Power distribution network multi-target scheduling method and device based on large simulation model, and medium

Abstract

The invention discloses a power distribution network multi-target scheduling method and device based on a large simulation model, and a medium, and the method comprises the steps: dividing a power distribution region into a plurality of designated regions according to the geographic position, dividing the power distribution region into a plurality of sub-regions according to the power utilization property, and building a region power utilization model corresponding to each sub-region; for each sub-region, carrying out statistics on historical electricity consumption through a statistical model, and carrying out estimation on future electricity consumption demands through an estimation model; establishing a central regulation and control model; acquiring power generation information corresponding to each power generation end through a power generation end monitoring model; and generating a multi-target cooperative scheduling scheme through the multi-target cooperative scheduling model, and performing cooperative scheduling on each power generation end according to the multi-target cooperative scheduling scheme. In the power distribution scheduling process, a mode of pre-estimation before power distribution and omnibearing monitoring in power distribution is adopted, so that multi-target collaborative power distribution can be more stably and reasonably realized, and the power distribution process is more intelligent, efficient and reliable.

Description

Technical Field

[0001] The present application relates to the field of target scheduling, and specifically to a distribution network multi-target scheduling method, equipment and medium based on a large simulation model. Background Art

[0002] The multi-objective optimal dispatching of distribution network refers to the optimization of multiple objectives such as economy, environmental protection and energy efficiency by optimizing the dispatching scheme under the premise of meeting the power demand. Specifically, the multi-objective optimal dispatching of distribution network aims to reduce the operating cost of the power system, reduce environmental pollution, improve energy efficiency and ensure the safety and stability of power supply through reasonable dispatching strategies.

[0003] At present, the multi-objective optimization scheduling method of distribution network is a complex and critical issue, which involves multiple optimization objectives and constraints.

[0004] Although the existing distribution network can achieve all-round power distribution during the power distribution process, it is difficult to conduct all-round real-time monitoring and feedback during power distribution, especially it is difficult to conduct all-round monitoring of each power generation end. Even if the monitoring is carried out, it is intermittent monitoring, which makes it difficult to achieve real-time monitoring and adjust the power distribution plan in real time. In addition, the power distribution plan is difficult to adjust in time according to the feedback of users and staff, so it is difficult to ensure the efficiency and reliability of power distribution. Summary of the invention

[0005] In order to solve the above problems, this application proposes a multi-objective dispatching method for distribution network based on a large simulation model, including:

[0006] The power distribution area is divided into multiple designated areas according to the geographical location, and each designated area is divided into multiple sub-areas according to the power consumption nature, and a regional power consumption model corresponding to each sub-area is established; wherein the type of power consumption nature corresponding to each sub-area is different; the regional power consumption model includes a statistical model and an estimation model;

[0007] For each sub-region, historical electricity consumption is counted by the statistical model, and future electricity demand is estimated by the estimation model, and a user monitoring model and a climate monitoring model corresponding to the sub-region are established to monitor abnormal electricity consumption information and climate information respectively;

[0008] Establishing a central control model, wherein the central control model includes a power generation end monitoring model and a multi-objective collaborative scheduling model;

[0009] The central control model receives historical power consumption, future power demand, and climate information fed back from each designated area; and the power generation information corresponding to each power generation end is obtained through the power generation end monitoring model;

[0010] Generate a multi-objective collaborative scheduling scheme based on the historical power consumption, the future power demand, the abnormal power consumption information, the climate information, and the power generation information through the multi-objective collaborative scheduling model, and perform collaborative scheduling on each power generation end according to the multi-objective collaborative scheduling scheme;

[0011] By optimizing the feedback model, feedback suggestions are obtained regularly, and the regional power consumption model and / or the central control model are adjusted and optimized according to the feedback suggestions.

[0012] In one example, each designated area is divided into multiple sub-areas according to the nature of electricity consumption, including:

[0013] The types of electricity consumption to be determined include: residential, commercial, and industrial areas;

[0014] Within each designated area, three sub-areas are obtained based on the division of residential areas, commercial areas, and industrial areas.

[0015] In one example, the historical power consumption is counted by the statistical model, specifically including:

[0016] For the residential area, when the historical electricity consumption is counted by the statistical model, the electricity load is divided into non-adjustable rigid demand load and adjustable load, so that the adjustable load can be involved in the coordinated scheduling;

[0017] For the industrial zone, the historical electricity consumption of each factory during production is counted by the statistical model, and the factory with the largest historical electricity consumption is identified, so as to participate in the coordinated scheduling in the form of interruptible loads and excitable loads;

[0018] For the commercial area, the historical power consumption of each merchant is counted by the statistical model, and the historical power consumption is obtained by the commercial aggregate load algorithm.

[0019] In one example, obtaining power generation information corresponding to each power generation end through the power generation end monitoring model specifically includes:

[0020] Determining that the types of the power generation end include wind power generation, photovoltaic power generation, hydropower generation and coal-fired power generation;

[0021] For the wind power generation, power generation information is obtained by calculating through a probability density function formula;

[0022] For the photovoltaic power generation, the power generation information is calculated by using the photovoltaic and load prediction error formula;

[0023] For the hydroelectric power generation, power generation information is obtained by calculating through a boundary formula;

[0024] The coal-fired power generation is calculated using a power generation efficiency formula and corrected using a correction formula to obtain power generation information.

[0025] In one example, a multi-objective collaborative scheduling scheme is generated through the multi-objective collaborative scheduling model according to the historical power consumption, the future power demand, the climate information, and the power generation information, specifically including:

[0026] Generate a power demand level corresponding to each power distribution terminal in each sub-area according to the historical power consumption, the future power demand, and the climate information;

[0027] Determine a corresponding power transmission distance according to the power demand level; wherein the higher the power demand level, the longer the power transmission distance;

[0028] Within the transmission distance, a corresponding power generation terminal is selected according to the power generation information to perform coordinated power distribution for each distribution terminal in each sub-area.

[0029] In one example, within the transmission distance, according to the power generation information, a corresponding power generation end is selected to perform coordinated power distribution for each distribution end in each sub-area, specifically including:

[0030] Through multiple rounds of power distribution, coordinated power distribution is performed for each distribution terminal in each sub-area until the power demand of all distribution terminals is met, or there is no power generation terminal that can distribute power within the transmission distance of all distribution terminals;

[0031] Among them, in a single round of power distribution, the corresponding power distribution terminals are selected in order from high to low according to the power demand level;

[0032] Within the transmission distance corresponding to the distribution end, according to the distance between the power generation end and the distribution end from low to high, the corresponding power generation end is selected in sequence to distribute power to the distribution end until a first preset proportion of the power demand of the distribution end is met, and the current round of power distribution to the distribution end is completed, and the current round of power distribution to the next distribution end is performed;

[0033] When a single power generating end distributes power to a single distribution end, the distributed power of the power generating end to the distribution end is determined according to the distribution capacity of the power generating end; wherein the distributed power is lower than a second preset proportion of the distribution capacity of the power generating end, and is not higher than the power demand of the distribution end, and the total amount of distributed power is lower than the distribution capacity of the power generating end.

[0034] In one example, a multi-objective collaborative scheduling scheme is generated according to the abnormal power consumption information through the multi-objective collaborative scheduling model, specifically including:

[0035] Determining whether to adopt a conventional dispatching plan or an emergency dispatching plan according to the abnormal power consumption information;

[0036] Among them, compared with the conventional scheduling scheme, the emergency scheduling scheme has a higher weight for manual scheduling and a lower weight for automatic scheduling.

[0037] In one example, by optimizing the feedback model, obtaining feedback suggestions regularly, and adjusting and optimizing the regional power consumption model and / or the central control model according to the feedback suggestions, specifically includes:

[0038] By optimizing the feedback model, the user's electricity consumption feedback information is collected at first preset time intervals to form a feedback database;

[0039] Collecting power distribution feedback information from staff at a second preset time interval to form a power distribution database; wherein the second preset time is shorter than the first preset time;

[0040] According to the abnormal information in the power consumption feedback information and the power distribution feedback information, the regional power consumption model and / or the central control model are adjusted and optimized in real time;

[0041] The regional power consumption model and / or the central control model are adjusted and optimized regularly according to other information in the power consumption feedback information and the power distribution feedback information.

[0042] On the other hand, the present application also proposes a multi-objective dispatching device for a distribution network based on a large simulation model, comprising:

[0043] at least one processor; and,

[0044] a memory communicatively connected to the at least one processor; wherein,

[0045] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the distribution network multi-objective scheduling method based on the simulation large model as described in any of the above examples.

[0046] On the other hand, the present application also proposes a non-volatile computer storage medium storing computer executable instructions, wherein the computer executable instructions are configured to be: a multi-objective scheduling method for a distribution network based on a large simulation model as described in any of the above examples.

[0047] The multi-objective dispatching method of distribution network based on large simulation model proposed in this application can bring the following results:

[0048] Beneficial effects:

[0049] In the process of power distribution dispatching, the use of pre-distribution estimation and all-round monitoring during power distribution can achieve multi-objective coordinated power distribution more stably and reasonably, making the power distribution process more intelligent, efficient and reliable. Through feedback, the power distribution process is continuously optimized so that the power distribution process can guarantee stability for a long time, avoiding the failure to adjust the power distribution plan according to the specific situation of the user during the power distribution process, resulting in errors and anomalies in the power distribution process, thereby leading to the severity of the abnormal problems, and avoiding unreasonable power distribution process. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0051] Figure 1 It is a flow chart of a multi-objective dispatching method for a distribution network based on a large simulation model in an embodiment of the present application;

[0052] Figure 2 A schematic diagram of a multi-objective dispatching method for a distribution network based on a large simulation model in one embodiment of the present application;

[0053] Figure 3 It is a schematic diagram of a multi-objective dispatching device of a distribution network based on a large simulation model in an embodiment of the present application. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0055] The technical solutions provided by various embodiments of the present application are described in detail below in conjunction with the accompanying drawings.

[0056] like Figure 1 and Figure 2 As shown, the embodiment of the present application provides a distribution network multi-objective scheduling method based on a large simulation model, including:

[0057] S101: A power distribution area is divided into a plurality of designated areas according to geographical location, and each designated area is divided into a plurality of sub-areas according to the nature of power consumption, and a regional power consumption model corresponding to each sub-area is established; wherein the type of power consumption nature corresponding to each sub-area is different; and the regional power consumption model includes a statistical model and an estimation model.

[0058] Specifically, the types of electricity usage include: residential areas, commercial areas, and industrial areas. At this time, within each designated area, three sub-areas are obtained according to the division method of residential areas, commercial areas, and industrial areas. Each sub-area includes all residential areas, commercial areas, and industrial areas.

[0059] For example, multiple designated areas are obtained based on administrative divisions, and it is assumed here that they are divided into Area A, Area B, Area C and Area D (this is only an exemplary description and does not represent the actual division situation), and each area is further divided according to the nature of electricity consumption, and is divided into residential areas, industrial areas and commercial areas, and residential area models, industrial area models and commercial area models are established respectively.

[0060] Furthermore, Area A, Area B, Area C and Area D are modeled respectively, among which Area A is the Area A planning model, Area B is the Area B planning model, Area C is the Area C planning model and Area D is the Area D planning model. The residential area model, industrial area model and commercial area model are established within the Area A planning model, Area B planning model, Area C planning model and Area D planning model respectively.

[0061] Specifically, the steps for building residential area models, industrial area models, and commercial area models are as follows:

[0062] For the residential area model, each community within the residential area is registered, and after the registration is completed, the information of each household in the community is collected. When the information of all residents is collected, the actual occupancy rate within the community is counted.

[0063] The family size is also counted. The family size refers to the size of the population in the family, usually including parents, children and the elderly. Families can be divided into one-person families, two-person families, three-person families... six-person families or more. The family size structure refers to the proportion of the number of families of various sizes to the total number of families, and the family statistics formula is: Among them, H is the total number of households, H n is the number of households with n members, h n is the proportion of households with n persons to the total number of households.

[0064] For the industrial zone model, the factory information within the industrial zone is counted, and in the process of statistics, it is divided according to the factory's production capacity and the types of products produced by the factory. For example, it is divided into three levels, namely, factory one, factory two and factory three. Among them, factory one has the highest production capacity and produces products that require the most electricity, and so on. Factory two is the second, and factory three is the smallest, and the products produced by factory three consume the least electricity.

[0065] For the commercial area model, statistics are taken on the number of businesses within the commercial area, as well as the exterior lights of buildings, new energy charging equipment and central air conditioners within the commercial area. After the statistics are completed, the abatement model is used to regulate electricity consumption for commercial users.

[0066] Taking central air conditioning as an example, the average power consumption of commercial users in central air conditioning is related to the set temperature. By changing the user's set temperature and setting the temperature used by the user in advance, the power used will vary within the set range. In order to ensure the user's comfort, a comfortable range value can also be set for the temperature used by the user, and it can be set between 22-26℃ for cooling or heating. At the same time, since the air conditioner can be turned off for a short period of time when the temperature control is in compliance, it will not affect the user's experience, and the power consumption can also be reduced to a certain extent.

[0067] S102: For each sub-region, historical electricity consumption is counted by the statistical model, and future electricity demand is estimated by the estimation model, and a user monitoring model and a climate monitoring model corresponding to the sub-region are established to monitor abnormal electricity consumption information and climate information respectively.

[0068] After the zoning statistics are completed, the demands within Area A, Area B, Area C and Area D are monitored and estimated respectively. The statistical model is used for estimation and statistics. The estimation model is set in advance to estimate the electricity consumption within each sub-area. At the same time, a climate monitoring model is established to monitor the climate within each sub-area, and abnormal electricity consumption of users is monitored by establishing a user monitoring model.

[0069] Specifically, for residential areas, when historical electricity consumption is counted through statistical models, the electricity load is divided into non-adjustable rigid demand load and adjustable load, so that the adjustable load can be involved in coordinated scheduling.

[0070] When counting residential users in residential areas, the electricity load is divided into non-adjustable rigid load and adjustable load. The non-adjustable rigid load includes refrigerators, security equipment (cameras and alarm systems, etc.), routers, smart home devices (smart light bulbs, smart sockets and smart door locks, etc.), medical equipment (ventilators and oxygen generators for special groups at home) and smart fish tank filters; adjustable loads include electric water heaters, rice cookers, air conditioners and televisions, etc., and the load proxy algorithm is used to calculate and formulate the demand contract for electricity demand, and the load proxy algorithm is: Among them, P min P is the minimum limit of adjustable capacity in the load evaluation area. max is the maximum limit of adjustable capacity in the load business evaluation area, r is the capacity price curve, C max is the maximum price for load compensation, C minThe lowest price for loaders.

[0071] Based on the load proxy algorithm, price contracts and incentive contracts are formulated respectively, and the price contract adopts the load call algorithm as follows: Among them, p i0 and Δp i are the base electricity price at time i and the adjusted electricity price after price incentives, ΔE i 、E i0 are the adjusted power and power reference value respectively, i is the time value, C ip is the final value of the adjusted electricity price, ε ii is the final value of the base electricity price.

[0072] The load cost formula used in incentive contracts is as follows: Among them, α is the compensation electricity price when the incentive load is lowered, β is the discount electricity price when the incentive load power is raised, and p i0 is the base electricity price of the system at time i, C iE is the final value after calculation.

[0073] For industrial areas, the historical electricity consumption of each factory during production is counted through statistical models, and the factories with the largest historical electricity consumption are identified to facilitate coordinated scheduling in the form of interruptible loads and excitable loads.

[0074] The power consumption of each factory in the industrial zone is collected during production. After the data is collected, the factory with the largest power consumption is marked. For example, electrolytic aluminum, cement and ice factories are the largest power consumers, and the factory producing colored aluminum has the lowest power consumption. The load is used in the form of interruptible and excitable loads to participate in coordinated control, and the specific formula is as follows: P i,ll =-η i,off μ i,off P i,t +η i,on μ i,on P i,l ; Among them, P i,ll , P i,l are the power and total power of the industrial high-energy load in the i-th power plant that can be dispatched, μ i,on , μ i,off is the excitation coefficient and interruption coefficient, η i,on , η i,off is a pair of decision coefficients that determine the load to participate in dispatch, P i,r , P i,load They are the output of new energy units and load power respectively.

[0075] Its complementary function is: Among them, c i,on 、ci,off is the load incentive compensation unit price and interruption compensation unit price, g i,l (P i,l ) supplement the function value.

[0076] For commercial areas, the historical electricity consumption of each merchant is counted through a statistical model, and the historical electricity consumption is obtained through a commercial aggregate load algorithm.

[0077] After collecting data on the electricity consumption of merchants in the commercial area, the electricity consumption is calculated using the commercial aggregate load algorithm: i,2l =-η i,Ban μ i,Ban K i,Ban P i,2 ; Among them, P i,2l , P i,2 are the power and total power of the commercial aggregated load in virtual power plant i that can be dispatched, μ i,Ban is the reduction coefficient, η i,Ban is the decision coefficient, K i,Ban To reduce the proportion.

[0078] Its complementary cost function is: g i,l (P i,l ) = η i,Ban c i,Ban |P i,2l |; where c i,Ban To reduce the load compensation unit price, g i,l (P i,l ) supplement the function value.

[0079] After the statistical model collects the data, an estimation model is established based on the statistical data. The estimation model mainly estimates the electricity consumption in industrial areas, commercial areas and residential areas in advance. The construction steps include:

[0080] 1. Collect statistics on electricity consumption data for three months in industrial areas, commercial areas and residential areas, and establish industrial databases, commercial databases and residential databases respectively after statistics. Use curve graphs to form curve graphs of the data in the industrial databases, commercial databases and residential databases.

[0081] 2. Collect statistics on the production capacity plan data for each factory building in the industrial zone for the next three months, as well as the daily business hours in the commercial zone for the next three months, and the scale, frequency and time of activities held during the business period. Collect statistics on weddings and funerals in the residential zone, as well as the number of holidays.

[0082] 3. According to the trend of the curve chart, the future trend is estimated through linear regression or polynomial regression, and corrections are made according to the number of holidays, etc. to obtain the estimated electricity consumption.

[0083] In addition, the steps for constructing the user monitoring model include: using smart sensors to monitor each electricity-consuming end in the industrial area, commercial area and within the industrial area, and forming an electricity consumption database with the data monitored by the electricity-consuming end. When it is found that too much electricity is consumed in a short period of time and no electricity is consumed for a long time, it is considered that there is abnormal electricity consumption information, and the staff will verify it as soon as possible.

[0084] The steps for building a climate monitoring model include: during the monitoring process, the climate in industrial areas, commercial areas and within industrial areas is monitored simultaneously, the data collected during monitoring is integrated to form a climate database, and the climate database is connected to the local meteorological bureau to enable real-time updates of the climate database.

[0085] S103: Establishing a central control model, wherein the central control model includes a power generation end monitoring model and a multi-objective collaborative scheduling model.

[0086] In the process of distribution network dispatching, a central control model is established, and the central control model includes a power generation end monitoring model and a multi-objective collaborative dispatching model. The data of the statistical model, the estimation model, the user monitoring model and the climate monitoring model are collected through the central control model, and each power generation end is monitored in real time, and mainly wind power, photovoltaic power, hydropower and coal-fired power generation are monitored. During the monitoring, a power generation end monitoring model is formed, and the monitoring data is transmitted to the dispatching end through the power generation end monitoring model. The dispatching end then adopts a multi-objective collaborative dispatching model for dispatching, and sets up conventional dispatching plans and emergency dispatching plans respectively within the multi-objective collaborative dispatching model.

[0087] S104: Receive historical electricity consumption, future electricity demand, and climate information fed back from each designated area through the central control model; and obtain power generation information corresponding to each power generation end through the power generation end monitoring model.

[0088] Specifically, the types of power generation terminals are determined to include wind power generation, photovoltaic power generation, hydropower generation and coal-fired power generation.

[0089] For wind power generation, the power generation information is calculated using the probability density function formula.

[0090] Wind power generation converts wind power into electrical energy through generators and is calculated using the probability density function formula, which is as follows: Among them, k is the shape parameter, c is the scale parameter, the unit is m / s, e is the reference parameter value, and f(v) is the probability density function value.

[0091] The power characteristic curve of a wind turbine is the relationship curve between output power and wind speed. The active output of the wind turbine is P W The functional relationship with wind speed v can be approximately described as follows: Among them, P WN is the rated power of the fan, in MW,v in 、v N 、v out are the cut-in wind speed, rated wind speed and cut-out wind speed of the fan, in m / s. When v<v in When the wind speed is too low, the mechanical torque generated is not enough to make the motor rotate. out When the wind speed is greater than the cut-out wind speed, the fan is cut out to prevent mechanical damage to the fan. in ≤v≤v N When v N ≤v≤v out When the wind power resources are sufficient, the wind turbine maintains stable operation at rated power, and the output power curve of the wind turbine is formed after calculation.

[0092] For photovoltaic power generation, the power generation information is calculated through the photovoltaic and load forecast error formula.

[0093] Photovoltaic power generation is the process of converting light energy into electrical energy through photovoltaic power generation equipment, and is calculated using the photovoltaic and load forecast error formula. The normal distribution is described as follows: Among them, σ PV,t Expressed as the standard deviation of period t, ε PV,t It is expressed as the percentage of the forecast error in period t to the forecast value, It represents the photovoltaic prediction value in period t, exp is the reference value, and f is the auxiliary value.

[0094] From the above formula, we can get The actual photovoltaic power The calculation formula is:

[0095] Load demand forecasting is not yet very accurate, so an error adjustment algorithm is needed, and the error adjustment algorithm formula is as follows: The adjusted value can be obtained The load value The calculation formula is: Among them, σ Loadt is the adjustment difference within the approved time period, ε Loadt To adjust the percentage of the difference, Adjust the differential for load.

[0096] For hydropower generation, the power generation information is calculated using the boundary formula.

[0097] Hydropower generation uses hydropower equipment to generate electricity by utilizing the pressure difference of water level. During the power generation process, the reservoir capacity and the power generation end need to be monitored. When monitoring the reservoir capacity, due to the large reservoir area, it is assumed that the water levels of the upstream and downstream reservoirs are constant during the transient period, and considering the loss of water head during power generation, the upstream and downstream water levels are constants, and the upstream reservoir outlet meets the negative characteristic line equation C - , the downstream reservoir inlet conforms to the positive characteristic line equation C + , the boundary formula is used for calculation, the specific formula is as follows: Among them, Q Pl and Q p2 are the flow rates at points p1 and p2, respectively, and H p1 and H p2 are the water heads at points p1 and p2, H const1 and H const2 are the upstream and downstream reservoir heads, C n1 ,C p2 ,C a1 and C a2 is an intermediate variable.

[0098] At the same time, the generator is also monitored and the generator power algorithm is used for monitoring. The specific formula is as follows: Among them, δ is the generator rotor angle, ω is the relative deviation of the generator electrical angle speed, D is the generator damping coefficient, m e is the electromagnetic torque of the generator. In the dynamic characteristics analysis of the generator model, it can be considered that the electromagnetic torque of the generator m e Equal to electromagnetic power, m t is the electromagnetic energy value of the generator, T is the energy value, that is: in, is the q-axis state potential of the generator, V s is the infinite bus voltage, x q are the d-axis transient reactance and q-axis reactance, respectively, P e is the electromagnetic power value and is expressed as: Among them, x L is the transformer short-circuit reactance, x T It is the reactance of the transmission line, and it will also monitor the working status of the speed regulator, and model the pressure water diversion pipeline through virtual technology. The model is a pressure water diversion pipeline model. When the power generation operation is carried out, the pressure water diversion pipeline model will run simultaneously and provide feedback on various parameters during the operation.

[0099] For coal-fired power generation, calculations are performed using the power generation efficiency formula and corrections are made using the correction formula to obtain power generation information.

[0100] In the process of coal-fired power generation, energy is generated by burning coal, and the energy is converted by the generator set to complete the power generation. The operating parameters of the boiler and the generator set are determined. The different power supply caused by different coal qualities can be corrected according to the weighted calorific value of the coal entering the furnace and the total coal feed, and the correction formula is used for correction processing. The specific correction formula is as follows: Among them, ES1 and ES2 are the supply amounts of the corresponding coal qualities, S1 and S2 are the coal feed amounts of the corresponding coal qualities, and Q1 and Q2 are the weighted average calorific values ​​of the corresponding coal qualities.

[0101] Online working condition judgment and dynamic threshold selection. In the actual power generation process, due to the frequent switching of working conditions and the influence of variable coupling, the current normal range is dynamic and time-varying, with passive mean and variance. Therefore, the process variables affected by working conditions and variables should be configured with real-time dynamic thresholds to adapt to the reasonable changes of the process and avoid false alarms and missed alarms of the DCS alarm system. The proposed dynamic alarm threshold design method is specifically described as follows:

[0102] Data preparation, for variables that need threshold optimization, add up the newly collected samples x 1， Through the ω-order time series expansion, we can get {x t-ω+1 , …, x t}, which can be expressed as X t ∈R nω×1 .

[0103] Step 1: Expand data X based on the time series of new samples t The classification model is obtained by the TICC algorithm, and the online working condition discrimination is performed. From the K sample time sets P = {P1, P2, ..., P K} to determine the category P i , the formula is as follows: Among them, ll(X t ,Θ i ) is X t Belongs to category Θ i The likelihood function of , β is the constraint coefficient for the time series consistency of clustering results, Whether it is consistent with the judgment result of the previous moment, if it is consistent, it is 0, otherwise it is 1.

[0104] Step 2: According to the category set P i Multivariate Gaussian model Select any variable from the n process variables as x1, the current values ​​of other variables and the historical values ​​of all ω-1 moments can be regarded as the conditions for judging the reference value, when the high and low alarm thresholds of the variable at the current moment.

[0105] Step 3: Select one of the n process variables as x1 without repetition, and repeat step 2 until all variables are calculated.

[0106] Step 4: Compare each process variable with its corresponding dynamic high and low alarm thresholds. If the threshold is exceeded, the variable will alarm, otherwise it will not alarm.

[0107] S105: Generate a multi-objective collaborative scheduling plan through the multi-objective collaborative scheduling model according to the historical electricity consumption, the future electricity demand, the abnormal electricity consumption information, the climate information, and the power generation information, and perform collaborative scheduling on each power generation end according to the multi-objective collaborative scheduling plan.

[0108] Specifically, based on historical power consumption, future power demand, and climate information, the power demand level corresponding to each distribution terminal in each sub-area is generated. According to the power demand level, the corresponding transmission distance is determined; wherein, the higher the power demand level, the longer the transmission distance.

[0109] In the process of power distribution dispatching, each power generation end and distribution end are adjusted through the central control model. In the process of control, the power demand of the user end is classified, and the distance between the power consumption end and the distribution end and power generation end is analyzed and processed.

[0110] The electricity demand is divided into four levels, namely level one, level two, level three and level four, among which level one is the highest and level four is the lowest. After grading each electricity consumer, the distance is set, with the farthest distance set to A, the farther distance set to B, the medium distance set to C and the closest distance set to D. For example, the farthest distance is set to 80-100 kilometers, the farther distance is set to 60-80 kilometers, the medium distance is set to 40-60 kilometers, and the closest distance is set to within 40 kilometers. Emergency plans and common plans are formulated based on the level of demand and comprehensive judgment of the transmission end.

[0111] Within the transmission distance, the corresponding power generation end is selected according to the power generation information to coordinate power distribution for each distribution end in each sub-area.

[0112] Specifically, multiple rounds of power distribution can be used to coordinate power distribution for each distribution terminal in each sub-area until the power demand of all distribution terminals is met, or there is no power generation terminal capable of power distribution within the transmission distance of all distribution terminals, that is, although the power demand is not met at the distribution terminal, there is no power generation terminal around that can provide power distribution.

[0113] In a single round of power distribution, the corresponding distribution terminals are selected in order from high to low according to the power demand level. When there are multiple distribution terminals of the same level, they can be selected randomly.

[0114] Within the transmission distance corresponding to the distribution end, the corresponding power generation end is selected in sequence according to the distance between the power generation end and the distribution end from low to high for power distribution to the distribution end, until a first preset proportion (for example, 20%) of the power demand of the distribution end is met, completing the current round of power distribution to the distribution end, and then performing the current round of power distribution to the next distribution end.

[0115] When the current round of power distribution at all distribution ends is completed, that is, all distribution ends have completed 20% of the power demand, the next round of power distribution will be carried out, thereby ensuring that each distribution end has at least a certain amount of power distribution available.

[0116] When a single power generating end distributes power to a single distribution end, the power distributed by the power generating end to the distribution end is determined according to the power distribution capacity of the power generating end; wherein the distributed power is lower than a second preset proportion of the power distribution capacity of the power generating end (for example, 30%, to prevent a single power generating end from distributing too much power to a single distribution end at one time, thereby making it difficult to take into account the needs of other surrounding distribution ends), and is not higher than the power demand of the distribution end, and the total amount of distributed power is lower than the power distribution capacity of the power generating end.

[0117] In addition, according to the abnormal power consumption information, determine whether to adopt the conventional dispatching plan or the emergency dispatching plan. If there is abnormal power consumption information, the emergency dispatching plan is adopted, otherwise, the conventional dispatching plan is adopted. Among them, compared with the conventional dispatching plan, the emergency dispatching plan has a higher weight for manual dispatching and a lower weight for automatic dispatching.

[0118] For example, the conventional dispatching plan uses a combination of intelligent equipment and manual processing to carry out power distribution processing. In the power distribution process, 80% is completed by intelligent equipment and only 20% is completed manually. In the emergency dispatching plan, 50% is operated manually and the remaining 50% is completed by intelligent equipment. The emergency plan is checked and adjusted at regular intervals (for example, one week) to be able to adapt to and handle unexpected situations.

[0119] S106: Obtain feedback suggestions regularly by optimizing the feedback model, and adjust and optimize the regional power consumption model and / or the central control model based on the feedback suggestions.

[0120] Specifically, by optimizing the feedback model, the user's power consumption feedback information is collected every first preset time (for example, 24 hours) to form a feedback database. The power distribution feedback information of the staff is collected every second preset time (for example, 12 hours) to form a power distribution database; wherein the second preset time is shorter than the first preset time.

[0121] The feedback processing model is divided into normal feedback process and abnormal feedback process. The normal feedback process can complete the feedback through normal channels, while the abnormal feedback process will provide feedback to the staff as soon as possible.

[0122] With respect to abnormal information in the power consumption feedback information and the power distribution feedback information, an abnormal feedback process is adopted to adjust and optimize the regional power consumption model and / or the central control model in real time.

[0123] For other information in the power consumption feedback information and distribution feedback information, the normal feedback process is adopted to regularly adjust and optimize the regional power consumption model and / or the central control model.

[0124] like Figure 3 As shown, in one embodiment, the present application also proposes a distribution network multi-objective dispatching device based on a large simulation model, including:

[0125] at least one processor; and,

[0126] a memory communicatively connected to the at least one processor; wherein,

[0127] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the distribution network multi-objective scheduling method based on the simulation large model as described in any of the above embodiments.

[0128] In one embodiment, the present application further proposes a non-volatile computer storage medium storing computer executable instructions, wherein the computer executable instructions are configured to be: a multi-objective scheduling method for a distribution network based on a large simulation model as described in any of the above embodiments.

[0129] Each embodiment in this application is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device and medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0130] The devices and media provided in the embodiments of the present application correspond one-to-one to the methods. Therefore, the devices and media also have similar beneficial technical effects as the corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the devices and media will not be repeated here.

[0131] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A multi-objective dispatching method for distribution network based on large simulation model, characterized in that: include: The power distribution area is divided into multiple designated areas according to the geographical location, and each designated area is divided into multiple sub-areas according to the power consumption nature, and a regional power consumption model corresponding to each sub-area is established; wherein the type of power consumption nature corresponding to each sub-area is different; the regional power consumption model includes a statistical model and an estimation model; For each sub-region, historical electricity consumption is counted by the statistical model, and future electricity demand is estimated by the estimation model, and a user monitoring model and a climate monitoring model corresponding to the sub-region are established to monitor abnormal electricity consumption information and climate information respectively; Establishing a central control model, wherein the central control model includes a power generation end monitoring model and a multi-objective collaborative scheduling model; The central control model receives historical power consumption, future power demand, and climate information fed back from each designated area; and the power generation information corresponding to each power generation end is obtained through the power generation end monitoring model; Generate a multi-objective collaborative scheduling scheme based on the historical power consumption, the future power demand, the abnormal power consumption information, the climate information, and the power generation information through the multi-objective collaborative scheduling model, and perform collaborative scheduling on each power generation end according to the multi-objective collaborative scheduling scheme; By optimizing the feedback model, feedback suggestions are obtained regularly, and the regional power consumption model and / or the central control model are adjusted and optimized according to the feedback suggestions.

2. The method according to claim 1, characterized in that Each designated area is divided into multiple sub-areas according to the nature of electricity consumption, including: The types of electricity consumption to be determined include: residential, commercial, and industrial areas; Within each designated area, three sub-areas are obtained based on the division of residential areas, commercial areas, and industrial areas.

3. The method according to claim 2, characterized in that The historical electricity consumption is counted by the statistical model, specifically including: For the residential area, when the historical electricity consumption is counted by the statistical model, the electricity load is divided into non-adjustable rigid demand load and adjustable load, so that the adjustable load can be involved in the coordinated scheduling; For the industrial zone, the historical electricity consumption of each factory during production is counted by the statistical model, and the factory with the largest historical electricity consumption is identified, so as to participate in the coordinated scheduling in the form of interruptible loads and excitable loads; For the commercial area, the historical power consumption of each merchant is counted by the statistical model, and the historical power consumption is obtained by the commercial aggregate load algorithm.

4. The method according to claim 1, characterized in that: The power generation information corresponding to each power generation end is obtained through the power generation end monitoring model, specifically including: Determining that the types of the power generation end include wind power generation, photovoltaic power generation, hydropower generation and coal-fired power generation; For the wind power generation, power generation information is obtained by calculating through a probability density function formula; For the photovoltaic power generation, the power generation information is calculated by using the photovoltaic and load prediction error formula; For the hydroelectric power generation, power generation information is obtained by calculating through a boundary formula; The coal-fired power generation is calculated using a power generation efficiency formula and corrected using a correction formula to obtain power generation information.

5. The method according to claim 1, characterized in that Through the multi-objective coordinated scheduling model, a multi-objective coordinated scheduling scheme is generated according to the historical power consumption, the future power demand, the climate information, and the power generation information, specifically including: Generate a power demand level corresponding to each power distribution terminal in each sub-area according to the historical power consumption, the future power demand, and the climate information; Determine a corresponding power transmission distance according to the power demand level; wherein the higher the power demand level, the longer the power transmission distance; Within the transmission distance, a corresponding power generation terminal is selected according to the power generation information to perform coordinated power distribution for each distribution terminal in each sub-area.

6. The method according to claim 5, characterized in that Within the transmission distance, according to the power generation information, a corresponding power generation end is selected to coordinate power distribution for each distribution end in each sub-area, specifically including: Through multiple rounds of power distribution, coordinated power distribution is performed for each distribution terminal in each sub-area until the power demand of all distribution terminals is met, or there is no power generation terminal that can distribute power within the transmission distance of all distribution terminals; Among them, in a single round of power distribution, the corresponding power distribution terminals are selected in order from high to low according to the power demand level; Within the transmission distance corresponding to the distribution end, according to the distance between the power generation end and the distribution end from low to high, the corresponding power generation end is selected in sequence to distribute power to the distribution end until a first preset proportion of the power demand of the distribution end is met, and the current round of power distribution to the distribution end is completed, and the current round of power distribution to the next distribution end is performed; When a single power generating end distributes power to a single distribution end, the distributed power of the power generating end to the distribution end is determined according to the distribution capacity of the power generating end; wherein the distributed power is lower than a second preset proportion of the distribution capacity of the power generating end, and is not higher than the power demand of the distribution end, and the total amount of distributed power is lower than the distribution capacity of the power generating end.

7. The method according to claim 1, characterized in that By means of the multi-objective collaborative scheduling model, a multi-objective collaborative scheduling scheme is generated according to the abnormal power consumption information, specifically including: Determine whether to adopt a conventional dispatching plan or an emergency dispatching plan according to the abnormal power consumption information; Among them, compared with the conventional scheduling scheme, the emergency scheduling scheme has a higher weight for manual scheduling and a lower weight for automatic scheduling.

8. The method according to claim 1, characterized in that By optimizing the feedback model, obtaining feedback suggestions regularly, and adjusting and optimizing the regional power consumption model and / or the central control model according to the feedback suggestions, specifically including: By optimizing the feedback model, the user's electricity consumption feedback information is collected every first preset time to form a feedback database; Collecting power distribution feedback information from staff at a second preset time interval to form a power distribution database; wherein the second preset time is shorter than the first preset time; According to the abnormal information in the power consumption feedback information and the power distribution feedback information, the regional power consumption model and / or the central control model are adjusted and optimized in real time; The regional power consumption model and / or the central control model are adjusted and optimized regularly according to other information in the power consumption feedback information and the power distribution feedback information.

9. A multi-objective dispatching device for distribution network based on a large simulation model, characterized in that: include: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the distribution network multi-objective scheduling method based on a simulation large model as described in any one of claims 1 to 8.

10. A non-volatile computer storage medium storing computer executable instructions, characterized in that: The computer executable instructions are configured as: a multi-objective dispatching method for a distribution network based on a large simulation model as described in any one of claims 1 to 8.

Images