Sustainability evaluation of regional microgrid interconnection system is conducive to a profound and comprehensive understanding of the impact of interconnection system projects. In order to realize the comprehensive and scientific intelligent evaluation of the system, this paper proposes an evaluation model based on combination entropy weight rank order-technique for order preference by similarity to an ideal solution (TOPSIS) and Niche Immune Lion Algorithm-Extreme Learning Machine with Kernel (NILA-KELM). Firstly, the sustainability evaluation indicator system of the regional microgrid interconnection system is constructed from four aspects of economic, environmental, social, and technical characteristics, and the evaluation indicators are explained. Then, the classical evaluation model based on TOPSIS is constructed, and the entropy weight method and rank order method (RO) are coupled to obtain the indicator weight. The niche immune algorithm is used to improve the lion algorithm, and the improved lion algorithm is used to optimize the parameters of KELM, and the intelligent evaluation model based on NILA-KELM is obtained to realize fast real-time calculation. Finally, the scientificity and accuracy of the model proposed in this paper are verified. The model proposed in this paper has the lowest RMSE, MAE and RE values, indicating that its intelligent evaluation results are the most accurate. This study is conducive to the horizontal comparison of the overall performance of regional microgrid interconnection system projects, helps investors to choose the most promising project scheme, and helps the government to find feasible project.
Different from the traditional centralized power supply mode, the microgrid power supply mode is more flexible, and its unique duality of electricity and power supply can play a very important role in the power system [
The definition of sustainability can be understood from four relevant dimensions: economy, environmental protection, technology, and society [
As a necessary link in the development and construction of regional microgrid interconnection system projects, sustainability evaluation is an important part of the whole project life cycle management process. Due to the development of regional microgrid interconnection systems in China is not yet mature, the research on the sustainability evaluation of regional microgrid interconnection systems is very limited. The performance of regional microgrid interconnection systems in environmental, social, economic, and technological aspects has become the core concern of local governments and potential investors [
The construction of the evaluation model has an important influence on the accuracy of evaluation. Ideal point order preference by similarity to ideal solution (TOPSIS) is a multi-objective decision-making method, which has been widely used in different areas of comprehensive evaluation research [
The extreme learning machine theory was proposed by Huang et al. in 2006, based on which the basic extreme learning machine, online sequential extreme learning machine, extreme learning machine with the kernel (KELM), and other related algorithms have been derived [
In summary, this paper constructs a regional microgrid interconnection system sustainability intelligent evaluation model based on combination entropy weight rank order-TOPSIS and NILA-KELM. The rest of the paper is arranged as follows. The second part designs the evaluation indicator system of regional microgrid interconnection system sustainability from four aspects of economic, environmental, social, and technical characteristics, and explains the evaluation indicators. The third part constructs the classical evaluation model based on the combination entropy weight rank order method and TOPSIS and constructs the intelligent evaluation model based on NILA optimized KELM. The fourth part selects practical cases to verify the accuracy and effectiveness of the model proposed in this paper. The fifth part summarizes the research results. The fifth part summarizes the research results.
Establishing a scientific and reasonable evaluation indicator system is the premise of regional microgrid interconnection system sustainability evaluation. The established indicator system should include various factors that have a key impact on the sustainability of regional microgrid interconnection system. The evaluation indicator system should be established on the basis of the following selection principles [ The comprehensive principle
The establishment of the evaluation indicator system aims to comprehensively measure the performance of various aspects involved in the regional microgrid inter-connection system project, so as to provide reference for potential investors to judge the advantages and disadvantages of the regional microgrid interconnection system project and the local government to formulate relevant policies. The Regional microgrid interconnection system project consists of multiple sub-microgrids, and each sub-microgrid contains a variety of distributed power, energy storage and other facilities. The operation management and related impacts of these system components will affect the benefit performance of regional microgrid interconnection system projects. The evaluation indicator system established in a specific environment should include economic impact, environmental impact and social impact. Therefore, to have a more profound and specific understanding of the research object, a more comprehensive evaluation indicator system should be established.
Hierarchical principle
The Hierarchical structure is one of the most widely used structures in both theory and practice among many benefit evaluation structures. Establishing an evaluation indicator system, hierarchical structure can systematically, accurately and comprehensively show the subordinate relationship and correlation between the sub-level evaluation indicator and each level.
Scientific principle
The selection of evaluation indicators should be based on the objective laws and related theories, supported by corresponding scientific theories, and studied from the perspectives of practical experience and theoretical knowledge. The selection of evaluation indicators should fully consider the objective environment, to scientifically express the essential characteristics of the research object.
Representative principle
The evaluation indicator system consists of a series of sub-level indicators at different levels. Using a large number of indicators to construct the evaluation indicator system may not necessarily improve the accuracy of system evaluation. Therefore, representative indicators closely related to the research object should be selected to describe the characteristics of various aspects, fully reflecting the purpose and significance of the selection of indicators. If excessive evaluation indicators are selected in the research process, it may reduce and weaken the influence of important indicators and reduce the reliability of evaluation results.
Independence principle
The evaluation system is a hierarchical expression set of multiple representative single indicators. In the selection of indicators, repeated or highly correlated indicators should be avoided. Each indicator should be independent, and each indicator can express all aspects of the research problem from different perspectives, so as to improve the scientificity and effectiveness of the evaluation results.
This section takes the selection criteria of indicators listed in
Target level | Criterion layer | Project hierarchy | Indicator number | Indicator type I | Indicator type II |
---|---|---|---|---|---|
Sustainability benefit evaluation of regional microgrid interconnected multi-microgrid system | Economic indicators | Investment cost | E1 | Quantitative | Cost |
Operation and maintenance costs | E2 | Quantitative | Cost | ||
Payback period | E3 | Quantitative | Cost | ||
The levelized cost of electricity | E4 | Quantitative | Cost | ||
Environmental protection indicators | CO2 emission reduction | P1 | Quantitative | Benefit | |
PM2.5 emission reduction | P2 | Quantitative | Benefit | ||
Polluted gas emissions | P3 | Qualitative | Cost | ||
Space occupancy | P4 | Quantitative | Cost | ||
Social indicators | Resident satisfaction | S1 | Quantitative | Benefit | |
Job opportunities | S2 | Qualitative | Benefit | ||
Poverty rate decreases | S3 | Quantitative | Benefit | ||
Technical indicator | Technological innovation | T1 | Qualitative | Benefit | |
Electricity consumption reliability | T2 | Quantitative | Benefit | ||
Renewable energy utilization ratio | T3 | Quantitative | Benefit | ||
System expansibility | T4 | Quantitative | Benefit |
The economic indicators mainly investigate the economic performance of regional microgrid interconnection system, including four sub-level indicators, namely construction cost, operation and maintenance cost, investment payback period and unit power cost.
Investment cost
The investment cost of the project mainly refers to the cost of obtaining land and related equipment in the initial construction process of the regional microgrid interconnection system project. Mainly includes land costs, equipment (such as wind turbines, photovoltaic panels, diesel generators, energy storage systems and related ancillary facilities) acquisition costs and labor costs.
Operation and maintenance costs
The operation and maintenance cost mainly refers to the cost needed to maintain the normal operation of the system and ensure the continuous operation of the equipment in the daily management of the regional microgrid interconnection system. Mainly includes the salary of operation management personnel, the cost of parts or spare parts for regular maintenance.
Payback period
The payback period of investment refers to the number of years used when the accumulated economic benefits of regional microgrid interconnection system projects are equal to the initial investment costs. Investment payback period is one of the most concerned economic indicators for project investors. By calculating the investment payback period, the time for capital recovery can be clarified, thereby reducing investment risks.
The levelized cost of electricity
The normalized electricity cost is an evaluation indicator for the net present value of unit electricity cost in the whole life cycle of the regional microgrid interconnection system project. From the perspective of the project life cycle, this indicator evaluates the economic benefits of regional microgrid interconnection system project, which is one of the important reference indicators in the energy industry.
The development of the regional microgrid interconnection system has a certain degree of complexity in its environmental impact. On the one hand, wind power and photovoltaic power generation in the regional microgrid interconnection system are clean power sources, which are of positive significance to environmental protection; On the other hand, when the system is in the peak load period, the use of diesel generators equipped in the system for energy supply will produce carbon dioxide and polluting gases, adversely affecting the environment.
CO2 emission reduction
Carbon dioxide emission reduction refers to the reduction of greenhouse gas emissions by the system through the use of renewable energy power plants in the life cycle of regional microgrid interconnection system, compared with the use of traditional power generation forms of thermal power plants. The reduction of carbon dioxide emissions is of positive significance to alleviate global warming. The reduction of carbon dioxide emissions is of positive significance to alleviate global warming.
PM2.5 emission reduction
PM2.5 emission reduction refers to the reduction of particulate matter less than 2.5 mm in diameter in the atmosphere through the use of renewable energy in the regional microgrid interconnection system compared with the use of traditional power generation in thermal power plants. This indicator is one of the key indicators to evaluate the quality of air quality.
Pollutant gas emission
The emission of polluting gases is mainly used to evaluate the emission of polluting gases generated by burning fossil fuels such as sulfur dioxide and nitrogen oxides during the life cycle of regional microgrid interconnection system when the load is in the peak period when the diesel generator in the system works. Pollutant gas emissions will have a bad impact on the environment, such as acid rain, which will damage the ecology and seriously affect the sustainability of the system.
Space occupancy
Space occupancy refers to the relevant equipment units in the regional microgrid interconnection system, such as wind turbines, photovoltaic panels, diesel generators, energy storage batteries, and the layout of corresponding transmission lines and control equipment will occupy a certain user area, which will have a certain impact on the residents and the surrounding environment and landscape.
The implementation of the regional microgrid interconnection system project can bring about positive social effects, such as increasing residents’ satisfaction, increasing employment opportunities, and reducing poverty rates.
Increase resident satisfaction
Residents’ satisfaction is the degree of satisfaction to local residents through the use of regional microgrid interconnection system for energy supply. The improvement of residents’ satisfaction is one of the necessary indicators for system implementation. The level of residents’ satisfaction has a direct impact on their enthusiasm to participate in regional microgrid interconnection system and project implementation.
Increase employment opportunities
Increased employment opportunities refer to direct and indirect employment opportunities arising from the implementation and management of regional microgrid interconnection system projects. Direct employment refers to the staff created by the project itself, such as those engaged in manufacturing, installation or operation and maintenance of the system. Indirect employment refers to the increase in job opportunities for relevant personnel who provide investment and knowledge for regional microgrid interconnection systems, such as banks providing funds for project construction and suppliers providing necessary materials.
Reduce poverty rate
The construction and implementation of regional microgrid interconnection system can not only create local employment opportunities, but also play a positive role in promoting local economic development. These impacts have a positive impact on poverty reduction, particularly in remote areas. Current society is paying more and more attention to human development. On 03 November 2013, the concept of “targeted poverty alleviation” first proposed poverty reduction as a policy. Therefore, taking poverty reduction as a project evaluation indicator has responded well to the Government's policy call.
As an emerging technology, the construction of a new regional microgrid interconnection system project can actively explore the actual practice of the project, promote the accumulation of experience and theory at the technical level, and provide effective reference for the gradual maturity of relevant technologies, which is conducive to the further development of intelligent power grid layout.
Technological innovation
The construction, operation and management technology of regional microgrid interconnection system is not fully mature. The development of new regional microgrid interconnection system will play a positive role in promoting the accumulation of relevant experience and is conducive to the formation of relevant technological innovation.
Electricity consumption reliability
The intermittent of renewable energy in regional microgrid interconnection system poses a serious challenge to the reliability of users’ electricity consumption and the frequent access of renewable energy will also threaten the stability of power system. The energy storage system in the regional microgrid interconnection system and the power mutual assistance mechanism in the system can effectively buffer the intermittent characteristics of renewable energy, improve the reliability of power supply, reduce the dependence of regional microgrid interconnection system on large power grid, and effectively improve the safety and stability of power system.
Renewable energy utilization ratio
Due to the randomness of renewable energy power output, a single microgrid has certain limitations on the consumption ability of renewable energy. Through the regional microgrid interconnection system, the excess renewable energy power that cannot be consumed in the sublevel microgrid can be traded to the sublevel microgrid with energy shortage in the form of mutual assistance within the system, so as to reduce the operation cost of the supply and demand sides, improve the utilization rate of renewable energy, and reduce the interference on the distribution network side.
System expansibility
Regional microgrid interconnection system is a joint optimal dispatching system composed of several separate sub-microgrids. With the continuous development of power system intelligence, there may be new sub-level microgrids merging into the regional microgrid interconnection system in the future, or realizing system-level interconnection between regional microgrid interconnection systems. In addition, in special cases, such as system failures, in order to ensure the stability of the overall operation of the interconnected microgrid system, the system structure needs to have the possibility of flexible integration and removal of the sub-level microgrid.
The specific steps of RO method are as follows [ Data normalization processing
There are various types of indicators, including maximal indicators, minimal indicators and interval indicators. Maximal indicators reflect the development trend of evaluation indicators. The larger the indicator, the higher the level of sustainable development; On the contrary, minimal indicators indicate that the larger the indicator, the lower the level of sustainable development. To facilitate the calculation and analysis, the values of each index must be consistent, so as to facilitate the comparison and selection of each scheme, and then get the evaluation results of each scheme. After analyzing the index system constructed in this paper, it is found that all evaluation indexes are maximal indicators.
Since the attribute and quantity level of the original data are quite different, it is necessary to normalize each index. Set the evaluation sample set as The average value of each evaluation index sample Calculate the weight of each index
Calculate the information entropy and weight of each indicator separately [
Combined weights
After the weight value of the evaluation index is obtained by the combination weight calculation method mentioned above, the sustainability of the regional microgrid interconnection system project can be evaluated and analyzed by the ideal point sorting method. The specific steps are as follows [ Weighted processing of evaluation data
According to the original evaluation data, the weighted normalization matrix needed for the operation of the evaluation model can be obtained by the following formula:
Obtaining positive ideal solution and negative ideal solution by operation
The positive ideal solution refers to the hypothetical value of the optimal value in the ideal situation, while the negative ideal solution refers to the virtual solution of the worst value. The specific calculation formula is:
In the above formula,
After the above steps, the Euclidean distance can be calculated according to the following formula:
Calculating relative approach degree
Relative approach degree is calculated according to the following formula:
In the calculation process discussed above, the distance between each index and positive and negative ideal solutions in the evaluation index system can be calculated, and the corresponding Euclidean distance represents the degree of closeness of the sustainability and ideal level of the regional microgrid interconnection system project. Therefore, the sustainability of regional microgrid interconnection system projects can be described by the operational distance and then sorted by the calculated distance. The smaller the distance is, the closer the sustainability of the regional microgrid interconnection system project is to the ideal degree. At the same time, it indicates that the regional microgrid interconnection system project has better sustainability and the rank ordering will be at the top.
LA is a bionic algorithm based on the social behavior of lions, which was proposed by Rajakumar of B. R. in 2012 [ Generate population
In the initial stage of the algorithm, the lion group is initialized as Mating
In the process of iteration and searching for the optimal solution, mating is a very effective method. It can generate new solutions through existing solutions. Mating operations achieve the purpose of updating the lion group and maintaining the stability of the lion group by crossing, mutation, clustering, killing sick, weak cubs and other steps.
The mating step introduces the crossover method based on double probabilities (crossover with two different probabilities).
By mating, the generated
Mutation operation is to use probability for random mutation to produce pups. After the crossover and mutation were completed, the number of cub species was 8. Clustering uses the K-means method to group the existing 8 solutions into gender groups: male cubs (
Finally, by testing the health status (or target), kill a larger group of thin individuals to ensure the balance of the two groups of cubs, and finally achieve the purpose of renewing the population. And after the population update is completed, the age of the cubs is initialized to 0.
Territorial defense
In the breeding process of lion groups, they will be attacked by nomadic lions. At this point, in order to protect the cubs and continue to occupy the territory, the lion will defend the attack of the nomadic lions, as shown in
In the process of territorial defense, first use the method of generating territorial lions to generate nomadic lions
Let Territorial takeover
In the territorial takeover stage, the best solution between the female lion and the male lion is searched for, replacing the inferior solution, and mating until the termination conditions are reached. The replacement process is as follows:
First follow the following criteria:
Choose the best male lion (
Let
After completing the above steps, return to Step 2 until the termination condition. The best lion in the repeating population of the whole process is selected as the optimal solution.
Lion swarm algorithm is a parallel search method that does not depend on specific problems. It has the characteristics of self-adaptability, group search and heuristic random search. However, in the process of multiple iterations, the individuals with large fitness in the population will form ‘inbreeding’, resulting in premature and reduced diversity. To solve this problem, the niche immune algorithm is introduced to limit the excessive replication of similar individuals, so as to ensure the diversity of population. The detailed steps of niche immune algorithm are shown in reference [ In the iteration, each certain algebra dl is centered on the position of each lion, and M lion is cloned according to the size of its objective function value. The cloning formula is as follows:
In the formula, After cloning, do single parent mutation for M lion, for the lion with lower objective function value, use monosexual reproduction lion to mutation operation, the mathematical formula is as follows:
In the formula, After cloning and mutating the lion,
First, briefly explain the neural network construction mechanism of the basic ELM algorithm, and its neural network function is as follows [
ELM ensures the accuracy of regression prediction by minimizing the output error, as shown below:
At the same time, ELM algorithm ensures the generalization ability of neural network by minimizing the output weight
In the formula:
For KELM algorithm, the kernel function is introduced to obtain better regression prediction accuracy. Mercer's condition is used to define the kernel matrix, as follows:
The kernel matrix
The parameter 1/C is added to the main diagonal in the unit diagonal matrix HHT so that the characteristic root is not zero, and then the weight vector
From the above formula, the output of the KELM model is:
In the kernel-based KELM algorithm, the specific form of the feature mapping function
Based on the establishment of the evaluation index system, this section proposes a new hybrid evaluation method based on combined entropy weight rank order-TOPSIS and NILA-KELM, which uses combined entropy weight rank order-TOPSIS to obtain classic evaluation results. Then use the NILA algorithm to optimize KELM, so as to obtain the optimal value of the KELM model parameters, and finally get the evaluation result and analyze the result. The proposed hybrid evaluation framework is shown in
The specific steps are as follows:
Step 1: Carry out initial input variable selection and data preprocessing. Based on the established evaluation index system, an initial set of input variables is formed, and the original data of each input factor is quantified and standardized.
Step 2: Classical evaluation calculation. The combined entropy weight rank order is used to obtain the evaluation index weight, and then it is substituted into the TOPSIS evaluation model to obtain the classic evaluation result.
Step 3: Use NILA algorithm optimization to obtain the optimal value of the KELM model parameters. The key parameters of the KELM model will have a significant impact on its final evaluation effect, and it is related to the accuracy of the sustainability evaluation of the regional microgrid interconnection system project. If the stop condition is not reached, the algorithm needs to be rerun to obtain the corresponding optimized solution set.
Step 4: Output intelligent evaluation results and compare and analyze the results. Based on the above-obtained intelligent evaluation model for the sustainability of the regional microgrid interconnection system project, the simulation calculation is carried out, and the obtained intelligent evaluation results are compared with the classical evaluation model calculation results.
Through field research and data collection, the relevant data of 30 regional microgrid interconnection system projects were collected and sorted out. At the same time, 25 experts were invited to score the indicators based on the interval scores, and then the scores were summarized and averaged.
According to the evaluation index preprocessing method in
Index | M1 | M2 | M3 | … | M15 | M16 | … | M28 | M29 | M30 |
---|---|---|---|---|---|---|---|---|---|---|
E1 | 1.0000 | 0.7143 | 0.5306 | … | 0.8163 | 0.4490 | … | 0.5714 | 0.1224 | 0.3265 |
E2 | 0.7400 | 0.7600 | 0.1000 | … | 0.2200 | 0.7000 | … | 0.8000 | 0.5200 | 0.3200 |
E3 | 0.9545 | 0.8182 | 0.6364 | … | 0.6818 | 0.8409 | … | 0.6136 | 0.0909 | 0.8409 |
E4 | 0.0000 | 0.3913 | 0.9783 | … | 0.1087 | 0.9565 | … | 0.7174 | 0.8261 | 0.0000 |
P1 | 0.8800 | 0.9800 | 0.8600 | … | 0.9600 | 0.0000 | … | 0.3600 | 0.5600 | 0.9600 |
P2 | 0.3231 | 1.0000 | 0.8000 | … | 0.8615 | 0.3692 | … | 0.0000 | 0.0154 | 0.9846 |
P3 | 1.0000 | 0.7872 | 0.5957 | … | 0.7660 | 0.3830 | … | 0.7234 | 0.3404 | 0.9149 |
P4 | 0.8200 | 0.4800 | 0.7800 | … | 0.2200 | 0.4200 | … | 0.4400 | 0.7200 | 0.8600 |
S1 | 0.0625 | 0.8333 | 0.2083 | … | 0.9792 | 0.2292 | … | 0.3750 | 0.8958 | 1.0000 |
S2 | 0.3556 | 0.7111 | 0.4222 | … | 0.4667 | 0.7556 | … | 0.0444 | 0.4000 | 0.5556 |
S3 | 0.0426 | 0.1702 | 0.1489 | … | 0.9362 | 0.6383 | … | 0.7234 | 0.8936 | 0.4894 |
T1 | 0.7708 | 0.8542 | 0.1875 | … | 0.2083 | 0.5625 | … | 0.0000 | 0.2708 | 0.6875 |
T2 | 0.7083 | 0.7083 | 1.0000 | … | 0.1667 | 0.0000 | … | 0.2708 | 0.8958 | 0.8958 |
T3 | 0.1042 | 0.3542 | 0.7292 | … | 0.8958 | 0.9583 | … | 0.3750 | 0.5417 | 0.0833 |
T4 | 0.2600 | 0.3800 | 0.8200 | … | 0.3400 | 0.0400 | … | 0.5400 | 0.7400 | 0.4200 |
According to the weight determination method described in
Index number | RO weight | Entropy weight | Combination weight |
---|---|---|---|
E1 | 0.0167 | 0.0733 | 0.0190 |
E2 | 0.0083 | 0.0693 | 0.0090 |
E3 | 0.1083 | 0.0608 | 0.1027 |
E4 | 0.0417 | 0.0686 | 0.0445 |
P1 | 0.1167 | 0.0653 | 0.1186 |
P2 | 0.1250 | 0.0540 | 0.1052 |
P3 | 0.0833 | 0.0684 | 0.0889 |
P4 | 0.0333 | 0.0690 | 0.0359 |
S1 | 0.0500 | 0.0891 | 0.0694 |
S2 | 0.1000 | 0.0513 | 0.0800 |
S3 | 0.0750 | 0.0786 | 0.0918 |
T1 | 0.0250 | 0.0693 | 0.0270 |
T2 | 0.0917 | 0.0637 | 0.0910 |
T3 | 0.0667 | 0.0663 | 0.0688 |
T4 | 0.0583 | 0.0530 | 0.0482 |
As can be seen from the data in
Then, the initial data are weighted to obtain the weighted normalization matrix, and the calculation results are shown in
Index | M1 | M2 | M3 | … | M15 | M16 | … | M28 | M29 | M30 |
---|---|---|---|---|---|---|---|---|---|---|
E1 | 0.0190 | 0.0136 | 0.0101 | … | 0.0155 | 0.0085 | … | 0.0109 | 0.0023 | 0.0062 |
E2 | 0.0067 | 0.0068 | 0.0009 | … | 0.0020 | 0.0063 | … | 0.0072 | 0.0047 | 0.0029 |
E3 | 0.0980 | 0.0840 | 0.0653 | … | 0.0700 | 0.0863 | … | 0.0630 | 0.0093 | 0.0863 |
E4 | 0.0000 | 0.0174 | 0.0436 | … | 0.0048 | 0.0426 | … | 0.0320 | 0.0368 | 0.0000 |
P1 | 0.1044 | 0.1163 | 0.1020 | … | 0.1139 | 0.0000 | … | 0.0427 | 0.0664 | 0.1139 |
P2 | 0.0340 | 0.1052 | 0.0841 | … | 0.0906 | 0.0388 | … | 0.0000 | 0.0016 | 0.1036 |
P3 | 0.0889 | 0.0700 | 0.0529 | … | 0.0681 | 0.0340 | … | 0.0643 | 0.0303 | 0.0813 |
P4 | 0.0294 | 0.0172 | 0.0280 | … | 0.0079 | 0.0151 | … | 0.0158 | 0.0258 | 0.0308 |
S1 | 0.0043 | 0.0578 | 0.0145 | … | 0.0680 | 0.0159 | … | 0.0260 | 0.0622 | 0.0694 |
S2 | 0.0284 | 0.0569 | 0.0338 | … | 0.0373 | 0.0604 | … | 0.0036 | 0.0320 | 0.0444 |
S3 | 0.0039 | 0.0156 | 0.0137 | … | 0.0860 | 0.0586 | … | 0.0664 | 0.0821 | 0.0449 |
T1 | 0.0208 | 0.0230 | 0.0051 | … | 0.0056 | 0.0152 | … | 0.0000 | 0.0073 | 0.0185 |
T2 | 0.0645 | 0.0645 | 0.0910 | … | 0.0152 | 0.0000 | … | 0.0246 | 0.0815 | 0.0815 |
T3 | 0.0072 | 0.0244 | 0.0502 | … | 0.0616 | 0.0660 | … | 0.0258 | 0.0373 | 0.0057 |
T4 | 0.0125 | 0.0183 | 0.0395 | … | 0.0164 | 0.0019 | … | 0.0260 | 0.0357 | 0.0202 |
The positive and negative ideal solutions are calculated as shown in
Index number | Positive | Negative |
---|---|---|
E1 | 0.0190 | 0.0000 |
E2 | 0.0090 | 0.0000 |
E3 | 0.1027 | 0.0000 |
E4 | 0.0445 | 0.0000 |
P1 | 0.1186 | 0.0000 |
P2 | 0.1052 | 0.0000 |
P3 | 0.0889 | 0.0000 |
P4 | 0.0359 | 0.0000 |
S1 | 0.0694 | 0.0000 |
S2 | 0.0800 | 0.0000 |
S3 | 0.0918 | 0.0000 |
T1 | 0.0270 | 0.0000 |
T2 | 0.0910 | 0.0000 |
T3 | 0.0688 | 0.0000 |
T4 | 0.0482 | 0.0000 |
The Euclidean distances from the microgrid interconnection system projects in each region to the positive and negative ideal solutions are calculated as shown in
Regional microgrid interconnection system project | Euclidean distance to positive ideal solution | Euclidean distance to negative ideal solution | Regional microgrid interconnection system project | Euclidean distance to positive ideal solution | Euclidean distance to negative ideal solution |
---|---|---|---|---|---|
M1 | 0.1666 | 0.1910 | M16 | 0.1929 | 0.1556 |
M2 | 0.1090 | 0.2232 | M17 | 0.1843 | 0.1711 |
M3 | 0.1256 | 0.2031 | M18 | 0.1705 | 0.1597 |
M4 | 0.1763 | 0.1987 | M19 | 0.1873 | 0.1603 |
M5 | 0.0741 | 0.2409 | M20 | 0.1846 | 0.1590 |
M6 | 0.1892 | 0.1524 | M21 | 0.1703 | 0.1623 |
M7 | 0.0963 | 0.2406 | M22 | 0.1921 | 0.1820 |
M8 | 0.2014 | 0.1442 | M23 | 0.1561 | 0.1608 |
M9 | 0.2077 | 0.1366 | M24 | 0.1957 | 0.1544 |
M10 | 0.1146 | 0.2156 | M25 | 0.1510 | 0.1872 |
M11 | 0.1759 | 0.1643 | M26 | 0.1393 | 0.1706 |
M12 | 0.1703 | 0.1686 | M27 | 0.1682 | 0.1909 |
M13 | 0.1696 | 0.1590 | M28 | 0.1884 | 0.1357 |
M14 | 0.1544 | 0.1974 | M29 | 0.1736 | 0.1687 |
M15 | 0.1152 | 0.2209 | M30 | 0.1046 | 0.2345 |
The relative proximity of regional microgrid interconnection system projects can be calculated, and the relative proximity is sorted by size. The greater the relative proximity is, the better the sustainability of regional microgrid interconnection system projects is. Finally, 30 regional microgrid interconnection system project sustainability comprehensive evaluation rank ordering results are shown in
Regional microgrid interconnection system project | Relative approach degree | Rank | Regional microgrid interconnection system project | Relative approach degree | Rank |
---|---|---|---|---|---|
M1 | 0.4660 | 20 | M16 | 0.5536 | 6 |
M2 | 0.3282 | 27 | M17 | 0.5187 | 9 |
M3 | 0.3821 | 24 | M18 | 0.5165 | 11 |
M4 | 0.4701 | 18 | M19 | 0.5389 | 7 |
M5 | 0.2352 | 30 | M20 | 0.5372 | 8 |
M6 | 0.5538 | 5 | M21 | 0.5121 | 14 |
M7 | 0.2857 | 29 | M22 | 0.5135 | 13 |
M8 | 0.5828 | 2 | M23 | 0.4926 | 17 |
M9 | 0.6033 | 1 | M24 | 0.5589 | 4 |
M10 | 0.3470 | 25 | M25 | 0.4464 | 22 |
M11 | 0.5171 | 10 | M26 | 0.4494 | 21 |
M12 | 0.5024 | 16 | M27 | 0.4684 | 19 |
M13 | 0.5162 | 12 | M28 | 0.5812 | 3 |
M14 | 0.4390 | 23 | M29 | 0.5072 | 15 |
M15 | 0.3428 | 26 | M30 | 0.3085 | 28 |
From the above, based on the combination entropy weight rank order-TOPSIS method, the sustainability of 30 regional microgrid interconnected system projects is comprehensively evaluated. The sustainability of M9 is the best, and the sustainability of M5 is the worst.
In order to verify the effectiveness and feasibility of NILA-KELM model, it is compared with KELM model, ELM model and BPNN model. BPNN, a concept introduced in 1986 by scientists led by Rumelhart and McClelland, is a multilayer feed-forward neural network trained according to an error back propagation algorithm and is one of the most widely used neural network models. Therefore, this model has been chosen for comparison in this paper. The specific test results are shown in
Relative error (RE) refers to the absolute error caused by the measurement and the ratio of the measured (agreed) true value multiplied by 100% of the resulting value, expressed as a percentage. In general, RE better reflects the degree of confidence in the measurement. Root mean squard error (RMSE) can eliminate the effect of the magnitude and describe the data better compared to mean squard error (MSE). Mean absolute error (MAE) is the average of the absolute values of the deviations of all individual observations from the arithmetic mean. The mean absolute error avoids the problem of errors canceling each other out and thus accurately reflects the magnitude of the actual prediction error. In summary, these three indicators are chosen in this paper.
Regional microgrid interconnection system project | Classic evaluation results | NILA-KELM | KELM | ELM | BPNN |
---|---|---|---|---|---|
M1 | 0.4660 | 0.4621 | 0.5000 | 0.4802 | 0.5162 |
M2 | 0.3282 | 0.3116 | 0.3058 | 0.3078 | 0.3469 |
M3 | 0.3821 | 0.3718 | 0.3947 | 0.4140 | 0.4153 |
M4 | 0.4701 | 0.4796 | 0.4823 | 0.4575 | 0.5037 |
M5 | 0.2352 | 0.2285 | 0.2236 | 0.2541 | 0.2461 |
M6 | 0.5538 | 0.5340 | 0.5792 | 0.6042 | 0.5834 |
M7 | 0.2857 | 0.2920 | 0.2928 | 0.2738 | 0.3050 |
M8 | 0.5828 | 0.5873 | 0.5996 | 0.5595 | 0.5615 |
M9 | 0.6033 | 0.5871 | 0.5866 | 0.6604 | 0.6442 |
M10 | 0.3470 | 0.3425 | 0.3532 | 0.3281 | 0.3684 |
M11 | 0.5171 | 0.5003 | 0.5311 | 0.4704 | 0.5542 |
M12 | 0.5024 | 0.5262 | 0.4863 | 0.4860 | 0.5294 |
M13 | 0.5162 | 0.5188 | 0.5293 | 0.4898 | 0.5331 |
M14 | 0.4390 | 0.4318 | 0.4467 | 0.4021 | 0.4108 |
M15 | 0.3428 | 0.3421 | 0.3687 | 0.3335 | 0.3078 |
M16 | 0.5536 | 0.5413 | 0.5938 | 0.5012 | 0.5801 |
M17 | 0.5187 | 0.5145 | 0.4794 | 0.5037 | 0.4820 |
M18 | 0.5165 | 0.4974 | 0.5474 | 0.5022 | 0.4757 |
M19 | 0.5389 | 0.5602 | 0.5161 | 0.5902 | 0.5709 |
M20 | 0.5372 | 0.5246 | 0.5095 | 0.5500 | 0.5001 |
M21 | 0.5121 | 0.5060 | 0.5217 | 0.4991 | 0.5394 |
M22 | 0.5135 | 0.5114 | 0.5216 | 0.4758 | 0.4557 |
M23 | 0.4926 | 0.4776 | 0.4715 | 0.4613 | 0.5290 |
M24 | 0.5589 | 0.5465 | 0.5168 | 0.5945 | 0.5856 |
M25 | 0.4464 | 0.4601 | 0.4193 | 0.4826 | 0.4633 |
M26 | 0.4494 | 0.4512 | 0.4258 | 0.4620 | 0.4146 |
M27 | 0.4684 | 0.4657 | 0.4342 | 0.5029 | 0.5005 |
M28 | 0.5812 | 0.5912 | 0.5615 | 0.5545 | 0.5236 |
M29 | 0.5072 | 0.4870 | 0.4978 | 0.4849 | 0.4810 |
M30 | 0.3085 | 0.3044 | 0.3024 | 0.3357 | 0.2926 |
RMSE | 1.22% | 2.27% | 3.06% | 3.29% | |
MAE | 1.02% | 2.01% | 2.73% | 3.09% |
As shown in
It can be seen from
Scientific and reasonable measurement of various benefits of regional microgrid interconnection system in terms of sustainability is conducive to the stable realization of multi-stakeholder mutual benefit and win-win situation, including power system and microgrid. Therefore, this paper designs a set of a sustainability evaluation system for regional microgrid interconnection system, which mainly includes an evaluation index system and a new hybrid intelligent evaluation method. The main research results and conclusions include the following four points:
The sustainability evaluation index system of regional microgrid interconnection system is constructed from four aspects of economic, environmental, social and technical characteristics, which solves the problems that the sustainability of regional microgrid interconnection system is mainly reflected in. The weight of evaluation index is obtained based on the combination entropy weight rank order method, and the TOPSIS evaluation model is designed, and the evaluation results are obtained from the perspective of classical evaluation methods. The niche immune algorithm is used to improve the lion algorithm, and a new niche immune lion swarm algorithm is formed. NILA is used to optimize KELM, and an intelligent evaluation model is constructed. The scientificity and accuracy of the evaluation model proposed in this paper are verified by example analysis. The classical evaluation model can obtain accurate reference results, while the modern intelligent evaluation model can achieve the purpose of fast calculation and support for relevant decisions.
In summary, the research results of this paper can provide some reference for local governments and potential investors to make investment decisions in regional microgrid interconnection system projects. However, the paper does not screen the input indicators, which is a shortcoming of the paper and a direction for future research.