GMM-HMM-Based Medium- and Long-Term Multi-Wind Farm Correlated Power Output Time Series Generation Method

Abstract

Medium-and long-term wind farm output scenarios are usually required for power system planning. However, with the traditional medium-and long-term wind power output time series generation method, it is difficult to simultaneously take into account the spatiotemporal correlation of multiple wind farms, leading to the failure of the output sequence to reflect their medium-and long-term statistical characteristics. Hidden Markov model (HMM) can simultaneously consider the temporal and spatial correlation of multiple wind farms, but the existing HMM method is not accurate in describing the output distribution of multiple wind farms, which leads to a large error between the model output and the actual wind power output. Therefore, this paper proposes a method for generating medium-and long-term correlated output time series of multiple wind farms based on the Gaussians mixture Hidden Markov model (GMM-HMM). The discrete state variable in the hidden Markov model is used to describe the meteorological state. The Markov chain between discrete state variables is used to describe the temporal correlation of wind power output. The wind power output vector of multiple wind farms is used as the observation variable, and the mixed Gaussian probability distribution mapping relationship between the state variable and the multidimensional wind power output vector is established. Based on the Monte Carlo sampling method, the multi-wind farm output series satisfying the spatiotemporal correlation of historical output series are generated monthly. In the calculation example, the monthly wind power output series generated by five wind farms in Jilin Province are analyzed. The results show that the main statistical characteristics of the multi-wind power output time series generated by the proposed method are generally superior to those obtained with the traditional wind power output modeling method, which proves the superiority of the proposed method. INDEX TERMSMultiple wind farms, spatiotemporal correlation, Gaussians mixture Hidden Markov model (GMM-HMM), time series generation NOMENCLATURE Most of the symbols and notations used throughout this paper are defined below for quick reference.

Survey of Time Series Data Generation in IoT

“…Li et al [ 39 ] proposed a method which uses the Gaussian mixture model hidden Markov model to generate medium- and long-term wind power generation data. The method uses the expectation-maximum (EM) [ 74 ] algorithm to estimate the parameters of the model, and then randomly samples from the initial state probability distribution to generate an initial hidden state.…”

“…Researchers in many fields have proposed a variety of time series data generation methods in their respective fields such as biology [ 8 ], database benchmark [ 29 , 30 ], electricity [ 31 ], energy [ 32 , 33 , 34 , 35 , 36 , 37 , 38 , 39 ], environment [ 40 , 41 , 42 ], finance [ 5 , 43 ], medicine [ 7 ], music [ 44 ], networks [ 45 ], remote sensing [ 46 , 47 , 48 , 49 ] and sensors [ 50 ]. Despite the abundance of research on time series generation, a comprehensive survey that systematically classifies and evaluates the previous work is lacking.…”

“…Based on the underlying algorithms and models used by the existing time series data generation methods, this paper divides these methods into four categories: rule-based methods [ 51 , 52 , 53 ], simulation-model-based methods [ 37 , 45 ], traditional machine-learning-based methods [ 29 , 32 , 39 ] and deep-learning-based methods [ 7 , 30 , 44 , 54 , 55 , 56 , 57 , 58 , 59 , 60 , 61 , 62 , 63 ]. Rule-based methods use a set of rules or constraints to specify the properties and structure of the data to be generated.…”

Real Time Congestion Management Using Plug in Electric Vehicles (PEV’s): A Game Theoretic Approach

“…and an equal amount of generation was reduced from conventional generators. Wind speed was generated using Auto Regressive Moving Average model (ARMA) [32], [43]. The tie-line power between bus 60 and 61 carrying power from area 2 to area 1 varies between −2.529pu to −4.6077pu due to variable wind power as shown in (Fig.…”

Human Activity Recognition with an HMM-Based Generative Model

“…Moreover, it has been applied in human activity recognition [ 63 , 64 , 65 , 66 , 67 , 68 , 69 , 70 , 71 , 72 , 73 , 74 , 75 , 76 , 77 , 78 , 79 , 80 , 81 ]. In HMM, it is commonly assumed that emission probabilities are raised from Gaussian mixture models (GMM) [ 82 , 83 , 84 , 85 , 86 , 87 ]. However, the assumption of Gaussianity could not be valid for all cases.…”