Probability-Based Power Network Control for NERC Compliance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Independent system operators of electrical power networks face challenges in matching changes in load requirements with power generation, leading to increased energy costs due to unanticipated demand shortages, and need a method to comply with regulatory performance standards such as NERC's ACE and CPS standards.
Innovation Solution
A real-time control strategy based on probability theory is employed to calculate network performance target values using historical and future terms, allowing for predictive control of power generation to achieve compliance with NERC's CPS1 and CPS2 standards by adjusting power production in response to instantaneous ACE values and frequency biases.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If operators react to power trends and schedule information, then they can meet scheduled power interchange requirements, but they cannot effectively respond to unanticipated demand shortages
Solution Approach 1:
The system calculates a performance target value using historical performance data before real-time operation occurs. This preliminary calculation establishes a proactive baseline that guides real-time control decisions, allowing operators to anticipate rather than merely react to performance requirements.
Solution Approach 2:
The system continuously monitors real-time performance data and compares it against the calculated target value. This feedback mechanism enables dynamic adjustment of power generation to maintain compliance with control performance standards while responding to unanticipated demand changes.
2Reliability
If operators meet unanticipated energy demand, then reliability is improved, but energy costs increase
Solution Approach 1:
By calculating the performance target value in advance using historical data, the system establishes an optimal performance baseline that accounts for typical demand patterns. This allows operators to prepare appropriate response strategies before unanticipated demands occur, minimizing costly reactive measures.
Solution Approach 2:
The system uses its own historical performance data to generate the target value, enabling self-optimization without requiring external intervention or expensive real-time analysis. This self-service capability reduces operational costs while maintaining reliability.
3Loss of energy
If operators operate with oversupply of energy, then energy costs decrease, but control performance standards may not be met
Solution Approach 1:
The continuous comparison of real-time performance against the calculated target value provides feedback that prevents excessive oversupply. Operators can maintain cost-effective energy levels while automatically adjusting to stay within control performance standards.
Solution Approach 2:
The system dynamically adjusts operational parameters based on the calculated target value, which is derived from historical performance data. This allows optimization of energy supply levels to balance cost efficiency with compliance requirements.
Data Source
AI summary
A system (400) and method for controlling an operation of an electrical power network (420) is described. The method includes configuring an allowable performance of an electrical power network over a predetermined time period as a probability expression comprising a historical term and a future term having an electrical power network operating condition variable. The method also includes calculating a network performance target value (300) according to the probability expression by using a historical electrical power network operating condition value for the electrical power network operating condition variable. The method further includes using the performance target value for controlling the electrical power network effective to achieve the allowable performance. The system includes a database (406), a processor (404) coupled to the database, and a monitoring and control module (410) coupled to the processor.