Many studies have been carried out on carbon emission reduction, and different studies are committed to different links of carbon emission reduction. This paper reviews the relevant literature from three aspects: the analysis of the influencing factors of carbon emission reduction, the prediction mo Contact online >>
Many studies have been carried out on carbon emission reduction, and different studies are committed to different links of carbon emission reduction. This paper reviews the relevant literature from three aspects: the analysis of the influencing factors of carbon emission reduction, the prediction model of carbon emission reduction, and the random uncertainty of carbon emission reduction.
The research and achievements of other scholars are shown in the Table 1.
Since the end of 2019, COVID-19 has swept the world, and the world situation has become more complex, which has profoundly affected and changed global and regional energy consumption and carbon emissions. Wang et al. analyzed the various impacts of COVID-19 on China''s environmental development, and discussed Chinese future economic development and its impact on the world economy by scenarios (Wang and Su 2020; Wang et al. 2021, 2022; Whang and Zhang 2021). It also needs to explore the development mode of carbon emission reduction of power grid enterprises to adapt to the new international environment.
Generally speaking, there are few studies on carbon emission reduction of power grid enterprises, and there are few studies on the evaluation method of carbon emission reduction contribution of power grid enterprises. In addition, the research on carbon emission reduction mostly focuses on one of the above aspects, and there is a lack of research that integrates the three aspects. Therefore, under the background of the national evaluation of carbon reduction contribution of energy industry (Su 2021), this research has important theoretical and practical significance.
As a link and platform connecting the upstream and downstream of the power industry, power grid enterprises will play an important strategic supporting role in realizing China''s carbon peak and carbon neutralization strategic objectives. Through the in-depth implementation of the new energy security strategy of "four revolutions and one cooperation," take measures to support clean energy consumption, realize fine line loss management, and improve the proportion of energy consumption and power efficiency of power terminals, so as to contribute to the goal of carbon peak and carbon neutralization.
In accordance with the carbon emission reduction plan, China Southern Power Grid has strengthened the allocation of clean energy and helped the five southern provinces build a low-carbon and clean energy structure. The installed capacity and electricity of non-fossil energy in the whole network have accounted for more than 50% for 5 consecutive years; continuously optimize the power grid structure, control the line loss rate at 5.59%, and save 3.66 billion kWh and 1.07 million kW of power on the demand side (China Southern Power Grid 2020).
Due to the difficulty of data collection of comprehensive energy service emission reduction, energy supply, and energy efficiency service emission reduction, this example only considers three parts: reducing line loss emission reduction, consuming clean energy emission reduction, and electric energy substitution emission reduction.
In order to systematically analyze the influencing factors of the contribution of power grid enterprises to carbon emission reduction (Lin and Qi 2018), starting from the three sub indicators constituting the contribution of power grid enterprises to carbon emission reduction, based on artificial experience and literature review, 26 influencing factors on the three sub-indicators of carbon emission reduction contribution are collected. The details are shown in Fig. 1 (note: the length of high-voltage transmission line refers to the length of ≥ 220 kV AC transmission line, and the availability factor of high-voltage transformer refers to the availability factor of transformer ≥ 110 kV and above).
Influencing factors of carbon emission reduction contribution
This paper studies a variety of influencing factor screening methods and compares the traditional principal component analysis method, grey correlation degree method, and PLS-VIP method. Compared with PLS-VIP method, the influencing factors screened by principal component analysis method and grey correlation degree method have a higher repetition rate of influencing factors among sub-indexes and poor performance in the preliminary trial calculation learning accuracy. Therefore, PLS-VIP method is determined to be used as the influencing factor screening method in this paper.
Based on the preliminary collection of influencing factors, this paper uses PLS-VIP screening method (Liu and Jiang 2017; Liu 2016) and considers the VIP value of influencing factors and the actual correlation. Thirteen main influencing factors are screened from three aspects: consumption of clean energy emission reduction, reduction of line loss emission reduction, and electric energy substitution emission reduction. The details are shown in Figs. 2 and 3 (note: the 13 influencing factors obtained by screening are shown in bold text in the figure).
PLS-VIP screening results of influencing factors of carbon emission reduction contribution
Proportion of influencing factors
In order to highlight the impact of the three sub-indicators of carbon emission reduction contribution on the contribution of carbon emission reduction, according to the selected influencing factors and combined with the reality, this paper sets up four scenarios: control type, strengthening the reduction of line loss type, strengthening the consumption of clean energy type, and strengthening the power substitution type. The four scenarios correspond to the change range combination of different influencing factors. The subsequent contribution analysis of carbon emission reduction in this paper is based on these four scenarios (Tian et al. 2021).
In this paper, the machine learning combination algorithm and Monte Carlo simulation method are combined to construct the carbon emission reduction contribution evaluation model of power grid enterprises. The specific process of the model is shown in Fig. 4.
Genetic algorithm (GA) is a heuristic intelligent search algorithm with good robustness and strong global search ability (Holland 1975). Its implementation path is shown in Fig. 5.
Basic flow chart of genetic algorithm
Limit learning machine (ELM) is a single hidden layer feedforward neural network learning algorithm (Huang et al. 2006). Before training, the algorithm needs to randomly set the input layer weight matrix and threshold, and keep the matrix and threshold unchanged during the training process. The prediction accuracy is improved by optimizing the number of hidden layer neuron nodes. It has the advantages of simple structure, few calculation parameters, fast training speed, and good generalization performance (Xu et al. 2019). Compared with the traditional neural networks, especially the single hidden layer feedforward neural networks (SLFNS), the limit learning machine is faster than the traditional learning algorithm on the premise of ensuring the learning accuracy.
H is the output of the hidden layer node, β is the output weight, and T is the expected output.
The output weight β can be determined accordingly
Setting of genetic algorithm parameters: the initial parameters of genetic algorithm include population number, crossover probability, mutation probability, chromosome length, selection rules, maximum iteration times, etc., different parameter combinations will have a great impact on the optimization results. For example, the number of populations will affect the calculation rate and the potential to produce better solutions, and the crossover probability and mutation probability will affect the transmission of excellent genes.
Population initialization: determine the neural network topology of ELM, including the number of input and output nodes of the prediction model, and set the number of hidden layer nodes; the weight matrix and threshold of ELM neural network are randomly generated and binary coded to form the chromosome of the initial population.
Individual fitness calculation: for each individual of each generation of population, ELM algorithm is used to obtain the output matrix, and the second norm of the difference between the calculated result and the expected matrix is used as the individual fitness.
Population iteration: the roulette method is used to select the cross population. All individuals in each generation will have a certain probability of variation and genotype change, which is conducive to a better solution. The optimal solution of each generation is recorded in the population iteration process. When the maximum number of iterations is reached, the optimal value in the optimal solution of each generation is the final prediction result.
Advantages of GA-ELM model: the neural network input parameters of elm are optimized through the process of genetic algorithm, including structural parameter optimization and weight threshold parameter optimization. On the basis of genetic algorithm, it can improve the accuracy of elm output prediction value, so as to achieve better prediction effect, and provide a basis for the subsequent assessment of carbon emission reduction contribution by scenario.
Monte Carlo method, also known as random simulation method, key is to generate random numbers that obey a certain distribution. The basic principle is as follows.
About 170 kWh energy saving and emission reduction
As the photovoltaic (PV) industry continues to evolve, advancements in 170 kWh energy saving and emission reduction have become critical to optimizing the utilization of renewable energy sources. From innovative battery technologies to intelligent energy management systems, these solutions are transforming the way we store and distribute solar-generated electricity.
When you're looking for the latest and most efficient 170 kWh energy saving and emission reduction for your PV project, our website offers a comprehensive selection of cutting-edge products designed to meet your specific requirements. Whether you're a renewable energy developer, utility company, or commercial enterprise looking to reduce your carbon footprint, we have the solutions to help you harness the full potential of solar energy.
By interacting with our online customer service, you'll gain a deep understanding of the various 170 kWh energy saving and emission reduction featured in our extensive catalog, such as high-efficiency storage batteries and intelligent energy management systems, and how they work together to provide a stable and reliable power supply for your PV projects.
