A multi-stage load forecasting method for industrial parks considering development uncertainty
By dividing functional plots, building load density saturation rate curves, and using Gaussian nuclear density and Markov models, the accuracy problem of multi-stage load prediction in industrial parks is solved, and adaptability and accurate prediction to dynamic development are achieved.
Patent Information
- Application Number
- CN202411088439.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-09
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2044-08-09
AI Technical Summary
The prior art is difficult to effectively predict the multi-stage load of industrial parks, especially in the case of large development uncertainty and dynamic changes, resulting in low energy waste and utilization.
By dividing industrial parks into plots of different functional types, a load density saturation rate S-shaped development curve is constructed, and combined with Gaussian core density and Markov prediction model, the load density saturation value and land area of each functional plot are calculated, and multiple predictions are predicted. Load changes.
It realizes accurate prediction of the diversified load of industrial parks, adapts to dynamic development and changes, and improves the scientificity and adaptability of planning and design, especially in newly built or lacking load data.
Smart Images

Figure CN118644105B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of industrial park load forecasting, and in particular relates to a multi-stage load forecasting method for an industrial park taking development uncertainty into account. Background Art
[0002] The integrated energy system of industrial parks is an on-demand energy supply system oriented to the demand side. Industrial parks have a large number of users and diversified energy consumption, and have both production and living scene attributes. It is necessary to coordinate user resources, natural resources and space resources to scientifically and rationally plan and design the integrated energy system. At present, the planning and design of the integrated energy system of traditional industrial parks mainly determines the system structure of the integrated energy system of the planned park, including equipment selection, placement, etc., and forms a planning and design scheme for the determined system structure and conducts environmental protection, economic and safety evaluation to obtain a planning and design guidance scheme. This method is prone to lead to advanced investment and idle assets, and is out of touch with the dynamic development and changes of the park in the later stage. Incomplete human thinking or untimely adjustment of the later planning and design scheme will cause problems such as energy waste and low energy utilization in the park; in addition, the planning and design scheme of the integrated energy system of the industrial park cannot meet the dynamic development energy demand of the park in one planning and design.
[0003] Combining the load change characteristics of industrial parks and then scientifically and reasonably predicting the load is an important part of the comprehensive energy planning and design of industrial parks, which can provide a scientific and reasonable basis for the formulation of comprehensive energy system planning and design schemes for industrial parks. Therefore, studying the complex load characteristics of the comprehensive energy system of industrial parks and establishing an accurate load prediction model are the key to carrying out scientific and reasonable comprehensive energy system planning and design of industrial parks. How to establish a multi-stage load prediction model that matches the dynamic development and changes of industrial parks in response to the load change characteristics caused by the dynamic development and changes of industrial parks is an urgent problem to be solved.
[0004] Based on the above technical problems, it is necessary to design a new multi-stage load forecasting method for industrial parks that takes into account development uncertainties. Summary of the invention
[0005] The technical problem to be solved by the present invention is to overcome the shortcomings of the existing technology and provide a multi-stage load forecasting method for industrial parks taking into account development uncertainty, effectively solving the problem of load forecasting accuracy in multiple stages of the planning period affected by development uncertainty of industrial parks, so that the multi-dimensional load forecasting of industrial parks matches the dynamic development and changes of the parks, especially for newly built industrial parks and industrial parks with severe lack of historical load data, and has strong adaptability.
[0006] In order to solve the above technical problems, the technical solution of the present invention is:
[0007] The present invention provides a multi-stage load forecasting method for an industrial park taking into account development uncertainty, which comprises:
[0008] S1. According to the development goal positioning and land planning of the industrial park, the industrial park is divided into a plurality of plots with different functional types during the planning and design period of the industrial park; wherein the planning and design period of the industrial park includes the following stages: the park formation stage, the park growth stage, the park maturity stage, and the park renewal or decline stage;
[0009] S2. Construct an S-shaped development curve of load density saturation rate and solve the load density saturation rate of each functional plot corresponding to different planning years;
[0010] S3. Considering the imbalance of the development of multiple loads in different functional plots in the industrial park, based on the load data samples of the existing limited plots, the load density saturation value of each functional plot is calculated by the load density saturation value estimation method based on Gaussian kernel density; wherein the multiple loads include at least electrical load, heat load and cooling load;
[0011] S4. Considering the impact of the evolution of the types of functional plots in industrial parks on the load forecast of industrial parks, a Markov prediction model for the mutual conversion of different functional plots is constructed to predict the area of each functional land at different stages during the planning and design period of industrial parks;
[0012] S5. Based on the load density saturation rate, load density saturation value and functional land area of each functional plot in the industrial park, the multivariate load prediction results at different stages during the planning and design period of the industrial park are calculated.
[0013] In the above technical solution, further, in step S1, the multiple plots of land with different functional types include at least: residential land, commercial service facility land, industrial land, and public administration and public service facility land.
[0014] Furthermore, the characteristics of the load of each functional plot changing with the dynamic development of the industrial park are analyzed as follows:
[0015] 1) Analyze the intra-day variation pattern of the multi-load on the demand side of each functional plot, and obtain the peak and valley characteristics and multi-dimensional influencing factors of the multi-load on each functional plot; the multi-dimensional influencing factors are related to the functional type of the plot and the behavior preference of the plot;
[0016] 2) Based on the daily variation patterns of the multi-loads on the demand side of each functional plot, combined with the dynamic development and change characteristics of the industrial park at various stages of the planning and design period, determine the multi-dimensional influencing factors of the multi-load changes of each functional plot; the multi-dimensional influencing factors include the scale expansion of each functional plot, the occupancy rate of enterprises, the adjustment of industrial structure, the fluctuations of policies and market economic factors, external climate and environmental factors, and the actual life differences of various equipment;
[0017] 3) Analyze the load change characteristics of different functional plots and the same functional plots: Due to the different types of user entities on the demand side, in the dynamic development and change process of the industrial park at different stages, the multi-dimensional influencing factors of the multiple loads of each functional plot and the different load change laws lead to different load growth rates, and the load peaks of different functional plots will not appear at the same time; and, based on the influence of the dynamic development and change characteristics of the park at different stages, there are also differences in the dynamic development and changes between the same functional plots, and the load density saturation value of the same functional plot is also different.
[0018] Furthermore, in step S2, the load density of each functional plot complies with the S-curve development; initially, the load density grows slowly, then the load density grows rapidly, then the load density continues to grow rapidly, and finally the load density maintains a steady growth until it grows to a saturation value. The construction of the load density saturation rate S-curve development curve specifically includes:
[0019] S201, use logistic curve to calculate load density saturation rate , expressed as:
[0020] ;
[0021] t is the planning period; k, t c is a constant; k determines the rising slope of the load growth curve of each functional plot. When the load growth rate of each plot is the same, the k value is the same; t c It indicates that the corresponding planning period when the load growth rate is the fastest has nothing to do with the k value;
[0022] S202, according to k, t c The value of is taken, and each functional plot is defined as two dimensions: the same functional plot and similar functional plot, and the load density saturation rate S-shaped development curve corresponding to the two dimensions is calculated and constructed respectively.
[0023] Furthermore, the same functional plots are defined as functional plots with the same multi-load growth rate, the same multi-load density saturation rate S-shaped development curve, the same k value and t c value;
[0024] The S-shaped development curve of the multivariate load density saturation rate of the same functional plot is expressed as:
[0025] ;
[0026] , , They are the S-shaped development curves of load density saturation rate of electrical load, thermal load and cooling load in the multiple loads of the same functional plot.
[0027] Furthermore, the similar functional plots are defined as functional plots with different construction times, similar dynamic development changes, and similar S-shaped development curves of multivariate load density saturation rate, with the same k value and different t c value;
[0028] For two similar functional plots, the multivariate load density saturation rate S-shaped development curve of one of the functional plots is defined as , the curve parameters are k, t c , the construction period is The multivariate load density saturation rate S-shaped development curve of another functional plot to be calculated is: , the construction period is , then the multivariate load density saturation rate S-shaped development curve of another functional plot is It is expressed as:
[0029] ξ ′ ( t ) = 1 1 + e − k [ t − t c − ( T ′ 0 − T 0 )] .
[0030] Furthermore, in step S3, based on the existing limited plot load data samples, the load density saturation value of each functional plot is calculated by a load density saturation value estimation method based on Gaussian kernel density, specifically including:
[0031] S301. For newly built or formative industrial parks, the sample of land load data is limited. Therefore, it is assumed that a functional land parcel in an industrial park has sample plots, the load density is , the probability density function of load density is , expressed as:
[0032] ;
[0033] is the sample plot capacity; W is the bandwidth; As the kernel function, the standard Gaussian function is selected; is the input load density parameter; For the Load density of sample plots;
[0034] S302, generating a probability density curve of the load density according to the formula in S301; dividing the load density into a number of density intervals based on the probability density curve of the load density, which can be specifically divided into a low load density interval, a medium load density interval and a high load density interval, or a low load density interval and a high load density interval;
[0035] S303, in each load density interval obtained in step S302, find the load density value of the maximum probability density, and use the value as the typical load density representative value of the interval, obtain the low load density typical value of the low load density interval, the medium load density typical value of the medium load density interval, and the high load density typical value of the high load density interval; if it is divided into two density intervals, the medium load density typical value is not included;
[0036] S304. The load density between the low load density typical value and the high load density typical value obtained in step S303 is taken as the concentrated area of the load density of the functional plot, and the high load density typical value is selected as the load density saturation value of the functional plot.
[0037] Furthermore, in step S4, in different planning and design stages of the industrial park, after the types of functional plots in the industrial park evolve and adjust with the dynamic development of the park, the land area and multivariate load values of each functional plot change accordingly; therefore, the Markov chain theory is introduced in the present invention, and the land planning status of each functional plot is represented by the land area, and the land planning status of each functional plot is converted to each other at different stages during the planning and design period.
[0038] In step S4, the influence of the evolution law of the functional plot types of the industrial park on the load forecast of the industrial park is considered, a Markov prediction model for the mutual conversion of different functional plots is constructed, and the area of each functional land at different stages during the planning and design period of the industrial park is predicted, which specifically includes:
[0039] S401. Define the initial state of land use planning in the first phase of industrial park planning and design as , then the expected land use planning status of the industrial park after s conversions in the sth stage is , expressed as , is the probability matrix after the initial state has been transformed s times;
[0040] S402. Based on the expected land use planning status of the converted industrial park , predict the functional land area at different stages during the planning and design period of the industrial park.
[0041] Furthermore, the initial state , For the industrial park The initial status of the functional plots; the expected land use planning status of the industrial park after the conversion , For the industrial park The land use planning state after the functional land parcel is converted s times; the probability matrix after the initial state is converted s times is a transition probability matrix The sth power, a conversion probability matrix It is defined as the transition probability from land use planning state a to land use planning state b, expressed as:
[0042] C ab = [ C 1 1 C 1 2 ⋯ C 1 n C 2 1 C 22 ⋯ C 2 n ⋮ ⋮ ⋮ ⋮ C n 2 C n 2 C nn ] ;
[0043] is the transition probability between the nth functional plot in land use planning state a and the nth functional plot in land use planning state b.
[0044] Furthermore, in step S5, the multivariate load forecasting results at different stages during the planning and design period of the industrial park are expressed as:
[0045] ;
[0046] is the predicted value of the j-th type of load in the industrial park in the planning year t; is the load density saturation value of the jth type of load of the zth type of functional land; is the load density saturation rate of the j-type load of the l-th plot in the z-th functional plot in the planning year t; is the area of the lth plot of the zth functional plot in the planning year t; is the plot ratio of the lth plot in the zth functional plot; It is the load simultaneous use coefficient corresponding to the lth plot in the zth functional plot.
[0047] The beneficial effects of the present invention are:
[0048] In the face of uncertain factors such as dynamic adjustment of industrial park production capacity, industrial transformation, and expansion, the present invention proposes a multi-stage load prediction method for industrial parks taking into account development uncertainty on the basis of an improved classification and zoning method. Based on the load growth characteristics of each functional plot, an S-shaped development curve of load density saturation rate is constructed to obtain the load density saturation rate of each functional plot; considering the imbalance of multi-load development of different functional plots in the industrial park, based on the existing limited plot load data samples, the load density saturation value of each functional plot is calculated by a load density saturation value estimation method based on Gaussian kernel density; considering the impact of the evolution law of the functional plot type in the industrial park on the industrial park, the load density saturation value of each functional plot is estimated by a Gaussian kernel density method. The impact of the industrial park's development uncertainty on load forecasting is solved, and the multi-factor load forecasting of the industrial park is matched with the dynamic development and changes of the park, especially for newly built industrial parks and industrial parks with severely lacking historical load data, which has strong adaptability.
[0049] Other features and advantages will be described in the following description, and partly become apparent from the description, or understood by practicing the invention. The purpose and other advantages of the invention are realized and obtained by the structures particularly pointed out in the description and the drawings.
[0050] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0051] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0052] Figure 1 A flow chart of a method for multi-stage load forecasting of an industrial park taking into account development uncertainty according to the present invention;
[0053] Figure 2 This is a schematic diagram of the principle of multi-stage load forecasting for an industrial park taking into account development uncertainty according to the present invention;
[0054] Figure 3 This is a schematic diagram of estimating the saturation value of heat load density of the present invention;
[0055] Figure 4 A structural diagram of an electronic device provided by the present invention;
[0056] Figure 5 A structural diagram of a multi-stage load forecasting device for an industrial park taking development uncertainty into account provided by the present invention. DETAILED DESCRIPTION
[0057] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are 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.
[0058] Example 1
[0059] like Figure 1 , Figure 2 As shown, this embodiment 1 provides a multi-stage load forecasting method for an industrial park taking into account development uncertainty, which includes:
[0060] S1. According to the development objectives and land planning of the industrial park, the industrial park shall be divided into multiple plots with different functional types during the planning and design period of the industrial park;
[0061] S2. Construct an S-shaped development curve of load density saturation rate and solve the load density saturation rate of each functional plot corresponding to different planning years;
[0062] S3. Considering the imbalance of the development of multiple loads in different functional plots in the industrial park, based on the load data samples of the existing limited plots, the load density saturation value of each functional plot is calculated by the load density saturation value estimation method based on Gaussian kernel density; the multiple loads include at least electrical load, heat load and cooling load;
[0063] S4. Considering the impact of the evolution of the types of functional plots in industrial parks on the load forecast of industrial parks, a Markov prediction model for the mutual conversion of different functional plots is constructed to predict the area of each functional land at different stages during the planning and design period of industrial parks;
[0064] S5. Based on the load density saturation rate, load density saturation value and functional land area of each functional plot in the industrial park, the multivariate load prediction results at different stages during the planning and design period of the industrial park are calculated.
[0065] In this embodiment, in step S1, the multiple plots of land with different functional types include at least: residential land, commercial service facility land, industrial land, and public management and public service facility land.
[0066] In actual applications, residential land refers to land used for residences and corresponding service facilities; commercial service facility land refers to land used for various commercial, business, entertainment and other facilities; industrial land refers to land used for production workshops, storage rooms and their ancillary facilities of industrial enterprises; public administration and public service facility land refers to land used for administrative, cultural, educational, sports, health and other institutions and facilities; in addition, it also includes land used for transportation facilities, land used for public facilities, green space and square land; land used for transportation facilities refers to land used for urban roads, transportation facilities and other facilities; land used for public facilities refers to land used for supply, environment, safety and other facilities; green space and square land refers to open space land such as park green space and protective green space.
[0067] In step S1, the various stages of the industrial park planning and design period include: the park formation stage, the park growth stage, the park maturity stage, and the park renewal or decline stage; the multidimensional factors include the scale expansion of each functional plot, the occupancy rate of enterprises, the adjustment of industrial structure, the fluctuations of policies and market economic factors, external climate and environmental factors, and the actual life span differences of various types of equipment.
[0068] In this embodiment, the characteristics of the load of each functional plot changing with the dynamic development of the industrial park are analyzed, specifically including:
[0069] 1) Analyze the intra-day variation pattern of the multi-load on the demand side of each functional plot, and obtain the peak and valley characteristics and multi-dimensional influencing factors of the multi-load on each functional plot; the multi-dimensional influencing factors are related to the functional type of the plot and the behavior preference of the plot;
[0070] 2) Based on the daily variation patterns of the multi-loads on the demand side of each functional plot, combined with the dynamic development and change characteristics of the industrial park at various stages of the planning and design period, determine the multi-dimensional influencing factors of the multi-load changes of each functional plot; the multi-dimensional influencing factors include the scale expansion of each functional plot, the occupancy rate of enterprises, the adjustment of industrial structure, the fluctuations of policies and market economic factors, external climate and environmental factors, and the actual life differences of various equipment;
[0071] 3) Analyze the load change characteristics of different functional plots and the same functional plots: Due to the different types of user entities on the demand side, in the dynamic development and change process of the industrial park at different stages, the multi-dimensional influencing factors of the multiple loads of each functional plot and the different load change laws lead to different load growth rates, and the load peaks of different functional plots will not appear at the same time; and, based on the influence of the dynamic development and change characteristics of the park at different stages, there are also differences in the dynamic development and changes between the same functional plots, and the load density saturation value of the same functional plot is also different.
[0072] It should be noted that industrial parks have a large number of users, a large total amount of energy consumption, and multiple energy consumption methods. The characteristics of multiple loads in industrial parks are diverse and are affected by time factors, weather factors, user types, etc. The change pattern of multiple loads in residential buildings is usually manifested as follows: in terms of electrical load, the peak occurs in the morning and evening, which is related to the residents' daily routine of going out early and returning late. The heat load increases during the midnight period, while the cold load has remained at a high level since noon. The demand for cold and hot loads during these periods is relatively increased; in addition, the cold and hot load curves have no obvious peak and valley characteristics, indicating that the cold and hot loads are not only related to user behavior, but also affected by multiple factors such as weather changes and seasonal demand. The change pattern of cold and hot loads presents complex disorder and randomness.
[0073] The variation pattern of multiple loads in commercial service facility land is usually as follows: around 10 a.m., the load growth rate accelerates, which is related to the start of commercial activities, including the opening of stores and the start of work of employees in office buildings. The small peak in energy consumption around noon is caused by concentrated commercial activities such as employee meals and peak customer flow in shopping malls. As time goes on, the load continues to rise. From 17:00 to 21:00 in the evening, the multiple load values fluctuate within a large range. Commercial activities are most frequent and intensive during this period. It drops sharply after 22:00, which is due to the end of commercial activities in commercial buildings and the departure of employees from office areas.
[0074] Formation stage: In this stage, the focus is on park planning and infrastructure construction, and methods such as attracting investment and reducing taxes are adopted, while relevant enterprises are introduced. The park's production capacity expands with the increase in the number of enterprises and total production. Growth stage: At this time, the park relies on the existing infrastructure and regional resource advantages to cultivate emerging industries and promote regional development. Through industrial agglomeration, the park achieves scale and clustering, forming an efficient industrial network, thereby rapidly enhancing production capacity. Maturity stage: In this stage, all work in the park is running well, and industrial development has entered a relatively stable period. Renewal or decline stage: Over time, some parks face the challenges of industrial technology upgrading and transformation due to market conditions, technical conditions and homogeneous competition, and enter a renewal or decline period.
[0075] The load density method predicts the future load demand of the industrial park by analyzing the land classification standards and detailed planning of the park planning and design, combined with the known empirical data of load density of industrial parks. Specifically, the load density method multiplies the load density of different land types in the park by the planned land area to estimate the load amount of different land types. This method can predict the load demand of industrial parks at different stages and provide a reference for planning and design. However, the load density method also has certain limitations. Due to the uncertainty of the time and scale of the settled enterprises, the prediction results of the load density method may be affected by these factors. In addition, the load density method also needs to rely on known load density data of industrial parks. Therefore, for newly built parks or industrial parks that lack relevant data, it is necessary to supplement the load density reference data to support the load forecast of industrial parks.
[0076] In this embodiment, in step S2, the load density of each functional plot complies with the S-curve development; the load density grows slowly at first, then the load density grows rapidly, then the load density continues to grow rapidly, and finally the load density maintains a steady growth until it grows to a saturation value. Constructing the load density saturation rate S-curve development curve specifically includes:
[0077] S201, use logistic curve to calculate load density saturation rate , expressed as:
[0078] ;
[0079] t is the planning year; k, t c is a constant; k determines the rising slope of the load growth curve of each functional plot. When the load growth rate of each plot is the same, the k value is the same; t c It indicates that the corresponding planning period when the load growth rate is the fastest has nothing to do with the k value;
[0080] S202, according to k, t c The value of is taken, and each functional plot is defined as two dimensions: the same functional plot and similar functional plot, and the load density saturation rate S-shaped development curve corresponding to the two dimensions is calculated and constructed respectively.
[0081] It should be noted that, with the dynamic development process of the industrial park, the load density growth curve of a certain area presents an S-shape, that is, the load density grows slowly in the initial development stage; as the load density grows faster, it enters a rapid development stage, and in a certain year the load density growth rate reaches a maximum value; then the load density continues to grow rapidly and enters the post-development stage; when it reaches the saturated development stage, with the restrictions of long-term urban population planning, land resources, energy utilization efficiency, etc., the load density finally maintains a steady growth until it grows to the saturation value.
[0082] In this embodiment, the same functional plots are defined as functional plots with the same multi-load growth rate, the same multi-load density saturation rate S-shaped development curve, the same k value and t c value;
[0083] The S-shaped development curve of the multivariate load density saturation rate of the same functional plot is expressed as:
[0084] ;
[0085] , , They are the S-shaped development curves of load density saturation rate of electrical load, thermal load and cooling load in the multiple loads of the same functional plot.
[0086] In actual applications, on the same functional plot, the total energy consumption is proportional to the frequency of production and living activities in the area, and the total growth rate of electricity, heat and cooling loads is the same, that is, the S-shaped development curve of the density saturation rate of electricity, heat and cooling multi-loads is the same, k, t c For industrial parks, we can refer to the electric load development curve obtained from a certain functional land (such as residential land, industrial land, public management service land) to infer the load density saturation rate S-shaped development curve of the heat load and cold load on this land.
[0087] In this embodiment, the similar functional plots are defined as functional plots with different construction times, similar dynamic development changes, and similar multi-load density saturation rate S-shaped development curves, with the same k value and different t c value;
[0088] For two similar functional plots, the multivariate load density saturation rate S-shaped development curve of one functional plot is defined as , the curve parameters are k, t c , the construction period is The S-shaped development curve of the multivariate load density saturation rate of another functional plot to be calculated is: , the construction period is , then the multivariate load density saturation rate S-shaped development curve of another functional plot is It is expressed as:
[0089] ξ ′ ( t ) = 1 1 + e − k [ t − t c − ( T ′ 0 − T 0 )] .
[0090] In practical applications, similar functional plots are defined as plots with different construction times but similar economic levels, industrial structures, population densities, etc. The multivariate load density saturation rate S-shaped development curves are similar and have the same k value.c The value depends on the construction time. The curves of two similar plots are translated in the plane coordinate system according to the difference in the initial construction time.
[0091] In this embodiment, in step S3, based on the existing limited plot load data samples, the load density saturation value of each functional plot is calculated by a load density saturation value estimation method based on Gaussian kernel density, specifically including:
[0092] S301. For newly built or formative industrial parks, the sample of land load data is limited. Assuming that a functional land parcel in an industrial park has sample plots, the load density is , the probability density function of load density is , expressed as:
[0093] ;
[0094] is the sample plot capacity; W is the bandwidth; As the kernel function, the standard Gaussian function is selected; is the input load density parameter; For the Load density of sample plots;
[0095] S302, generating a probability density curve of the load density according to the formula in S301; dividing the load density into a number of density intervals based on the probability density curve of the load density, which can be specifically divided into a low load density interval, a medium load density interval and a high load density interval, or a low load density interval and a high load density interval;
[0096] S303, in each load density interval obtained in step S302, find the load density value with the maximum probability density, and use the value as the typical load density representative value of the interval, to obtain the low load density typical value of the low load density interval, the medium load density typical value of the medium load density interval, and the high load density typical value of the high load density interval;
[0097] S304. The load density between the low load density typical value and the high load density typical value obtained in step S303 is taken as the concentrated area of the load density of the functional plot, and the high load density typical value is selected as the load density saturation value of the functional plot.
[0098] It should be noted that the kernel function needs to meet the following three conditions, including:
[0099] ;
[0100] ;
[0101] ;
[0102] The standard Gaussian function is expressed as:
[0103] ;
[0104] Parameters for the kernel function input.
[0105] In practical applications, by analyzing the probability density distribution of load density, the minimum and maximum values of all sample load densities under a specific land use type and the probability density size under a specific load density value can be obtained. In order to reduce the impact of extreme value samples on the data and to more accurately understand the distribution trend of load density, the lowest and highest 5% load density data are excluded, and the lower and upper limits of load density are redefined accordingly.
[0106] like Figure 3 As shown in the figure, the heat load density of the industrial land in the park presents a typical probability density curve distribution characteristic, showing a "double peak" feature. This shows that the heat load density of the industrial land in the park can be divided into two categories: low-segment density and high-segment density. After removing 5% of the data at the head and tail (i.e., the highest and lowest) of the probability density graph, the upper and lower limits of the heat load saturation density of the industrial land are obtained to be (7.21~124.15) W / m 2 The heat load density corresponding to the maximum probability density of the high-density segment is taken as the reference value of the heat load saturation density of this land type, that is, the heat load density saturation value of the industrial land in the park is 114.84 W / m 2 .
[0107] In this embodiment, in step S4, in different planning and design stages of the industrial park, the types of functional plots in the industrial park evolve and adjust with the dynamic development of the park, and the land area and multivariate load values of each functional plot change accordingly; the Markov chain theory is introduced, and the land planning status of each functional plot is represented by the size of the land area, and the land planning status of each functional plot is converted to each other at different stages during the planning and design period.
[0108] In step S4, the influence of the evolution law of the functional plot types of the industrial park on the load forecast of the industrial park is considered, and a Markov prediction model for the mutual conversion of different functional plots is constructed to predict the land area of each function at different stages during the planning and design period of the industrial park, including:
[0109] S401. Define the initial state of land use planning in the first phase of industrial park planning and design as , then the expected land use planning status of the industrial park after s conversions in the sth stage is , expressed as , is the probability matrix after the initial state has been transformed s times;
[0110] S402. Based on the expected land use planning status of the converted industrial park , predict the functional land area at different stages during the planning and design period of the industrial park.
[0111] It should be noted that, with the rapid development of social economy and the continuous advancement of urbanization, the adjustments in land use planning brought about by changes in land, industry, finance, population and other policies have increased the uncertainty of the dynamic development and changes of industrial parks, and brought difficulties to the load forecasting of the planning and design of the comprehensive energy system of industrial parks. Therefore, it is unreasonable to directly carry out load forecasting without considering the land evolution characteristics at different development stages in the dynamic development and changes of industrial parks. In order to study the impact of the evolution of land use types in industrial parks on the load forecasting of industrial parks, this application will construct a Markov prediction model for the mutual transformation of different functional land types in industrial parks to predict the area changes of various functional land at different stages during the comprehensive energy planning period of industrial parks. By analyzing the initial state probability matrix of different land use types in industrial parks and the transition probability matrix between various functional land types, the Markov process is used to study the state evolution trend of land use types to achieve the purpose of predicting future trends.
[0112] In this embodiment, the initial state , For the industrial park The initial status of the functional plots; the expected land use planning status of the industrial park after the conversion , For the industrial park The land use planning state after the functional land parcel is converted s times; the probability matrix after the initial state is converted s times It is the sth power of a conversion probability matrix. It is defined as the transition probability from land use planning state a to land use planning state b, expressed as:
[0113] C ab = [ C 1 1 C 1 2 ⋯ C 1 n C 2 1 C 22 ⋯ C 2 n ⋮ ⋮ ⋮ ⋮ C n 2 C n 2 C nn ] ;
[0114] is the transition probability between the nth functional plot in land use planning state a and the nth functional plot in land use planning state b.
[0115] In this embodiment, in step S5, the multivariate load forecasting results at different stages during the planning and design period of the industrial park are expressed as:
[0116] ;
[0117] is the predicted value of the j-th type of load in the industrial park in the planning year t; is the load density saturation value of the jth type of load of the zth type of functional land; is the load density saturation rate of the j-type load of the l-th plot in the z-th functional plot in the planning year t; is the area of the lth plot of the zth functional plot in the planning year t; is the plot ratio of the lth plot in the zth functional plot; It is the load simultaneous use coefficient corresponding to the lth plot in the zth functional plot.
[0118] It should be noted that load forecasting can adopt the classification and zoning method. The main idea of the classification and zoning method is to set the load saturation density of various functional blocks according to the land use plan to comprehensively calculate the load during the planning period. The advantage of this method is that it can better consider the impact of land planning on the load, but it also ignores the impact of dynamic development and changes in the park during the planning period. The present application adopts an improved classification and zoning method to carry out multi-dimensional load forecasting in industrial parks. According to the detailed control planning map of the area issued by the planning department, the land use nature, land area, floor area ratio and other information of each plot are counted, and the plots with the same functional positioning in the planning area are further integrated into functional zones. Then, according to building simulation, relevant planning specifications or building energy consumption standards and other methods, the multi-dimensional load saturation density indicators of different functional land use planning are calculated. According to the different attributes of the plots, load forecasting is classified and expressed as:
[0119] ;
[0120] is the predicted value of the industrial park regional load, W; It is the load density index of different functional plots in the industrial park; For different functional land types; is the area of the corresponding functional land; is the plot ratio of the corresponding functional land; It is the simultaneous use coefficient of the load for the corresponding functional land.
[0121] The load density of the actual land in the industrial park shows a dynamic growth trend with the entry of enterprises, economic development, and population growth. Therefore, the load density is a function of time and changes over time. To simplify the calculation, the load density saturation rate will be used to indirectly represent the development and change of the park's load density over time. Therefore, the load density saturation rate for:
[0122] ;
[0123] represents the function of load density changing with time, is the load saturation density value.
[0124] Example 2
[0125] like Figure 4 The present invention provides an electronic device, including: one or more processors; a storage medium for storing one or more programs; and an internal bus and a network interface; when the one or more programs are executed by the one or more processors, the one or more processors implement the multi-stage load forecasting method for industrial parks taking into account development uncertainty. In addition, the electronic device may also include an internal bus and a network interface, and the network interface and the memory are connected via the internal bus; the memory may specifically include a memory and a non-volatile memory.
[0126] Example 3
[0127] like Figure 5 The present invention provides a multi-stage load forecasting device for an industrial park taking into account development uncertainty, the device comprising:
[0128] The land parcel division module is used to divide the industrial park into multiple land parcels with different functional types according to the development target positioning and land planning of the industrial park during the planning and design period of the industrial park;
[0129] The load density saturation rate calculation module is used to construct the load density saturation rate S-shaped development curve and solve the load density saturation rate of each functional plot corresponding to different planning years;
[0130] The load density saturation value calculation module is used to consider the imbalance of multi-load development of different functional plots in the industrial park. Based on the load data samples of the existing limited plots, the load density saturation value estimation method based on Gaussian kernel density is used to calculate the load density saturation value of each functional plot;
[0131] The land area calculation module is used to consider the impact of the evolution of the types of functional plots in industrial parks on the load forecast of industrial parks, build a Markov prediction model for the mutual conversion of different functional plots, and predict the land area of each function at different stages during the planning and design period of industrial parks;
[0132] The load forecasting module is used to calculate the multivariate load forecasting results at different stages during the planning and design period of the industrial park based on the load density saturation rate, load density saturation value and the area of each functional land plot in the industrial park.
[0133] In several embodiments provided in the present application, it should be understood that the disclosed system and method can also be implemented in other ways. The system embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems, methods and computer program products according to multiple embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and a part of the module, program segment or code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.
[0134] In addition, the functional modules in the various embodiments of the present invention may be integrated together to form an independent part, or each module may exist independently, or two or more modules may be integrated to form an independent part.
[0135] If the function is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.
[0136] Based on the above ideal embodiments of the present invention, the relevant staff can make various changes and modifications without departing from the technical concept of the present invention through the above description. The technical scope of the present invention is not limited to the contents of the specification, and its technical scope must be determined according to the scope of the claims.
Claims
1. A multi-stage load forecasting method for industrial parks taking into account development uncertainty, It is characterized in that It includes: S1. According to the development objectives and land planning of the industrial park, the industrial park shall be divided into multiple plots with different functional types during the planning and design period of the industrial park; S2. Construct an S-shaped development curve of load density saturation rate and solve the load density saturation rate of each functional plot corresponding to different planning years; S3. Considering the imbalance of multi-load development of different functional plots in the industrial park, based on the load data samples of the existing plots, the load density saturation value of each functional plot is calculated by the load density saturation value estimation method based on Gaussian kernel density; S4. Considering the impact of the evolution of the types of functional plots in industrial parks on the load forecast of industrial parks, a Markov prediction model for the mutual conversion of different functional plots is constructed to predict the area of each functional land at different stages during the planning and design period of industrial parks; S5. Based on the load density saturation rate, load density saturation value and functional land area of each functional plot in the industrial park, calculate the multivariate load prediction results at different stages during the planning and design period of the industrial park; In S2, an S-shaped development curve of load density saturation rate is constructed, which specifically includes: S201. Use the logistic curve to calculate the load density saturation rate ξ(t), expressed as: t is the planning period; k, t c is a constant; k determines the rising slope of the load growth curve of each functional plot. When the load growth rate of each plot is the same, the k value is the same; t c It indicates that the corresponding planning period when the load growth rate is the fastest has nothing to do with the k value; S202, according to k, t c The values of are used to define each functional plot into two dimensions: the same functional plot and the similar functional plot, and the load density saturation rate S-shaped development curves corresponding to the two dimensions are calculated and constructed respectively; The same functional plots are defined as functional plots with the same multi-load growth rate, the same multi-load density saturation rate S-shaped development curve, and the same k value and t c value; The S-shaped development curve of multivariate load density saturation rate of the same functional plot is expressed as: x Ele (t)=ξ Heat (t)=ξ Cold (t); ξ Ele (t), ξ Heat (t), ξ Cold (t) are the S-shaped development curves of load density saturation rate of electric load, heat load and cooling load in the multi-load of the same functional plot; The similar functional plots are defined as functional plots with different construction time, similar dynamic development changes, and similar S-shaped development curves of multi-load density saturation rate, with the same k value and different t c value; For two similar functional plots, the multivariate load density saturation rate S-shaped development curve of one of the functional plots is defined as ξ(t), and the curve parameters are k, t c , the construction period is T 0 The multivariate load density saturation rate S-shaped development curve of another functional plot to be calculated is ξ′(t), and the construction period is T 0 ′, then the multivariate load density saturation rate S-shaped development curve of another functional plot is ξ′(t) and is expressed as: In S3, based on the existing plot load data samples, the load density saturation value of each functional plot is calculated by a load density saturation value estimation method based on Gaussian kernel density, specifically including: S301, Assume that there are N functional plots in the industrial park l The corresponding load densities are The probability density function of load density is Γ(ρ), which is expressed as: N l is the sample plot capacity; W is the bandwidth; Θ(·) is the kernel function, and the standard Gaussian function is selected; ρ is the input load density parameter; ρ e is the load density of the e-th sample plot; S302, generating a probability density curve of load density according to the formula in S301; dividing the load density into a plurality of density intervals based on the probability density curve of load density and its peak and valley values; the density intervals at least include a low load density interval and a high load density interval; S303, in each load density interval obtained in step S302, find the load density value with the maximum probability density, and use the value as the typical load density representative value of the interval, to obtain the low load density typical value of the low load density interval and the high load density typical value of the high load density interval; S304, taking the load density between the low load density typical value and the high load density typical value as the concentrated area of the load density of the functional plot, and selecting the high load density typical value as the load density saturation value of the functional plot; In S4, the influence of the evolution law of the functional plot types of the industrial park on the load forecast of the industrial park is considered, and a Markov prediction model for the mutual conversion of different functional plots is constructed to predict the land area of each function at different stages during the planning and design period of the industrial park, specifically including: S401. Define the initial state of land use planning in the first phase of the industrial park planning and design period as D (0) , then the expected land use planning status of the industrial park after s conversions in the sth stage is D (s) , denoted as D (s) =D (0) C (s) , C (s) is the probability matrix after the initial state has been transformed s times; S402, based on the expected land use planning status of the converted industrial park D (s) , predict the land area of each function at different stages during the planning and design period of the industrial park; The initial state is the initial state of the lth functional plot in the industrial park; The expected land use planning status of the industrial park after the conversion is the land use planning state of the lth functional plot in the industrial park after s conversions; the probability matrix C after the initial state is converted s times (s) is a transition probability matrix C ab The sth power, the first conversion probability matrix C ab It is defined as the transition probability from land use planning state a to land use planning state b, expressed as: C nn is the transition probability between the nth functional plot in land use planning state a and the nth functional plot in land use planning state b.
2. The multi-stage load forecasting method for an industrial park taking into account development uncertainty according to claim 1, It is characterized in that In step S1, the multiple plots of land with different functional types include at least: residential land, commercial service facility land, industrial land, and public management and public service facility land.
3. The multi-stage load forecasting method for industrial parks taking into account development uncertainty according to claim 1, It is characterized in that In S5, the multivariate load forecasting results at different stages during the planning and design period of the industrial park are expressed as: L j (t) is the predicted value of the j-th load in the industrial park in the planning period of year t; ρ j,z is the load density saturation value of the jth type of load of the zth type of functional land; j,z,l (t) is the load density saturation rate of the j-type load of the l-th plot in the z-th functional plot when the planning period is t years; S z,l (t) is the area of the lth plot of the zth functional plot when the planning period is t years; V z,l is the plot ratio of the lth plot in the zth functional plot; U z,l It is the load simultaneous use coefficient corresponding to the lth plot in the zth functional plot.
4. An electronic device, It is characterized in that include: one or more processors; A storage medium for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 3.
