Load regulation and control method for building demand response potential evaluation based on non-privacy data
By building a multi-attribute evaluation index system and a fuzzy comprehensive evaluation method, the building demand response evaluation model is optimized, and the problems of information leakage and evaluation deviation in the existing technology are solved, and the prediction accuracy and response management capabilities of building load regulation are improved.
Patent Information
- Application Number
- CN202411031998.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-30
- Publication Date
- 2024-11-26
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing technology has information leakage risks and deviations in the evaluation of building demand response potential, and relying solely on mathematical statistics and fuzzy theory is difficult to meet the needs of smart grids and complex electricity use behaviors, and the prediction accuracy is insufficient.
Define building demand response decisions as multi-attribute decision-making problems, build a regulatory potential evaluation index system including load, user, economy and management levels, adopt fuzzy comprehensive evaluation and entropy weight method combined with game theory method, optimize weight combination, and train through PSO algorithm Evaluate the model to improve the accuracy and versatility of the evaluation.
It improves the accuracy of scheduling task allocation, enhances the prediction accuracy and response management capabilities of building load regulation, and meets the evaluation needs of smart grids and complex power consumption behaviors.
Smart Images

Figure CN119026980A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of power grid demand response, and in particular to a load regulation method for building demand response potential assessment based on non-privacy data. Background Art
[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.
[0003] Due to the randomness and uncertainty of renewable energy, the power system faces huge pressure to balance supply and demand. With the decreasing adjustable resources on the power generation side, it is difficult to cope with the supply and demand balance problem in the operation of the power grid by relying solely on the adjustment capacity of the power generation side. In order to cope with the increasingly severe imbalance between supply and demand, load-side regulation has received more and more attention. Building systems have also become the focus of attention due to their high operating energy consumption, and power users have gradually participated in various businesses of power grid companies. Considering the high controllability of user loads, adjusting loads is of great significance for changing user load curves and achieving peak load reduction and valley filling of power grids. Therefore, through the overall regulation of building loads, effective measures to cope with the rapid growth of loads are formed, which can not only further supplement and enrich the existing load control means, improve the utilization efficiency and overall economy of energy, but also ensure the safe and stable operation of the power grid, significantly reduce the investment in installed capacity and transmission and distribution network capacity to meet peak load demand, and achieve optimal allocation of power resources and sustainable development.
[0004] In the process of implementing demand response, evaluating the user's response potential in advance can better improve the accuracy of scheduling task allocation. Direct load control is the main way for user load to participate in demand-side regulation. The direct load control method belongs to top-down control. On the one hand, the real-time information of resources on the user side cannot be obtained in time, making it difficult to fully tap its flexible regulation potential; on the other hand, this intrusive control is prone to information leakage problems, reducing the number of times users actively participate in the operation of the distribution network. Therefore, the way and number of times users choose to participate in demand response are important factors affecting the building demand response potential. Most of the existing research focuses on modeling users' orderly electricity consumption behavior and grid dispatching strategies considering load response, while research on the selection of users participating in demand response and the effective acquisition of high-potential electricity users is relatively limited.
[0005] Regarding the quantitative evaluation system of demand response, existing studies usually establish an indicator system for the corresponding potential of user demand from three aspects, namely, user historical behavior and user information, as well as incentives from aggregators to users. However, they ignore indicators such as the number of user adjustments and the timeliness of adjustments. These factors will affect the quality of demand response and lead to deviations in the evaluation results.
[0006] At present, the most widely used technologies for building demand response potential assessment methods are mathematical statistics and fuzzy theory. Most of the evaluations are subjective, and subjective factors will have a certain negative impact on the evaluation results. The use of multivariate statistical methods such as principal component analysis has high requirements for sample size and the distribution form of evaluation index data, which limits its application field. In addition, with the development of smart grids and the increasing complexity of building electricity consumption behavior, the model composed of these potential assessment methods alone to assess the building demand response potential can no longer meet the needs and cannot achieve the prediction accuracy required for current response management. Summary of the invention
[0007] In order to solve at least one technical problem existing in the above-mentioned background technology, the present invention provides a load control method for building demand response potential evaluation based on non-privacy data. Based on the current status of building demand response, the building demand response decision is defined as a typical multi-attribute decision-making problem, and then based on the principle of multi-attribute, the influencing factors of the decision are mined and classified, and a building control potential evaluation index system is constructed. The evaluation values are trained and tested to improve the prediction accuracy and versatility of the results, and finally a control plan for building load control is obtained.
[0008] In order to achieve the above object, the present invention adopts the following technical solution:
[0009] A first aspect of the present invention provides a load control method for evaluating building demand response potential based on non-privacy data, comprising the following steps:
[0010] Access historical data that influences demand response decisions in commercial buildings;
[0011] Based on historical data that affects commercial building demand response decisions and a building demand response potential evaluation model, a potential value of building demand response is obtained; wherein the construction process of the building demand response potential evaluation model includes:
[0012] The building demand response decision is defined as a multi-attribute decision problem. Based on the principle of multi-attribute, the influencing factors of the decision are mined and classified, and an evaluation index system for building regulation potential is constructed.
[0013] Based on historical data and the building regulation potential evaluation index system, the non-privacy data value is calculated;
[0014] The building demand response potential is quantified to obtain the weight of the evaluation index; the fuzzy comprehensive evaluation method is combined with the weight of the evaluation index to calculate the potential value of the building demand response;
[0015] The comprehensive indicator weight is used as the real risk value to determine the number of training samples and test samples. The potential value of building demand response is used as a priori sample to train and test the constructed evaluation model. The building demand response potential evaluation result is obtained based on the trained evaluation model.
[0016] The task scheduling allocation is guided by the building demand response potential assessment results to obtain the task scheduling allocation control strategy, and the corresponding equipment operation is controlled according to the task scheduling allocation control strategy.
[0017] Furthermore, the building regulation potential evaluation index system includes four aspects: load level, user level, economic level, and management level;
[0018] The load level includes five indicators: peak-to-valley ratio, average load, weekly load reduction rate, peak power consumption ratio, and load fluctuation rate;
[0019] The user level includes four indicators: capacity ratio of load used for dispatch response, average annual load reduction, smart meter coverage, and user comfort requirements;
[0020] The economic level includes five indicators: local economic level, user social benefits, income per unit response capacity, and pollutants per unit electricity;
[0021] The management level includes three indicators: number of user adjustments, actual degree of cooperation, and timeliness of adjustments.
[0022] Furthermore, the building demand response potential is quantified to obtain the weight of the evaluation index, including:
[0023] For the indicators at different levels of the building regulation potential evaluation index system, the fuzzy analytic hierarchy process is used to quantify the building demand response potential and obtain the subjective weights;
[0024] Combined with the non-privacy data value, the entropy weight method is used to obtain the objective weight;
[0025] Combining subjective weights and objective weights, the game theory method is used to optimize the combined weights and the linear combination coefficients between the two weights, with the goal of minimizing the deviation to obtain the most reasonable weights.
[0026] Furthermore, the indicators at different levels of the building regulation potential evaluation index system are quantified using the fuzzy hierarchical analysis method to obtain subjective weights, including:
[0027] Compare the indicators at different levels in pairs to determine an importance score, fill in the matrix, and establish the judgment matrix R = (r ij ) n×n , where r ij Indicates the index x i Xj The relative importance of ij +r ji =1, r ij =r ij -r jk +0.5;
[0028] The fuzzy scale value corresponding to the scale of 0.1-0.9 is used, 0.1 means that the indicator x i With x j Compared with the above, they are of equal importance; 0.3 means that the index x i With x j In comparison, x i With x j Slightly important; 0.5 means that the index x i With x j In comparison, x i With x j Obviously important; 0.7 means that the index x i With x j In comparison, x i With x j Strongly important; 0.9 means that the indicator x i With x j In comparison, x i With x j Extremely important; 0.2, 0.4, 0.6, 0.8 means that the index x i With x j In comparison, the median of the above two adjacent judgments;
[0029] Sum the original judgment matrix row by row Then perform data transformation b ij =(r i -r j ) / 2n+0.5, construct the consistency matrix B=(b ij ) n×n ;
[0030] The ranking method is used to calculate the weight, obtain the relative importance weight of each indicator to the indicator of the previous level, and calculate the subjective weight of the indicator.
[0031] Furthermore, the non-privacy data value is combined to obtain an objective weight using an entropy weight method, including:
[0032] According to the principle of combining qualitative and quantitative methods, the indicator matrix R′=(r ij ') m×n , where potential has n evaluation indicators, each indicator has m evaluation levels, r ij ' represents the evaluation value of the jth evaluator for the i-th indicator;
[0033] r ij 'Do standardization. The index system consists of positive and negative indicators. Standardization is required to obtain r ij ;
[0034] Calculate the entropy value of the ith indicator, and combine it with the entropy value of the ith indicator to obtain the entropy weight of the jth indicator, that is, the objective weight of the indicator.
[0035] Furthermore, the comprehensive indicator weight is used as the real risk value, the number of training samples and test samples is determined, and the potential value of building demand response is used as a priori sample to train and test the constructed evaluation model, including:
[0036] Initialize PSO parameters and variable optimization range, including population size N, maximum number of iterations T, learning factors c1 and c2, inertia weight w, i-th particle, position X i ;
[0037] Based on particle position X i Calculate fitness with training set;
[0038] Calculate the fitness of the PSO algorithm particle speed and position after updating;
[0039] If the current number of iterations t>T, the optimal position of the particle is output, and the optimal position is used as the optimal initial weight and threshold of the ELM model. It is substituted into the ELM model and the test set is used to verify and analyze the building demand response potential.
[0040] A second aspect of the present invention provides a load control system for evaluating building demand response potential based on non-privacy data, comprising:
[0041] A data acquisition module for acquiring historical data that influences demand response decisions for commercial buildings;
[0042] A demand response potential assessment module is used to obtain a potential value of building demand response based on historical data that affects commercial building demand response decisions and a building demand response potential assessment model; wherein the construction process of the building demand response potential assessment model includes:
[0043] The building demand response decision is defined as a multi-attribute decision problem. Based on the principle of multi-attribute, the influencing factors of the decision are mined and classified, and an evaluation index system for building regulation potential is constructed.
[0044] Based on historical data and the building regulation potential evaluation index system, the non-privacy data value is calculated;
[0045] The building demand response potential is quantified to obtain the weight of the evaluation index; the fuzzy comprehensive evaluation method is combined with the weight of the evaluation index to calculate the potential value of the building demand response;
[0046] The comprehensive indicator weight is used as the real risk value to determine the number of training samples and test samples. The potential value of building demand response is used as a priori sample to train and test the constructed evaluation model. The building demand response potential evaluation result is obtained based on the trained evaluation model.
[0047] The load control module is used to guide the task scheduling allocation according to the building demand response potential assessment result to obtain the task scheduling allocation control strategy, and control the corresponding equipment operation according to the task scheduling allocation control strategy.
[0048] A third aspect of the present invention provides a computer-readable storage medium.
[0049] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps in the load regulation method for evaluating building demand response potential based on non-privacy data as described above.
[0050] A fourth aspect of the present invention provides a computer device.
[0051] A computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in the load regulation method for evaluating building demand response potential based on non-privacy data as described above are implemented.
[0052] Compared with the prior art, the present invention has the following beneficial effects:
[0053] The present invention constructs a building demand response potential index system and provides a comprehensive building demand response potential evaluation system based on non-privacy data. It takes the current orderly electricity consumption scheduling method as the basis, fully considers the load characteristics of the building, and establishes a demand response potential evaluation index system through four aspects: load level, user level, economic level, and management level. Combined with the demand response potential evaluation index system, the accuracy of scheduling task allocation can be better improved.
[0054] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0055] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.
[0056] Figure 1 It is a flow chart of a load regulation method for evaluating building demand response potential based on non-privacy data provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0057] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.
[0058] It should be noted that the following detailed descriptions are all illustrative and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.
[0059] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.
[0060] Embodiment 1
[0061] like Figure 1 As shown, this embodiment provides a load control method for evaluating building demand response potential based on non-privacy data, including the following steps:
[0062] Step 1: Obtain power data information of multiple commercial buildings;
[0063] In this embodiment, the power data information of the commercial building includes various basic evaluation data, including vector data and table data formats;
[0064] After obtaining the power data information of multiple commercial buildings, the original data is pre-cleaned and converted, including data screening, outlier identification and correction, and dimensionless data processing.
[0065] Step 2: Build an assessment model for the demand response potential of buildings;
[0066] The construction of an evaluation model for building demand response potential includes the following steps:
[0067] Step 201: Based on the principle of multiple attributes, factors affecting demand response potential are mined and classified, and power information is acquired, counted and summarized;
[0068] Existing research focuses on load information such as orderly electricity consumption behavior of users and historical electricity consumption information in constructing an index system. Based on the analysis of the actual operation process of the power grid, it can be found that the evaluation of the load response potential of a building needs to consider the impact of economy, user load characteristics, user management, intelligence level and information characteristics. Therefore, the building demand response decision is defined as a typical multi-attribute decision problem. Based on the principle of multi-attribute, the influencing factors of the decision are mined and classified. Starting from the load level, user level, economic level and management level, an index system for evaluating the building regulation potential is proposed.
[0069] According to the different characteristics of user load information, the first level of demand response potential is constructed: load level B1, which consists of five indicators: peak-to-valley ratio, average load, weekly load reduction rate, peak power consumption ratio, and load fluctuation rate;
[0070] Information such as the technical transformation capabilities of building users, user comfort requirements, and historical average annual peak load reduction are all affected by the decisions of building power system managers. Therefore, a second level is constructed: the user level B2, which consists of four indicators: the capacity proportion of load that can be used for dispatch response, the average annual load reduction, the smart meter coverage rate, and the user comfort requirements;
[0071] Different types of power users will suffer different losses during the load response process due to their different business models and economic benefits. In addition, under normal circumstances, if the user's electricity pollutant emissions are large, the load response of this type of power user will cause a certain social impact, thereby indirectly affecting the demand response potential. Based on this, the third level is constructed: the economic level B3, which consists of five indicators: local economic level, user social benefits, income per unit response capacity, and pollutants per unit of electricity;
[0072] The building demand response and regulation potential are affected by the actual number of users and the management level. The effect of managing demand response is mainly reflected in the number of regulation itself, the actual degree of cooperation, the timeliness of regulation, etc. Therefore, the fourth level is constructed: management level B4, which consists of three indicators: the number of user regulation, the actual degree of cooperation, and the timeliness of regulation.
[0073] Step 202: Based on the acquired power data information of multiple commercial buildings and the constructed evaluation model suitable for building demand response potential, a corresponding calculation method is called, and the calculated data is used as a non-privacy data value;
[0074] The load level B1 calculation includes:
[0075] Peak-to-valley ratio C1: the difference between the maximum daily load of the user and the average load during the non-peak period;
[0076] Average load C2: the average value of the user's load when not participating in peak load regulation;
[0077] Weekly load reduction rate C3: the degree to which the weekend load decreases compared to the weekday load during normal working hours;
[0078] Peak power consumption ratio C4: the proportion of power consumption during peak hours to the total power consumption during the day in the typical load curve of the user on a working day;
[0079] Load fluctuation rate C5: the ratio of the standard deviation of the user's load curve on that day to the average value.
[0080] User-level B2 calculations include:
[0081] The capacity proportion of load that can be used for dispatch response C6: the capacity of equipment that users can reduce or stop using when responding, reflecting how much customers have invested in energy-saving technology transformation and reflecting the current level of electricity utilization of customers;
[0082] Annual average load reduction C7: the average annual load reduction after users participate in the response;
[0083] Smart meter coverage C8: The proportion of smart meters in buildings, reflecting the user's ability to respond automatically;
[0084] User comfort requirements C9: User comfort requirements, reflecting the user's electricity usage behavior.
[0085] The economic level B3 calculation includes:
[0086] Local economic level C 10 :The economic level of users and electricity tariff levels are inversely related. The higher the economic level of users, the more capable they are of coping with peak electricity prices, but at the same time, they will pay less attention to compensation. The lower the economic level of users, the more likely they will actively respond to compensation and gain benefits;
[0087] The benefit obtained per unit response capacity C 11 : When buildings actively participate in regulation, the power grid should give users certain compensation according to relevant policies, and different participation modes have different compensation methods and degrees;
[0088] Pollutants per unit of electricity C 12 : Reflect the impact of building air conditioning system use on the ecological environment;
[0089] User social benefits: This indicator refers to the benefits that electricity users generate for society during the production process.
[0090] Management-level B4 calculations include:
[0091] User control times C 14: Measured by the number of times a user has been scheduled to participate in peak load shaving in the past year.
[0092] Actual degree of cooperation C 15 :When a building user receives a scheduling task, the user's enthusiasm and completion of the task, the aggregator will record the work completed at each task for subsequent scoring.
[0093] Regulation timeliness C 16 : Demand response requires the timeliness of information and communication to ensure that the information obtained can be updated in real time so that timely judgments can be made.
[0094] Step 203: Combined with the construction of an evaluation model suitable for building demand response potential, the fuzzy analytic hierarchy process is used to quantify the building demand response potential and obtain the subjective weight of the indicator; specifically, it includes:
[0095] Step 2031: Compare the indicators at different levels in pairs to determine an importance score, fill in the matrix, and establish a judgment matrix R = (r ij ) n×n ;
[0096] Among them, r ij Indicates the index x i X j The relative importance of ij +r ji =1, r ij =r ij -r jk +0.5;
[0097] The fuzzy scale value corresponding to the scale of 0.1-0.9 is used, 0.1 means that the indicator x i With x j Compared with the above, they are of equal importance; 0.3 means that the index x i With x j In comparison, x i With x j Slightly important; 0.5 means that the index x i With x j In comparison, x i With x j Obviously important; 0.7 means that the index x i With x j In comparison, x i With x j Strongly important; 0.9 means that the indicator x i With x j In comparison, x i With x j Extremely important; 0.2, 0.4, 0.6, 0.8 means that the index xi With x j In comparison, the median of the above two adjacent judgments;
[0098] Step 2032: Sum the original judgment matrix row by row Then perform data transformation b ij =(r i -r j ) / 2n+0.5, construct the consistency matrix B=(b ij ) n×n ;
[0099] Step 2032: Use the ranking method to calculate the weight, and obtain the relative importance weight of each indicator to the indicator of the previous level;
[0100] That is, use the sum of each row of the judgment matrix i=1,2,…,n; and the number of factors n is solved. The specific calculation is:
[0101]
[0102] Step 204: Combine the non-privacy data value, use the entropy weight method to obtain the objective weight of the indicator, construct the indicator matrix, and perform different standardized processing methods for positive indicators and negative indicators;
[0103] Step 2041: Construct an indicator matrix R′=(r ij ') m×n ;
[0104] Among them, potential has n evaluation indicators, each indicator has m evaluation levels, r ij ' represents the evaluation value of the jth evaluator for the i-th indicator.
[0105] Step 2042: Standardization of evaluation matrix:
[0106] Since different indicators have different dimensions, for the convenience of calculation, r ij 'Do standardization. The index system consists of positive and negative indicators. Standardization is required to obtain r ij ;
[0107] Step 2043: Calculate the entropy value of the i-th indicator;
[0108] in,
[0109] Step 2044: Calculate the entropy weight of the j-th indicator:
[0110]
[0111] Step 205: Combining the subjective weight and the objective weight, using the game theory method, calculate the comprehensive weight of the indicator; including:
[0112] Step 2051: Combine the weights and the linear combination of the two weights:
[0113] w=a1w1+a2w2,a1+a2=1,
[0114] Step 2052: According to the idea of the game aggregation model, the two linear combination coefficients a1 and a2 in the above formula are optimized to minimize the deviation and obtain the most reasonable weight;
[0115] Step 2053: By using the matrix differential property, a derivative is taken to obtain the following equation:
[0116]
[0117] Step 2054: normalize the combination coefficients to obtain the index weights;
[0118]
[0119] Step 206: Calculate the potential value of building demand response using the fuzzy comprehensive evaluation method, which specifically includes the following steps:
[0120] Step 2061: First, determine the commercial building load demand response potential value level table:
[0121] Table 1 Regulation potential value level table
[0122] Regulation potential value Regulation potential level 0≤P<20 Difference 20≤P<40 Poor 40≤P<60 medium 60≤P<80 better 80≤P<100 good
[0123] Step 2062: Then establish a fuzzy comprehensive evaluation factor set U: U = {U1, U2, U3, U4}
[0124] Among them, U represents the building control potential, U1 represents the load level capability, U2 represents the user level capability, U3 represents the economic level capability, and U4 represents the management level capability.
[0125] U1={U11, U12, U13, U14, U15}, which respectively represent the five factors of peak-to-valley ratio, average load, weekly load reduction rate, peak power consumption ratio and load fluctuation rate at the load capacity level.
[0126] U2={U21,U22,U23,U24}, respectively represent the capacity proportion of load that can be used for dispatch response, annual average load reduction, smart meter coverage, and user comfort requirements at the load capacity level. U3={U31,U32,U33,U34}, respectively represent the economic effect, income per unit response capacity, pollutants per unit electricity, and social benefits for users. U4={U41,U42,U43}, respectively represent the number of user regulation, actual cooperation degree, and regulation timeliness at the management capacity level.
[0127] Step 2063: Create a comment set V for evaluation factors
[0128] According to the five levels of commercial building demand response potential, namely, "good, relatively good, medium, relatively poor, and poor", that is, V = {good, relatively good, medium, relatively poor, and poor}, the evaluation value of the qualitative indicator is expressed by the numerical value [0, 1] to represent the degree and implementation of each indicator, and the evaluation interval of the qualitative indicator is also divided into five levels, namely, (0.8, 1], (0.6, 0.8], (0.4, 0.6], (0.2, 0.4], (0, 0.2], representing the five levels of "good, relatively good, medium, relatively poor, and poor".
[0129] Step 2064: Establish the fuzzy relationship matrix R
[0130] Combined with the commercial building load demand response potential value grade table, the fuzzy comprehensive evaluation factor set U and the evaluation factor comment set V, the membership degree of each qualitative indicator is determined through expert scoring. The number of experts who make the same grade selection for the same evaluation factor is added up, the frequency is counted, the membership degree is determined, and the fuzzy relationship matrix R is established. Finally, the demand response potential score of each building is obtained based on the fuzzy relationship matrix R.
[0131] Step 3: Automated assessment of building demand response potential based on indicator weights and measurement values combined with application of assessment models;
[0132] Step 301: using the evaluated potential value as a priori sample to train and test the extreme learning machine;
[0133] Step 302: Using the indicator weight obtained in step 2 as the true risk value, determine the number of training samples and test samples;
[0134] Step 303: Initialize PSO parameters and variable optimization range: population size N, maximum number of iterations T, learning factors c1 and c2, inertia weight w, i-th particle, position X i =[W 11 ,W 21 ,...,W h1 ,…,W 1n,b1,b2,b3,…,b h ];
[0135] Step 304: Set the particle position X i Substitute the training set into the formula to calculate the fitness:
[0136]
[0137] Among them, y i and i are the predicted value of the comprehensive score of regulation potential and the comprehensive score of regulation potential, respectively, and n is the number of samples in the training set;
[0138] Step 305: Find the individual optimal solution and the global optimal solution; calculate the fitness of the PSO algorithm particle speed and position after updating;
[0139] The updated position and velocity of the particle are:
[0140]
[0141] in, is the position and velocity of the i-th particle in the k+1th iteration; g ibest and g gbest are the current global optimal solution and the current individual optimal solution respectively;
[0142] Step 306: If the current number of iterations t>T, then output the optimal particle position, which is the optimal initial weight and threshold of the ELM model. Substitute it into the ELM model and use the test set to verify and analyze the building demand response potential; otherwise, return to step 304.
[0143] Step 4: Guide task scheduling allocation based on building demand response potential assessment results. After receiving the grid demand response signal and determining whether to participate in demand response, the commercial building demand response system can finely allocate scheduling tasks based on the calculated commercial building demand response potential, and use strategies to control adjustable devices to participate in the grid company's demand response. The specific allocation control strategy is as follows:
[0144] First, sort the buildings according to their demand response potential from high to low, give priority to buildings with greater potential, and maximize the overall demand response volume.
[0145] Secondly, for buildings with equal potential, a balance is made based on factors such as real-time electricity prices, electricity usage, and energy consumption, and priority is given to scheduling buildings with higher cost-effectiveness.
[0146] In addition, in special circumstances, such as important facilities and peak electricity consumption periods, manual intervention is adopted for scheduling, and different control strategies are formulated specifically, such as temperature setting adjustment, equipment switch control, etc., and differentiated control solutions are adopted according to the characteristics of building equipment.
[0147] Finally, monitor the scheduling effect in real time and adjust the allocation strategy in time according to feedback information to ensure that the overall demand response volume reaches the expected target.
[0148] Embodiment 2
[0149] This embodiment provides a load control system for evaluating building demand response potential based on non-privacy data, including:
[0150] A data acquisition module for acquiring historical data that influences demand response decisions for commercial buildings;
[0151] A demand response potential assessment module is used to obtain a potential score of building demand response based on historical data that affects commercial building demand response decisions and a building demand response potential assessment model; wherein the construction process of the building demand response potential assessment model includes:
[0152] The building demand response decision is defined as a multi-attribute decision problem. Based on the principle of multi-attribute, the influencing factors of the decision are mined and classified, and an evaluation index system for building regulation potential is constructed.
[0153] Based on historical data and the building regulation potential evaluation index system, the non-privacy data value is calculated;
[0154] The building demand response potential is quantified to obtain the weight of the evaluation index; the fuzzy comprehensive evaluation method is combined with the weight of the evaluation index to calculate the potential value of the building demand response;
[0155] The comprehensive indicator weight is used as the real risk value to determine the number of training samples and test samples. The potential value of building demand response is used as a priori sample to train and test the constructed evaluation model. The building demand response potential evaluation result is obtained based on the trained evaluation model.
[0156] The load control module is used to guide the task scheduling allocation according to the building demand response potential assessment result to obtain the task scheduling allocation control strategy, and control the corresponding equipment operation according to the task scheduling allocation control strategy.
[0157] Embodiment 3
[0158] This embodiment provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the steps in the load regulation method for evaluating building demand response potential based on non-privacy data as described above are implemented.
[0159] Embodiment 4
[0160] This embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in the load control method for evaluating building demand response potential based on non-privacy data as described above are implemented.
[0161] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of hardware embodiments, software embodiments, or embodiments combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer-usable program code.
[0162] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0163] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0164] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0165] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.
[0166] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.
Claims
1. A load control method for evaluating building demand response potential based on non-privacy data, characterized in that: The steps include: Access historical data that influences demand response decisions in commercial buildings; Based on historical data that influences commercial building demand response decisions and a building demand response potential assessment model, a building demand response potential score is obtained; wherein the building demand response potential assessment model is constructed in the following steps: The building demand response decision is defined as a multi-attribute decision problem. Based on the principle of multi-attribute, the influencing factors of the decision are mined and classified, and an evaluation index system for building regulation potential is constructed. Based on historical data and the building regulation potential evaluation index system, the non-privacy data value is calculated; The building demand response potential is quantified to obtain the weight of the evaluation index; the fuzzy comprehensive evaluation method is combined with the weight of the evaluation index to calculate the potential value of the building demand response; The comprehensive indicator weight is used as the real risk value to determine the number of training samples and test samples. The potential value of building demand response is used as a priori sample to train and test the constructed evaluation model. The building demand response potential evaluation result is obtained based on the trained evaluation model. The task scheduling allocation is guided by the building demand response potential assessment results to obtain the task scheduling allocation control strategy, and the corresponding equipment operation is controlled according to the task scheduling allocation control strategy.
2. The load control method for building demand response potential assessment based on non-privacy data as claimed in claim 1, characterized in that: The building regulation potential evaluation index system includes four aspects: load level, user level, economic level, and management level; The load level includes five indicators: peak-to-valley ratio, average load, weekly load reduction rate, peak power consumption ratio, and load fluctuation rate; The user level includes four indicators: capacity ratio of load used for dispatch response, average annual load reduction, smart meter coverage, and user comfort requirements; The economic level includes five indicators: local economic level, user social benefits, income per unit response capacity, and pollutants per unit electricity; The management level includes three indicators: number of user adjustments, actual degree of cooperation, and timeliness of adjustments.
3. The load control method for building demand response potential assessment based on non-privacy data as claimed in claim 1, characterized in that: The quantification of the building demand response potential and the weights of the evaluation indicators are obtained, including: For the indicators at different levels of the building regulation potential evaluation index system, the fuzzy analytic hierarchy process is used to quantify the building demand response potential and obtain the subjective weights; Combined with the non-privacy data value, the entropy weight method is used to obtain the objective weight; Combining subjective weights and objective weights, the game theory method is used to optimize the combined weights and the linear combination coefficients between the two weights, with the goal of minimizing the deviation to obtain the most reasonable weights.
4. The load control method for building demand response potential assessment based on non-privacy data as claimed in claim 3, characterized in that: The indicators at different levels of the building regulation potential evaluation index system are quantified using the fuzzy hierarchical analysis method to obtain subjective weights, including: Compare the indicators at different levels in pairs to determine an importance score, fill in the matrix, and establish the judgment matrix R = (r ij ) n×n , where r ij Indicates the index x i X j The relative importance of ij +r ji =1, r ij =r ij -r jk +0.5; The fuzzy scale value corresponding to the scale of 0.1-0.9 is used, 0.1 means that the indicators xi and xj are equally important; 0.3 means that the indicators xi and xj are slightly important; 0.5 means that the indicators xi and xj are obviously important; 0.7 means that the indicators xi and xj are strongly important; 0.9 means that the indicators xi and xj are extremely important; 0.2, 0.4, 0.6, 0.8 mean that the indicators xi and xj are the median of the above two adjacent judgments; Sum the original judgment matrix row by row Then transform the data Construct the consistency matrix B = (b ij ) n×n ; The ranking method is used to calculate the weight, obtain the relative importance weight of each indicator to the indicator of the previous level, and calculate the subjective weight of the indicator.
5. The load control method for building demand response potential assessment based on non-privacy data as claimed in claim 3, characterized in that: The non-privacy data value is combined and the objective weight is obtained by using the entropy weight method, including: According to the principle of combining qualitative and quantitative methods, the indicator matrix R′=(r ij ') m×n , where potential has n evaluation indicators, each indicator has m evaluation levels, r ij ' represents the evaluation value of the jth evaluator for the i-th indicator; r ij 'Do standardization. The index system consists of positive and negative indicators. Standardization is required to obtain r ij ; Calculate the entropy value of the ith indicator, and combine it with the entropy value of the ith indicator to obtain the entropy weight of the jth indicator, that is, the objective weight of the indicator.
6. The load control method for building demand response potential assessment based on non-privacy data as claimed in claim 1, characterized in that: The fuzzy comprehensive evaluation method is combined with the weight of the evaluation index to calculate the potential value of the building demand response, including the following steps: Determine the commercial building load demand response potential value rating table; Establish the factor set U of fuzzy comprehensive evaluation and the comment set V of evaluation factors; Combined with the commercial building load demand response potential value grade table, the fuzzy comprehensive evaluation factor set U and the evaluation factor comment set V, the membership degree of each qualitative indicator is determined through expert scoring. The number of experts who make the same grade selection for the same evaluation factor is added up, the frequency is counted, the membership degree is determined, and the fuzzy relationship matrix R is established. Finally, the demand response potential score of each building is obtained based on the fuzzy relationship matrix R.
7. The load control method for building demand response potential assessment based on non-privacy data as claimed in claim 1, characterized in that: The method of taking the comprehensive indicator weight as the real risk value, determining the number of training samples and test samples, and taking the potential value of building demand response as a priori sample to train and test the constructed evaluation model includes: Initialize PSO parameters and variable optimization range, including population size N, maximum number of iterations T, learning factors c1 and c2, inertia weight w, i-th particle, position X i ; Based on particle position X i Calculate fitness with training set; Calculate the fitness of the PSO algorithm particle speed and position after updating; If the current number of iterations t>T, the optimal position of the particle is output, and the optimal position is used as the optimal initial weight and threshold of the ELM model. It is substituted into the ELM model and the test set is used to verify and analyze the building demand response potential.
8. A load control system for building demand response potential assessment based on non-privacy data, characterized in that: include: A data acquisition module for acquiring historical data that influences demand response decisions for commercial buildings; A demand response potential assessment module is used to obtain a potential score of building demand response based on historical data that affects commercial building demand response decisions and a building demand response potential assessment model; wherein the construction process of the building demand response potential assessment model includes: The building demand response decision is defined as a multi-attribute decision problem. Based on the principle of multi-attribute, the influencing factors of the decision are mined and classified, and an evaluation index system for building regulation potential is constructed. Based on historical data and the building regulation potential evaluation index system, the non-privacy data value is calculated; The building demand response potential is quantified to obtain the weight of the evaluation index; the fuzzy comprehensive evaluation method is combined with the weight of the evaluation index to calculate the potential value of the building demand response; The comprehensive indicator weight is used as the real risk value to determine the number of training samples and test samples. The potential value of building demand response is used as a priori sample to train and test the constructed evaluation model. The building demand response potential evaluation result is obtained based on the trained evaluation model. The load control module is used to guide the task scheduling allocation according to the building demand response potential assessment result to obtain the task scheduling allocation control strategy, and control the corresponding equipment operation according to the task scheduling allocation control strategy.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps in the load regulation method for evaluating building demand response potential based on non-privacy data as described in any one of claims 1 to 7 are implemented.
10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps in the load regulation method for evaluating building demand response potential based on non-privacy data as described in any one of claims 1 to 7 are implemented.
