The ability to accurately predict and prevent power fluctuations is of considerable importance to solar PV (photovoltaic) plant operators in terms of sustaining profitability, estimating revenue returns and ensuring customer quality of service. Variations in solar irradiance can cause rapid fluctuations in power generation, reducing the quality and reliability of the power generated by large grid-connected PV plants.
Such inconsistency in solar irradiance can be caused by a number of factors, including shadowing as a result of cloud cover and dust gathering on PV panels. For intervals of less than ten minutes, these fluctuations are directly absorbed by PV electricity systems which results in variations in power frequency. Utility operators are powerless to correct these imbalances which can ultimately result in electrical power systems failure.
Battery Storage Systems
Typically, PV power fluctuations are counter-balanced by the use of battery storage systems within conventional PV power plants. However, battery systems not only increase the size and cost of PV power systems but power fluctuations also have the effect of reducing the useful life-time of the battery storage system.
Simple Moving Average Method
A new and effective technique combining a PV energy storage system with a novel smoothing strategy known as the Single Moving Average (SMA) has been proven to not only reduce PV power fluctuations but also optimise battery storage systems by reducing their usage within PV power plants.
This approach calculates the SMA of past PV energy production over a certain time period — the longer the running average, the smoother the PV Power fluctuations. The key advantage of this technique is that the battery state of charge (SOC) will always return to its initial condition and therefore does not need a specific control. This means decisions do not need to be made about when to conserve power to avoid shortages, but more importantly it helps avoid damage to the battery bank, prolonging the useful life of the battery.
It also calculates the SMA even on clear days when irradiance is not a factor in order to measure a longer and more complete operating period. This is done to ensure that the maximum possible power fluctuation is below PV plant operator requirements. Combined with an empirical technique, as discussed in the next section, this process enables the prediction of PV power fluctuation on a day-ahead basis. The use of the battery storage system can therefore be fully optimised, minimising storage requirements.
Optimising Battery Storage
The SMA technique has been evaluated over a twelve-month period on 5-s registered power output from a 1.1-MW PV plant operated by Acciona Energy in Tuleda, Spain. Given the normal operating characteristics of the plant, the energy storage requirement for smoothing maximum power fluctuations below 2.5 percent per minute was calculated. Simulated over an annual period, it was determined that 2,400 seconds of storage per day was required in order to reduce the power fluctuations. This level of power storage was to be applied every day regardless of weather conditions in order to ensure the required level of power fluctuation smoothing while optimising battery storage requirements. In terms of battery storage capacity, this translates into 312 kWh per annum.
Acciona Energy PV Plant in Tudela, Spain where the Single Moving Average (SMA) Technique was evaluated. Images are subject to copyright.
Mitigating Predictive Uncertainty
The degree of predictive uncertainty was also measured and was found to have a positive correlation with the minimum daily energy storage required to reduce PV power fluctuations below 2.5 percent per minute over this period. This means that the storage time required could be accurately scheduled one day in advance to minimise the use of the energy storage system. In extraordinary situations, where greater than predicted PV fluctuations are unavoidable, a ramp rate limiter can be used to smooth power fluctuations.
The SMA smoothing technique has been simulated over a one-year period and has proven to be an effective means of decreasing power fluctuations in large PV plants. Importantly, its predictive accuracy has demonstrated an ability to optimise the usage of energy storage systems over a prolonged time period. During its evaluation period, the technique was shown to reduce average PV energy storage requirements by more than 15 percent, requiring less than 263 kWh capacity of an average PV battery storage system per annum.