REALTIME STRING FAILURE DETCTION FOR PV PPRJCET WITH ZONE MONITORING

ABSTRACT

?Efficient monitoring of solar photovoltaic (PV) projects is essential for maximizing performance and minimizing revenue loss due to system failures. Traditionally, manual string monitoring using clamp meters has been the norm, requiring significant human resources and prone to errors. Moreover, the average detection time for failures ranges from 15 to 45 days, leading to substantial financial and operational setbacks.

This white paper proposes a real-time string failure detection method utilizing zone monitoring, which has become increasingly popular for its capital expenditure optimization benefits. By integrating advanced monitoring systems, this method slashes the detection time from 45 days to just half an hour, representing an impressive 99% reduction in detection time. The proposed approach not only enhances the efficiency and accuracy of failure detection but also eliminates the need for manual intervention, resulting in an estimated savings of up to 80% in manpower resources.

Through a comprehensive discussion of the methodology and its implementation in solar PV projects, this paper offers a practical solution to revolutionize monitoring practices, ensuring timely identification and mitigation of string failures, thereby enhancing system performance and profitability.

1.??? INTRODUCTION

?As project sizes grow and the number of string combiner boxes increases, the feasibility of installing string monitoring systems diminishes due to higher capital and operational expenditures, as well as the complexity of data communication networks and SCADA system capabilities. Consequently, many project developers are turning to Zone (Combiner Box) monitoring systems within inverters, enabling the monitoring of individual combiner box currents.

However, Zone monitoring introduces its own set of challenges. Each combiner box may possess different DC capacities, and for larger projects, the sheer volume of combiner boxes—often reaching into the thousands, such as approximately 2000 for a 500 MWp project—adds complexity. Moreover, the constantly varying Maximum Power Point Tracking (MPPT) operating points hinder operators' ability to identify string breakdowns effectively.

In the absence of intelligent monitoring systems, the industry has resorted to a manual practice: human operators performing individual string current measurements using clamp meters at regular intervals to detect and address string breakdowns. Yet, this manual method is labour-intensive and time-consuming.

This paper outlines a method for identifying string failures in real-time, streamlining the process and allowing for prompt corrective action by deploying manpower directly to rectify issues as they arise.

2.? SAMPLING PERIOD

To avoid spurious calculation due to limitation on refresh rate of different measuring equipment’s and cancel our transient weather impact, 5-minute average/integrated samples has been used for calculation of this method.

3.? STEPS TO IDETIFY STRING FAILURE

Following procedure to be utilized to identify staring failure.

I.?Calculate 5-minute average power and convert it to energy by multiplying current of each combiner box and MPPT operating voltage of inverter in which particular combiner box is connected. Refer below formula for calculation of Energy.

II.?Calculate Specific Yield of particular combiner box by integrating over the monitoring period (Usually monitoring period could be current day or last one week or last one month).

III. Take Average of Specific yield of combiner box connected with same inverter.

IV.?Calculate specific yield deviation of each individual combiner box specific yield with respect to average calculated in above step-3.

V.?Create alert that whenever Specific yield deviation is negative and more than 80% of least string capacity, it should be considered as string failure. (Fore example if in one combiner box there are 24 strings each of capacity 7 KWp than deviation of 3.3% shall be considered as string failure).

4. VISULIZATION AND DETECTION

Visualization of Specific Yield deviation of individual combiner box can be done in multiple ways. It can be monitored though Excel sheet as well as it can be visualized in form for status map using tiles in central monitoring system (CMS) or site SCADA. Refer below Picture-1 of Excel version of visualization and Picture-2 example of visualization in form of Status Map in central monitoring system.

5.? USE CASES AROUND THIS

?AI-ML-enabled central monitoring systems can leverage recorded Specific Yield deviation for various use cases such as real-time string failure identification, degradation analysis, annual testing (EL/IV) sample selection, PV module cleaning effectiveness monitoring, and soiling pattern determination

6. ADVANTAGES OF USE

Recorded Specific Yield deviation offers immediate advantages, including Opex reduction of approximately Rs. 1250/MW/Year and annual revenue increase ranging from 0.5% to 3.5%, depending on the plant's characteristics.

7.? CONCLUSION

This paper presents a method for real-time string failure detection with a 99% confidence level, resulting in Opex reduction and revenue increase. It offers a comprehensive solution to enhance monitoring practice