Research on Risk Assessment Model of Light Pollution and Countermeasure Based on AHP Combined with TOPSIS

: With the development of human society, light pollution has gradually become one of the major problems faced by today's society. In order to improve human awareness of the impact of light pollution and reduce the impact of light pollution, we constructed a "light pollution risk assessment" model and proposed relevant strategies to mitigate the impact of light pollution. We established a 3-level Analytical Hierarchy Process (AHP) combined with Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) model by dividing into three major factors: economic, social, and natural, and selected each index to take data collection and data analysis, then the determination of weights among factors in the model was carried out by entropy weighting method, and the light pollution risk level assessment model was obtained by K-clustering analysis. The model can systematically assess and classify the risk level of light pollution in the investigated area, which is conducive to the establishment of a scientific system for the evaluation and prevention of urban light pollution in the future, and propose effective countermeasures for how to reduce urban light pollution.


Introduction
From the end of the 19th century to the present, more than a century, modern urban lighting technology and urban night environment construction has developed rapidly, urban night artificial light illumination has experienced incandescent, mercury, sodium, and LED energy-saving lamps four stages. Especially in recent years, the introduction of new technologies has made the urban nighttime light environment rich and colorful, thus urban light pollution has become more and more serious, and the problems caused by it have become increasingly prominent [1] . Light pollution is a form of environmental degradation, excessive artificial outdoor lighting affects the natural environment and ecosystem of the earth. It is not only a waste of energy and earth's resources, but also indirectly contributes to global environmental problems. And importantly, the sky light produced by these artificial lighting sources causes a decrease in the quality of the night sky, reducing the number of observable stars and preventing us from seeing the beautiful night sky anymore [2] .
One of the first models of light pollution was introduced by Treanor in 1973 [3] . [In 2016, the scholar Yali Katz et al. obtained images with different spatial resolutions by EROS-B satellite to compare the quantitative study of urban light pollution [4] . In 2021, Szombathely proposed to record the night lighting brightness through satellite observation and suggested that people perceive a brighter environment when orange high-pressure sodium lights turn into white LED lights, but the intensity of artificial light generated through satellite observation is reduced [5] .
As light pollution has not been studied for a long time, there are still many unresolved problems. In October 2013, the Journal of Integrative Environmental Sciences published a survey article on the Finnish public's perception of light pollution, and the respondents generally agreed that light pollution has become a distraction for people's outdoor recreation and relaxation. Therefore, relatively mandatory policy measures are still needed to curb the light pollution problem [6] . There is an article in the journal Humanities that shows that light pollution is gradually becoming the most serious problem worldwide and we need to pay attention to it and take measures accordingly [7] . Therefore, the prevention and control of light pollution problems requires not only continuous exploration in the field of science, but also an increase in public awareness of light pollution prevention and control.
We established the light pollution risk level assessment model, which can assess the light pollution risk level of the study site, and can determine which light pollution risk area the study site belongs to, in the application of the real society can be established through the model, people's understanding of light pollution, as well as for the prevention and control of light pollution problems, legislation and other work to provide a theoretical basis and play a guiding role. This paper is divided into five parts. The second part is the establishment of the light pollution risk level evaluation model, which is used to determine whether the study site is affected by light pollution. The third part is the establishment of risk assessment model for light pollution based on AHP, which is used to determine the type of light pollution risk area of the study site. The fourth part is the simulation application of the model, taking the Zhongshan District of Dalian City as an example, to test the rationality of the model. The fifth part is the conclusions, which writes about the characteristics and roles of our model, as well as the conditions of applicability.

Establishment of the light pollution risk level evaluation model
To evaluate the light pollution risk level of a location, we developed the Light Pollution Risk Level Evaluation (LPRLE) model and the Topsis model based on the entropy weight method. We come to quantitatively and comprehensively represent and calculate the light pollution risk level of a location. With the emphasis on a healthy light environment, research on the ecological impact of light pollution has been intensively promoted in many countries. China still lacks research on light pollution, but its pollution hazards have long been an important issue that we cannot ignore, and it seriously affects personal health, ecological balance and astronomical observation. The ecological health of lighting will become the trend of research and application in the field of lighting under the premise of whether it is suitable for human habitation and whether it is compatible with the city [8] . Combining theory and practice, paying attention and importance to the harmony between human and nature in the development process, we propose a new index to determine the risk level of light pollution in a place, to accurately determine the degree of light pollution in an area with wide adaptability, and to remind human beings to take remedial measures against existing ecological changes that are detrimental to human survival.

Determination of indicators
The Light Pollution Risk Level Evaluation (LPRLE) model is an assessment model that is widely applicable in the field of environmental quality. By thinking about human life and natural resources, various aspects are considered that affect the level of risk of light pollution. We will measure the level of light pollution risk at a site based on a combination of social, economic, and natural dimensions, as can be seen in Figure 1:

Calculation of the Topsis model based on the entropy weight method
Considering the relevant factors affecting the light pollution in the area, our evaluation model first uses the entropy weighting method to determine the weights of each index, and then uses the Topsis method for specific calculations, and finally evaluates the risk level of light pollution in the area. (Entropy weighting method is an objective weighting method.) The formula is as follows: We first put the indicator data into the corresponding research object to form the matrix X.
The indicator variables are normalized. In this model, we first normalize the very small model, then normalize the matrix X after the forward processing, and finally obtain the matrix Z Formula for converting very small indicators to very large indicators: max x  .
Each element is divided by the arithmetic square root of the sum of the squares of the elements in its column.
The matrix is then processed computationally to standardize the specification of the data.
We define the maximum value as The weights of each index are calculated by the entropy weighting method.
Calculate the probability matrix P, where each element in The calculation formula of information entropy is   Information utility value (the larger the utility value is, the more information it corresponds to) Define the distance between the i-th evaluation object and the maximum and minimum values, respectively Finally, the distance between the study object and the maximum and minimum values is defined, and then the light pollution risk assessment degree score is obtained.
In building the model, considering the generality and applicability of the model, only specific target countries can determine its parameters. Therefore, we selected 15 places with different geographical locations, climates and economic development, The entropy weighting method enables the weights of each indicator to be obtained, as shown in Table 1 below:

Light pollution risk level classification
In 2009, Cosmopolitan Australia reported that according to a recent US survey, 70% of the world's population lives in light pollution and about 75% of the world's countries are affected by light pollution. In these cities in China, we will define a critical point as the dividing line between these light-polluted and non-polluted cities. We use the K-Means clustering algorithm to minimize the sum of distances to the classification center .
where is the observed value in the kth star cluster Ck. The critical point can be calculated as 0.786. The model simulation results are shown in Fig.2

Establishment of risk assessment model for light pollution based on AHP
Considering the different types of sites examined, we further evaluated the light pollution risk areas by hierarchical analysis on the basis of the original weights obtained, and divided the light pollution risk areas into two major categories: economic light pollution risk areas, and environmental light pollution risk areas. Outdoor lighting is an indispensable element of modern society, urban development, residential entertainment and security cannot be achieved without light, however, excessive and excessive lighting levels and unreasonable use of lighting equipment can lead to light pollution [9,10] . Economic light pollution risk areas need to take more into account the economic development, while environmental light pollution risk areas take more into account the quality of the environment. The AHP model we constructed can classify the types of light pollution risk areas in the study site and can provide theoretical guidance for the prevention and control of light pollution in the study site. pollution risk areas take more into account the quality of the environment. The AHP model we constructed can classify the types of light pollution risk areas in the study site and can provide theoretical guidance for the prevention and control of light pollution in the study site.The constructed AHP model is shown in Figure 3:

Further assessment of indicators
Economic (A1) light pollution risk areas and environmental (A2) light pollution risk areas are both affected by indicators sunshine hours (ST), gross product (GDP), population density (PD), atmospheric visibility (AY), number of lighting fixtures (NLF), nighttime traffic flow (NTF), and power resource consumption (PRC) to a different extent, so after the above model has been used to determine the indicators through the entropy weighting method equation (3)--(5) to determine the weights of each indicator, the association of the indicator with each type of light pollution risk area is then further evaluated. This is shown in Figure 4 below:

Establishment of light pollution risk area types
The corresponding weights are calculated by the special value method and multiplied with the weights calculated by the entropy weight method respectively, and then the final score is obtained, and the type of light pollution risk area is determined by comparing the high and low scores.

Application of the model
Using Zhongshan District of Dalian City as the study site, the indexes established by the LPRLE model were used to find the corresponding data. Then, through the entropy-based Topsis model, we calculated the score of 0.712 and concluded that Zhongshan District of Dalian City is a city affected by light pollution, and finally, through the AHP model, we concluded that the study site belongs to the economic light pollution risk area. Based on the results obtained from our model simulation, the local government of Dalian can implement corresponding light pollution prevention strategies for Zhongshan District according to the criteria of economic light pollution risk area and provide data and theoretical guidance for light pollution prevention legislation.

Conclusions
In order to explore how to judge whether an area suffers from light pollution, to provide regional officials with a wide range of practical judgment indicators as well as to provide theoretical guidance for the prevention and control of light pollution, we will select appropriate evaluation indicators, give them weights, and combine some small indicators to achieve some comprehensive indicators. Thus, the whole evaluation system will be formed. the study area through the model. For this research direction, we can continue to go deeper into the proposal of specific decisions and the simulation study of the effect of the implementation of the decisions, so that we can prevent and control the existing light pollution problems more comprehensively from the assessment and classification of light pollution risks to the proposal of solutions to light pollution problems and the prediction of their effectiveness.