Design and Optimisation of a Microgrid for Improved Efficiency of the Volta River Authority (Navrongo) Solar Power Plant

In ensuring proper energy mix and reducing the number of emissions from traditional thermal plants for power generation, the Energy Commission of Ghana built the Solar Power Plant at Navrongo. This was to help cut down the cost of crude fuel imports and also play a part in mitigating global warming which results from the continuous emission of carbon dioxide (CO2) at the thermal power. Over the years the plant has been faced with inconsistent power generation. This research paper sought to investigate the power losses at the Navrongo Volta River Authority (VRA) Solar Power Plant and come out with measures to improve its efficiency. Power production downtime and power transmission losses were identified as the major constraints of the solar power plant. General evaluation and review of the grid design and transmission system of the plant were considered and a microgrid technology was proposed to eliminate generation downtime and power transmission losses. The output of the proposed microgrid system was predicted using R-studio statistical simulations, also the plant was optimised to ascertain the gains in power generation and the merits of this system were discussed. Finally, conclusions and recommendations where made to ensure energy security and economic competitiveness of the plant.


Introduction
The Energy Commission of Ghana in ensuring proper energy mix and reducing the number of emissions from traditional thermal plants for power generation, built the Solar Power Plant at Navrongo. This was to aide Ghana cut down the cost of crude fuel imports and also play a part in mitigating global warming which results from the continuous emission of carbon dioxide (CO 2 ) at the thermal power generating stations in the country [1].
It has come to the notice of the management of the plant that there is inconsistent power generation from the plant, this is because the DC inverters are power sensitive, and they depend on the Navrongo-Pungu grid lines for power supply. Again there are transmission losses as Power is generated at Pungu and transmitted to the Navrongo Substation and back to Pungu community for usage. Furthermore, module cracking and high cell temperatures have been also identified as contributing factors to the losses. Consequently, there is a need to improve the power output of the installed solar plants and eliminate the associated power losses.
The Navrongo VRA Solar Plant has an installed capacity of 2.6 MW. However, the actual output is averaged to be 1.5 MW. Subsequently, the output power deficit stands at 1.1 MW in terms of installed capacity. This loss is significant in terms of power supply to the people of Ghana, and when translated into money is about GH¢ 9,367.74 (USD 1622.46) loss of income per day. To avert all these, this paper analyses the power generation process, performance indexes and ways of improving power efficiency at the solar plant.

Materials and Procedures Used
Prefeasibility study was carried out using the master-writing picture: 18 th Edition and the multi-functional power meters to measure the amount of energy generated per solar day, amount of energy loss, the energy delivered by the grid system and average energy consumed at the site. Secondary data of the performance of the plant was also obtained from the site for diagnostic analysis. A review of the design criteria and evaluation of the existing grid design and layout of the Solar Power plant was done ( Figure 1).  Figure 1 depicts the circuit diagram of the Navrongo VRA Solar Power Plant which is made up of 120 arrays, 32 combiner boxes and 5 D. C inverters. The combiner boxes transfer a composite voltage between 700 to 900 volts of DC to the inverters which transform it into 415 volts of AC. The AC transformer receives the 415 volts from the inverters and steps it up to 34.5 kV. Power from the transformer unit is delivered into the Pungu line for transmission into the Navrongo Substation which also receives power from Bolgatanga. The Navrongo Substation supplies energy to the Navrongo town, Burkina Faso, Tumu, Sandema, Tono and Spare for consumption. Again, power is transmitted back to Pungu township for consumption and this to and fro of power transmission between Pungu and Navrongo increases transmission power losses coupled with the grid-power sensitive nature of the D. C inverters which gets the plant shut down whenever the grid is off. It is imperative that a solution be found to improve efficiency of the solar power plant and supply of power to the Pungu Community.

Proposed Microgrid Design and Description
The losses encountered during transmission and the power inconsistencies during grid decoupling or grid-tying conditions has resulted in the need for design of a microgrid system for continuous power supply.
The advantage of a Microgrid is that it operates autonomously within the grid power network system. Power consumers derive a steady and dependable source of energy as against being tied to the grid for power supply [2]. Generating power by the microgrid technology system and small-scale renewable energy resources improves energy efficiency and also has environmental advantages over the large-scale grid generation system [3].
The Microgrid technology further strengthens the efforts of the concept of the smart grid for locally produced power and focuses on power supplies which satisfy the needs of the consumers served [4] Moreover, the Microgrid system power generation is not affected by downtime, transmission losses and power rationing. Microgrid power generation is unique and allows consumers the choice to generate power based on their needs [5].
Consequently, the design of a microgrid was proposed which is depicted in Figure 2. The proposed Microgrid for Pungu community is composed of a group of interconnected loads and power supply resource in a clearly defined electrical network that acts as a single regulated unit with respect to the grid. The Microgrid can be connected and disconnected from the grid relative to the rotation of solar day and night [6].

Description of Design
The circuit diagram in Figure 2 consists of the PV arrays wired into combiner boxes that deliver a DC voltage into the DC inverters. An AC circuit breaker is stationed between the DC inverter and the multi-purpose transformer. There are two lightning arresters connected at the output of the inverters by a circuit breaker and also at the output of the transformer. Two earthing wires are each fixed at the two output paths. The potential transformer maintains the value of the power to be transmitted into the Pungu community for consumption.
A start-up power generator is incorporated to initiate the inverters into action during generation time. The automatic change over switch makes it possible for the pungu community to still draw power from the national grid when the solar plant is not generating energy to maintain continuity of power supply. The power consumption of the Pungu community is measured to be 2.8 MW.
However, the plant has an installed capacity of 2.6 MW and generates 1.5 MW which is 1.1 MW less than the power requirements of the community. Therefore, there is the need to: Segregate consumers with low energy consumption and give them solar panels; and Design a mircogrid for the community to ensure that whatever amount of power generated is utilised by communities. This will curb energy losses due to transmission and downtime.

Performance Analysis of Power Output of the Microgrid
It was determined from data at the plant that, about 21% of the energy loss is attributable to production downtime, and also 26% by transmission losses as similar findings by Asmus and Stimmel [7].
In the old system, when generation downtime occurs the PV arrays continue to generate between 700 V to 900 V of DC power, which goes waste at the solar cells because the inverters stop procession for lack of power in the grid transmission lines whilst, the microgrid does not depend on power in the grid. The R-studio statistical software was used to analyse the proposed system and the results yielded.
In deed to predict the effectiveness of the microgrid, simulations were done using R-studio and about 47% loss via transmission and downtime in the current setup was eliminated. This step was essential because for a standalone PV system the losses due to transmission and downtime are void since the energy generated would be the exact energy consumed. The grid connected power output readings were recorded alongside the simulated output of the microgrid, to help evaluate the gains of the microgrid system over the existing solar power plant.
The percentage of power gained by the microgrid technology is determined as: The theory of solar cells explains the process by which light energy in photons is converted into electric current when photons strike a suitable semiconductor device [8].
The theory foretells the functional limits of PV cell, and offer guidance on the constraints that lead to losses and also indicates the solar cell efficiency [9]. Mismatch losses arise if the PV modules in use have different physical and chemical compositions. But are wired together in series and in parallel. [10].
It should be stated that, the electrical output from a single cell is small, so multiple cells are connected and encapsulated to form a module [11]. Therefore, the selection of modules is important in the overall performance of the plant and it involves the knowledge of the solar cell theory [12]. Usually, the more sunlight hitting the modules, the more current they will produce [13].
The temperature of PV cells is one of the most important variables used in assessing the performance of PV systems and their electrical energy production [14]. This temperature depends on parameters such as the thermal properties of materials used in PV module encapsulation, type of PV cells, configuration of PV modules' installation and local climatic conditions which are governed by the theory of solar cells [15].

PV Optimisation Model Equation, Resuts and Dicussuion
Microgrid has been designed for the plant, however, for improved efficiency of the system, there is the need to look at some parameters that influence the magnitude of power generation and how to optimise it.
Consequently, a mathematical model was developed to determine the optimum power that can be generated taking cognizance of the following constraints; general efficiency, cell efficiency, dust deposition, cell and ambient temperatures, peak sun hours, irradiance and total PV module area.
The objective function of the mathematical optimisation can be expressed as: Constraints C : X ≤ 67 OR X ≤ b From Figure 3, the month of June had the lowest optimised energy generated because it recorded the lowest rang of irradiance of 872 Wh/m 3 to 78 Wh/m 3 as indicated in Table 1.     From Figure 4, at a cell temperature of 50.3°C and a minimum ambient temperature of 32.0°C, an optimised energy of 512,890.5 kWh was estimated which occurred in July. Thus, maximum power is produced in July at cell temperature of 52.7°C with a peak irradiance of 1127 Wh/m 2 . Also, February had a lower energy generated and optimized energy readings than January even though the constraint values were almost the same. This was due to the fact that there was dust accumulation form November 2013 to February 2014 due to the Harmattan climate condition and cleaning was done only in March 2014.
It is observed that in Figure 5, the highest power was obtained in January 2015 with a cell temperature of 53°C and a maximum irradiance of 956 Wh/m 2 .   Again, in Figure 6 optimised energy of 594561.0 kWh in October which is the highest, with cell temperature of 48.9°C and ambient temperature of 32.7°C. Furthermore, from Tables 1 to 4, the optimised Power values for each month exceeds the nominal Power generated which depicts a shortfall in the existing power generation system.
To determine the threshold of cell temperatures that resulted in the maximum optimised energy, the optimised energy, and the cell temperatures were stimulated as indicated in   It could be observed from figure 8 that for the maximum optimized energy to be achieved, the cell temperatures should range from 48.1°C to 53°C.
The plant produces maximum power at peak sun hours with an average cell temperature of 50.8°C, approximated to 51°C.
The study could, therefore, conclude that there is a combination of threshold cell temperature and ambient temperature for which maximum power is delivered. As the cell temperature decreases from 53°C, the power generated increases until a minimum temperature of 48°C is reached within which power generation starts decreasing.
Percentage increase after optimisation could be expressed as: The optimised power value difference (O. V. D) is also given as: Table 5 shows the computed optimised values.

Verification of Percentage Increase of Power Generation
The strength of the PV optimisation module equation was tested further using Rstudio as an alternative method of simulation besides Microsoft excel. The system produced an average boost of 29.8% of the power output of the solar plant at peak sun hours as compared to 30.4% boost by the Microsoft excel simulation of the module.
There is therefore convergence in terms percentage power gained, using the two statistical tools in the module simulation.

Conclusions
Results of the Performance Analysis of the Navrongo VRA solar Plant shows that the major factors associated with power losses at the plant are generation downtime, transmission inefficiencies, module cracking and high cell temperatures.
A new microgrid service system has been designed which will eliminate downtime and power transmission losses.
It was determined from Microsoft Excel and Rstudio that maintaining an optimum cell temperature of 51.0°C will boost power output by an average of 30.1% at peak sun hours and also prevent PV module cracking. From the optimisation, power output will increase on average by 1,285,191.4 kWh per year which translates to GH¢ 965,178.74 (USD 165,789.69) per annum.

Recommendations
The following are the recommendations from the study: Micro-grid design technology should be used in place of the grid transmission system in future investments in solar energy in order to prevent transmission losses and production downtime; Automatic mist blower systems should be integrated into the solar power plant to maintain the threshold cell temperature of 51°C for optimum cell delivery; Solar trackers are recommended for receiving maximum radiation at any point in time for optimum power delivery of the plant; and further studies should be carried out on the rate of dust deposition on the PV modules.