On-Line HEV Energy Management Using a Fuzzy Logic Yacine Gaoua 1,2,3 - - PowerPoint PPT Presentation

β–Ά
on line hev energy management using a
SMART_READER_LITE
LIVE PREVIEW

On-Line HEV Energy Management Using a Fuzzy Logic Yacine Gaoua 1,2,3 - - PowerPoint PPT Presentation

Aug 25, 2022 556 likes β€’671 views

On-Line HEV Energy Management Using a Fuzzy Logic Yacine Gaoua 1,2,3 , Stphane Caux 1 , Pierre Lopez 2,3 and Josep Domingo Salvany 4 1. Institut National Polytechnique de Toulouse, INPT 2. Laboratoire PLAsma et Conversion d'Energie, LAPLACE 3.

slide-1
SLIDE 1

Yacine Gaoua 1,2,3, StΓ©phane Caux 1, Pierre Lopez 2,3 and Josep Domingo Salvany 4

  • 1. Institut National Polytechnique de Toulouse, INPT
  • 2. Laboratoire PLAsma et Conversion d'Energie, LAPLACE
  • 3. Laboratoire d'Analyse et d'Architecture des Systemes, LAAS-CNRS
  • 4. Nexter Electronics, NE

ygaoua@laplace.univ-tlse.fr caux@laplace.univ-tlse.fr, lopez@laas.fr, j.domingo@nexter-group.fr

On-Line HEV Energy Management Using a Fuzzy Logic

slide-2
SLIDE 2

Outline of the presentation I. Introduction to HEV energy chain II. Sources characteristics

  • III. Modeling
  • IV. Solving method

V. Off-line optimization

  • VI. Results and performance
  • VII. Conclusion
slide-3
SLIDE 3

Hybrid Electrical Vehicle Battery Super-capacitor Fuel cell HEV energy chain

  • I. HEV energy chain
slide-4
SLIDE 4

Parameter Meaning π‘±π’…π’Š Demand of the powertrain (A) 𝑱𝒃

𝒏𝒋𝒐,𝑱𝒃 π’π’ƒπ’š

Min/Max current exiting the PCube converter (A) 𝑱𝒕𝒅

𝒏𝒋𝒐,𝑱𝒕𝒅 π’π’ƒπ’š

Min/Max current provided by the super-capacitor (A) 𝑽𝒕𝒅

𝒏𝒋𝒐,𝑽𝒕𝒅 π’π’ƒπ’š, 𝑽𝒕𝒅(0)

Min/Max/Initial voltage of the super-capacitor (V) 𝑻𝑷𝑫𝒄𝒃𝒖

𝒏𝒋𝒐,𝑻𝑷𝑫𝒄𝒃𝒖 π’π’ƒπ’š, 𝑻𝑷𝑫𝒄𝒃𝒖(0)

Min/Max/Initial energy level in the battery pack (%) 𝑫𝒃𝒒𝒄𝒃𝒖 Battery capacity (Ah) πœ π’– Time stepsize (s) 𝑺𝒕𝒅 Super-capacitor internal resistance (Ξ©) 𝑫𝒕𝒅 Super-capacitor capacity (F) 𝑭𝑴𝒑𝒕𝒕𝒄𝒃𝒖 Battery energy losses (kW) Eπ‘΄π’‘π’•π’•π’…π’˜π’• Energy losses of the PCube converter (kW)

Input parameters. Battery efficiency. Convertor efficiency.

  • II. Sources characteristics
slide-5
SLIDE 5
  • 𝑱𝒄𝒃𝒖

𝑺: Real battery current

  • 𝑱𝒄𝒃𝒖: battery current
  • 𝑻𝑷𝑫𝒄𝒃𝒖: Battery State of charge
  • 𝑽𝒄𝒃𝒖: Battery voltage
  • 𝑱𝒕𝒅: Super-capacitor current
  • 𝑽𝒕𝒅: Super-capacitor voltage
  • 𝑱𝒃: Convertor current

(nlp)

𝐽𝑐𝑏𝑒 + 𝐽𝑏 = π½π‘‘β„Ž π½π‘‘β„Ž > 0 π½π‘‘β„Ž ≀ 𝐽𝑐𝑏𝑒 + 𝐽𝑏 ≀ 0 π½π‘‘β„Ž ≀ 0 𝐽𝑏

π‘π‘—π‘œ ≀ 𝐽𝑏≀ 𝐽𝑏 𝑁𝑏𝑦

𝐽𝑑𝑑

π‘π‘—π‘œ ≀ 𝐽𝑑𝑑 ≀ 𝐽𝑑𝑑 𝑁𝑏𝑦

𝑉𝑑𝑑

π‘π‘—π‘œ ≀ 𝑉𝑑𝑑≀ 𝑉𝑑𝑑 𝑁𝑏𝑦

𝑇𝑃𝐷𝑐𝑏𝑒

π‘π‘—π‘œ ≀ 𝑇𝑃𝐷𝑐𝑏𝑒 ≀ 𝑇𝑃𝐷𝑐𝑏𝑒 𝑁𝑏𝑦

𝑄

𝑐𝑏𝑒 𝑆 = 𝑄 𝑐𝑏𝑒 + πΉπ‘šπ‘π‘‘π‘‘π‘π‘π‘’(𝑄 𝑐𝑏𝑒 )

𝑄

𝑑𝑑 = 𝑄 𝑏 + πΉπ‘šπ‘π‘‘π‘‘π‘‘π‘€π‘‘ 𝑄 𝑏 + 𝑆𝑑𝑑 𝐽𝑑𝑑 2

𝑇𝑃𝐷𝑐𝑏𝑒 = 𝑇𝑃𝐷𝑐𝑏𝑒 0 +

100.𝐹𝑐𝑏𝑒 π·π‘π‘žπ‘π‘π‘’

βˆ†π‘’ 𝑉𝑑𝑑 = 𝑉𝑑𝑑 0 + 𝐽𝑑𝑑 + 𝑆𝑑𝑑 +

βˆ†π‘’ 𝐷𝑑𝑑

𝑉𝑐𝑏𝑒 = 𝑔 𝑇𝑃𝐷𝑐𝑏𝑒 0 𝐹𝑐𝑏𝑒 = 𝑕 𝐽𝑐𝑏𝑒

𝑆

Decision variables: Mathematical modeling

Goal: Minimize battery discharge Under constrains of (system functioning, sources design, safety limitation),

𝒉: Computation of electrical quantity π’ˆ: Computation of battery voltage

  • III. Modeling
slide-6
SLIDE 6
  • IV. Solving method using fuzzy logic

Powertrain demand. Super-capacitor voltage. Battery current. Rules engine.

π’‹π’ˆ π‘±π’…π’Š = . 𝒃𝒐𝒆 𝑽𝒕𝒅 = . π’–π’Šπ’‡π’ 𝑱𝒄𝒃𝒖 = . 𝒑𝒔

Rules generation. Decision surface (centroid method).

Parameters setting: Genetic algorithm (off-line - GPS) Control and correction algorithm

slide-7
SLIDE 7
  • V. Off-line optimization

Mission profile NE (176s).

(nlp)

𝑡𝒋𝒐 𝟐𝟏𝟏 βˆ’ 𝑻𝑷𝑫𝒄𝒃𝒖 𝑼 = π‘΅π’ƒπ’š 𝑻𝑷𝑫𝒄𝒃𝒖 𝑼 𝐽𝑐𝑏𝑒(𝑒) + 𝐽𝑏(𝑒) = π½π‘‘β„Ž(𝑒) π½π‘‘β„Ž(𝑒) > 0 π½π‘‘β„Ž ≀ 𝐽𝑐𝑏𝑒 + 𝐽𝑏 ≀ 0 π½π‘‘β„Ž(𝑒) ≀ 0 𝐽𝑏

π‘π‘—π‘œ ≀ 𝐽𝑏 (𝑒) ≀ 𝐽𝑏 𝑁𝑏𝑦

𝐽𝑑𝑑

π‘π‘—π‘œ ≀ 𝐽𝑑𝑑(𝑒) ≀ 𝐽𝑑𝑑 𝑁𝑏𝑦

𝑉𝑑𝑑

π‘π‘—π‘œ ≀ 𝑉𝑑𝑑 𝑒 ≀ 𝑉𝑑𝑑 𝑁𝑏𝑦

𝑇𝑃𝐷𝑐𝑏𝑒

π‘π‘—π‘œ ≀ 𝑇𝑃𝐷𝑐𝑏𝑒(𝑒) ≀ 𝑇𝑃𝐷𝑐𝑏𝑒 𝑁𝑏𝑦

𝑄𝑐𝑏𝑒

𝑆 𝑒 = 𝑄𝑐𝑏𝑒 𝑒 + πΉπ‘šπ‘π‘‘π‘‘π‘π‘π‘’ 𝑄𝑐𝑏𝑒 𝑒

𝑄

𝑑𝑑 (𝑒) = 𝑄 𝑏 (𝑒) + πΉπ‘šπ‘π‘‘π‘‘π‘‘π‘€π‘‘ 𝑄 𝑏 (𝑒) + 𝑆𝑑𝑑 𝐽𝑑𝑑(𝑒) 2

𝑇𝑃𝐷𝑐𝑏𝑒 𝑒 = 𝑇𝑃𝐷𝑐𝑏𝑒 𝑒 βˆ’ 1 +

100.𝐹𝑐𝑏𝑒 𝑒 π·π‘π‘žπ‘π‘π‘’

βˆ†π‘’ 𝑉𝑑𝑑(𝑒) = 𝑉𝑑𝑑 𝑒 βˆ’ 1 + 𝐽𝑑𝑑(𝑒) + 𝑆𝑑𝑑 +

βˆ†π‘’ 𝐷𝑑𝑑

𝑉𝑐𝑏𝑒 = 𝑔 𝑇𝑃𝐷𝑐𝑏𝑒 𝑒 βˆ’ 1 𝐹𝑐𝑏𝑒 = 𝑕 𝐽𝑐𝑏𝑒

𝑆(𝑒)

Global optimization

Optimization using Operations Research methods: AMPL+ IpOpt algorithm (Interior Points)

slide-8
SLIDE 8
  • VI. Results and performance

HEV sources/ Method Number of cycles / Battery discharge HEV battery alone 30 Cycles – 88.3143% HEV with PCube - FL 34 Cycles – 88.8872% HEV with PCube - GAFL 35 Cycles – 87.6296% HEV with PCube βˆ’ IpOpt 39 Cycles – 88.7396% HEV with PCube Battery discharge (1 cycle) GAFL 2.546% IpOpt 2.29315%

NE Mission profile 176s. Mission profile 3h 50min.

HEV sources/ Method Number of cycles / Battery discharge HEV battery alone 1 Cycle – 52.3566% HEV with PCube - FL 2 Cycles – 85.7596% HEV with PCube - GAFL 2 Cycles – 71.9029% HEV with PCube βˆ’ IpOpt 3 Cycles – 89.896% HEV with PCube Battery discharge (1 cycle) GAFL 36.1712% IpOpt 30.49%

slide-9
SLIDE 9
  • VII. Conclusions and perspectives
  • Genetic algorithm improve the solution by setting FL parameters off-line,
  • Good quality of the results (in regard to the global optimization),
  • Development of decision support tool in C + + (implementation in a dsp target).
  • Validation of results on a real prototype.

Conclusions: Perspectives:

slide-10
SLIDE 10

Thank you for your attention

ygaoua@laplace.univ-tlse.fr caux@laplace.univ-tlse.fr, lopez@laas.fr, j.domingo@nexter-group.fr