe With a zero-order hold, the discrete equivalent transfer function G(z) can be obtained using the formula for converting a first-order system under ZOH sampling: eal G(0) i Here: + G(s) = 3,with K = 3 anda ee ee Bei) ee 2. Discrete-Time Transfer Function: Using the formula, we find the discrete-time transfer function: rie ae 3, Stability Condition: For the discrete-time system to be stable, all poles of Gz Dee eee nea ne Ea ome Ste a an Since e~8? is always positive, the inequality e~®? < 1 must hold. Taking the Pee ees yar) This condition is always true for positive T, but for stability, we need to ensure that Ga 4. Range of T: In this case, any positive T' will result in a stable discrete-time system eee eee eek a T>0 GIeueic The range of T that makes the system stable is: yaa) ‘This means any positive sampling period will yield a stable discrete-time system for the given continuous-time transfer functionCompare lag and lead compensator with respect to application, pole- zero plot and circuit for implementation. onus Ea ‘compensators from their respective circuits err Cem ets Ur Bs Re ule Deut See + Purpose: A lag compensator is primarily used to improve the steady-state Ce RCM ec ere emo sO Maa minimal changes to the transient response. It provides low-frequency gain and Poe ec Cee I sean ey Cree cea} + Typical Use: Lag compensators are often used when there is a need to improve the steady-state performance (e.g., reducing position error for step inputs) ST ern RCs ee sets) Bee eco Oe een oie ea US ee Cy (e.g, increases the phase margin, reduces overshoot, and settling time) by re Ce ou e on Cuma La stability and speed. It does this by increasing the high-frequency gain and enhancing the systemis speed of response. + Typical Use: Lead compensators are used when there is a need to improve the Gee ee nee eo RCE Cu) speed of response, without significantly increasing the steady-state error. Pee edad Seeley cing Peper i Cie ean inn eee enc nonce the origin) and a pole at a higher frequency. The pole is typically much farther away from the zero, creating a low-frequency dominant response. Cece * The zero is located close to the origin (s = —2). + The pole is located at a higher value (s = —p), with p > 2. Sti Mo ee eee aa aie se adding gain and improving steady-state accuracy without influencing the Tee ena ene Pa eee eePac ecu ce Pe ee coe me ou om Ene Mme To and a pole at a lower frequency, with the zero being closer to the imaginary axis than the pole. This configuration adds positive phase and increases the phase Cr mu Ec Cee elas Skee RUT I) Se ea an eae Cee ML ed Ree oR Mn Ui ca cei Un performance by adding phase at higher frequencies. eRe uct + Lag Compensator Circuit: ‘+ Alag compensator can be realized using a resistor-capacitor (RC) network. See ert CU ar ea Re ae ae implemented by using a series RC circuit. The capacitor introduces a phase Tem Ue Sm ei Rec ie RRUC Run cats CUT Ro aca caw D(s) 77 where z is the zero, and pis the pole, with p > 2: Circuit diagram: A simple implementation would consist of a resistor and capacitor tok a Cues caiod See Dore omeiee + Alead compensator can be implemented with a series RC network but with the Mc ues eR RS te Seen Ur ae Cee ES ee oho EEL) De mie aUC Un eae Ce Rieu og ers Pere Rue Run ae Moe Ree cao ew Oe a7 where z is the zero, and pis the pole, with z > p. Circuit diagram: A simple implementation would consist of a resistor and capacitor ete a Cuenca PMU mateo Bicol mel fel iosCUT Cm ate Con cele folic Se ei Ra Re een UCC ICs eR Ren ue Ca D(s) =**# ra SVU a aed a CoS ate Re OS Cun ue ect cae aru DE eco ea Sa ec ee oe + Lead Compensator Transfer Function: From the RC circuit configuration, the transfer eeu Ree ue od ew This transfer function describes a system with faster dynamics, improving the transient response by increasing the phase margin and reducing overshoot Summary Table CT ie ae eo Cec ed ee Ce ac) (reduce error) Ceo eo Crd Cron Ce ee ee ee Rg DO Eset es Pea) eet eee arco Sy fe eon cis foe Ween Coren) Zero closer to the imaginary axis, pec RC network, capacitor and resistor in rd WO ete eee ed eee In summary, lag compensators improve steady-state accuracy with minimal impact on TCU eR en ae CUTE) stability) without significantly affecting steady-state accuracy. Both can be implemented Po eet e ne ee ua. eee Can Me the system acts as a lag or lead compensator. oOo6 lo sdoCS MUR eis aac OEIC Cees) Clces 5)(s+7)) Design a controller to yield a 10% overshoot and a seconds by assuming that the plant is represented in the Pre nay To design a controller for the given open-loop plant G(s) CY 10% overshoot and a settling time of 2 seconds. Welll approach the design by first analyzing the plant and then designing a controller that meets the desired performance Bee Ie eee ce Aaa Cara) This is a second-order system in parallel form. To achieve the desired overshoot and. Ser RR ea ee mre Ec Reece ace function model, such as a Proportional-Derivative (PD) controller, which typically shapes ROC ue Cas ee SMR ee Co Roe en + Overshoot (03): 10% overshoot corresponds to a damping ratio (¢) of approximately OCU ncaa aie ae * Settling time (Ts): For a second-order system, the settling time can be approximated oy, Teme hc nee a Cue Given that T, = 2 seconds, we can solve for the product of ¢ and wp Pee nee Best 7 deal 0) Gaetan Es Sera eee te cel To meet the desired specifications, we can use a PD controller, which has the form: O(s) = K, + Kus This will add a zero to the open-loop transfer function, influencing the transient response eeu eet eT) Step 3: Find the closed-loop transfer function The closed-loop transfer function T(s) for the system with feedback can be written asby: C3 Paes a eee te eee Given that T, = 2 seconds, we can solve for the product of ¢ and wn nat om one) We already know ¢ = 0.591, so: ae Cd dil ry Rakes Step 2: Choose a suitable controller Bae ee alle ue ne Re el RR RCL ol ee ena er) SURE PR RU Ee RC Cae Cue RRC US Teh (overshoot and settling time). Step 3: Find the closed-loop transfer function Dee ca PaO Rue can Ue Cas G(s)C(s) A ea wmrerPyI61¢) Inserting the given G(s) and the PD controller C'(s) Step 4: Analyze the poles and zeros To achieve the desired damping ratio and natural frequency, we will need to analyze the poles and zeros of the closed-loop system and adjust the controller gains K’, and Dee eae a Cu ec Cone) achieve the desired transient response. Oe nares Re ee Re Ee Oe end fine-tune the location of the poles and achieve the desired overshoot and settling NR eR Retna eee een ek Python (with control systems libraries) to simulate and tune the controller. Rr Sn Ce eS ecient eR eu nee eed wOAG®cy Use frequency response methods to design a lead compensator for a 10 unity feedback system where G(s) overline K (5#7) s(s + 5)(s + 15) and the following Bree ee ee eee cee) Settling time 0.1seo, and Kv = 1000 To design a lead compensator using frequency response methods for the given system, we need to meet the following specifications: ‘+ Percent Overshoot: 15% (which translates to a desired damping ratio C) Set eae ee a Cee te Rega + Velocity constant Ky: 1000 (which helps us determine the system's steady-state error for a ramp input) SCRE See cree Ty K(s+7) (8 + 5)(s + 15) Gs) = = ‘The closed-loop system has unity feedback, so the closed-loop transfer function T'(s) is: 10} Boe a + Percent Overshoot: 15% (which corresponds to a damping ratio ¢ of approximately 0.516) + Settling Time T, = 0.1 sec: The settling time for a second-order system can be Ere ES cy Ver (an Meee ene na gent rt 4 cI Se lea ell yaad ors Tae om 0.1¢ ~ 0.1 x 0.516 SOc ecu tacit ts Oe ti nee LI) Be Ta as) Se Lut) SVU RUA eee CUle Le The velocity constant K., fora system is related to the low-frequency gain for a ramp a ue ee Co K(s+7) 0 array ce Ty) Ser ney aes UCC Cen ea eee Cancers ica isOry eRe ecu ea frequency gain. The velocity constant Ky for this system is: K, =lims G(s) eee eee K(s+7)_ K+? a = = TCE) CesT) 15 CRs Cie UL Te We ee Pres [eae ETE) 75 So, we set K = 10714.29 to satisfy the velocity constant requirement. Se eae Te Resa ic ‘Now that we have the desired performance specifications, we can design a lead Coa rl) P Where = is the zero and pis the pole, with p > 2. The purpose of the lead compensator isto increase the phase margin, which will help achieve the desired overshoot and Een De ae ee Dee Cn ha eo em Se a) higher frequencies, thus increasing the bandwidth and improving transient co 2. Achieve the desired phase margin: The phase margin is related to the damping ratio, EE erry Bee ely et Co ea) To design the lead compensator, we need to: DR ese ee ee frequency corresponding to the desired natural frequency w, ~ 77.52 rad/s. Pe oe eg eee gee mus eed placed at a frequency lower than the pole p, and the compensator is designed so CES ee oe ee oe een ee eet anys eee ea ene RC eee Pe ee ee ae) Ces Score uS ancl eS ee ee ae (using control system libraries). The compensators parameters z and p can be fine tuned to achieve the required phase margin, bandwidth, and transient response.ee ee ey compensator fora 10 unity feedback system where G(s) Bern ea ete Dee ee ee ea Suet ead © To design a lead compensator for the given system using frequency response methods ‘and meet the specified requirements, let's break down the problem step-by-step, MSC pena See Cun eg eee eae ag Cea 1(s) = G(s)O(s) where C(s) is the compensator. BS ere ey Bere + From the standard second-order system, we know that the percent overshoot ( Coenen eee not ores For OS = 15%, solving for ¢, we get: + Settling time: T, = 0.1 sec Aa ee ea py SRL Uses Bae ee eee) I eno Ey SAC aes K, = lim sL(s) = lim sG(s)C(s) ene cake Om Or Rs veR oy We want the system to have a velocity constant K’, = 1000, which influences the Cramster eae cikel 6) ee RA enn yee eae oz Since we are designing a lead compensator, we will choose a lead compensator of the cr (oleae 7 where z is the zero and pis the pole. The compensator will add a zero and a pole, shaping the frequency response to meet the desired specifications. See RRO aa Re act mcd The compensator should be designed to shift the open-loop pole locations to meet the desired ¢ and w,, as well as satisfy the velocity constant requirement. + Tomeet the overshoot requirement (¢ = 0.516), the poles should be placed close to the desired natural frequency w, = 77.52 rad/s. This corresponds to a damping PERE cis nome ct Sn Ronse Sem Ceo et CON eta + Toachieve the setting time of T, = 0.1 sec, we want the closed-loop poles to be cr len oe ne eal Te neh aeRO Cr Sc LE $= ~0.516 x 77.52 + j77.52V1 — 0.516 which corresponds to the poles of the closed-loop system. SCOR Rae ce gcc a ued To finalize the compensator design, you would typically use frequency response tools to Peau gue ee nm es uk Rene the gain, zero, and pole locations until the system meets the performance criteria. Key Objectives: ce og ee ae cae ca ct Cee oe een GG 2. Natural frequency: Place the poles at a location corresponding to w, = ae Reece e Cee eee (LUD US In practice, these calculations would be followed by using MATLAB or another control DCU ie UU neg sdGiven the unity feedback system with G(s) = k/(s(s +3)(s + 9) use frequency response methods to determine the value CaN Ces au R ey eet To determine the gain K for the given unity feedback system G(s) = RSC RES Sc Cs ie MI Sc) Coney ee COMMU cU RUE nn rca meee Dra iT Ratio BO ee Oe) eee cei One enc nO) Pe RC Given that the desired overshoot is 15%, we can solve for ¢ Coe 0.15 Solving for ¢, we can use the inverse of the overshoot formula. This equation can be RCTS een fea y Sera cma Ce nace sg om) ce eee ee eC er nu ur enc ages PCR en al rt = (en 7 Cee ena ee nee ee) system. Welll first use the damping ratio to determine the desired closed-loop pole eo Step 3: Analyze the Plant and Find the Open-Loop Transfer Function The open-loop transfer function is: k tO ah Re Le ie pare yr ara See ee ei) RC cava eo Ec lene L(s) Gx) TG) et GIG aes 7O mee eHThe closed-loop transfer function T'(s) is: aC) Ecce (Oia te aie a con See a eetart Meoie acon Mente ne ea ne eae eC Co oem eco ee es eee To achieve the desired damping ratio ¢ = 0.516, we need to find the value of K' that will place the poles at the correct location. The poles are governed by the characteristic equation of the closed-loop transfer function fe ee ease edt eee ea) ecg toms een Oe Ihe Cre ae ecr Bee enc eae ice uy response, and adjusting k will help achieve the desired overshoot and damping. Step 5: Simulation and Fine-Tuning Reco OU ease Cena) system and plot the poles for different values of k. You would adjust k: iteratively to SCE eR eee eens MATLAB Example (for Simulation) acl ol Mae ROR a eG rat oe % Given plant transfer function G(s) ED) PCIe CoE Re ee ee eee scum cme NEC) eo Co ee) coe acy Re CCR CU aCe) Po ee OCC MAE eC ee OTC or RLS) Peer ee ae ee moc eee etD) ee we he Ree ness ee ec CRE) Ge eee a aoe eee BeneCS OUP U Ba ML en R UCR Seer euler Ca STS) Step 5: Simulation and Fine-Tuning cea eI eC UR a eee MRO Ca OCU n UM See Ce RCM EC ee aceon) match the desired damping ratio and overshoot. NTO Neha Clams Ute) Here's how you could use MATLAB to solve for K: RCM Cec om as CIES) gS tf("s"); ea A Cl CLD Peer ee eC Re Cm asC mcm ese cc ras Ee tC Paste sean aS) desired_os = 0.15; Pee mec ree eee) zeta = -log(desired_os) / sqrt(pi*2 + log(desired_os)42); % Compute the natural frequency omega_n (example calculation) omega_n = 2 / (zeta * Ts); % 1s = desired settling time, example Ts = 0. ea eR aed Benn This script can help you determine the value of K that achieves the desired overshoot CYS Rese ee Step 6: Conclusion Bena Se eee eee ae ee uC kek 3 op BS ate Oc Pace ete numa cay Ce enon cee SRLS eee oe cen ec orn Ron CAE Rei eCn r ae Gr enn PoECee Te eee) ee a ec nea is transfer functions. Also develop the circuits for their or) ee eae Ue Re ecu ory Let's first break down the two compensators (PI and Lag) and compare them with respect to their pole-zero locations, transfer functions, and their circut realizations Dea Gc Re ‘API compensator has the following transfer function: Cpr(s) = K, (14+ 2 710 ead Cat or oes + K,,s the proportional gain. Oe ec kn Ue This can also be written as: Vereen 1 ae aro Oe eda ea ee Cee eee Seton ee eek cm Ec} the steady-state error (for a step input, this results in zero steady-state error). Restate SUC eee eC Cus * The pole at s = 0 increases low-frequency gain, improving the systems ability to handle steady-state error (ike position error constant K), Pole-Zero Locations for Pl Compensator Cerro ad aR RR Oe ey 2. Lag Compensator A.Lag compensator is designed to improve the steady-state response (specifically to reduce the steady-state error), while it has minimal effect on the transient response (the poles are placed far from the imaginary axis). It has the following transfer function: Ps Oe acca Pea ecec Oe eee ec Seer eee The relationship between the pole and zero for a lag compensator is |p| > |z|, $0 the POSS ae ene OSGeo oe ee eee AE + Pole of Lag compensator: At s = —p, where pis much farther from the origin Coe Ce? Effect of Lag Compensator: RU eget a me en eo Bee eure en weer en een CS response. This compensator is often used when the goal is to reduce the steady state error without drastically affecting the speed of the systern. Pole-Zero Locations for Lag Compensator Oecd ese eee naman ae ad cra Cane Re ein coy ead eed Soy Cpx(s) = Kp (1+ 4.) ea ra as=-} wher oa eet cre) ee ees Eee Penna ae eee Transient time and reduces overshoot Dee ey Seed Eg en ee ee ees See inputs ee a rey See een Response __improves steady-state accuracy _frequencies without much change in eos Cee ea eee cee Cees De ee ee ec response Pea rca Dee kc Pee ewe ae ue ne eg enn Fosse Meeferred cri Dee uc ogee ‘API compensator can be realized using an operational amplifier (Op-Amp) in the oes Renee et caer Peat Se err ee err Kea ern «The resistor R and capacitor C set the 7. term in the transfer function SRC ne AS na Ruy Realization See na ro Ld Sa an ae Se a ene ne es ie Ca Canale Pere Caen’ OE igen ee Een ae CR CL aes ad choices for the resistor and capacitor values to control the location of the zero and pole. EERE To cete eee eC Eee Cait eae ORCS n ee Reo ios act Corer Oe ae en sca eee ee eo a eens Oe oc mea ec ny OMe an + Pole at s = —pis set by the Rp, C2 combination, where p is much greater than 2, Eee ee eee naan oe ce fore TATE) + Pl Compensator: Best for improving the transient response, particularly to reduce OTe ie Oa eC esate ae be eee eee Uc Sete a em acer er Sau Ay significantly affecting the transient response. It has a zero near the origin and a pole Den eae ee cae ee sms ay+ The resistor R and capacitor C’ set the 71, term in the transfer function ee ere siete caehery Realization: Cr aa an A Oe resin aes Ona Rte Renee eu te UR RR un al nl Pa en ieee LU ett ene en ee es ceo ere em nee Maen Cera er Reco + The zero Is realized by adding a resistor and capacitor in series. Oe eee eee neal) Corre SR OSC RU Re cur ci eC) a Ge ae a en uE te Deo era eet a OR Ly * Pole at s = —pis set by the Ry, C2 combination, where p is much greater than z. Core et oee eu Re RIE ae TU eer Conclusion Re meee RUC US neem Cen UAC ted overshoot and improve steady-state error for systems with position error. It has a Ce eee ERLE UIC a Oe Re ee am RR ase anu g significantly affecting the transient response. It has a zero near the origin and a pole much farther away, typically used to improve accuracy without compromising Reset Re ee neg ne a cn ee Re steady-state error and improve transient performance, a Pl compensator is preferred. If Neus ee ees Cec eM mM eS Mec Re tetal Ra CUSCUs Pt ee Te Ree San ome LT aaee On en ea G(s) = K/(s + 1)(s + 5)(s + 8) has 15% overshoot. Evaluate the Cee nee ei Cee Re Se esd 6 Problem Overview: Cee cuca ech 1. Evaluate the current dominant poles using the Root Locus (R.L.) method. Pee Renn eg eet ee SR cr Ruy Reece ae rer cues i G(s) Peas Crs CEs) This isa third-order system with three poles at s = —1, 8 = ~5, and s = ~8. BSCR eeu ae RSet The operioop transfer function with unity feedback is iis sn a) +5)(¢ +8) (Cran For a unity feedback system, the characteristic equation is: i He r f Siete baa CE SNES) CESS) Se SS a on Suna Cn Ce ae} + The system has three poles at s = —1, 5 = —5, ands = —8 Sete nen ee ee + Since the poles are real and distinct, the root locus will be drawn along the real axis between the poles. Dea ORT d ‘The dominant poles of the system are the ones closest to the imaginary axis, For a third order system, the dominant poles typically determine the transient behavior, including Risees Pee ue oe at ue eet csi Yap relationship: (ors ete ee aren agenteeesti pe rere eats The natural frequency w, can be estimated using the following relationship for the eet ge ce ee aris Overshoot = 15% => ¢=0.516 We can then proceed to find the value of wand locate the dominant poles. The ‘dominant poles will be located near the locations where the root locus tends to move when is increased, typically near the real axis and associated with the desired ¢ Step 3: PD Controller Design to Reduce Peak Time by a Factor of 2 Dae terry The peak time T, of a second-order system is given by: T= re Neo SO ee a os time is, then the desired peak time would be: Since Tis inversely related to, halving implies doubling w,. Therefore, we need CM cea Ros ey ‘APD (Proportional-Derivative) controller has the following transfer function: O(s) = K, (1+ Ts) es Be eae re See uu tence De ao eee plane to add damping tothe system and shift the dominant poles. Cee Tee cates + The PD controller affects the root locus by adding a zero, which will cause the root locus to shift towards the zero, changing the location of the dominant poles. Oe eu eR oe eM en) pas The PD controler will help shift the potes tothe rght in the complex plane, increasing Se nee ne See eucarRiSaeocg See Re eee ac Cee et eee + We can adjust the derivative time constant 7, to achieve the desired frequency ea Crete Re