Abstract
Power transmission and transformation projects (PTTPs) under new energy grid connections are different from ordinary engineering construction projects. With large investment amounts, various supporting facilities, and high safety and quality requirements, PTTPs are vulnerable to the volatility of new energy power generation, climate, geological conditions, geopolitical environment, technological changes, and other kinds of uncertain factors. Therefore, the investment risk of PTTPs under new energy grid connection is particularly considerable in the project management. For the owner of PTTPs, adopting the engineering-procurement-construction (EPC) mode is an effective attempt to solve the project construction problems faced by the owners. Thus, this paper deeply excavates the key risk points of PTTPs in the initial investment phase under the EPC mode and constructs the novel risk evaluation index system from the perspectives of economy, management, policy, society, and environment. An assessment model is established based on triangle fuzzy number-hesitant fuzzy linguistic term sets (HFLTS-TFNs), the entropy method, and the fuzzy comprehensive evaluation method, which is used to evaluate the investment risk of PTTPs to provide reference for power grid corporations to control project risks. Finally, a case study in the Hexi Corridor region, Gansu Province, China, is illustrated to demonstrate the rationality of the decision model and find management risks are the most vital risk factors for PTTPs under EPC mode.
1. Introduction
As the infrastructure of the power system, the power grid plays an important role in national economic and social development [1]. Therefore, lots of countries attach great importance to the investment and construction of the power grid. Taking China as an example, the total planned investment in China’s power grid during the 14th Five-Year Plan period is expected to be close to 3 trillion yuan, which is an obvious increase compared with 2.57 trillion yuan during the 13th Five-Year Plan period. In addition, with the introduction of China’s national energy strategies for the development of new energy resource, such as carbon peak and carbon neutralization strategies, the investment in the power grid will continue to increase due to energy transformation [2]. PTTPs, as basic projects in power grid construction, play a key role in the stable operation of the power grid [3]. Different from other types of engineering construction projects, PTTPs are capital-intensive industries with large investment, complex technology, many project participants, and easy to be affected by the external environment. These factors make the construction and operation of PTTPs face various risks. Therefore, analyzing the investment risks of PTTPs plays a very important role in promoting the progress and improving economic benefits in the initial stage.
With more and more attention paid to low-carbon emission reduction, the installed scale of new energy gradually is increasing all over the world [4]. And lots of new energy projects will be connected to the power grid in the future. Large-scale new energy grid connections are accompanied by the large number of investment and construction of PTTPs. And there are still some problems in the new energy grid connection, such as single function and low equipment utilization rate. The existing problems cause the social and economic benefits of PTTPs to suffer losses, which increases the investment pressure and risk of power grid corporations. However, the traditional risk assessment of PTTPs mainly focuses on the risks of the economy, society, and environment, which pay little attention to the risks brought by the new energy grid connection.
With the development of new energy and the improvement of power demand, the construction conditions of power engineering projects have become more complex. The traditional construction and management modes have been difficult to meet the current needs. The EPC mode, as shown in Figure 1, can effectively coordinate participants and strengthen cooperation between various departments, so the EPC mode has gradually become the development trend of power engineering project construction mode in recent years [5], and Table 1 shows the differences between the traditional project contracting mode and the EPC mode in PTTPs. The application of EPC mode to PTTPs can promote project subdivision and strengthen cooperation between participants, which not only liberates the owners but also makes the project construction more holistic, thus effectively improving the construction efficiency and reducing construction risks [6]. It is worth noting that the owner does not directly participate in the project construction, so this situation makes the owner need to pay more attention to the risk control in the initial stage of project investment planning.
Based on the above background, this paper conducts the research on investment risk assessments of initial PTTPs stage under new energy grid connection from the owner’s perspective of EPC mode, in order to provide decision support for power grid corporations. This paper identifies the key factors affecting the investment risk of PTTPs by referring to relevant research and consulting experienced experts and then constructs a characteristic risk index system for the PTTPs investment. In addition, the corresponding evaluation model is established based on HFLTS-TFN, the entropy method, and the fuzzy comprehensive evaluation method, which provides theoretical support for the construction and development planning of PTTPs for relevant investors to a certain extent.
2. Literature Reviews
2.1. Research on Project Risk Management under the EPC Mode
Internationally, risk management research has experienced the establishment period of risk management research, the improvement period of risk management research, and the accelerated development period of risk management research. So far, risk management theory has become a relatively mature theoretical system and has been applied to many fields on a large scale. The research on project risk management under EPC mode, as an important branch, has also been greatly enriched and developed.
After investigating 350 senior executives in real estate and construction, Grynbaum [7] found that risk management often has a better risk resistance effect when joint contracting is adopted in general contracting projects. Mukilan et al. [8] proposed a modified particle swarm optimization algorithm-based claim management system to minimize project cost and time of EPC projects. Pal et al. [9] adopted logistic regression and neural networks to identify four critical factors which impacted international EPC projects’ success, including service provided by suppliers and/or subcontractors, continuous improvement, supplier and/or subcontractor delivery reliability, and effective problem. Wang et al. [10] put forward that partnering with project participants and enhancing organizational capability can improve contractors’ risk management, and they research causal relationships among them systematically. Shang et al. [11] pointed all parties have synergistic risks in the process of forming a cooperative alliance among stakeholders of guaranteed savings EPC projects, and the allocation based on risk coefficient is more in line with the actual situation in China.
On the one hand, we can find that risk management is extremely necessary through the above research. On the other hand, due to PTTP under EPC mode with huge engineering investment capital, a number of project participants, and various materials, the implementation of PTTPs will contribute to the occurrence of varying degrees of risks. However, there is little research on investment risk assessment of PTTPs under EPC mode currently. If effective risk management measures cannot be taken for the potential risks in the early stage of PTTPs, it may bring various degrees of losses to the contractor, the owner, and subcontractors. Thus, improving risk management ability and reducing risk events are essential issues to be studied urgently in the PTTPS under EPC mode.
2.2. Research on Risk Assessment Methods
In the risk assessment of PTTPs, the experts’ linguistic description of specific variables is descriptive rather than computable, making conducting the quantitative analysis difficult. Fortunately, the fuzzy set theory solved this problem proposed by Zadeh. But the fuzzy set theory has the disadvantage that a single linguistic value cannot fully describe the ambiguity of information. With the further advancement of predecessors’ research, HFLTS proposed can solve this problem well, which describes multiple variables at the same time and obtains more realistic results. As it has good data collection characteristics, it has been widely used in many fields. For example, Liao et al. [12] proposed that the theory of HFLTS was quite useful in objectively dealing with situations in which people were hesitant in providing linguistic assessments. Erol et al. [13] used the HFLTS method to investigate the true potential of blockchain to address the circular economy adoption barriers. Zhang et al. [14] applied the HFLTS method to two-sided matching decision-making problems. Yao et al. [15] adopted a synthetical assessment process based on HFLTS to select the current and potential optimal alternative for missile design.
As the risk assessment environment is relatively complicated, real values are too precise to describe fuzziness. TFNs with the upper limit, the lower limit, and the most probable value can be used as a numerical form to express experts’ uncertainty and be introduced into various fields. For example, Ma and Xiao [16] put forward an improved method to obtain basic belief assignments based on TFNs and the k-means algorithm. Rusu [17] introduced TFNs to the critical path method in practical engineering. Koirala et al. [18] thought TFNs could be used to overcome the uncertainty of insufficient information and applied them to select the proper site for a battery swapping station. And TFNs had been used to recognize uncertainties in the estimation of health risks proposed by the United States Environmental Protection Agency in Yinchuan city, northwest China [19].
The multicriteria decision-making method (MCDM) portfolio is a common strategy to measure project risk. Assigning different weights to each variable is a key link in the risk evaluation process, and general weight determination methods are susceptible to the influence of subjectivity, such as the analytic hierarchy process [20] and the analytic network process [21]. The entropy method, which determines the weights of variables according to their own information [22], has outstanding advantages and has been applied in so many fields. For example, Oluah et al. [23] used the weight method to select the phase change materials for optimum Trombe wall performance. Alipour et al. [24] adopted the entropy method for solving the selection problem of fuel cell combined with hydrogen supplier. Wang et al. [25] provided a model based on the entropy method for investors to select the most appropriate battery supplier for the battery swapping station. For related evaluation methods, fuzzy comprehensive evaluation has great superiority in solving MCDM problems, so it has been widely used in various research, such as the energy-saving rating of green bed and breakfast [26], the research on project postevaluation of wind power [27], and the coupled thermal-hydraulic-mechanical model of the geothermal system [28].
HFLTS, TFNs, the entropy method, and fuzzy comprehensive evaluation have been successfully applied in various fields, which further illustrate the rationality and practicality of these methods. And it is reasonable and novel to introduce these MCDM methods into the investment risk assessment of PTTP under EPC mode. The above makes our research more meaningful.
3. Risk Evaluation Index System of PTTPs
The expert survey method is adopted to identify the risk factors of PTTPs in this paper, which diverts experts’ opinions to collect a wider range of risk factors. Four experts are visited to brainstorm and collect expert opinions fully. Basing the principle of representativeness and feasibility, the following risk evaluation index system is established as shown in Table 2.
3.1. Economic Risks (C1)
(i)Financing risk (C11): the EPC owner’s funds of PTTPs are mainly composed of project capital and project financing funds, in which the proportion of project financing funds is generally large. The financing environment depends on the national macro monetary policy, and the financing cost fluctuates with the monetary market [29].(ii)Revenue and expenditure plan risk (C12) [30]: the EPC owner arranges the own cost plan according to the PTTPs’ progress. If the EPC owner fails to pay the project costs on time, it will delay the normal progress and cause losses to the investment.(iii)Price risk (C13): considering the long construction period of PTTPs, it is likely to face price fluctuations during the construction. For example, the rising price of building materials leads to more project costs and increases the investment risk.(iv)Interest rate risk [31] (C14): the fluctuation of interest rate directly affects the project construction cost, which is one of the typical economic risks the EPC owner faces.
3.2. Management Risks (C2)
(i)Performance risk of EPC general contractor (C21): the performance ability of EPC general contractor has a direct and important impact on construction progress [32], project quality, and construction period.(ii)Performance risk of EPC supervision team (C22): the EPC supervision engineers are different from simple construction supervision under EPC mode [32]. Their supervision work is more comprehensive, and the equivalent is to increase the requirements of responsibilities and abilities. The performance ability of the EPC supervision team has a significant impact on PTTPs, which is a typical risk factor for the EPC owner.(iii)Contract management risk (C23): under the EPC project management mode, the owner’s contract management is mainly to use the contract terms to protect own interests and effectively reduce the investment cost [33]. If the contract content is not comprehensive and clear, it will bring contract management risks to the EPC owner.
3.3. Policy Risks (C3)
(i)Change risk of policies and regulations (C31): the construction period of PTTPs is long, and the EPC owner’s benefit is likely affected by the changes in various policies and regulations during the long construction period, such as improvement of environmental protection standards, increase in project reserves, and collection of fixed asset investment adjustment tax. These factors will have a certain impact on the future income of the EPC owner.(ii)Risk of imperfect EPC contractor qualification management (C32): on the one hand, since China's power engineering EPC management mode has just started, there are no separate relevant provisions on EPC contractor qualification management currently in China. On the other hand, only a few large power design institutes have the ability to become EPC general contractors in China. These make the PTTPs’ investors have great limitations and risks in the selection of EPC general contractor.(iii)Risk of mismatched professional qualification management (C33): at present, there is no management system for the professional qualification of EPC project managers in China. Most of the employees engaged in EPC project management have transferred from management personnel, supervision engineers, or other teams. Although this personnel has certain management practical experience, they have great limitations. This situation brings certain risks to the EPC owner.
3.4. Social and Environmental Risks (C4)
(i)Downstream grid acceptance risk (C41): the instability of new energy power generation may affect the safe operation and maintenance of the power system and lead to the downstream grid acceptance risk(ii)Social attitude risk (C42) [34]: PTTPs usually involve multiple regions. So obtaining the support from the regional governments and local people will be beneficial to the implementation of PTTPs, which is also the responsibility and risk that the EPC owner needs to bear(iii)Force majeure risk (C43) [35]: due to the occurrence of natural and nonnatural factors of force majeure, PTTPs cannot operate normally or seriously affect the new energy grid connection, which may reduce the investors’ income and even loss of principal
4. Investment Risk Evaluation Model of PTTPs
Investment risks of PTTPs in the EPC mode under a new energy grid connection refer to the hazards that may occur in the phases of design, construction, procurement, and commissioning. The participants involved in these risks are mainly the EPC owner, EPC general contractor, and EPC supervision team. In order to fully collect the opinions of experts and scientifically calculate the risk level of PTTPs, this paper uses the following methods to construct the investment risk assessment model.
4.1. HFLTS-TFN Method
Due to the complexity and fuzziness of investment risk assessment, experts may hesitate to determine one index among several evaluation terms in the process of analysis and evaluation. So HFLTS is introduced to accurately reflect experts’ judgments to solve this problem. However, the qualitative linguistic terms of HFLTS cannot directly reflect the evaluation values and perform mathematical operations. Thus, this paper combines the TFN method and the HFLTS method to process evaluation data quantitatively. The HFLTS-TFN method is constructed as follows.
Definition 1. Let be a set of linguistic terms. If consists of a set of finite sequential linguistic terms from S, the is a hesitant fuzzy linguistic term on S. The mathematical form of HFLTS is where function represents the possible membership degree of to set [36].
Definition 2. These linguistic terms satisfy the following conditions:(i)Order: if , then (ii)Maximum operator: if then (iii)Minimum operator: if then
Definition 3. When experts make an initial evaluation of one index, these evaluation values may be between two or multiple linguistic terms. In this case, the operation of HFLTS can be expressed by function .
Definition 4. Set the membership function. Suppose a mapping from U to [0, 1] is defined on the universe U, and is called a fuzzy set on U. And is called the membership function. TFN can be expressed as , and the membership function is shown in the following equation:
Definition 5. and are both TFNs, and the related operations are as follows:The linguistic terms are evenly distributed into seven finite ordered sets in this paper, but the qualitative linguistic terms cannot directly reflect the evaluation values and mathematical operations. Therefore, this paper converts linguistic information into TFNs, and linguistic variables and their corresponding TFNs are shown in Table 3. For further understanding, an example of the corresponding distribution of HFLTS-TFN is shown in Figure 2.
4.2. Determination of Index Weights by the Entropy Method
The entropy method is an available decision-making method to determine the index weights from the data themselves of PTTP indexes. The smaller the information entropy is, the larger the index weight [37]. The detailed steps are shown as follows: Step 1: calculate the entropy of the jth index in equation (10), where represents the jth index valuation of PTTP from the ith expert. Let is the mean value of the jth index, it satisfies the equation (5), where is the ith expert’s weight. Step 2: calculate the jth index weight of PTTP by the following equation:
4.3. Investment Risk Assessment of PTTPs
Fuzzy comprehensive evaluation provides a way to evaluate fuzzy variables so as to satisfy the uncertainty and difficulty in quantifying the influence degree of different risk factors [38]. In this paper, the fuzzy comprehensive evaluation method based on HFLTS-TFN is adopted to realize the comprehensive assessment of PTTP investment risk by aggregating expert evaluation opinions. The process is mainly divided into the following steps: Step 1: A first-level index evaluation vector is established. The TFNs of all second-level indexes under each first-level index constitute the evaluation vector . The is expressed as , which is an ordered finite subset of the continuous linguistic set. Among them, the variable represents the possible membership degrees of belonging to the set . Step 2: calculate the investment risk of each first-level index. The expression is the total weight of the second-level index under the first-level index, and the expression is the weight of the jth second-level index under the ith first-level index. The calculation steps are as follows: Step 3: sort out investment evaluation risk value of each first-level index Step 4: calculate the overall investment risk level, where the expression Wi is the weight of the first-level index, and the expression is the weight of the ith first-level index. The calculation is as follows: Step 5: calculate the investment risk level of PTTP. According to equation (10), the higher the similarity between one judgment result and one risk level, the closer the project is to the level. and are two TFNs.
5. A Case Study
5.1. Case Statement
Gansu power grid is located in the centre of the northwest power grid and operates with Xinjiang Province, Qinghai Province, Ningxia Province, and Shaanxi Province, so it is the power exchange hub of the northwest power grid. The exploitable amount of new energy in Gansu Province ranks at the forefront of China, especially wind power generation and photovoltaic power generation. And the new energy resources in Gansu Province are mainly concentrated in the Hexi Corridor region, including Jiuquan City, Zhangye City, Wuwei City, Jinchang city, and Jiayuguan City, as shown in Figure 3. Thus, the Hexi Corridor region has good development prospects.
Despite the rapid development of new energy in Gansu Province, the consumption contradiction is prominent due to the insufficient consumption capacity and the limited transmission capacity. For solving this dilemma, it is proposed to adopt the EPC mode to carry out PTTPs under a new energy grid connection in the Hexi Corridor region, Gansu Province, so as to further improve the delivery capacity and the utilization rate of new energy resources. In order to achieve the expected investment objectives and returns, the EPC owner needs to conduct the risk assessment of PTTPs in the initial stage.
5.2. Process and Results
In order to ensure the accuracy of data collection, this paper refers to the method from existing literature. Four experts in relevant fields were invited, who have the following characteristics: (i) having relevant research on new energy grid connection or PTTPs; (ii) studying or engaging in EPC projects; (iii) having certain knowledge of the risk assessment field. According to the knowledge level and work experience of the four experts, the experts’ evaluation weights are 0.3, 0.3, 0.2, and 0.2, respectively. For the 13 risk evaluation indexes of PTTPs, experts are invited to evaluate and score separately. Next, interview and collect experts’ opinions, and convert the experts’ evaluation linguistic terms into HTLTS results according to Section 4.1, as shown in Table 4. Then, according to the HTLTS-TFN calculation rules in Table 3, the experts’ HTLTS results are numerically calculated, as shown in Table 5.
This paper uses the entropy method in order to scientifically measure the weight of each index. First, we calculate the corresponding index values into equation (5) to obtain the average value of each index. Second, the entropy of each index is calculated by equation (4). Third, the weight of each index is calculated according to equation (6), as shown in Table 6. It can be found that C21 is the index with the largest weight value, and C12 is the index with the smallest weight value.
The first-level index of PTTPs in EPC mode is evaluated according to equation (10), and the calculation process is as follows:
Based on the above results, the risk level of the first-level index RC2 is the highest, and the risk level of the first-level indicator RC1 is the lowest. And the overall investment risk value of PTTPs is calculated below. And calculate the similarity of each investment risk level value, and the result is shown in Figure 4. Obviously, the overall investment risk evaluation of PTTPs is the closest to the L; that is, the investment risk of PTTPs in Hexi Corridor is at a low level.
5.3. Sensitivity Analysis
This paper conducts sensitivity analysis from index weight fluctuations to test the stability of the risk assessment result. Four first-level indexes, namely economic risks (C1), management risks (C2), policy risks (C3), and social and environmental risks (C4), all take 10%, 20%, and 30% less and more weight than the initial weights. And the purpose is to observe if the results change significantly when index weights change. The variation results of each investment risk level similarity are shown in Figure 5.
(a)
(b)
(c)
(d)
When the C1 or C3 weight fluctuation is from small to large, the similarity values of L gradually increase. On the contrary, the similarity values of ML decrease as C2 increases. And the calculated similarity values of VL and ML are nearly equal, when the weight of C1 increases to about 10% of the original value, the weight of C3 increases to about 20% of the original value, or the weight of C2 decreases to 10% of the original value.
Taking a closer look at the sensitivity analysis results, L always has the highest similarity value; VH always has the lowest similarity value. The similarity values from large to small, M, MH, and H are always the fourth, fifth, and sixth. The change of first-level index weights has no significant influence on the final assessment result to a certain extent. Therefore, the results in Section 5.2 are relatively scientific by using this model.
5.4. Suggestions
Although the overall risk of PTTPs project is at a low level, management risks (C2) are still severe. Therefore, corresponding preparation measures should be taken in advance for the performance risk of the EPC general contractor (C21), the performance risk of the EPC supervision team (C22), and the contract management risk (C23).
As for the performance risk of the EPC general contractor (C21), the owner should treat it with caution due to its great impact. The qualification, reputation, and performance of the general contractor can reflect the performance ability of the general contractor to a certain extent. Therefore, the owner shall have a comprehensive understanding of the bidder’s qualification conditions in terms of qualification, reputation, and performance in the bidding documents, so as to reduce the general contract performance risk. In addition, the owner shall fully communicate with the general contractor when setting the contract terms. The owner shall clearly stipulate the default behavior of the general contractor and its corresponding penalty classification and consider using the contract mechanism to realize risk sharing.
For performance risk of the EPC supervision team (C22), the owner shall first ensure full communication with the supervision team in the whole process of project implementation. Under the EPC mode, the supervision team should have all-around professional construction management capabilities such as design, construction, and electromechanical equipment. The owner shall make targeted agreements on the qualification conditions of the bidder in the supervision bidding documents, including qualification, reputation, and performance. The owner can also make statistics according to the previous work contents and effects from the members of the supervision team, so as to judge whether the bidding team is competent for the corresponding work.
In terms of contract management risk (C23), the owner shall invite a full-time project team to be responsible for the negotiation of the contract and the formulation of terms. In addition, relevant contract experts in the industry are also employed to carry out special training on contract management for contract management personnel to improve awareness and professionalism of contract management. When encountering difficulties, relevant contract experts can be invited to assist the EPC owner in solving contract management problems.
6. Conclusions
With the rapid development of new energy grid connection, the construction of PTTPs becomes more difficult, and the cost tends to be higher. The limitations of the traditional management mode of PTTPs become more prominent. This paper holds that adopting the EPC mode is an effective attempt to develop the PTTPs for power grid corporations. The cognition of power grid corporations on risks of EPC mode is a necessary factor in whether they will adopt EPC mode. This paper proposes that the risks faced by the EPC owner in PTTPs cannot be ignored by analyzing the typical risk factors of PTTPs. Thus, the EPC owner should carry out the risk assessment as soon as possible in the initial study stage of the PTTPs under the new energy grid connection. The main contributions of this paper are as follows:(i)Considering the characteristics of the new energy grid connection and EPC mode, this paper establishes the novel evaluation index system of PTTP risk assessment(ii)This paper establishes a comprehensive model of PTTP risk assessment from the perspective of the owner based on the HFLTS-TFN, the entropy method, and fuzzy comprehensive evaluation(iii)A case Hexi Corridor, Gansu Province testified using sensitivity analysis supports the applicability of the proposed risk assessment model. And the model can provide a reference for investors to make more reasonable investment decisions in PTTPs using EPC mode.
Finally, this paper still has some limitations. Due to the limitation of the research topic, this paper only considers the risks that EPC owners could face in PTTPs. In the future, we will also pay attention to the risks of PTTPs from the perspective of EPC contractors in China.
Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Acknowledgments
This research was supported by the National Key R&D Program of China (No. 2021YFE0102400), the China Postdoctoral Science Foundation (No. 2020M680488), and the Fundamental Research Funds for the Central Universities (Nos. 2021MS022 and 2021PT013).