10-09-2016, 11:16 AM
1454339586-1454314392731714411010finalreport1.docx (Size: 910.4 KB / Downloads: 23)
ABSTRACT
Security-Constrained Unit Commitment (SCUC), as one of key components in power system operation, is being widely applied in vertically integrated utilities and restructured power systems. The efficient solution framework is to implement iterations between a master problem unit commitment and sub-problems network security evaluations. In industrial applications, both Edge disintegration, Matrix Relaxation and Mixed-Integer Programming are commonly applied for the unit commitment problem, and both linear sensitivity factor and Benders cut methods are used to generate additional constraints in the phase of network security evaluations. This paper evaluates capabilities and performances of each algorithm through technical discussion and numerical testing. Special topics on the large-scale SCUC engine development are also discussed in this paper, such as input data screening, inactive constrains elimination, contingency management, infeasibility handling, parallel computing, and model simplification. This paper will benefit academic researchers, software developers, and system operators when they design, develop and assess effective models and algorithms for solving large-scale SCUC problems.
INTRODUCTION
1.1 GENERAL
Security-Constrained Unit Commitment (SCUC), as one of key components in power system operation, is being widely applied in perpendicularly incorporated utilities and updated power systems. The proficient resolution scaffold is to execute iterations between a master quandary unit commitment and sub problems network security evaluations. In industrial applications, both medium recreation and mixed-integer encoding are commonly applied for the unit assurance problem, and both linear compassion issue and edge cut methods are used to make extra constraints in the segment of association protection evaluations.
This evaluates capabilities and performances of each algorithm through methodological conversation and statistical trying. Special topics on the large-scale SCUC engine expansion are also discussed, such as input data show, dormant constrains removal, possibility executive, infeasibility management, comparable computing, and representation generality. This benefit scholarly researchers, software developers, and system operators when they design, develop and evaluate efficient models and algorithms for solving large-scale SCUC problems.
As SCUC is a key decision-making tool for power systems operation, careful attention has to be paid to its solution speed and performance, especially for large-scale power systems.
1.1.1 Security-Constrained Unit Commitment (SCUC)
SCUC as one of key components in power system operation is being widely applied in vertically integrated utilities and restructured power systems. The efficient solution framework is to implement iterations between a master problem (unit commitment) and sub-problems (network security evaluations). In industrial applications, both Lagrangian relaxation and mixed-integer programming are commonly applied for the unit commitment problem, and both linear sensitivity factor and Benders cut Methods are used to generate additional constraints in the phase of network security evaluations. This paper evaluates capabilities and performances of each algorithm through technical discussion and numerical testing. Special topics on the large-scale SCUC engine development are also discussed in this paper, such as input data screening, inactive constrains elimination, contingency management, infeasibility handling, parallel computing, and model simplification. This paper will benefit academic researchers, software developers, and system operators when they design, develop and assess effective models and algorithms for solving large-scale SCUC problems.
In general, SCUC satisfies the prevailing constraints including:
• Power balance.
• System spinning and operating reserve requirements.
• Minimum ON and OFF time limits.
• Maximum number of unit startups.
• Generating unit ramping limits.
• Minimum and maximum generation capacity.
• Must-on and area protection constraints.
• Forbidden operating region of generating units.
• Startup and shutdown characteristics of generating units.
• Fuel and emission constraints in vertically integrated utility or energy constraints in restructured power systems.
• Transmission flow and bus voltage limits.
• Limits on state and control variables including real and reactive power generation, shunt capacitors, controlled voltages, and tap-changing and phase-shifting transformer adjustments.
• System operation constraints for credible contingencies. In order to build the SCUC problem, the hourly load forecast and system reserve requirements over a time period should be determined by system operators. The physical characteristics and operating conditions of generating units (e.g., coal, oil, gas, nuclear, hydro, combined cycle, pumped storage, wind, and solar) and transmission equipment’s (e.g., tap-changing transformers and phase shifters, controllable series capacitors, switchable lines and high-voltage current transmissions) should also be considered as inputs. The SCUC solution provides the hourly generation schedule for supplying the system load and satisfying the transmission network security at the normal state. The scheduling of hourly generating reserve could also withstand sudden disturbances in power systems, such as unexpected changes in hourly load and transmission line/generating unit contingencies.
1.1.2 Methodologies for SCUCSub-Problems
In SCUC sub problems, the network evaluation for the base case and contingency cases are performed based on the generation schedules from the SCUC master problem. The network security constraints under base case and contingencies include real and reactive nodal power balances, transmission flow (MW or MVA) limits, flow gate MW flow limits, bus voltage limits, reactive power generation limits, limits on capacitor adjustment, limits on transformer taps and phase shifter angles, and corrective actions for contingencies.
The network security evaluation often adopts a dc model. The serious shortcoming of the DC model is that it provides no explicit information on bus voltages and reactive power, even that it has a poor accuracy in computing the real power flows on the lines with high ratios. The dc model is used in the SCUC problem only because of its simplicity and speed. A hybrid approach is another choice for the network security evaluation. For example, the full ac model is for base case due to its accuracy, the dc model for all contingencies because of its rapidity. However, a perfect network model should take into account full ac constraints for the base case and contingencies. Two main methods, the Linear Sensitivity Factor (LSF) method and Benders cut method are utilized to solve the hourly network security check sub problems and generate additional transmission security constraints for resolving the UC problem.
In restructured power systems, Independent System Operators (ISOs) utilize security-constrained unit commitment (SCUC) to clear the day-ahead and real-time markets. In general, the SCUC is a constrained optimization problem that minimizes the system operating cost including units' generation and startup/shutdown costs. The constraints include unit commitment constraints and transmission network security constraints.
4
1.2 NEED FOR THE STUDY
The challenge of committing reserves lies in optimizing the Tradeoff between system security and economic operation of the system. Any level of security can be achieved in the system given a sufficient amount of reserves. The challenge rests in choosing the level of reserves that satisfies certain operational criteria in an uncertain environment at least cost.
Traditional reserve commitment approaches on reserve requirements and security constraints that are meant to mitigate continuous fluctuations in demand and renewable supply as well as discrete disturbances such as generator and transmission line failures.
However, these models often fail to capture the full range of complexity in an uncertain environment and rely instead on heuristic practices adopted by operators through experience. The power system operations literature has proposed four fundamental paradigms for representing uncertainty and optimizing the commitment of reserves at least cost: stochastic optimization, security-constrained approaches, robust optimization and probabilistic constraints.
In this paper, a distributed SCUC (D-SCUC) algorithm isproposed for large-scale power systems by incorporating the advantages of both LR method (e.g., decomposition and scalability) and MIP method (e.g., easiness to include additional constraints and optimality). The proposed D-SCUC algorithm is faster than conventional centralized SCUC and hence it accelerates the generation scheduling process in large-scale power systems. The basic idea of the D-SCUC is to reduce the size of the optimization problem to be solved by decomposing the system into several zones and applying the analytical target cascading technique for solution. Each zone, which includes a portion of the generators, loads and network elements, is connected to other zones through tie-lines
5
1.3 OBJECTIVES OF THE STUDY
Security-Constrained Unit Commitment (SCUC) refers to the economic scheduling of generating units for serving the hourly load demand while satisfying temporal and operational limits of generation and transmission facilities in power systems. In a vertically integrated utility, regulatory departments apply SCUC for minimizing the operating cost. However, SCUC is utilized by Independent System Operators (ISOs) or Regional Transmission Organizations (RTOs) to clear real-time and day-ahead markets.
The objective of SCUC is to maximize the social welfares based on energy and price bids submitted by market participants, generation suppliers and load demanders. As a key decision-making component for the current power system operation, the modeling and solution of SCUC, especially for the large-scale power systems, should be seriously recognized and analyzed. In addition, the development of future power systems is bringing new challenges into the SCUC.
The next-generation SCUC decision-making tool should effectively integrate various sophisticated features including active demand response participation in both energy and ancillary service markets, increased accommodation of intermittent and volatile renewable energy, and multiple energy storage options. With increasing complexity of functionality and market participation in future power systems, a stochastic and/or robust SCUC engine might be promising to handle such uncertainties due to alternative sources of power such as demand response, green generators and energy storages.
6
In order to accommodate such needs, it is also necessary to pay more attentions to the SCUC modeling and solution and study their important aspects, which are fundamental issues for the development of the next-generation SCUC. In order to build the SCUC problem, the hourly load forecast and system reserve requirements over a time period should be determined by system operators.
The physical characteristics and operating conditions of generating units (e.g., coal, oil, gas, nuclear, hydro, combined cycle, pumped storage, wind, and solar) and transmission equipment’s (e.g., tap-changing transformers and phase shifters, controllable series capacitors, switchable lines, and high-voltage current transmissions) should also be considered as inputs.
The SCUC solution provides the hourly generation schedule for supplying the system load and satisfying the transmission network security at the normal state. The scheduling of hourly generating reserve could also withstand sudden disturbances in power systems, such as unexpected changes in hourly load and transmission line/generating unit contingencies.