• Energy Research

AbstractPhotovoltaic (PV) generation has high impact on the decarbonization pathways of power systems. Accuracy of day‐ahead PV power forecasting has become crucial in the operation and control of power system with high PV penetration. This paper develops a hybrid approach based on generative adversarial network (GAN) combined with convolutional autoencoder (CAE) to improve PV power forecasting accuracy. Self‐organizing map method is first utilized as data pre‐processing to classify target days into different weather types based on solar irradiance. With the ability of GAN to reduce the burden of loss and the advantages of CAE to extract multi‐scale effective features from the weather and PV power, PV power forecasting model consisting of GAN and CAE is proposed. The developed method has been tested on a real dataset in a Chinese PV station and compared with base reference PV forecasting methods. Numerical testing results demonstrate the effectiveness of our method with high accuracy.

AbstractPhotovoltaic (PV) generation has high impact on the decarbonization pathways of power systems. Accuracy of day‐ahead PV power forecasting has become crucial in the operation and control of power system with high PV penetration. This paper develops a hybrid approach based on generative adversarial network (GAN) combined with convolutional autoencoder (CAE) to improve PV power forecasting accuracy. Self‐organizing map method is first utilized as data pre‐processing to classify target days into different weather types based on solar irradiance. With the ability of GAN to reduce the burden of loss and the advantages of CAE to extract multi‐scale effective features from the weather and PV power, PV power forecasting model consisting of GAN and CAE is proposed. The developed method has been tested on a real dataset in a Chinese PV station and compared with base reference PV forecasting methods. Numerical testing results demonstrate the effectiveness of our method with high accuracy.

Abstract The expected increase of intermittent electricity production imposed by renewable energy sources in Northern Europe poses rising challenges to the operation of the power system. The Nordic hydro based power system, particularly the Norwegian, is in a favourable position to provide necessary production flexibility to the Continental European power system. For this provision sufficient transfer capacity is required. In order to assess the provision of power production flexibility, a detailed operation optimisation model is necessary. Moreover, investments in the power system have to be set in relation to the economic outcome of the system operation. As a simultaneous system optimisation is not possible due to the size of the operation optimisation model, a two-step investment algorithm containing an iteration loop is proposed in this paper. The proposed investment algorithm identifies profitable investments based on the results of the operation optimisation model, where the effect of the investments on the power market outcome is taken into account through the iterative process. The presented investment algorithm is used for a transmission expansion analysis for a 2030 scenario of the Northern European power system. The results for the simulation of the scenario and the transmission expansion analysis are presented as well.

Abstract The expected increase of intermittent electricity production imposed by renewable energy sources in Northern Europe poses rising challenges to the operation of the power system. The Nordic hydro based power system, particularly the Norwegian, is in a favourable position to provide necessary production flexibility to the Continental European power system. For this provision sufficient transfer capacity is required. In order to assess the provision of power production flexibility, a detailed operation optimisation model is necessary. Moreover, investments in the power system have to be set in relation to the economic outcome of the system operation. As a simultaneous system optimisation is not possible due to the size of the operation optimisation model, a two-step investment algorithm containing an iteration loop is proposed in this paper. The proposed investment algorithm identifies profitable investments based on the results of the operation optimisation model, where the effect of the investments on the power market outcome is taken into account through the iterative process. The presented investment algorithm is used for a transmission expansion analysis for a 2030 scenario of the Northern European power system. The results for the simulation of the scenario and the transmission expansion analysis are presented as well.

Abstract As the penetration of variable renewable power generation increases in power systems around the world, system security is challenged. It is crucial to coordinate the available flexible generating resources, such as hydropower, to meet the need for system balancing. However, reserved capacity on hydropower plants should only be activated if there is sufficient energy or storage capacity to either increase or decrease production. The potential change in production will also affect all reservoirs and plants connected by the cascaded topology. These issues are largely ignored or simplified in hydropower reserve scheduling models. To properly account for the possible activation of reserved capacity, several two-stage model formulations based on stochastic and robust optimization are presented and compared in this paper. The uncertainty in net load deviations due to forecasting errors in renewable power generation is considered the source of reserve capacity activation. The case study based on a real Norwegian watercourse clearly shows the benefit of using any of the two-stage model solutions over the standard deterministic reserve procurement. A novel hybrid stochastic-robust model formulation is presented and shown to efficiently increase the robustness of the solution without notably increasing the reserve procurement cost compared to the stochastic and robust models.

Abstract As the penetration of variable renewable power generation increases in power systems around the world, system security is challenged. It is crucial to coordinate the available flexible generating resources, such as hydropower, to meet the need for system balancing. However, reserved capacity on hydropower plants should only be activated if there is sufficient energy or storage capacity to either increase or decrease production. The potential change in production will also affect all reservoirs and plants connected by the cascaded topology. These issues are largely ignored or simplified in hydropower reserve scheduling models. To properly account for the possible activation of reserved capacity, several two-stage model formulations based on stochastic and robust optimization are presented and compared in this paper. The uncertainty in net load deviations due to forecasting errors in renewable power generation is considered the source of reserve capacity activation. The case study based on a real Norwegian watercourse clearly shows the benefit of using any of the two-stage model solutions over the standard deterministic reserve procurement. A novel hybrid stochastic-robust model formulation is presented and shown to efficiently increase the robustness of the solution without notably increasing the reserve procurement cost compared to the stochastic and robust models.

Integrating balancing markets across borders bears a cost-reducing potential but requires the availability of transmission capacity for the eventual activation of committed reserves. Using scarce transmission capacity for this purpose hence incurs opportunity cost. In this paper, the utilization of the yet to be built interconnector N ordlink for exchange on the spot electricity and reserve market between Norway and Germany is investigated. An optimized utilization underlines the importance of the spot market but also results in partial reserve exchange from Norway to Germany in 77 % of the hours and yields a cost reduction of 45 Ma compared to pure spot exchange.