Research on Technology Foresight of Terminal Energy Consumption Based on Delphi Survey

. In view of China's "carbon peak, carbon neutral" strategy, especially the important point of reaching the carbon peak by 2035 and achieving socialist modernisation, it is important to select a number of priority end-use energy consumption technology topics to promote the company's technological innovation. Based on the results of the two rounds of the Delphi survey, eight end-use energy consumption technologies with a "dual carbon" focus should be prioritised, namely: virtual power plant technology, electric vehicle technology, impact assessment and costing of carbon emissions and reduction, and carbon dioxide capture and storage technology. The results of the Delphi survey show that R&D investment, policy mechanisms and market viability are the main constraints to the completion of these technologies. In response to these constraints, relevant recommendations are made in terms of funding input, institutional development and market development.


Introduction
In the context of the "carbon peak, carbon neutral" energy strategy, the construction of a "new power system with new energy as the mainstay" and a "clean, low-carbon, safe and efficient energy system" has become an important guiding direction for the future of China's energy sector [1][2][3]. On March 22, 2022, China's National Development and Reform Commission and Energy Bureau issued the "14th Five-Year Plan for Modern Energy System", which aims to comprehensively and deeply expand electric energy substitution, enhance the level of low-carbon and electrification of end-use energy, and further significantly increase the proportion of non-fossil energy consumption from 25% by 2030. The proportion of non-fossil energy consumption will be further increased on the basis of 25% by 2030 [4]; with the gradual increase of the proportion of electric energy in the final energy consumption since the 13th and 14th Five-Year Plan, electric ships, offshore oil platform shore-based power supply, steam heat pump system, electric hydrogen production, electric vehicle charging and discharging equipment, high-efficiency storage equipment, and electric car charging and discharging equipment, New technologies for end-use electricity consumption, such as electric vehicle charging and discharging equipment and high-efficiency ice storage, have emerged continuously, which not only put forward new requirements for the safe and efficient operation of the power grid, but also posed new challenges for the company to explore and innovate the management mode and promotion and application mechanism that are compatible with them, and there is an urgent need to clarify the key technical issues and provide a basis for decision-making in the formulation of science and technology policies in this field. In March 2021, the "Outline of the 14th Five-Year Plan of the National Economic and Social Development of the People's Republic of China and the Vision 2035" once again pointed out that it is important to make technological self-sufficiency and self-improvement a strategic support for national development, integrate and optimise the allocation of scientific and technological resources and, in particular, to focus on key core technologies from the perspective of the country's urgent needs and long-term requirements [5]. Technology foresight [6] helps guide companies to explore and discover the priority of technology development in the field of end-use energy consumption around the strategic goal of "double carbon", in order to select strategic research areas and new technologies that may generate the greatest economic and social benefits. In terms of technology foresight, the Delphi method, contextual analysis and technology roadmaps are commonly used in practice in Japan, the UK, Germany and Russia [7]. Among them, the Delphi method has been the core method for technology foresight activities in various countries as an expert group workshop method to collect opinions from all sectors of the society, which is simple and easy to operate [8].
In summary, this paper will use the Delphi method as the main method and "double carbon" as the technology target, based on the results of the survey to filter out the technology topics in the field of end-use energy consumption that companies should prioritise, and propose relevant mechanisms and research outlooks in response to the constraints on the implementation of the technology topics as reflected in the survey.

2
The basic process of foreseeing new technologies for end-use energy consumption.
The technology foresight study follows the methodology of the "China Technology Foresight Study for the Next 20 Years" [9] and focuses on the technology needs for 2035 in the field of end-use energy consumption. The initial technology topics were identified using the expert discussion method, and the most important technology topics for 2035 in the field of end-use energy consumption were selected after two rounds of large-scale Delphi questionnaires, integrating the collective wisdom of experts,the steps as shown in figure1. The first step is to form an expert group. Considering the need for the head of the expert group to be able to take a national strategic perspective, with a high degree of responsibility and a very high level of expertise.
The second step is to research and propose technical topics. The Delphi questionnaire asks questions on a range of technical topics, and through several rounds of meetings and discussions, and on the basis of the opinions and suggestions of a wide range of technical experts, a list of technical topics and their specific content are proposed, following the principles of uniqueness, foresight, strategy, feasibility, consistency and completeness.
The third step is a large-scale Delphi survey. The Delphi questionnaire was designed following the format of the "China Technology Foresight Study for the Next 20 Years". Instead of using the classic 4 rounds of survey, this survey was based on two rounds of survey, and the frequency of expert review was increased to improve the credibility of the results. The first round of the survey was conducted in paper format + electronic format, while the second round of the Delphi survey was conducted online using mobile phones and the internet, taking into account the popularity of mobile phones and the internet.
In the first round of the questionnaire, part of the guiding questions were designed to allow experts to evaluate each technology list based on a list of technologies obtained by mining the technology hotspots in the patent literature in the field of end-use energy consumption, such as table1. The specific question set features a matrix of multiple choice questions with multidimensional options, in addition to standardised multiple choice, fill-in-the-blank and quiz questions, as shown (in the technical areas of virtual power plants and electric hydrogen production, for example in table3). The columns are technical items and the rows are set up according to the options of the multiple choice questions, which are clear and easy to answer at a glance. The expert user can simply select the corresponding option in each box (six items for 4 options ABCD, one for 7 options ABCDEFG and the last one for two options AB). For each line the description of the relevant question is displayed centrally, the questionnaire page is compact and concise, and the expert user can make an intuitive assessment.
Through two rounds of large-scale Delphi questionnaires, the project has collected feedback from more than 100 experts, which guarantees the reliability of the survey. In terms of expert selection, the questionnaire survey covered different institutions such as universities, research institutes, enterprises and governments as far as possible to guarantee the comprehensiveness of the survey respondents. In order to ensure the professionalism of the respondents, the survey required that the experts had associate or higher titles or were recommended by other experts to obtain them. The final list of technical topics was formed by summarising and collating the expert opinions from the two rounds of the Delphi survey and after the expert group's validation.

Treatment of foreseen results of new technologies for end-use energy consumption
The data to be processed is a series of frontier technologies in the field of end-use energy consumption, which are evaluated to find the level of importance of each technology and to analyse each technology specifically. The classical k-means clustering algorithm allows an intuitive distinction between the importance levels of the technologies and can show the various metrics of the technology items.
Clustering was performed according to each frontier technology in different areas of end-use energy consumption technology. For the frontier technology items, a value of k is chosen as the seed of k clusters, all technology items are divided into k clusters, and the dissimilarity of each technology to the k centroids is calculated, using Euclidean distance as a measure [10,11] : In which, d(X,Y) represents the distance of the two technologies in Euclidean space. According to the distance of each technology to each centroid, each technology is divided into the nearest centroid to obtain k families, and the final clustering result is obtained by several iterations, the calculation process is shown in figure 2.

Figure2
Flowchart for determining the optimal number of clusters based on the K-means clustering algorithm

Basic results of new technologies foreseen for end-use energy consumption
The two rounds of the Delphi survey 3rganized a combined total of 300 experts to participate in the survey. In the first round of the Delphi survey, 35.2%, 41.3%, 12.4% and 11.1% of the respondents were from universities, research institutes, enterprises and government respectively, and the average number of respondents for each technical topic was about 49, with a valid return rate of 29.52%. In the second round of the Delphi survey, 36.4%, 34.1%, 18.3% and 12.2% of the respondents were from universities, research institutes, enterprises and government respectively, with a valid return rate of 31.84%.
The average number of respondents per technology topic in the second round of the survey increased significantly to 75, as the project team increased the number of telephone follow-up calls to respondents and the 14 th Five-Year Plan process made more respondents aware of the importance of technology foresight. Based on the results of the survey, the expert group identified more than 100 technology needs for 2035 in the area of end-use energy consumption, and selected the eight most important technology topics as follows.
In order to determine the importance of the technical topics concerned, three criteria were proposed in the design of the Delphi questionnaire to promote the growth of the company's efficiency, improve the quality of the electricity supply and guarantee the security of the energy supply, to obtain an overall ranking of the importance of the technical topics.
The second round of Delphi survey results were processed using the single-factor importance index calculation method and the three-factor composite importance index calculation method to identify the eight most important technical topics for promoting the company's efficiency growth, improving the quality of power supply and ensuring the security of energy supply, as well as the eight most important technical topics taking into account the above three indicators. Eight topics are shown in table4. Overall, most of the eight technology topics appear to be achievable by around 2030. In terms of the main R&D countries for the eight technology topics, they are relatively behind the US, EU and Japan. The company needs to significantly increase its support for the corresponding technology topics to ensure that it does not fall behind in the new round of technology catch-up and take a dominant position in the technology market.

5
Recommendations for the development of technological developments in the field of end-use energy consumption.
According to the Delphi survey, in order of frequency, the constraints on the development of the eight technology topics are technical feasibility, R&D investment, infrastructure, regulations, policies and standards, and commercial viability. In view of the above, the development of technologies for 2035 in the field of enduse energy consumption to support the "double carbon" needs to clarify technical feasibility after extensive consultation, increase stable investment in R&D, focus on building scientific and technological infrastructure, promote international cooperation and exchange, and ensure the accumulation of basic research and breakthroughs in applied research. In addition, attention also needs to be paid to the two-wheel drive of policy and market to continuously provide impetus for technological progress and form a long-term value-added cycle. To this end, the following recommendations are made in terms of science and technology investment, mechanism guarantee and market pull: First, increase stable R&D investment. Combined with the results of technology foresight, we will fully demonstrate good technical feasibility assessment, increase R&D investment in the direction of new charging and discharging technology for electric vehicles, electric hydrogen production technology and virtual power plant technology, and deploy a number of forward-looking and strategic key R&D plans, headquarters science and technology projects, and provincial and municipal science and technology projects. Secondly, we will strengthen all-round and multichannel international scientific and technological cooperation. Extensive scientific and technological cooperation with advanced enterprises in the field of advanced customs in Europe, America and Japan, etc., to jointly promote the process of technological development in the field of terminal energy consumption.
Third, to create a mechanism to support technological progress. Accelerate the establishment of a sound technical planning and all-round support system in the field of the company's terminal energy consumption, focus on the evaluation of the effectiveness of technological innovation and application, accelerate the supporting mechanism system for the support of scientific and technological projects, and provide financial support and support for demonstration application conditions in areas such as gridconnected electric vehicles, virtual power plants and integrated energy systems, so as to provide sufficient protection and motivation for relevant subjects to participate in technological research and development.
Fourth, to establish a market-pulled mechanism for technology development. Accelerate the transfer and transformation of scientific and technological achievements, promote the localisation and market development of technologies, equipment and equipment generated by scientific and technological projects, use subsidies, funding and other means to stimulate product and service innovation in the field of end-use energy consumption, and promote companies to gradually become the main body of technological innovation in this field.

Summary
Using a technology foresight approach, the study explores the selection of key technologies for end-use energy consumption, and summarises experts' judgements on the selection of end-use energy consumption technologies with the goal of "peak carbon and carbon neutrality". The findings of the study provide a reference for the company's decision to invest in science and technology in related areas and to develop supporting management mechanisms. The shortcomings are that, firstly, the technology topics prioritised in the conclusion are selected from the technology foresight in the whole field of end-use energy consumption, which makes it difficult to cover the relevant theories and technologies comprehensively. Second, according to the logic of mutual construction in technology foresight, to improve the accuracy of technology foresight, it is necessary to consider the influence of technology, society, consensus and time dimensions, and other methods such as contextual planning and bibliometrics will be used in the future to more accurately identify the technologies that should be prioritised for development in this area.