Reducing total harmonic distortion with adaptive filters

• A Malaysian study has compared the techniques used to reduce total harmonic distortion caused by PV systems. The paper, which considers the benefits and drawbacks of the approaches studied, suggests the use of adaptive filters.

Researchers in Malaysia have published a study of total harmonic distortion (THD) reduction techniques in PV systems and claim to have found an holistic and systematic approach.

The paper Advances in reduction of total harmonic distortion in solar photovoltaic systems: A literature review, published in the International Journal of Energy Research, highlights the undesirable effects of THD, which is triggered by variations in solar irradiance and temperature as well as by the use of inverters, a major source of harmonics due to constant switching on and off.

“THD can be generated from low irradiance, fluctuation in the output DC signal of converters, [the] switching strategy of inverters and converters and [the] non-linearity of the consumer’s load,” stated researchers from the school of electrical and electronic engineering at the Universiti Sains Malaysia. The results can include heating the PV system, heating load, malfunctioning electronic equipment, incorrect meter readings, disruption of protective relays and communication interference.

Common approaches

Although THD remains a complex phenomenon, the researchers said one of its evident causes in solar power generation was the connection of numerous non-linear loads in the power system, which has an impact on load characteristics in distribution networks. That, according to the paper, can be easily rectified by demand-side management.

The study explained how at PV module level, techniques such as advanced perturbation and observation-based maximum power point tracking (MPPT) and hill climbing MPPT represent the most common solutions.

“Using this method guarantees steady precision, doubles tracking speed and converts considerable energy into electricity,” the study noted. Such techniques, however, are said to lack accuracy and to lead to misjudgment and oscillation around the maximum power point, which can result in an efficiency decrease of PV systems. MPPT technologies are also known to be unsatisfactory on partially shaded PV arrays.

More modern techniques, incorporating artificial intelligence, adaptation algorithms and machine learning are now being proposed, the study said, but still lack the desired accuracy.

As for power electronics, the research highlights how both DC-DC and DC-AC converters are a major source of harmonics, and how research is focusing on the switching control of the inverter or on using filters to remove harmonics from the power signal.

Filters

The use of active, passive or hybrid filters – which are cheap and easy to install – has been widely adopted for harmonic reduction in PV systems, due to the increase of non-linear loads in power systems. However, warned the researchers: “Numerous filters might also affect system stability because they can push the whole system to the resonance region of the S-plane. If more inductors or capacitors are used, the circuit might become bulky.”

A more valid alternative, according to the literature reviewers, may be the adoption of adaptive filters, which are very complex, self-designing digital devices with time-varying parameters adjusted repeatedly in accordance with input data. “The output signal from the PV system is the distorted input signal for the adaptive filter, and the output from the adaptive filter should be a clean power signal,” stated the study. “These filters have effectively reduced noise and disturbance in other systems.”

Adaptive filters have gained in popularity for grid synchronization in recent years, due their ability to reduce disturbance and extract the fundamental component from a polluted signal under grid frequency variations. Adaptive filtering-based energy management strategies can be applied to smooth renewable power fluctuations for solar and wind energy.