Low-carbon digital economic development evaluation system and method based on big data

Abstract

The invention belongs to the technical field of energy consumption monitoring, and discloses a low-carbon digital economic development evaluation system and method based on big data, and the system comprises a data collection module which collects the state data, collection time and regional energy consumption of different regions of a shopping mall; the data processing module is used for preprocessing the state data to obtain effective state data, and performing regional evaluation on the effective state data to obtain regional state description; the feature optimization module is used for performing state association evaluation on the effective state data based on regional state description to obtain an optimal data set; the model construction module is used for constructing an energy consumption prediction model based on the optimized data set and the regional energy consumption, and realizing accurate monitoring on the energy consumption of different regions of the shopping mall based on the energy consumption prediction model; and the accuracy of monitoring the energy consumption of different areas of the shopping mall is greatly improved.

Description

Technical Field

[0001] The present invention relates to the field of energy consumption monitoring technology, and more specifically, to a low-carbon digital economic development evaluation system and method based on big data. Background Art

[0002] The patent with application publication number CN116933951A discloses a low-carbon park carbon emission monitoring system and method based on big data, which belongs to the field of carbon emission monitoring technology. In order to solve the problem that carbon emissions are difficult to predict effectively and the effect display of control measures is not intuitive, the net carbon emission curve function is derived to obtain the change trend function of the net carbon emissions, which can predict and display the carbon emission change trend of the park in the future, help the system to better monitor and grasp the carbon emissions of the park, and ensure the low-carbon development of the park. Based on different park carbon reduction data under different park carbon reduction measures options, the net carbon emission curve is adjusted, and different carbon reduction emission data curves under different park carbon reduction measures options are generated. After the staff selects different carbon reduction measures options, the system will automatically display the carbon emission data curve after the carbon reduction measures are taken, which is convenient for the staff to conduct manual monitoring and overall regulation.

[0003] Under the current trend of digital economy and low-carbon development, the relationship between the energy consumption of air-conditioning systems and commercial activities in commercial buildings such as large shopping malls has gradually become a research focus. However, existing technologies have some limitations in evaluating the low-carbon operation effects of these buildings. For example, the energy consumption of air-conditioning systems is not only related to the control of indoor temperature and humidity, but also affected by multiple factors such as crowd density, activity type, and regional functions. When conducting a comprehensive evaluation of these factors, existing technologies are often unable to accurately capture the complex interactive relationship between energy consumption and commercial activities, and are unable to effectively model and analyze the relationship between air-conditioning energy consumption and commercial activities in different time periods and different regions, thereby affecting the effect of low-carbon operations.

[0004] In view of this, the present invention proposes a low-carbon digital economy development evaluation system and method based on big data to solve the above problems. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solution: a low-carbon digital economic development evaluation system based on big data, comprising:

[0006] Data collection module: collects status data, collection time and regional energy consumption in different areas of the mall;

[0007] Data processing module: pre-process the status data to obtain valid status data, perform regional evaluation on the valid status data, and obtain regional status description;

[0008] Feature optimization module: performs state correlation evaluation on valid state data based on regional state description to obtain the optimal data set;

[0009] Model building module: Build an energy consumption prediction model based on the preferred data set and regional energy consumption, and accurately monitor the energy consumption of different areas of the shopping mall based on the energy consumption prediction model.

[0010] Furthermore, the mall status data includes: regional temperature, regional humidity, human flow and regional coordinates.

[0011] Furthermore, the method of preprocessing the status data includes:

[0012] A filter and a cutoff frequency of the filter are preset, and the status data is input into the set filter. The frequency components of the regional temperature, regional humidity and flow of people in the shopping mall status data are separated by the filter, and the high-frequency noise higher than the cutoff frequency is filtered out to obtain the filtered status data. An anomaly detection algorithm is used to identify anomalies in the filtered status data, and the filtered parameters identified as abnormal are marked as abnormal. The filtered status data marked as abnormal are eliminated, and the remaining filtered status data constitute the valid status data.

[0013] Furthermore, the method of performing regional evaluation on the valid status data includes:

[0014] A clustering algorithm is used to perform cluster analysis on the flow of people in the valid state data to obtain flow clusters; based on the flow clusters, flow probability is allocated to the flow of people, and the amount of data in the flow cluster where the flow of people is located is divided by the total amount of data of the flow of people to obtain the flow probability corresponding to each person's flow of people; based on the flow probability, the environmental comfort of the regional temperature and regional humidity is evaluated to obtain the environmental comfort entropy;

[0015] The mall space radius is preset, and the regional centrality evaluation of the regional coordinates is performed based on the mall space radius. The formula for evaluating the regional centrality of the regional coordinates is: Among them, Imp represents regional centrality, R represents spatial radius, Cha represents regional coordinates, Org represents shopping mall center coordinates, dic() represents distance measurement function, and environmental comfort entropy and regional centrality constitute the regional status description.

[0016] Furthermore, the formula for evaluating the environmental comfort of the regional temperature and regional humidity is: Among them, f represents the environmental comfort entropy, T opt Represents the optimum ambient temperature, H opt represents the optimum environmental humidity, T represents the regional temperature, H represents the regional humidity, σ 1 represents the temperature tolerance factor, σ 2represents the humidity tolerance factor and p represents the flow probability.

[0017] Furthermore, the method of performing status association evaluation on the valid status data includes:

[0018] Each type of data in the valid state data is used as the data to be analyzed, and the value of the data to be analyzed is used as the grouping standard. The data to be analyzed with the same value are grouped into a group to obtain the same-value data group. Based on the same-value data group, probability evaluation is performed on each data to be analyzed. The data volume of the same-value data group where the data to be analyzed is located is divided by the total data volume of the data to be analyzed to obtain the independence probability; the environmental comfort entropy in the regional state description is used as the correlation index, and the statistical analysis method is used to perform data statistics on the correlation index to obtain the occurrence frequency of each correlation index. The occurrence frequency of each correlation index is divided by the total data volume of the correlation index to obtain the regional independence probability of the correlation index; the data to be analyzed and the corresponding regional state description constitute a joint vector, and the statistical analysis method is used to perform data statistics on the joint vector to obtain the occurrence frequency of each joint vector. The occurrence frequency of each joint vector is divided by the total data volume of the joint vector to obtain the regional joint probability of the joint vector;

[0019] Based on the independent probability, regional independent probability and regional joint probability, regional correlation evaluation is performed on each type of data to be analyzed to obtain the regional correlation degree; based on the regional correlation degree, feature time series evaluation is performed on each type of data to be analyzed to obtain the feature contribution degree; the acquisition time is used as the time scale in the feature time series evaluation formula for calculation to obtain the feature contribution degree of each type of data to be analyzed at different times, the mean of the feature contribution degree is used as the feature evaluation value, and a time series evaluation threshold is preset. The acquisition time with a feature evaluation value less than or equal to the time series evaluation threshold is used as the invalid acquisition time, and the data at the invalid acquisition time in the valid state data is eliminated to obtain the preferred data set.

[0020] Furthermore, the formula for performing regional association evaluation on each type of data to be analyzed is: Among them, I represents the regional correlation, X i represents the i-th data to be analyzed, Y i represents the correlation index corresponding to the i-th data to be analyzed, P(X i ,Y i ) represents the regional joint probability of the joint vector composed of the i-th data to be analyzed and the corresponding correlation index, P(X i ) represents the independent probability of the i-th data to be analyzed, P(Y i ) represents the regional independent probability of the correlation index corresponding to the i-th data to be analyzed;

[0021] The formula for evaluating the characteristic time series of each type of data to be analyzed is:

[0022] Among them, TQ represents the characteristic contribution, γ represents the dynamic parameter, v represents, v represents the scale parameter, I represents the regional correlation, π represents the pi, t represents the time scale, and dt represents the integration of time.

[0023] Furthermore, the energy consumption prediction model is constructed in the following manner:

[0024] The spline analysis method is used to perform gradient analysis on the regional temperature and acquisition time in the preferred data set to obtain the temperature gradient at each acquisition time. The spline analysis method is used to perform gradient analysis on the regional humidity and acquisition time in the preferred data set to obtain the humidity gradient at each acquisition time. The fluid characteristic function is constructed based on the temperature gradient and humidity gradient.

[0025] Preset the disturbance factor and B parameter spaces, where B is an integer greater than zero. Each parameter space contains: a weight matrix and a matrix bias. Initialize the optimal particle to be empty and preset the initial fluid neural network model.

[0026] The optimal data set and regional energy consumption are used as data samples. All data samples constitute a sample training set. The sample training set is evenly divided into M sample subsets. Each time, M-1 sample subsets are used to train the initial fluid neural network model, and the remaining sample subset is used for verification, which is repeated M times. Each parameter space is regarded as a particle, and each particle is evaluated with the sample training set to obtain the particle fitness.

[0027] Based on the particle fitness of each particle, the particle with the smallest particle fitness is selected as the particle to be selected, and the particle fitness of the particle to be selected is compared with that of the optimal particle. When the particle fitness of the particle to be selected is less than that of the optimal particle, the particle to be selected is used as the new optimal particle, and the state of each particle is updated based on the disturbance factor; this process is repeated until the optimal particle no longer changes, and the optimal particle at this time is output as the optimal parameter space; the optimal parameter combination is placed into the initial fluid neural network model to obtain the energy consumption prediction model.

[0028] Furthermore, the formula of the fluid characteristic function is: Among them, ε represents the gradient adjustment parameter, su represents the air flow velocity, t represents the time scale, Tve represents the mean value of the temperature gradient, and Hve represents the mean value of the humidity gradient;

[0029] The formula for evaluating each particle is:

[0030] Among them, Fit represents the particle fitness, M represents the number of sample subsets, represents the predicted energy consumption of the e-th data in the sample subset, Q erepresents the actual energy consumption of the e-th data in the sample subset, Size represents the size of the sample subset, and D k represents the kth sample subset;

[0031] The formula for updating the state of each particle is:

[0032] in, represents the state of the g-th particle at the n-th iteration, represents the state of the g-th particle when its fitness is the lowest in the iteration, w represents the disturbance factor, Represents the state of the g-th particle at the n-1th iteration, and New represents the optimal particle.

[0033] The evaluation method of low-carbon digital economy development based on big data includes:

[0034] S1. Collect status data, collection time and regional energy consumption of different areas of the mall;

[0035] S2. Preprocess the status data to obtain valid status data, perform regional evaluation on the valid status data, and obtain regional status description;

[0036] S3, performing status association evaluation on valid status data based on regional status description to obtain a preferred data set;

[0037] S4. Build an energy consumption prediction model based on the preferred data set and regional energy consumption, and accurately monitor the energy consumption of different areas of the shopping mall based on the energy consumption prediction model.

[0038] The technical effects and advantages of the low-carbon digital economy development evaluation system and method based on big data of the present invention are as follows:

[0039] The present invention pre-processes the status data to effectively remove high-frequency noise, ensure the accuracy and reliability of the input data, avoid the interference of outliers on the evaluation results, and improve the stability and accuracy of the evaluation model; by performing regional evaluation on the effective status data, it can help optimize the energy use of the shopping mall, improve the comfort of customers, and reduce unnecessary energy waste, helping to improve the operating efficiency of the shopping mall; by performing status association evaluation on the effective status data, the system can accurately perform time series analysis on the data to be analyzed, monitor environmental changes in real time, so that the system can quickly adapt to environmental changes and make corresponding adjustments, thereby improving the dynamic response capability of the system; by adopting a combination of particle swarm optimization algorithm and fluid neural network, it can achieve adaptive optimization when dealing with complex energy consumption prediction problems, and perform training based on actual energy consumption data, thereby enhancing the applicability and stability of the model in different shopping mall environments, thereby improving the accuracy of energy consumption prediction; through accurate environmental comfort evaluation and energy consumption prediction, the system can help shopping malls achieve refined management, optimize energy use, reduce energy consumption, and ultimately contribute to the development of a low-carbon digital economy. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a schematic diagram of a low-carbon digital economy development evaluation system based on big data of the present invention;

[0041] Figure 2 It is a schematic diagram of the low-carbon digital economy development evaluation method based on big data of the present invention. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] Embodiment 1;

[0044] See also Figure 1 As shown, the low-carbon digital economy development evaluation system based on big data described in this embodiment includes:

[0045] Data collection module: collects status data, collection time and regional energy consumption in different areas of the mall;

[0046] Data processing module: pre-process the status data to obtain valid status data, perform regional evaluation on the valid status data, and obtain regional status description;

[0047] Feature optimization module: performs state correlation evaluation on valid state data based on regional state description to obtain the optimal data set;

[0048] Model building module: build an energy consumption prediction model based on the preferred data set and regional energy consumption, and realize accurate monitoring of energy consumption in different areas of the shopping mall based on the energy consumption prediction model;

[0049] Each module is connected by wired and / or wireless means to achieve data transmission between modules;

[0050] The mall status data includes: regional temperature, regional humidity, human flow and regional coordinates; the regional temperature is obtained through a temperature sensor, the regional humidity is obtained through a humidity sensor, the human flow is the number of people within a specific time and in an area, and the human flow is obtained through an infrared sensor; the center of the first floor of the mall is used as the coordinate origin, a three-dimensional coordinate system is preset based on the coordinate origin, the center points of different areas of the mall are used as coordinate recording points, the horizontal distance, longitudinal distance and vertical distance from the coordinate recording point to the coordinate origin are used as the coordinate values ​​of the regional coordinates, and the regional coordinates of different areas of the mall are obtained; the time scale is the time point for data collection; the regional energy consumption is the overall resource consumption, including electricity resources, water resources and gas resources, etc., which is obtained by recording the data of electricity meters through electricity resources, water resources are obtained by recording water meters, and gas resources are obtained by recording gas meters.

[0051] Mall status data such as regional temperature, humidity and pedestrian flow may contain high-frequency noise, which affects the accuracy of the data and further affects subsequent analysis and decision-making. By using preset filters (such as low-pass filters and mean filters) to remove noise above the cutoff frequency, the data is ensured to be smoother and more accurate. Then, anomaly detection algorithms (standard score method or median absolute deviation method) are applied to identify and eliminate abnormal data to ensure that subsequent data analysis is based on valid status data. Specifically:

[0052] Preset the filter and the filter cutoff frequency, input the status data into the set filter, separate the frequency components of regional temperature, regional humidity and human flow in the mall status data, filter out the high-frequency noise above the cutoff frequency, and obtain the filtered status data; common filters include: low-pass filter and mean filter; use anomaly detection algorithm to identify anomalies in the filtered status data, mark the filtered parameters identified as abnormal as abnormal, remove the filtered status data marked as abnormal, and the remaining filtered status data constitutes valid status data. Common anomaly detection algorithms include standard score method and median absolute deviation method;

[0053] The pedestrian flow data presents different distribution patterns, which requires reasonable grouping and probability evaluation to conduct effective environmental comfort evaluation. The clustering algorithm is used to analyze the pedestrian flow, and the flow probability of each person's flow is calculated based on the flow clustering cluster. This method can accurately reflect the passenger flow characteristics of each area and evaluate the environmental comfort based on the distribution of pedestrian flow, thereby providing a basis for low-carbon operation. The environmental comfort of the shopping mall is affected by the regional temperature and humidity. The environmental comfort is evaluated based on the regional temperature and humidity based on the flow probability. The stability of the environment is quantified by constructing the environmental comfort entropy, which provides a basis for optimizing energy consumption and improving user experience. Specifically:

[0054] A clustering algorithm is used to perform cluster analysis on the flow of people in the valid state data to obtain flow clusters; based on the flow clusters, flow probability is allocated to the flow of people, and the flow probability corresponding to each person's flow is obtained by dividing the amount of data in the flow cluster where the flow of people is located by the total amount of data of the flow of people; based on the flow probability, the environmental comfort of the regional temperature and regional humidity is evaluated, and the formula for evaluating the environmental comfort of the regional temperature and regional humidity is: Among them, f represents the environmental comfort entropy, T opt Represents the optimum ambient temperature, H opt represents the optimum environmental humidity, T represents the regional temperature, H represents the regional humidity, σ 1 represents the temperature tolerance factor, which is used to control the impact of ambient temperature fluctuations on environmental comfort, σ 2 represents the humidity tolerance factor, which is used to control the impact of ambient humidity on ambient comfort. p represents the flow probability. The optimum ambient temperature and humidity are obtained through international comfort standards. Common international comfort standards include ASHRAE standards. The environmental comfort entropy represents the degree of environmental disorder. The larger the environmental comfort entropy, the more unstable the environmental conditions, the lower the comfort level, and the higher the energy consumption.

[0055] The mall space radius is preset. The mall space radius is the radius value of a circle drawn with the mall center as the center and covering the entire range of the mall with the minimum radius. The regional centrality evaluation of the regional coordinates is performed based on the mall space radius. The formula for evaluating the regional centrality of the regional coordinates is: Among them, Imp represents regional centrality, R represents spatial radius, Cha represents regional coordinates, Org represents shopping mall center coordinates, dic() represents distance measurement function. Common distance measurement functions include Euclidean distance function and Manhattan distance function. Environmental comfort entropy and regional centrality constitute the description of regional status.

[0056] There are complex correlations between different types of data (such as regional temperature, humidity, and pedestrian flow). By conducting regional correlation evaluation on each type of data to be analyzed and combining it with time series feature evaluation, we can find out the correlation and time series change trend between different data items and environmental comfort. This process can effectively identify the key factors affecting energy efficiency and comfort, and provide valuable reference for system optimization. Specifically:

[0057] Each type of data in the valid state data is used as the data to be analyzed, and the value of the data to be analyzed is used as the grouping standard. The data to be analyzed with the same value are grouped into a group to obtain the same-value data group. Based on the same-value data group, probability evaluation is performed on each data to be analyzed. The data volume of the same-value data group where the data to be analyzed is located is divided by the total data volume of the data to be analyzed to obtain the independence probability; the environmental comfort entropy in the regional state description is used as the correlation index, and the statistical analysis method is used to perform data statistics on the correlation index to obtain the occurrence frequency of each correlation index. The occurrence frequency of each correlation index is divided by the total data volume of the correlation index to obtain the regional independence probability of the correlation index; the data to be analyzed and the corresponding regional state description constitute a joint vector, and the statistical analysis method is used to perform data statistics on the joint vector to obtain the occurrence frequency of each joint vector. The occurrence frequency of each joint vector is divided by the total data volume of the joint vector to obtain the regional joint probability of the joint vector;

[0058] Based on the independent probability, regional independent probability and regional joint probability, regional association evaluation is performed on each type of data to be analyzed. The formula for regional association evaluation on each type of data to be analyzed is: Among them, I represents the regional correlation, X i represents the i-th data to be analyzed, Y i represents the correlation index corresponding to the i-th data to be analyzed, P(X i ,Y i ) represents the regional joint probability of the joint vector composed of the i-th data to be analyzed and the corresponding correlation index, P(X i ) represents the independent probability of the i-th data to be analyzed, P(Y i ) represents the regional independent probability of the correlation index corresponding to the i-th data to be analyzed; based on the regional correlation, the characteristic time series evaluation of each type of data to be analyzed is performed, and the formula for the characteristic time series evaluation of each type of data to be analyzed is: Among them, TQ represents feature contribution, γ represents the dynamic parameter, which is used to control the sensitivity of feature contribution to time, v represents, v represents the scale parameter, which is used to control the range of feature contribution, I represents regional correlation, π represents pi, t represents time scale, and dt represents integration of time. The acquisition time is used as the time scale in the feature timing evaluation formula to calculate, and the feature contribution of each type of data to be analyzed at different times is obtained. The mean value of the feature contribution is used as the feature evaluation value, and the timing evaluation threshold is preset. The acquisition time with a feature evaluation value less than or equal to the timing evaluation threshold is taken as the invalid acquisition time, and the data at the invalid acquisition time in the valid state data is eliminated to obtain the preferred data set.

[0059] The parameter optimization of the energy consumption prediction model requires an efficient and accurate method to ensure that the error between the energy consumption prediction and the actual energy consumption is minimized. The particle swarm optimization (PSO) algorithm is used to optimize the parameters of the fluid neural network model. Through multiple training and verification on the training set and the verification set, the parameter space is gradually adjusted to find the optimal prediction model. This method can improve the accuracy of energy consumption prediction and provide more reliable data support for the development of low-carbon economy. Specifically:

[0060] The spline analysis method is used to perform gradient analysis on the regional temperature and acquisition time in the preferred data set to obtain the temperature gradient at each acquisition time. The spline analysis method is used to perform gradient analysis on the regional humidity and acquisition time in the preferred data set to obtain the humidity gradient at each acquisition time. The fluid characteristic function is constructed based on the temperature gradient and humidity gradient. The formula of the fluid characteristic function is:

[0061] Among them, ε represents the gradient adjustment parameter, which is used to control the influence of humidity and temperature, su represents the air flow speed, which is obtained by the wind speed sensor, t represents the time scale, Tve represents the mean value of the temperature gradient, and Hve represents the mean value of the humidity gradient;

[0062] Preset disturbance factors and B parameter spaces, B is an integer greater than zero, the range of disturbance factors is: [0,1], each parameter space contains: weight matrix and matrix bias, initialize the optimal particle to be empty, preset the initial fluid neural network model, the formula of the initial fluid neural network model is: Q(t) = F(W×In+b)×Fuc; where Q(t) represents the predicted energy consumption at time t, F() represents the activation function, common activation functions include Relu function, In represents input data, W represents weight matrix, b represents matrix bias, and Fuc represents fluid characteristic function;

[0063] The optimal data set and regional energy consumption are used as data samples. All data samples constitute a sample training set. The sample training set is evenly divided into M sample subsets. Each time, M-1 sample subsets are used to train the initial fluid neural network model, and the remaining sample subset is used for verification, which is repeated M times. Each parameter space is taken as a particle, and each particle is evaluated with the sample training set. The formula for evaluating each particle is: Among them, Fit represents the particle fitness, M represents the number of sample subsets, represents the predicted energy consumption of the e-th data in the sample subset, Q e represents the actual energy consumption of the e-th data in the sample subset, Size represents the size of the sample subset, and D k represents the kth sample subset;

[0064] The particle fitness represents the deviation between the predicted energy consumption and the actual energy consumption. The smaller the particle fitness, the better the performance of the particle for the parameter combination of the initial fluid neural network model. Based on the particle fitness of each particle, the particle with the smallest particle fitness is selected as the particle to be selected, and the particle fitness of the particle to be selected is compared with the particle fitness of the optimal particle. When the particle fitness of the particle to be selected is less than the particle fitness of the optimal particle, the particle to be selected is used as the new optimal particle. The state of each particle is updated based on the disturbance factor. The formula for updating the state of each particle is:

[0065] in, represents the state of the g-th particle at the n-th iteration, represents the state of the g-th particle when its fitness is the lowest in the iteration, w represents the disturbance factor, Represents the state of the g-th particle at the n-1th iteration, New represents the optimal particle, and each state of the particle represents a parameter space. Repeat until the optimal particle no longer changes, and output the optimal particle at this time as the optimal parameter space; put the optimal parameter combination into the initial fluid neural network model to obtain the energy consumption prediction model.

[0066] This embodiment pre-processes the status data to effectively remove high-frequency noise, ensure the accuracy and reliability of the input data, avoid the interference of outliers on the evaluation results, and improve the stability and accuracy of the evaluation model; by performing regional evaluation on the effective status data, it can help optimize the energy use of the mall, improve the comfort of customers, and reduce unnecessary energy waste, helping to improve the operating efficiency of the mall; by performing status association evaluation on the effective status data, the system can accurately perform time series analysis on the data to be analyzed and monitor environmental changes in real time, so that the system can quickly adapt to environmental changes and make corresponding adjustments, thereby improving the dynamic response capability of the system; by adopting a combination of particle swarm optimization algorithm and fluid neural network, it can achieve adaptive optimization when dealing with complex energy consumption prediction problems, and train based on actual energy consumption data, thereby enhancing the applicability and stability of the model in different mall environments, thereby improving the accuracy of energy consumption prediction; through accurate environmental comfort evaluation and energy consumption prediction, the system can help the mall achieve refined management, optimize energy use, reduce energy consumption, and ultimately contribute to the development of a low-carbon digital economy.

[0067] Embodiment 2;

[0068] See also Figure 2 As shown, the part not described in detail in this embodiment is described in Example 1, and a method for evaluating the development of a low-carbon digital economy based on big data is provided, including:

[0069] S1. Collect status data, collection time and regional energy consumption of different areas of the mall;

[0070] S2. Preprocess the status data to obtain valid status data, perform regional evaluation on the valid status data, and obtain regional status description;

[0071] S3, performing status association evaluation on valid status data based on regional status description to obtain a preferred data set;

[0072] S4. Build an energy consumption prediction model based on the preferred data set and regional energy consumption, and accurately monitor the energy consumption of different areas of the shopping mall based on the energy consumption prediction model.

[0073] Embodiment 3;

[0074] This embodiment discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the operation mode of the low-carbon digital economy development assessment system method based on big data provided above is implemented.

[0075] Since the electronic device introduced in this embodiment is an electronic device used to implement the low-carbon digital economy development evaluation system method based on big data in the embodiment of this application, based on the low-carbon digital economy development evaluation system method based on big data introduced in the embodiment of this application, the technical personnel of this field can understand the specific implementation method of the electronic device of this embodiment and its various variations, so how the electronic device implements the method in the embodiment of this application is not introduced in detail here. As long as the technical personnel of this field implement the electronic device used in the low-carbon digital economy development evaluation system method based on big data in the embodiment of this application, it belongs to the scope of protection of this application.

[0076] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0077] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technical users in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A low-carbon digital economy development evaluation system based on big data, characterized by: include: Data collection module: collects status data, collection time and regional energy consumption in different areas of the mall; Data processing module: pre-process the status data to obtain valid status data, perform regional evaluation on the valid status data, and obtain regional status description; Feature optimization module: performs state correlation evaluation on valid state data based on regional state description to obtain the optimal data set; Model building module: Build an energy consumption prediction model based on the preferred data set and regional energy consumption, and accurately monitor the energy consumption of different areas of the shopping mall based on the energy consumption prediction model.

2. The low-carbon digital economy development evaluation system based on big data according to claim 1 is characterized in that: The mall status data includes: regional temperature, regional humidity, passenger flow and regional coordinates.

3. The low-carbon digital economy development evaluation system based on big data according to claim 2 is characterized in that: The method of preprocessing the status data includes: A filter and a cutoff frequency of the filter are preset, and the status data is input into the set filter. The frequency components of the regional temperature, regional humidity and flow of people in the shopping mall status data are separated by the filter, and the high-frequency noise higher than the cutoff frequency is filtered out to obtain the filtered status data. An anomaly detection algorithm is used to identify anomalies in the filtered status data, and the filtered parameters identified as abnormal are marked as abnormal. The filtered status data marked as abnormal are eliminated, and the remaining filtered status data constitute the valid status data.

4. The low-carbon digital economy development evaluation system based on big data according to claim 3 is characterized in that: The method of performing regional evaluation on the effective status data includes: A clustering algorithm is used to perform cluster analysis on the flow of people in the valid state data to obtain flow clusters; based on the flow clusters, flow probability is allocated to the flow of people, and the amount of data in the flow cluster where the flow of people is located is divided by the total amount of data of the flow of people to obtain the flow probability corresponding to each person's flow of people; based on the flow probability, the environmental comfort of the regional temperature and regional humidity is evaluated to obtain the environmental comfort entropy; The mall space radius is preset, and the regional centrality evaluation of the regional coordinates is performed based on the mall space radius. The formula for evaluating the regional centrality of the regional coordinates is: Among them, Imp represents regional centrality, R represents spatial radius, Cha represents regional coordinates, Org represents shopping mall center coordinates, dic() represents distance measurement function, and environmental comfort entropy and regional centrality constitute the regional status description.

5. The low-carbon digital economy development evaluation system based on big data according to claim 4 is characterized in that: The formula for evaluating the environmental comfort level of regional temperature and regional humidity is: Among them, f represents the environmental comfort entropy, T opt Represents the optimum ambient temperature, H opt represents the optimum environmental humidity, T represents the regional temperature, H represents the regional humidity, σ1 represents the temperature tolerance factor, σ2 represents the humidity tolerance factor, and p represents the flow probability.

6. The low-carbon digital economy development evaluation system based on big data according to claim 5 is characterized in that: The method of performing status association evaluation on the valid status data includes: Each type of data in the valid state data is used as the data to be analyzed, and the value of the data to be analyzed is used as the grouping standard. The data to be analyzed with the same value are grouped into a group to obtain the same-value data group. Based on the same-value data group, probability evaluation is performed on each data to be analyzed. The data volume of the same-value data group where the data to be analyzed is located is divided by the total data volume of the data to be analyzed to obtain the independence probability; the environmental comfort entropy in the regional state description is used as the correlation index, and the statistical analysis method is used to perform data statistics on the correlation index to obtain the occurrence frequency of each correlation index. The occurrence frequency of each correlation index is divided by the total data volume of the correlation index to obtain the regional independence probability of the correlation index; the data to be analyzed and the corresponding regional state description constitute a joint vector, and the statistical analysis method is used to perform data statistics on the joint vector to obtain the occurrence frequency of each joint vector. The occurrence frequency of each joint vector is divided by the total data volume of the joint vector to obtain the regional joint probability of the joint vector; Based on the independent probability, regional independent probability and regional joint probability, regional correlation evaluation is performed on each type of data to be analyzed to obtain the regional correlation degree; based on the regional correlation degree, feature time series evaluation is performed on each type of data to be analyzed to obtain the feature contribution degree; the acquisition time is used as the time scale in the feature time series evaluation formula for calculation to obtain the feature contribution degree of each type of data to be analyzed at different times, the mean of the feature contribution degree is used as the feature evaluation value, and a time series evaluation threshold is preset. The acquisition time with a feature evaluation value less than or equal to the time series evaluation threshold is used as the invalid acquisition time, and the data at the invalid acquisition time in the valid state data is eliminated to obtain the preferred data set.

7. The low-carbon digital economy development evaluation system based on big data according to claim 6 is characterized in that: The formula for performing regional association evaluation on each type of data to be analyzed is: Among them, I represents the regional correlation, X i represents the i-th data to be analyzed, Y i represents the correlation index corresponding to the i-th data to be analyzed, P(X i ,Y i ) represents the regional joint probability of the joint vector composed of the i-th data to be analyzed and the corresponding correlation index, P(X i ) represents the independent probability of the i-th data to be analyzed, P(Y i ) represents the regional independent probability of the correlation index corresponding to the i-th data to be analyzed; The formula for evaluating the characteristic time series of each type of data to be analyzed is: Among them, TQ represents the characteristic contribution, γ represents the dynamic parameter, v represents, v represents the scale parameter, I represents the regional correlation, π represents the pi, t represents the time scale, and dt represents the integration of time.

8. The low-carbon digital economy development evaluation system based on big data according to claim 7 is characterized in that: The energy consumption prediction model is constructed in the following manner: The spline analysis method is used to perform gradient analysis on the regional temperature and acquisition time in the preferred data set to obtain the temperature gradient at each acquisition time. The spline analysis method is used to perform gradient analysis on the regional humidity and acquisition time in the preferred data set to obtain the humidity gradient at each acquisition time. The fluid characteristic function is constructed based on the temperature gradient and humidity gradient. Preset the disturbance factor and B parameter spaces, where B is an integer greater than zero. Each parameter space contains: a weight matrix and a matrix bias. Initialize the optimal particle to be empty and preset the initial fluid neural network model. The optimal data set and regional energy consumption are used as data samples. All data samples constitute a sample training set. The sample training set is evenly divided into M sample subsets. Each time, M-1 sample subsets are used to train the initial fluid neural network model, and the remaining sample subset is used for verification, which is repeated M times. Each parameter space is regarded as a particle, and each particle is evaluated with the sample training set to obtain the particle fitness. Based on the particle fitness of each particle, the particle with the smallest particle fitness is selected as the particle to be selected, and the particle fitness of the particle to be selected is compared with that of the optimal particle. When the particle fitness of the particle to be selected is less than that of the optimal particle, the particle to be selected is used as the new optimal particle, and the state of each particle is updated based on the disturbance factor; this process is repeated until the optimal particle no longer changes, and the optimal particle at this time is output as the optimal parameter space; the optimal parameter combination is placed into the initial fluid neural network model to obtain the energy consumption prediction model.

9. The low-carbon digital economy development evaluation system based on big data according to claim 8 is characterized in that: The formula of the fluid characteristic function is: Among them, ε represents the gradient adjustment parameter, su represents the air flow velocity, t represents the time scale, Tve represents the mean value of the temperature gradient, and Hve represents the mean value of the humidity gradient; The formula for evaluating each particle is: Among them, Fit represents the particle fitness, M represents the number of sample subsets, represents the predicted energy consumption of the e-th data in the sample subset, Q e represents the actual energy consumption of the e-th data in the sample subset, Size represents the size of the sample subset, and D k represents the kth sample subset; The formula for updating the state of each particle is: in, represents the state of the g-th particle at the n-th iteration, represents the state of the g-th particle when its fitness is the lowest in the iteration, w represents the disturbance factor, Represents the state of the g-th particle at the n-1th iteration, and New represents the optimal particle.

10. A method for evaluating the development of a low-carbon digital economy based on big data, which is implemented based on a system for evaluating the development of a low-carbon digital economy based on big data as described in any one of claims 1 to 9, characterized in that: include: S1. Collect status data, collection time and regional energy consumption of different areas of the mall; S2. Preprocess the status data to obtain valid status data, perform regional evaluation on the valid status data, and obtain regional status description; S3, performing status association evaluation on valid status data based on regional status description to obtain a preferred data set; S4. Build an energy consumption prediction model based on the preferred data set and regional energy consumption, and accurately monitor the energy consumption of different areas of the shopping mall based on the energy consumption prediction model.

Images