Permanent magnet synchronous motor (PMSM) is widely used in home appliance applications due to its high power density, high efficiency, and wide speed range. The PMSM includes two major types: the surface-mounted PMSM (SPM), and the interior PMSM (IPM). SPM motors are easier to control due to their linear relationship between the torque and q-axis current. However, the IPMSM has electromagnetic and reluctance torques due to a large saliency ratio. The total torque is non-linear with respect to the rotor angle. As a result, the MTPA technique can be used for IPM motors to optimize torque generation in the constant torque region. The aim of the field weakening control is to optimize to reach the highest power and efficiency of a PMSM drive. Field weakening control can enable a motor operation over its base speed, expanding its operating limits to reach speeds higher than rated speed and allow optimal control across the entire speed and voltage range.

The voltage equations of the mathematical model of an IPMSM can be described in d-q coordinates as shown in Equation 66 and Equation 67.

The dynamic equivalent circuit of an IPM synchronous motor is shown in Figure 2-22.

The total electromagnetic torque generated by the IPMSM can be expressed as Equation 68 that the produced torque is composed of two distinct terms. The first term corresponds to the mutual reaction torque occurring between torque current $i_q$ and the permanent magnet ${\psi }_m$, while the second term corresponds to the reluctance torque due to the differences in d-axis and q-axis inductance.

In most applications, IPMSM drives have speed and torque constraints, mainly due to inverter or motor rating currents and available DC link voltage limitations respectively. These constraints can be expressed with the mathematical equations Equation 69 and Equation 70.

Where ${\mathrm{V}}_{\mathrm{m}\mathrm{a}\mathrm{x}}$and ${\mathrm{I}}_{\mathrm{m}\mathrm{a}\mathrm{x}}$ are the maximum allowable voltage and current of the inverter or motor. In a two-level three-phase Voltage Source Inverter (VSI) fed machine, the maximum achievable phase voltage is limited by the DC link voltage and the PWM strategy. The maximum voltage is limited to the value as shown in Equation 71 if Space Vector Modulation (SVPWM) is adopted.

Usually the stator resistance $R_s$ is negligible at high speed operation and the derivate of the currents is zero in steady state, thus Equation 72 is obtained as shown.

The current limitation of Equation 69 produces a circle of radius ${\mathrm{I}}_{\mathrm{m}\mathrm{a}\mathrm{x}}$ in the d-q plane, and the voltage limitation of Equation 71 produces an ellipse whose radius ${\mathrm{V}}_{\mathrm{m}\mathrm{a}\mathrm{x}}$ decreases as speed increases. The resultant d-q plane current vector must be controlled to obey the current and voltage constraints simultaneously. According to these constraints, three operation regions for the IPMSM can be distinguished as shown in Figure 2-23.

1. Constant Torque Region: MTPA can be implemented in this operation region to ensure maximum torque generation.
2. Constant Power Region: Field weakening control must be employed and the torque capacity is reduced as the current constraint is reached.
3. Constant Voltage Region: In this operation region, deep field weakening control keeps a constant stator voltage to maximize the torque generation.

In the constant torque region, according to Equation 68, the total torque of an IPMSM includes the electromagnetic torque from the magnet flux linkage and the reluctance torque from the saliency between $L_d$ and $L_q$. The electromagnetic torque is proportional to the q-axis current $i_q$, and the reluctance torque is proportional to the multiplication of the d-axis current $i_d$, the q-axis current $i_q$, and the difference between $L_d$ and $L_q$.

Conventional vector control systems of a SPM motors only utilizes electromagnetic torque by setting the commanded $i_d$ to zero for non-field weakening modes. But an IPMSM will utilize the reluctance torque of the motor, d-axis current should be controlled as well. The aim of the MTPA control is to calculate the reference currents $i_d$ and $i_q$ to maximize the ratio between produced electromagnetic torque and reluctance torque. The relationship between $i_d$ and $i_q$, and the vectorial sum of the stator current ${\mathrm{I}}_{\mathrm{s}}$ is shown in the following equations.

Where $\beta$ is the stator current angle in the synchronous (d-q) reference frame. Equation 68 can be expressed as Equation 76 where ${\mathrm{I}}_{\mathrm{s}}$ substituted for $i_d$ and $i_q$.

Equation 76 shows that motor torque depends on the angle of the stator current vector; as such

, the maximum efficiency point can be calculated when the motor torque differential is equal to zero. The MTPA point can be found when this differential, $\frac{\mathrm{d}{\mathrm{T}}_{\mathrm{e}}}{\mathrm{d}\mathrm{\beta }}$is zero as given in Equation 77.

Following, the current angle of the MTPA control can be derived as in Equation 78.

Thus, the effective d-axis and q-axis reference currents can be expressed by Equation 79 and Equation 80 using the current angle of the MTPA control.

However, as shown in Equation 78, the angle of the MTPA control, ${\beta }_{mtpa}$ is related to d-axis and q-axis inductance. This means that the variation of inductance will impede the ability to find the optimal MTPA point. To improve the efficiency of a motor drive, the d-axis and q-axis inductance should be estimated online, but the parameters $L_d$ and $L_q$ are not easily measured online and are influenced by saturation effects. A robust Look-Up Table (LUT) method ensures controllability under electrical parameter variations. Usually, to simplify the mathematical model, the coupling effect between d-axis and q-axis inductance can be neglected. Thus, assumes that $L_d$ changes with $i_d$ only, and $L_q$ changes with $i_q$ only. Consequently, d- and q-axis inductance can be modeled as a function of their d-q currents respectively, as shown in Equation 81 and Equation 82.

To reduce the ISR calculation burden by simplifying Equation 78. The motor-parameter-based constant, $K_{mtpa}$ is expressed instead as Equation 84, where $K_{mtpa}$ is computed in the background loop using the updated $L_d$ and $L_q$.

A second intermediate variable, $G_{mtpa}$ described in Equation 85, is defined to further simplify the calculation. Using $G_{mtpa}$, the angle of the MTPA control, ${\beta }_{mtpa}$ can be calculated as Equation 86. These two calculations are performed in the ISR to achieve a real current angle ${\beta }_{mtpa}$.

In all cases, the magnetic flux can be weakened to extend the achievable speed range by acting on the direct axis current $i_d$. As a consequence of entering this constant power operating region, field weakening control is chosen instead of the MTPA control used in constant power and voltage regions. Since the maximum inverter voltage is limited, PMSM motors cannot operate in such speed regions where the back-electromotive force, almost proportional to the permanent magnet field and motor speed, is higher than the maximum output voltage of the inverter. The direct control of magnet flux is not an option in PM motors. However, the air gap flux can be weakened by the demagnetizing effect due to the d-axis armature reaction by adding a negative $i_d$. Considering the voltage and current constraints, the armature current and the terminal voltage are limited as Equation 69 and Equation 70. The inverter input voltage (DC-Link voltage) variation limits the maximum output of the motor. Furthermore, the maximum fundamental motor voltage also depends on the PWM method used. In Equation 72, the IPMSM has two factors: one is a permanent magnet value and the other is made by inductance and current of flux.

Figure 2-24 shows the typical control structure is used to implement field weakening. ${\beta }_{fw}$ is the output of the field weakening (FW) PI controller and generates the reference $i_d$ and $i_q$. Before the voltage magnitude reaches its limit, the input of the PI controller of FW is always positive and therefore the output is always saturated at 0.

Figure 2-13 and Figure 2-15 show the implementation of FAST or eSMO based FOC block diagram. The block diagrams provide an overview of the FOC system's functions and variables. There are two control modules in the motor drive FOC system: one is MTPA control and the other one is field weakening control. These two modules generate current angle ${\beta }_{mtpa}$ and ${\beta }_{fw}$ respectively based on input parameters as show in Figure 2-25.

The switching control module is used to decide which angle should be applied, and then calculate the reference $i_d$and $i_q$ as shown in Equation 74 and Equation 75. The current angle is chosen as following Equation 87 and Equation 88.

Figure 2-26 is the flowchart that shows the steps required to run InstaSPIN-FOC with FW and MPTA in the main loop and interrupt.