This example is intended to show relevant modeling and simulation capabilities of SystemVision Cloud, for kinetic energy harvesting (EH) systems. It is not necessarily a practical EH design itself, but rather demonstrates the tool's ability to support knowledgeable users who are creating practical designs.

The design contains mechanical, magnetic and electronic circuit elements, with energy conservation and cross-discipline dynamic interactions automatically included in the system model. The user can directly specify the physical or behavioral characteristics of many of the components. This includes the mass of the armature, the stiffness of the resonant spring, the cross sectional area of the magnetic core, the number of winding turns, the resistor and capacitor values, as well as the drop-out voltage of the linear regulator.

In the nominal simulation results displayed on the schematic, the upper right waveform viewer shows the initial start-up of the system, with the amplitude of the external vibration source of 0.07 mm peak at 60 Hz, equivalent to a peak acceleration of 1 g. The nominal armature spring-mass resonance frequency is 60 Hz, and the armature displacement is seen to reach the frame's travel limit of 10 mm peak-to-peak!

In the upper left, the waveform viewer is zoomed-in near the 1 sec. simulation time mark. It shows the time-varying core flux density and winding voltage, as the two air-gaps expand and contract with armature displacement, rapidly changing the magnetic flux path's effective reluctance value. In the lower right waveform viewer, the DC output voltage from the Schottky diode full-wave rectifier and the linear regulator output voltage are observed. The periodic disturbance is caused by the switched load being applied to the system.

This battery charging example compares direct solar battery charging vs. an MPPT algorithm combined with a buck converter.

The solar panel model in this example allows the user to specify not only the "boilerplate" electrical characteristics (i.e. the open circuit voltage, short circuit current and peak power output capability), which are always given at the full or nominal irradiance level, but also to specify the reduced output and shift in the peak power point at lower irradiance levels. This shift in the load voltage at which peak power transfer occurs gives rise to the performance improvement that can be achieved with MPPT (maximum power point tracking) when the irradiance level varies over time.

The "electronics" section of this design contains a few passive analog circuit elements, but consists mainly of abstract "math block" models. These are used to represent the state-average (non-switching) behavior of the converter. The sampled-data MPPT algorithm dynamically adjusts the buck duty-cycle, to keep the solar panel operating at its peak power output. The user can adjust various "tuning" parameters of that algorithm (e.g. sample rate, the duty-cycle "delta" or perturbation used for tracking, etc.). The simple algorithm is visible in the open-source "MPPT-Solar" model, just right click and select "View/Copy Model".

The simulation results show clear improvement in both panel power output (magenta vs. light blue waveforms) and battery input or charging current (red vs. dark blue waveforms), for MPPT vs. direct charging, respectively. Also, the actual duty-cycle "hunting" or peak power tracking operation is visible in green waveform, as the irradiance level varies sinusoidally (brown waveform).

A companion design is available, which shows a circuit implementation of the buck converter power stage. That design can be used to analyze the efficiency of the MPPT approach. Efficiency can be a key factor in the trade-off assessment vs. simple direct charging. That circuit design can be found here: https://www.systemvision.com/design/solar-energy-harvesting-implementation-mppt-battery-charging

This example is intended to show relevant modeling and simulation capabilities of SystemVision Cloud for Electrothermal Energy Harvesting (EH) systems. It is not necessarily a practical EH design itself, but rather demonstrates the tool's ability to support knowledgeable users who are creating practical designs. The example also illustrates using a sampled-data algorithm for maximum power point tracking (MPPT), to optimize the energy harvest for changing operating temperatures.

The design includes a thermoelectric generator (TEG) that is supplied on the "hot" side by a sinusoidally time varying temperature between 75 degC and 100 degC. The "cold" side is held at a fixed 25 degC. The thermal resistance and heat capacitance of the hot-side heat-sink are shown in the schematic. The electronics section includes a mix of analog circuit elements, including an inductor, 1.0 F super-capacitor, LDO regulator and a periodically switched load resistor. It also includes abstract or "math block" models to represent the state-average (non-switching) behavior of a buck-boost converter.

The goal of the design is to extract sufficient power from the TEG, to provide a 2.5-Watt/1-second duration power burst once every 10 seconds. This burst is presumably to supply power for a periodic data transmission. The simple MPPT algorithm that helps achieves this is visible in the open-source MPPT-TEG model shown. The MPPT algorithm dynamically adjusts the load current draw from the TEG, to keep it operating at its maximum power output capability. That capability varies with the differential operating temperature. That shift can more easily be seen in the followingTEC/TEG calibration test schematic:

https://www.systemvision.com/design/calibrate-tecteg-energy-harvesting

This example is intended to show relevant modeling and simulation capabilities of SystemVision Cloud, for kinetic energy harvesting (EH) systems. It is not necessarily a practical EH design itself, but rather demonstrates the tool's ability to support knowledgeable users who are creating practical designs.

The design contains mechanical, magnetic and electronic circuit elements, with energy conservation and cross-discipline dynamic interactions automatically included in the system model. The user can directly specify the physical or behavioral characteristics of many of the components. This includes the mass of the armature, the stiffness of the resonant spring, the cross sectional area of the magnetic core, the number of winding turns, the resistor and capacitor values, as well as the drop-out voltage of the linear regulator.

In the nominal simulation results displayed on the schematic, the upper right waveform viewer shows the initial start-up of the system, with the amplitude of the external vibration source of 0.07 mm peak at 60 Hz, equivalent to a peak acceleration of 1 g. The nominal armature spring-mass resonance frequency is 60 Hz, and the armature displacement is seen to reach the frame's travel limit of 10 mm peak-to-peak!

In the upper left, the waveform viewer is zoomed-in near the 1 sec. simulation time mark. It shows the time-varying core flux density and winding voltage, as the two air-gaps expand and contract with armature displacement, rapidly changing the magnetic flux path's effective reluctance value. In the lower right waveform viewer, the DC output voltage from the Schottky diode full-wave rectifier and the linear regulator output voltage are observed. The periodic disturbance is caused by the switched load being applied to the system.

This circuit is an implementation of a buck converter power stage that could be used in the MPPT solar battery charger reference design found here:

https://explore.partquest.com/node/661856

That reference example is appropriate for "system level" trade-off assessment, where the MPPT (maximum power point tracking) approach can be compared with direct solar panel battery charging. It uses a state-average (non-switching) buck converter model for much faster simulation, so that power tracking performance over long time periods can be observed. However, it does not account for some important aspects of a real converter implementation, such as the predicted power loss in the switching components, or the switching noise on the current and voltage measurements that are needed for power detection.

This circuit confirms the performance of the system at full solar panel irradiance, and with the converter duty-cycle set to a fixed value of 0.8. That value was shown to provide maximum power under full irradiance. The results with this design confirm those of the reference design: The steady-state panel power output is just under 60 Watts (light blue waveform); The corresponding voltage (16.7 V) and current (3.58 A) into the converter (magenta and green waveforms, respectively); The current into the battery is 4.45 A (red waveform), as expected. Note also that the input L-C filter removes most of the switching noise that would be visible on these signals without that filtering effect.

The average power loss in the P-channel MOSFET is just under 1.8 Watts (brown waveform). This is important for assessing overall system efficiency, a key factor in the trade-off analysis with direct charging. The user can move the waveform probes around and observe the power loss in all of the other components too, as part of the efficiency analysis process.

This battery charging example compares direct solar battery charging vs. an MPPT algorithm combined with a buck converter.

The solar panel model in this example allows the user to specify not only the "boilerplate" electrical characteristics (i.e. the open circuit voltage, short circuit current and peak power output capability), which are always given at the full or nominal irradiance level, but also to specify the reduced output and shift in the peak power point at lower irradiance levels. This shift in the load voltage at which peak power transfer occurs gives rise to the performance improvement that can be achieved with MPPT (maximum power point tracking) when the irradiance level varies over time.

The "electronics" section of this design contains a few passive analog circuit elements, but consists mainly of abstract "math block" models. These are used to represent the state-average (non-switching) behavior of the converter. The sampled-data MPPT algorithm dynamically adjusts the buck duty-cycle, to keep the solar panel operating at its peak power output. The user can adjust various "tuning" parameters of that algorithm (e.g. sample rate, the duty-cycle "delta" or perturbation used for tracking, etc.). The simple algorithm is visible in the open-source "MPPT-Solar" model, just right click and select "View/Copy Model".

The simulation results show clear improvement in both panel power output (magenta vs. light blue waveforms) and battery input or charging current (red vs. dark blue waveforms), for MPPT vs. direct charging, respectively. Also, the actual duty-cycle "hunting" or peak power tracking operation is visible in green waveform, as the irradiance level varies sinusoidally (brown waveform).

A companion design is available, which shows a circuit implementation of the buck converter power stage. That design can be used to analyze the efficiency of the MPPT approach. Efficiency can be a key factor in the trade-off assessment vs. simple direct charging. That circuit design can be found here: https://www.systemvision.com/design/solar-energy-harvesting-implementation-mppt-battery-charging

This battery charging example compares direct solar battery charging vs. an MPPT algorithm combined with a buck converter.

The solar panel model in this example allows the user to specify not only the "boilerplate" electrical characteristics (i.e. the open circuit voltage, short circuit current and peak power output capability), which are always given at the full or nominal irradiance level, but also to specify the reduced output and shift in the peak power point at lower irradiance levels. This shift in the load voltage at which peak power transfer occurs gives rise to the performance improvement that can be achieved with MPPT (maximum power point tracking) when the irradiance level varies over time.

The "electronics" section of this design contains a few passive analog circuit elements, but consists mainly of abstract "math block" models. These are used to represent the state-average (non-switching) behavior of the converter. The sampled-data MPPT algorithm dynamically adjusts the buck duty-cycle, to keep the solar panel operating at its peak power output. The user can adjust various "tuning" parameters of that algorithm (e.g. sample rate, the duty-cycle "delta" or perturbation used for tracking, etc.). The simple algorithm is visible in the open-source "MPPT-Solar" model, just right click and select "View/Copy Model".

The simulation results show clear improvement in both panel power output (magenta vs. light blue waveforms) and battery input or charging current (red vs. dark blue waveforms), for MPPT vs. direct charging, respectively. Also, the actual duty-cycle "hunting" or peak power tracking operation is visible in green waveform, as the irradiance level varies sinusoidally (brown waveform).

A companion design is available, which shows a circuit implementation of the buck converter power stage. That design can be used to analyze the efficiency of the MPPT approach. Efficiency can be a key factor in the trade-off assessment vs. simple direct charging. That circuit design can be found here: https://www.systemvision.com/design/solar-energy-harvesting-implementation-mppt-battery-charging