This research addresses critical efficiency limitations in the production of adhesive-bonded motor cores for electric vehicles (EVs). Traditional lamination methods (e.g., welding, riveting) cause localized fixation, loose stacking, and increased core losses, hindering the power density and lightweight requirements of EV motors. Adhesive bonding offers superior insulation and tighter stacking but suffers from low production efficiency due to suboptimal drive system selection and parameter settings in rotary lamination mechanisms. This study develops a grey theory-based multi-objective optimization method to resolve these issues, significantly enhancing equipment efficiency and reducing costs for EV motor manufacturing.
1. Background and Challenges
1.1 EV Motor Core Losses
Iron cores constitute the dominant source of energy loss (>50%) in EV motors. Conventional cores use thousands of non-oriented electrical steel sheets stacked via welding/riveting, leading to:
- Reduced effective core area
- Increased eddy current paths and iron loss
- Lower power factor and torque output
- Incompatibility with high-power-density EV motor designs.
1.2 Adhesive Bonding Technology
Adhesive bonding involves spraying adhesive onto steel sheets, compressing them axially, and curing at room temperature. Benefits include:
- Tight lamination (thin, uniform adhesive layers)
- Insulation blocking eddy currents between sheets
- 15–30% lower core losses vs. traditional methods.
Table 1: Performance Comparison of Lamination Methods
| Method | Eddy Current Loss | Stacking Density | Production Speed (spm) |
| Welding/Riveting | High | Low-Medium | 200–300 |
| Adhesive Bonding | Low | High | 150–200 |
However, adhesive core production efficiency remains 25–40% lower due to uncoordinated process rhythms and empirical (“trial-and-error”) drive system selection.
2. Rotary Lamination Mechanism and Drive System
2.1 Mechanism Design
The rotary lamination mechanism integrates punching, adhesive dispensing, and stacking in a progressive die (Figure 1). Key functions:
- Rotating electrical steel sheets to precise angular positions
- Axial compression for adhesive bonding.
2.2 Drive System Selection
Servo motors drive the mechanism via transmission systems. Synchronous belt transmission is preferred for:
- Zero backlash
- Low cost
- High efficiency (η₁ = 0.98).
Gear reducers (e.g., PLN115 series) augment torque but reduce efficiency (η₂ = 0.98) and increase cost.
2.3 Motion Profiles
Three motion modes were evaluated:
- Triangular: High acceleration, excessive mechanical shock (discarded)
- Rectangular: Constant max speed, unrealistic inertia (discarded)
- Trapezoidal: Acceleration → Constant speed → Deceleration (adopted).
Equation: Trapezoidal Motion Time
t = t_1 + t_2 + t_3
where:
- t1t1: Acceleration time (s)
- t2t2: Steady-state time (s)
- t3t3: Deceleration time (s), t1=t3t1=t3.
3. Grey Theory-Based Multi-Objective Optimization
3.1 Optimization Workflow
\begin{aligned}
&\text{1. Define load parameters (Table 2)} \\
&\text{2. Calculate load inertia } J_L \\
&\text{3. Determine servo motor speed } (n_2), \text{ transmission ratio } (i), \text{ acceleration time } (t_1) \\
&\text{4. Compute } T_{max}, P_{max}, T_{rms} \\
&\text{5. Apply grey relational analysis for optimal combination}
\end{aligned}
Table 2: Key Load Parameters for EV Motor Core Processing
| Parameter | Symbol | Value |
|---|---|---|
| Strokes per minute (spm) | mm | 180 |
| Rotation angle | αα | π/3π/3 rad |
| Min. transmission ratio | iminimin | 1 |
| Max. transmission ratio | imaximax | 3 |
| Time coefficient | ff | 4 |
3.2 Load Inertia Calculation
Load inertia JLJL is critical for servo motor selection:
J_L = k_j \left( \frac{J_r}{\eta i^2} + J_f \right)
where:
- JrJr: Mechanism inertia (e.g., 0.3944 kg·m²)
- JfJf: Reducer inertia (2.5×10⁻⁴ kg·m² if used)
- kjkj: Inertia amplification factor (1.1).
Inertia matching principle: For optimal dynamic response, JL/JM<3JL/JM<3 (ideally ≈1), where JMJM is motor rotor inertia.
3.3 Servo Motor Parameter Calculation
- Peak torque:
T_{a2} = \left( 2k_j \frac{J_r}{\eta i^2} + 2J_f \right) \frac{\pi n_2}{30t_1}
- RMS torque (for thermal validation):
T_{rms} = \sqrt{\frac{2T_{a2}^2 t_1}{t_0}}
- Peak power:
P = \frac{T_{a2} n_2}{9550}
3.4 Grey Relational Analysis
847 parameter combinations were evaluated. Grey theory normalized data and computed relational grades:
- Data normalization:
y_j(g) = \frac{x_j(g) - \min x_j}{\max x_j - \min x_j}
- Grey relational coefficient:
\gamma_j = \frac{\min \Delta_j + 0.5 \max \Delta_j}{\Delta_j(g) + 0.5 \max \Delta_j}
- Grey relational grade (higher = better):
\gamma = \frac{1}{n} \sum_{j=1}^{n} \gamma_j \quad (n=3 \text{ objectives})
Table 3: Optimal Parameters for EV Motor Core Production
| Parameter | Value |
|---|---|
| Motor speed n2n2 | 560 rpm |
| Transmission ratio ii | 3 |
| Acceleration time t1t1 | 30 ms |
| TmaxTmax | 193.1 N·m |
| TrmsTrms | 81.9 N·m |
| PmaxPmax | 11.3 kW |
4. Validation and Implementation
4.1 Motor Selection and Testing
A 15-kW servo motor (15KLM(B)) was selected:
- , → JL/JM=1.57<3JL/JM=1.57<3
- (motor capability)
- (rated torque).
4.2 Efficiency Gains
Optimization enabled:
- Increased strokes/minute: 180 → 200 spm
- 11.1% higher productivity
- Validated for stator cores (Table 4).
Table 4: Performance in EV Stator Core Production
| Stator Model | Optimal n2n2 (rpm) | i1/i2i1/i2 | Efficiency Gain |
|---|---|---|---|
| 1 | 2,788 | 2.5 / 7 | 10.2% |
| 2 | 2,350 | 1.8 / 5 | 9.7% |
4.3 Adhesive Core Quality
Cores met all EV motor specifications:
- Dimensional accuracy: ±0.05 mm
- Glue distribution: Uniform, no overflow
- Core loss reduction: 22% vs. welded cores.
Table 5: Adhesive Core Quality Metrics (EV Rotor)
| Parameter | Nominal (mm) | Avg. (mm) | Std. Dev. | Pass? |
| Outer diameter | 149.0 | 149.0401 | 0.01345 | Yes |
| Thickness | 60.0 | 60.3354 | 0.14054 | Yes |
| Parallelism | ≤0.05 | 0.0250 | 0.01186 | Yes |
5. Software Implementation
A MATLAB-based tool automates servo motor selection:
- Inputs: m,α,Jr,ηm,α,Jr,η
- Outputs: Optimal n2,i,t1n2,i,t1, motor specifications.
This reduces selection time from 2–3 days to <10 minutes.
6. Conclusion
- Grey theory optimization resolved drive system inefficiencies in EV adhesive core production, boosting efficiency by 11.1% and reducing costs.
- Servo motor selection software standardized parameter calibration, eliminating trial-and-error methods.
- Validated core quality confirms suitability for high-power-density EV motors.
This methodology enhances the competitiveness of electric vehicle motor manufacturing by addressing a critical bottleneck in adhesive core production. Future work will extend optimization to multi-axis synchronization for higher stroke rates.